<4D F736F F D20BDC9C6F7C1F6BEF62DB3BBC1F6C3D6C1BE2E646F6378>

Transcription

1 Proceedings of International Symposium International Forest Fire Symposium Commemorating the International Year of Forests 2011 Date June 7-8, 2011 Place Sol Beach, Gangwon-do, Republic of Korea Hosted by Supported by KOREA FOREST RESEARCH INSTITUTE

2 International Forest Fire Symposium on Commemorating the International Year of Forests ii -

3 June 7-8, 2011 International Forest Fire Symposium Commemorating the International Year of Forests 2011 Background Discussion on reducing forest fire damage Public relations for the progression in the prevention and suppression Strategies since 2000 s mega fires in Republic of Korea Reviews on situation of ecosystem due to restoring activity in burned area Exchanging information on preventing mega fires in the world Subjects of Symposium The progression in the strategies about fire prevention and suppression since 2000 s mega fires in Republic of Korea - Improvement of the fire prevention and suppression techniques by IT - Characteristics of forest fires and countermeasures in each country - Strategies for reducing the forest fire damages The stabilization of ecosystem and restoring progress for 10-year after mega fire at Korea s East Coast in 2000 Schedule Dates: June 7~8, 2011 Venue: Sol Beach, Yangyang-gun, Gangwon-do, Republic of Korea Program Schedule Dates Program Location Place 7 June Oral Presentation 8 June Poster Session & Awarding Sol Bearch resort Conference room Yangyang-gun, Gangwon-do

4 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 PROGRAMME Oral Presentation, June 7, 2011 (Tuesday) Time Activity/Seminar 09:00~10:00 Participants Registration 10:00~10:10 Opening address Dr. Gil Bon Koo (Director General, KFRI, Korea) 10:10~10:35 Congratulatory address I Dr. Don Koo Lee (Minister, KFS, Korea) Congratulatory address II Mr. Moonsoon Choi (Governor, Gangwon-do) Congratulatory address III Prof. Johann G. Goldammer (Chair UN_ISDR/GFMC) 10:35~10:45 Welcome address Dr. Sun Do Choi (Vice President of Kangwon Univ. Korea) 10:45~10:50 Photo time 10:50~11:00 Coffee Break 11:00~11:20 Keynote Speech I International protocols and agreements on cooperation in wildland fire management: needs, current status, and the way ahead in the asian region and at global level Prof. Johann G. Goldammer (Chair UN_ISDR/GFMC) 11:20~11:35 Forest fires and management in China Dr. Xiaorui Tian (CAF, China) 11:35~11:50 Forest fires suppression strategy and management in Korea Mr. Hyunbok Lee (KFS, Korea) 11:50~12:05 The report of forest fire organization, system and skill in Korea Prof. Siyoung Lee (Kangwon Univ. Korea) 12:05~12:20 WRF-fire simulation of the 2005 Yangyang, Gangwon province forest fire Dr. Sangok Han (NIMR, Korea) 12:20~14:00 Lunch

5 June 7-8, 2011 Time Activity/Seminar 14:00~14:15 Soap-based firefighting foam for forest fire, current forest fire situation in Japan Prof. Kazuya Uezu (Kitakyushu Univ. Japan) 14:15~14:30 Spatial pattern of human caused forest fire in Korea Prof. Wookyun Lee (Korea Univ. Korea) 14:30~14:45 Indonesian forest fires moratorium and greenhouse gas emission reduction Prof. Bambang Saharjo (BA Univ. Indonesia) 14:45~15:05 Keynote Speech II Climate change impacts on forest fire characteristics and restoration strategy in Korea Dr. Myungbo Lee (KFRI, Korea) 15:05~15:20 Effects of disturbance intensity on arthropod communities in burned pine forests in Korea Dr. Taesung Kwon (KFRI, Korea) 15:20~15:30 Coffee Break 15:30~15:45 Impacts of forest fires on ecosystems functioning including biodiversity and climate change: a perspective from Hindu-Kush Himalayan (HKH) region Mr. Sundar P. Sharma (MWR, Nepal) 15:45~16:00 Assessment of the crown fire hazard of Pinus densiflora based on the crown fuel characteristics in Korea Prof. Youngjin Lee (Kongju Univ. Korea) 16:00~16:15 Bush fire management and cooperation Dr. Jinjoo Jae (NEMA, Korea) 16:15~16:30 The status of forest fires and management in Morgolia Dr. Ganbaatar Jamiyansuren (Fire Department Agency(NEMA), Mongolia) 16:30~16:45 Projection of climate change over East Asia based on global downscaling with high resolution AGCM Prof. Jaiho Oh (Bukyung Univ. Korea) 16:45~17:00 Vegetation regeneration and soil erosion for ten years following forest fires in Korea Prof. Yeonsook Jung (Kangwon Univ. Korea) 17:00~17:30 Question and Discussion Dr. Kyongha Kim (KFRI, Korea)

6 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Poster Session, June 8, 2011(Wednesday) Time Activity/Seminar 09:00~09:30 Attaching Posters (June 7, 2011) 09:00~11:30 Poster Presentation & Appraisal 11:30~11:50 Awarding of Poster Presentation 1 st : 1 Poster 2 nd :1 Poster 3 rd : 1 Poster 11:50~13:00 Lunch Exhibition Forest Fire Management and Suppression System - Forest Fire Danger Rating System - Device for Measuring Fire Weather - Forest Fire Chemical - Forest Fire Management System using Smartphone - Korean Fire Pump - MOLMOS : Forest fire and vegetation change detection program based on MODIS images

7 June 7-8, 2011 CONTENTS Opening Session i Keynote Speech I 1 Intenational protocols and agreements on cooperation in wildland fire management: needs, current status, and the way ahead in the Asian region and at global level Johann G. Goldammer 3 Oral Presentation I 51 Forest fires and management in China Xiaorui Tian 53 Forest fires suppression strategy and management in Korea Hyunbok Lee The report of forest fire organization, system and skill in Korea Siyoung Lee 61 WRF-fire simulation of the 2005 Yangyang, Gangwon province forest fire Sangok Han 69 Soap-based firefighting foam for forest fire, current forest fire situation in Japan Kazuya Uezu 72 Spatial pattern of humen caused forest fire in Korea Wookyun Lee 77 Indonesian forest fires moratorium and greenhouse gas emission reduction Bambang Saharjo 87

8 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Keynote Speech II 93 Climate change impacts on forest fire characteristics and restoration strategy in Korea Myungbo Lee 95 Oral Presentation II 103 Effects of disturbance intensity on arthropod communities in burned pine forests in Korea Taesung Kwon 105 Impacts of forest fires on ecosystems functioning including biodiversity and climate change: a perspective from Hindu-Kush Himalayan (HKH) region Sundar P. Sharma 126 Assessment of the crown fire hazard of Pinus densiflora based on the crown fuel characteristics in Korea Youngjin Lee 150 Bush fire management and cooperration Jinjoo Jae 154 The status of forest fires and management in Mogolia Ganbaatar Jamiyansuren Projection of climate change over East Asia based on globar downscaling with high resolution AGCM Jaiho Oh 170 Vegetation regeneration and soil erosion for ten years following forest fires in Korea Yeonsook Jung 187 System for the reduction of CO 2 emitted from forest fire Jun-ichi Kudoh and Masanao Ito 192 Poster Session 197

9 OPENING SESSION

10 International Forest Fire Symposium on Commemorating the International Year of Forests ii -

11 June 7-8, 2011 OPENING ADDRESS I would like to sincerely welcome all distinguished participants for attending the International Forest Fire Symposium in the International Year of Forests 2011 designated by United Nations in the acknowledgement of the importance of forests as the solution to overcome environmental crisis such as climate change. Especially, I would like to acknowledge Dr. Johann G. Goldammer, chair of the Global Fire Monitoring Center (GFMC) under the UN International Strategy for Disaster Reduction (UNISDR), Dr. Tian Xiaorui, a researcher at the Chinese Academy of Forestry, Dr. Bambang Hero Saharjo, a professor at Bogor Agriculture University in Indonesia, Sundar P. Sharma, the representative of South Asian Networks, Kazuya Uezu, a professor at the Department of International Environmental Engineering, the University of Kitakyushu, and Ganbaatar Jamiyansuren, the senior officer at the National Emergency Management Agency in Mongolia. Also, I want to thank the presenters, professors who will serve as the chairmen and participate in the discussion session for today's symposium, who are, Dr. Gyuho Lim, a professor at Seoul National University, Dr. Siyoung Lee, Dr. Jaeseon Yi, Dr. Yeonsook Jung, professors at Kangwon National University, Dr. Youngjin Lee, a professor at Kongju National University, Dr. Jaeho Oh, a professor at Pukyong University, Dr. Woogyun Lee, a professor at Korea University, Dr. Sangok Han of the National Institute of Meteorological Reserach, and Dr. Jinjoo Jae of the National Emergency Management Agency. There have been several major forest fires of unprecedented size, that have recently broken out from place to place due to environmental issues, such as climate change. Forest fires are no longer an issue limited to only one country, rather they have become a global issue. From the tropics to Far-East Asia, ecosystems are becoming more vulnerable to forest fires. Large-scale forest fires have become a reality as well. The international cooperation, which shares expertise experiences, and - iii -

12 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 knowledge from other countries, as well as resources for extinguishing fires is urgently needed in order to deal with such forest fires effectively. The Fifth International Wildland Fire Conference(IWFC), held last May in South Africa, brought about regional agreement through bilateral and multilateral cooperation regarding international cooperation in the forest fire control sector. Additionally, the Fifth IWFC adopted a written agreement on cross-border cooperation at the worldwide level. I am confident that today's the International Forest Fire Symposium and tomorrow's the International Conference of the Pan-Asia Wildland Fire Network are going to contribute to develop improved technology and find the most effective international cooperation. So, this will ultimately contribute to the wise use of forest fires. The Korea Forest Research Institute (KFRI) is fulfilling its duties and fully committed to joint action towards forest fires occuring in Asia as the chair of the Reigonal North East Asia Forest Fire Network and Pan-Asia Wildland Fire Network as well as one of members of Global Wildland Fire Network. Last but not least, I would like to acknowledge everyone's efforts for preparing the symposium with deep gratitude, especially to Mr. Moon Soon Choi, the governor of Kangwon Province, Dr. Sun Do Choi, the vice president of Kangwon National University, and Dr. Don Koo Lee, the minister of Korea Forest Service for their greatly spared supports. I, once again, hope the symposium to be held most successfully, and extend many thanks to all distinguished guests for your commitment and interest in forest fire control. Gil Bon Koo Director General Korea Forest Research Institute - iv -

13 June 7-8, 2011 CONGRATULATORY ADDRESS I would like to offer my congratulations on the opening of the International Forest Fire Symposium to discuss effective ways for improving extinguishment technology, the forest ecosystem restoration technology, and international cooperation on large-scale forest fire. I believe it is very meaningful to hold the Symposium here in Yangyang, on the east coast of Korea where major forest fires broke out in 2000 and The global change resulting from climate change, population growth, and land use change makes the world more vulnerable to forest fire and the frequency as well as the intensity has become increasingly serious over the past decades. For the resolution of these issues, establishment of integrated management strategy on forest fire, local community-based development and adaptation of forest fire management policy, and practical actions of international cooperation have become essential at this juncture. Beginning with its response to a series of forest fires on the east coast of Korea in 2000, the Korea Forest Service as the government body in charge of forest fire has notably developed the whole process of forest fire prevention, extinguishment and restoration. The Korea Forest Service is affectively achieving the objectives of early fire detection and rapid initial attack through various measures, in particular by adopting cutting-edge information technologies to the field of forest fire management. Additionally, the Korea Forest Service is providing training and education on forest fire control and the resources of extinguishment through bilateral and multilateral international cooperation in the forestry sector. The Korea Forest Service is fulfilling its duties as a major partner of international community also by hosting the sixth International Wildland Fire Conference (IWFC) in 2015, the largest international conference on forest fire. - v -

14 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 The International Symposium on Forest Fire Control and the International Conference of the Pan-Asia Wildland Fire Network to be held tomorrow are going to be a solid platform for the successful meeting of the sixth International Wildland Fire Conference (IWFC). I hope strategies and technologies of forest fire management of every country and international organizations could be shared and ways of enhancing these technologies could be discussed, thereby contributing to forging effective countermeasures on worsening wildland fires. And last but not least, I would like to express my gratitude to everyone who has been greatly supportive to bringing about this Symposium and would like to extend special thanks to Dr. Johann Goldammer, the chair of the Global Fire Monitoring Center (GFMC), many other distinguished guests from abroad, and Mr. Moon Soon Choi, Governor of Kangwon Province, Dr. Sun Do Choi, the vice president of Kangwon National University. I hope the Symposium builds up very successful and fruitful outcomes. Don Koo Lee Minister Korea Forest Service - vi -

15 June 7-8, 2011 CONGRATULATORY ADDRESS Excellencies, Mr. Governor Moonsoon CHOI of Gangwon Province, Republic of Korea, Minister Dr. Donkoo LEE, Head of the Korea Forest Service Dr. Gilbon KOO, Director General of the Korea Forest Research Institute Dr. Sundo CHOI, Vice President of Gangwon University Let me take the opportunity of the opening of the International Forest Fire Symposium Commemorating the International Year of Forests 2011 and the Second Pan-Asia & 7 th North-East Asia Wildland Fire Network Meeting to reflect on the current situation of forest fires in the Asian Region and globally. In today s conference forest fire specialists from Korea and neighboring countries of Asia are coming together to exchange recent experiences and views on fires affecting ecosystems throughout the region. At the Global Fire Monitoring Center, which is based in Germany, we are monitoring fires in the region regularly using satellite imageries. What we can see from space are the high-temperature signals indicating active fires. We can see the spread of smoke plumes in countries, and often crossing borders of countries and spreading thousands of kilometers. And we can see the black scars in the landscapes of Asia that are indicating the area burned in the different ecoregions. This spring we have monitored fires all over the region. Major vegetation fires have been detected burning in Japan, on Kyushu Island on 13 March 2011, two days after the earthquake and tsunami hit the country. These fires were probably ignited by the many structural fires that started during the earthquake and spread to surrounding lands. At this time, however, the attention was given to save the lives of people affected by the disaster and to control the failure of the Fukushima Daiichi Nuclear Power Plant. During the same week extensive wildfires were monitored in the highlands of Nepal, including fires burning in the Annapurna Conservation Area. You will - vii -

16 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 hear more about the fire situation by the contribution of Mr. Sundar Sharma from Nepal and the Regional South Asia wildland Fire Network during this conference and the regional network meeting. One month later extensive fires burning in the North of the Korean peninsular were depicted by satellites, indicating widespread agricultural burning activities producing smoke plumes that reached Japan. In neighboring Southeast Asia the amount of fires set for conversion of native vegetation to other land use decreased as compared to earlier years. By monitoring not only the active fires in Southeast Asia but also the political environment we could see that leaders of the ASEAN member states did their best to meet the objectives of the regional ASEAN Agreement on Transboundary Haze Pollution and to reduce illegal burning activities throughout the region. You will hear more about recent developments by the representative of Indonesia, Professor Bambang Hero Saharjo from Bogor Agricultural University. In the middle of May 2011 extensive wildfires started to affect the Central- Northern Asian region, in Yakutia, Russian Federation. The daily progress of wildfires affecting more than 100,000 hectares of forests could be seen from space until clouds and rainfall moved in towards the end of May and obscured the view from space. In Europe, where the Global Fire Monitoring Center is located, we are experiencing an extreme drought. It looks like 2011 will become one of the driest years since long time possibly a consequence of regional climate change. Many western European countries like the United Kingdom, Ireland, The Netherlands and Belgium experienced large wildfires phenomena that are usually uncommon in these countries. The same happened in the United States of America where wildfires burning in dry grasslands, agricultural lands and forests caused major economic and humanitarian problems during the recent weeks. Wildland fire specialists from throughout the world discussed the current and probable future situation four weeks ago at the 5 th International Wildland Fire Conference in South Africa. The International Wildland Fire Conferences are taking place every four years and provide a major platform and forum of the - viii -

17 June 7-8, 2011 international community of wildland fire specialists globally, especially those who are organize in the Global Wildland fire Network. In an opening statement conveyed by the Secretary General of the United Nations,Mr. Ban Ki-moon, he encouraged the participants of the conference to identify real solutions that will help communities and nations to better handle the adverse impacts of fires and to build safer, more sustainable societies for all. In the closing statement of the conference the Vice President of the Korean Forest Service, Mr. Yougho Ha, announced that the Republic of Korea with host the 6 th International Wildland Fire Conference in With this Korea has taken over the torch to bring the spirit of global cooperation of wildland fire scientists, managers and policy makers to Asia. I would like to congratulate the Republic of Korea, notably the Korea Forest Service and the Korea Forest Research Institute, to become host of the next global wildland fire conference. During this meeting and forthcoming consultations with the International Liaison Committee of the International Wildland Fire Conferences, as well as the Pan-Asia Wildland fire Network meetings this week and in 2013, the Republic of Korea will be a major contributor towards building global partnerships in fire management. May I congratulate you, Mr. Governor Moonsoon CHOI, and Minister Dr. Donkoo LEE, as well as Dr. Gilbon KOO and Dr. Sundo CHOI, and the dedicated staff of the Department of Forest Conservation of the Korea Forest Research Institute, to host this conference and network meeting. I wish all of us a productive work atmosphere and a good spirit of international collaboration Johann Georg Goldammer Chairperson Global Fire Monitoring Center (GFMC) - ix -

18 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 WELCOME ADDRESS It is a great pleasure to host the International Forest Fire Symposium, here in Yang-yang, Kang-won-do, for the International Year of Forests Also, I would like to welcome you all; forest experts, government officials, and other honoured guests from both home and abroad. Here, the eastern cost area of Kang-won-do, has been susceptible to natural disasters having forest fires and drought during the spring and fall, flood and typhoons in the summer, and snowfall in the winter. The consequences of forest fires have not just been forest damage but have extended to the loss of domiciles and forest crops which threaten the basic livelihood of local people. Here in Yangyang, during the forest fire which occurred on the 4 th of April, 2005, a thousand year old Buddist temple, Naksan-sa sustained severe damage and the huge bronze bell, one of our national treasures was also destroyed. As well as Yangyang forest fire, small to mega forest fires, such as Go-sung forest fire in 1996, Dong-hae-an forest fire in 2000, have continuously occurred in the eastern area of Kang-won-do, which has caused great damage to forests and have resulted in local people suffering as a consequence. However, since the Yangyang forest fire in 2005, no such big fire has occurred. I believe that this is the result of the continuous efforts made by Korea Forest Service, nation, the government, and military, towards successful forest management. This symposium, held here in Yangyang today, is more meaningful as it is the 10 th year since the Dong-hae-an forest fire occurred in I hope that studies on forest management, including its current situation, fire prevention, fire control techniques, and rehabilitation of forests presented today can contribute greatly to forest fire prevention. - x -

19 June 7-8, 2011 I d like to thank you all again for being here today and I hope that through this international symposium, we can foster forest fire prevention techniques and reduce damages from the fire in order to preserve our forests. Lastly, I also would like thank those of you who helped us to prepare today s event. Thank you. Sun Do Choi Vice president Kangwon National University - xi -

20 International Forest Fire Symposium on Commemorating the International Year of Forests xii -

21 KEYNOTE SPEECH I

22 International Forest Fire Symposium on Commemorating the International Year of Forests

23 June 7-8, 2011 International Protocols and Agreements on Cooperation in Wildland Fire Management: Needs, Current Status, and the Way Ahead in the Asian Region and at Global Level Johann Georg Goldammer* Global Fire Monitoring Center, Secretariat of UNISDR Global Wildland Fire Network / Wildland Fire Advisory Group and International Fire Aviation Working Group Georges-Koehler-Allee 75, D Freiburg, Germany info@gfmc.org Abstract In the endeavor of protecting global vegetation resources from wildfires and excessive fire application in land use and land-use change, including the protection of human and environmental security from the adverse direct impacts or secondary effects of wildland fires such as air pollution and emissions affecting the global atmosphere and climate, countries are increasingly willing to share expertise and resources in fire management, including transboundary cooperation and mutual assistance in wildland fire emergencies. Attempts to further develop transboundary cooperation includes sharing of scientific knowledge and practical expertise in integrated fire management, i.e. the integration of natural fires and managed prescribed fires in sustainable ecosystem management and land use. In order to realize systematic and efficient international cooperation the international wildland fire community needs agreements and internationally agreed guidelines that provide protocols and procedures and thus enhance governance, efficiency and efficacy for international cooperation. Keywords: Wildfire, wildland fire, fire management, international cooperation, legal agreements, voluntary agreements, UNISDR Global Wildland Fire Network, UNISDR Wildland Fire Advisory Group 1. Introduction Increasing vulnerability of societies and the environment to wildfires is noted globally. The severity and destructivity of wildfire disasters, which in many cases exceed the response capabilities of a country affected, and the transboundary - 3 -

24 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 consequences of fires are prompting governments to develop cooperative mutual (reciprocal) agreements on emergency assistance at bilateral level, and in some cases at multilateral level. Several international (global) conventions, such as the three Rio Conventions (CBD, CCD and FCCC), are examples of international legal agreements that provide rationale and catalogues of commitments for signatory countries to protect the global environment. However, none of the international legally binding conventions or any informal or voluntary international instrument, such as the Hyogo Framework for Action : Building the Resilience of Nations and Communities to Disasters, is addressing the wildland fire problem explicitly. There are also no protocols in place that are providing internationally accepted standard methods and procedures for countries providing and receiving assistance in wildland fire emergencies that would ensure efficiency and inter-operability of cooperating parties. In preparation of and following up the International Wildland Fire Summit of the international wildland fire community had taken steps to develop first concepts, templates and guidelines with widely agreed principles and best practices in fire management and incident command. Detailed operational standards are needed now to facilitate the exchange of fire fighting resources, including aviation, management personnel, and equipment. The UN Disaster Assessment and Coordination (UNDAC) and the International Search and Rescue Advisory Group (INSARAG) Guidelines 2 and the principles laid down in the Rosersberg Initiative Improving the international environmental emergency response system may serve as examples for developing interoperable standards, protocols, Standard Operating Procedures (SOPs) and rules of engagement. At the level of multilateral bodies, e.g. under the umbrella of the Association of South East Asian Nations (ASEAN), the UN Economic Commission for Europe (UNECE), the Asia-Pacific Economic Cooperation (APEC), the European Union (EU) or the Southern African Development Community (SADC) recent developments have revealed an interest of countries to enhance capability in regional transboundary cooperation in fire management. Experience gained in bilateral (reciprocal) agreements, including the common use of the Incident Command System (ICS) as practiced in exchanges under bilateral agreements and

25 June 7-8, 2011 between North American countries (U.S.A., Canada and Mexico) and between the U.S.A. and Canada on the one side, and Australia and New Zealand on the other side, may serve as examples for developing regional agreements or protocols. 3 The Global Wildland Fire Network (GWFN) 4, a voluntary network borne as an activity of the UNECE/FAO Team of Specialists on Forest Fire 5 and operating under the United Nations International Strategy for Disaster Reduction (UNISDR), facilitated by the UNISDR Wildland Fire Advisory Group (WFAG) and the Global Fire Monitoring Center (GFMC) 6, is promoting international cooperation in wildland fire management, notably through capacity building in wildfire prevention, preparedness and suppression, and the development of standardized procedures for use in international wildfire incident response. Long-term the GWFN is also aiming at the development of an International Wildland Fire Accord (voluntary or binding under international law), which would be based on the rationale that there is a common international interest in protecting of global vegetation cover against degradation or destruction and that common endeavors in fire management will contribute to disaster risk reduction, e.g., direct fire damages to human assets and ecosystems, fire-generated smoke pollution affecting human health and security, release of greenhouse gases, secondary disasters such as landslides, erosion, and floods, as well as threats to biodiversity. In the following sections examples of achievements and some currently ongoing activities are highlighted, which reflect some progress in the dialogue to enhance international cooperation in wildland fire management. Within the Asia Cluster of the Global Wildland Fire Network, consisting of the four UNISDR Regional Wildland Fire Networks of North East Asia, South East Asia, South Asia and Central Asia this dialogue has been initiated during the first regional meeting in Busan, South Korea, February 2009, and will be continued at this International Forest Fire Symposium commemorating the International Year of Forests See special issue of UNECE/FAO International Forest Fire News (IFFN) No. 29, with examples of agreements and Annual Operating Plans: Note: An earlier version of this paper has been presented at the 5 th International Wildland Fire Conference, South Africa, 9-13 May

26 International Forest Fire Symposium on Commemorating the International Year of Forests Progress in regional cooperation in fire management 2.1 Association of South East Asian Nations (ASEAN) As a consequence of extended fire and smoke episodes since the early 1980s and especially in the 1990s ten member states of the Association of Southeast Asian Nations (ASEAN) started to negotiate an agreement at regional ASEAN level to address regional air pollution caused by land-use fires and wildfires. In June 2002 the Agreement on Transboundary Haze Pollution was adopted and came into force on 25 November 2003, with nine states currently participating (Brunei Darussalam, Cambodia, Lao PDR, Malaysia, Myanmar, Philippines, Singapore, Thailand, and Viet Nam). The Agreement is the first regional arrangement in the world that binds a group of contiguous states to tackle transboundary haze pollution resulting from land and forest fires. 8 The Agreement requires the Parties: - cooperate in developing and implementing measures to prevent, monitor, and mitigate transboundary haze pollution by controlling sources of land and/or forest fires, development of monitoring, assessment and early warning systems, exchange of information and technology, and the provision of mutual assistance; - respond promptly to a request for relevant information sought by a State or States that are or may be affected by such transboundary haze pollution, with a view to minimizing the consequence of the transboundary haze pollution; and - take legal, administrative and/ or other measures to implement their obligations under the Agreement. The Agreement establishes an ASEAN Coordinating Centre for Transboundary Haze Pollution Control to facilitate cooperation and coordination in managing the impact of land and forest fires in particular haze pollution arising from such fires. Pending the establishment of the Centre, ASEAN Secretariat and ASEAN Specialized Meteorological Centre (ASMC) co-performed the interim functions of the Centre. Quite noticeably, Indonesia is missing from the list of state parties. This presents a particularly acute problem for the region since Indonesia is by far the biggest source of the fires and haze (Khee-Jin Tan, 2005; Nguitragool, 2011). In 8 See ASEAN website Haze Online : and the full text of the agreement at:

27 June 7-8, 2011 early 2011 Indonesia indicated its willingness to push the ratification European Union (EU) In the 1980s and 1990s there was some exchange of expertise on firefighting within the EU but little in the way of formal cooperation. The European Union Civil Protection Mechanism, was established in 2001 and further strengthened in It provided a new capacity for coordination for Europe. It now plays a central role in the EU forest fire risk prevention and forest firefighting coordination at EU level. There are currently 31 countries participating in the Mechanism ( Participating States ): The 27 Member States of the European Union (EU) together with Iceland, Liechtenstein, Norway and Croatia. The Mechanism, which is managed by the European Commission, has tools to cope with wildfires in three phases of the disaster management cycle. The main responsibilities and the tools allocated to the European Commission are outlined as follows 10 : Monitoring and prevention The core body in the operation of the European Union Civil Protection Mechanism is the Monitoring and Information Centre (MIC). The MIC s three major roles are: - to provide a coordination platform for exchange of requests for assistance and offers of resources among Participating States; - to be an agent for information exchange and dissemination on natural and man-made disasters worldwide and the Mechanism interventions; - to be a coordinator that identifies gaps and develops solutions on the basis of the information it receives, and facilitates the pooling of common resources where possible, and supplies expert teams in the field of disaster to tackle the problems more effectively. The Common Emergency Communication and Information System (CECIS) Note: This general description of the EU Mechanisms has been taken from the introduction to the report Study on wild fire fighting resources sharing models, prepared by GHK Consultants to DG ECHO, European Commission (October 2010), in which the author participated (the study is not yet published)

28 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 facilitates coordination between the MIC and national authorities. The main tasks of CECIS include hosting a secure and reliable database on potentially available assets for assistance; handling requests for assistance on the basis of these data; and facilitating the exchange of information and documenting all action and message traffic. The MIC receives fire risk assessment information from the European Forest Fire Information System (EFFIS). This web-based platform, which consists of a scientific and technical infrastructure, was developed jointly by the European Commission Joint Research Centre and Directorate General Environment (European Commission). Preparedness The EU Civil Protection Mechanism intermediates dissemination activities and exchange of best practice among Participating States and provides training programmes and exercises to intervention teams. It organizes informative activities, seminars, conferences and pilot projects on the main aspects of interventions. It also provides access to the assets in the European Union Forest Fire Tactical Reserve (EUFFTR), a pilot project designed to step up cooperation between Member States on combating forest fires during high risk seasons. The project, to which two Canadair CL-215 aircraft were allocated during the summers of 2009 and 2010, was activated in the cases where Member States were not in a position to provide assistance to a requesting country due to their aerial resources being needed in their own territory or because they could not reach the fire site quickly enough. The Mechanism develops implementing rules for module development and administers the CECIS module database. Response Through the EU Civil Protection Mechanism, the European Commission is able to: - Mobilize small teams of experts to the site of an emergency; - Provide and distribute information during an emergency/intervention; - Play a facilitating role in the coordination of assistance requests and offers - 8 -

29 June 7-8, 2011 from Participating States; - Coordinate with other actors at the international level and with other EU services; and - Provide co-financing for the transport of assistance to the affected areas, on the request of the offering Participating States. The EU Civil Protection Mechanism is well-accepted and increasingly used by Participating States. Twenty-eight requests for assistance were received by the MIC in from the EU and 18 from third countries, compared with just three in total in The Mechanism coordinates and facilitates voluntary efforts each Participating State decides on its contribution on an ad hoc basis. 2.3 Asia-Pacific Economic Cooperation (APEC) In October 2010 the first International Conference on Forest Fires: Management and International Cooperation in Preventing Forest Fires in APEC Region was convened at the initiative of the Russian Federation and aimed to strengthen cooperation between the emergency services of the APEC member economies in order to express readiness of the region to reduce the risks of disasters. 11 Through an in-depth and comprehensive analysis of the problem of forest fires in the APEC region and the world, the conference identified the urgent necessity and importance of joint efforts, mutual help and cross-border cooperation in forest fire risk reduction. The conference released the Khabarovsk Recommendations on Management and International Cooperation in Preventing Forest Fires in the APEC Region. Among other the following priority directions of international cooperation under APEC were proposed: - Development of an international mechanism to monitor and enhance responsibility of the APEC member economies to ensure forest fire protection on their territories, and coordinate action under APEC with the existing institutions of international cooperation, such as UNISDR Global Wildland Fire Network, ASEAN, UNECE and others. - Promotion of economic cooperation in projects that aim to reduce the degree of fire risk and restoration of forests on lands degraded by fire and nonsustainable forest management; - Development of bilateral agreements on cooperation in fire management,

30 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 particularly between APEC economies sharing common borders, and a regional voluntary agreement on cooperation in fire management, aiming at harmonizing cooperation with neighboring regional entities such as the UNECE and ASEAN, particularly in the light of overlapping membership of some economies. - Development of long-term fire management strategies in each economy that allow for the mitigation of the consequences of climate change. - Improvement of strategic and operational early warning of forest fire risk in the APEC region as a regional activity to be coordinated with the Global Wildland Fire Early Warning system. - Conduct regular consultations, and exchange knowledge and best practices informed by a high level of expertise within directions of APEC. - Reconvene and contributing to the 5 th International Wildland Fire Conference scheduled for 2011 (South Africa), and the following conference scheduled for 2015 (South Korea). 2.4 Southern African Development Community (SADC) In the last two decades, vegetation fires have become a major concern in the region of the Southern African Development Community (SADC) with regard to the negative impacts they have on the environment and humans welfare. 12 Uncontrolled (un-prescribed) wildfires cause forest and vegetation degradation and related biodiversity loss resulting in immediate and long-term impacts on the livelihoods of local communities and upstream impacts on national and regional economies. Fires in the tropical environment are a major contributor to tropical forest degradation, where over time frequent fires lead to savannization in these areas. However, fires are also needed to maintain healthy ecosystems and biodiversity of African savannah and grassland vegetation types, many of them being adapted to regular fire influence. Prescribed or controlled burning is used to meet objectives often essential to sustaining livelihoods. Fire is used also for conservation reasons, removal of old growth, suppression of bush encroachment and stimulation of the growth of grazing grass as well as the removal of fuel with the aim of pre-empting dangerous wildfires at the peak of the fire season. The SADC region of 14 Member States is home to 238 million people of 12 The information provided in the chapter are taken from the SADC Regional Fire Management Programme Document, draft proposal (June 2010). Web source: Africa/SADC%20Regional%20Fire%20Management%20Programme%20Document-Final-6.pdf

31 June 7-8, 2011 which approximately 75% are rural based. The perceived rise in the number of wildfires negatively affects these rural communities, many living near forests that provide them with their basic needs. The on-going process of climate change has the potential to exacerbate this situation by altering the frequency, intensity, severity and seasonality of fires in the SADC region. A SADC regional fire management programme has been proposed in It provides a framework for cooperation on fire management issues across national boundaries. Fire management is a technical, socio-cultural and political challenge that requires an effective network of willing partners that include governments, the private sector, local communities and international partners to find the appropriate balance between developing and conserving natural resources and managing unwanted fires while at the same time promoting the safe use of beneficial fires. The programme intends to foster cooperation and collaboration on fire management on a regional basis to move towards integrated environmental policies and fire management practices. The programme pursues a multiple stakeholder approach working closely with regional and international organization to support five areas of fire management: legal and regulatory aspects of fires, community based fire management, institutional strengthening and establishment of a fire management coordination centre, generation and dissemination of relevant fire information for detection and early warning as well as lastly associated capacity building in the respective areas. The envisaged programme is based on the SADC Protocol on Forestry of 2002, which forms the policy framework for sustainable forest management in the SADC member states. One of its objectives is to achieve effective protection of the environment, and safeguard the interests of both the present and future generations. The SADC Forestry Strategy of 2010 is based on the vision is to develop and maintain a vibrant and evolving forest sector that contributes to rural development, poverty reduction and, industrial progress, while retaining the vital ecosystem services of forests such as, water supply, climate change mitigation, and protecting biological diversity and thereby providing the motivation for countries to cooperate for their protection, management, and sustainable use. The purpose of the strategy is to provide a framework for both regional cooperation and international engagement on forest issues; paying special attention to issues that transcend national boundaries and to encourage concerted action by SADC Member States in the management, conservation and sustainable use of their forests. The strategies mission aims to facilitate cooperation among member states

32 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 to promote the active protection, management and sustainable use of forest resources, through, sound policy guidance and the application of requisite skills and the best available technology, in order to enjoy the multiple benefits of forests in perpetuity. There is increased willingness by SADC member states to cooperate on fire management on a regional and international basis. There is also recognition that a regional framework, based on cross border, regional and international cooperation is required to address issues of national, regional and transboundary fire management. Member states have expressed the need for a regional agency or centre to foster and coordinate such cooperation and information exchange in fire management. Establishing a regional agency or centre responsible for collection and analysis of fire related data and formulating standardised rules, guidelines and procedures will ensure reliability and coordinated dissemination of relevant information as well as guide policy processes and development. Furthermore it would spearhead the promotion of integration of Community-Based Fire Management (CBFiM) into national policies and fire management strategies. The expected activities and outcomes of the SADC Regional Fire Management Programme include the establishment of a Regional Fire Management Coordination Center. This center will also facilitate and coordinate international and regional cooperation in fire management providing a framework within one country may request and receive wildfire suppression resources from another country and to encourage cooperation and exchanges on other fire management activities such as training and lessons learnt. A SADC fire management programme will facilitate the development of a SADC Memorandum of Understanding (MoU) outlining cross border cooperation to combat transboundary fires by fire teams from different Member States as well as operational guidelines for the regional coordination centre. The programme intends to foster regional level interaction by developing guiding policy frameworks and procedures for several aspects of fire management. During the Consultative Workshop on the Development of a SADC Regional Fire Management Programme (January 2010, Maputo) participants from all SADC member states identified and compiled capacity development measure to be carried out by all SADC member states to ensure success of a regional fire management program. The project includes the following components: - Establishment of a Regional Fire Management Coordination Center. Objective: To promote the establishment of a regional fire management coordination

33 June 7-8, 2011 centre for improved stakeholder cooperation and collaboration - Reform and Harmonization of policies and Procedures. Objective: To secure essential policy harmonization at national and regional level to provide the basis for controlling harmful fires and promoting the safe use of beneficial fires within SADC - Community based fire management. Objective: To secure essential policy harmonization at national & regional level to provide the basis for controlling harmful fires and promoting the safe use of beneficial fires within SADC - Fire information. Objective: To improve production, access, dissemination and application of fire information within the region - Capacity development. Objective: To increase awareness of and knowledge in balanced and integrated fire management and its elements In late 2010 a trilateral project between South Africa, Tanzania and Germany (Trilateral Cooperation Fund TRI-CO Fund) has been launched to develop improved coordination and exchange of techniques, resources, science and other capacity building measures related to fire and fire management amongst contributing parties and SADC member states (duration: ). 2.5 UN Economic Commission for Europe (UNECE) Within the last decade the region of the United Nations Economic Commission for Europe (UNECE) has experienced a number wildfire episodes that have resulted in severe environmental damages, high economic losses and considerable humanitarian problems. 13 Reasons and underlying causes of extreme wildfires may be quite different throughout the region a region which is extending over the whole temperate-boreal northern hemisphere, bearing a large variety of ecosystems and land-use systems. However, there are issues in common. For instance, in the Eurasian part of the region many neighboring countries are sharing similar natural, cultural and social conditions that are determining the occurrence and impacts of wildfires. Countries are also confronted with newly arising or escalating fire problems resulting from changes in the political, social and economic environment, land-use change and climate change. 13 This section has been derived from the planning document for the UNECE/FAO Regional Forum on Cross-boundary Fire Management, to be organized by the UNECE/FAO Teams of Specialists on Forest Fire through its Coordinator the Global Fire Monitoring Center (GFMC), Germany, and supported by the Council of Europe as a contribution to the UN International Year of Forests in For details and updates: see website of the UNECE/FAO Teams of Specialists on Forest Fire:

34 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 On the other hand progress is noted in advancing multinational cooperative efforts in wildland fire research and related sciences and in transboundary cooperation in fire management. At national levels increased openness of sectoral mandated institutions to overcome historic division of roles of agencies in fire management is noted. There is a high interest of governments, national agencies, international organizations and civil society in the UNECE region to address the increasing threats or newly arising problems by fire management solutions that could be developed collectively, and thus economically, allowing inter-operability in fire management between nations and regions. In early 2012 the UNECE / Council of Europe Regional Forum on Crossboundary Fire Management will be organized at the United Nations in Geneva by the Global Fire Monitoring Center (GFMC) and the Secretariat of the UNECE/FAO Forestry and Timber Section, and co-sponsored by the Secretariat of the Euro-Mediterranean Major Hazards Agreement (EUR-OPA), Council of Europe. The conference will elaborate recommendations to UNECE and CoE member states to take advantage of recent insights and solutions of contemporary and expected future wildfire problems. Main focus of the conference will address the situation in countries in which progress of enhancing fire management capabilities is limited, e.g. as a consequence of political and administrative transition as well as difficult economic conditions, or countries with significant or extraordinary fire situations, which would benefit from the experience of their neighbor countries. The outcome of the conference will build on existing and already proposed initiatives, e.g.: - New approaches in integrated vegetation management with regards to renewable energy concepts and carbon storage, some related to the UNFCCC endeavor to reduce deforestation and forest degradation in the frame of the international climate and forest regimes, with relevance to wildfire hazard reduction and opportunities for fire management to contribute; - Wider application of prescribed fire in nature conservation, forestry and landscape management, with encouraging progress of countries cooperating under the Eurasian Fire in Nature Conservation Network 14 and similar

35 June 7-8, 2011 initiatives; - Exploitation of the results of successfully completed international fire research projects for developing adequate public policies affecting fire management and operational fire management, e.g. the latest accomplished multi-national Fire Paradox 15 project, or the ongoing development of the multinational Alpine Forest Fire Warning System (ALPF FIRS) 16 ; - Introduction and further development of competency-based fire management training standards for achieving qualifications for fire fighters, foresters and land managers, e.g. the EuroFire Competency Standards 17 ; - Strengthening dedicated networks of wildland fire specialists, agencies and other representatives of civil society, e.g. the six Regional Wildland Fire Networks of the Global Wildland Fire Network that are covering the UNECE region; - Application and further development of existing and development of new bilateral agreements on reciprocal transboundary assistance in wildfire emergencies all over the ECE region; - Endeavor to enhance governance of UNECE member states to provide and receive assistance in wildfire (and other environmental) emergency situations by setting up standards, protocols and agreements, in cooperation with procedures evolving under the lead of the UNEP/OCHA Joint Environment Unit and the UN Advisory Group on Environmental Emergencies, e.g., the proposed creation of an Environmental Emergencies Center (EEC); - Follow up of the recommendations of regional groups, projects and programmes, or earlier regional conferences to enhance international cooperation in fire management in the UNECE region and adjoining regions, e.g. the recommendations for the development of a Regional Strategy for Cooperation in Fire Management in Southeast Europe of , the outcomes of expert meetings such as the recommendations of the workshop Assessment of Forest Fire Risks and Innovative Strategies for Fire Prevention, an activity of the Ministerial Conference for the Protection of Forests in Europe (MCPFE) of 2010, or the European Commission network of fire experts and its information system, the European Forest Fire Information System (EFFIS) 19 ; or the recommendations from projects supported by the Environment and Security Initiative (ENVSEC) addressing wildland fire, human security

36 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 and peace in the EECCA region 20, and the outcomes of the International Conference on Cross-Boundary Fire Management (Irkutsk, Russia, 2010) 21 and the APEC Conference on Forest Fire Management and International Cooperation in Fire Emergencies of the Asia Pacific (Khabarovsk, Russia, 2010) 22. The outcomes of the forum shall be regarded as complementary and valueadded to existing agreements and mechanisms. The conference will be held as a contribution to the UN International Year of Forests (2011) to further the objectives of the international forest and climate regimes and shall contribute to the evolving of an international wildland fire regime as envisaged by the UNISDR Global Wildland Fire Network. A large number of countries of the UNECE region are members of the Council of Europe, member states of the European Union and signatory states of the MCPFE, and are all concerned of the impact of climate change on forests and forest destruction by fire. There is a collective demand for robust forest policies. However, wildland fires are not only impacting the protection and functioning of forest ecosystems. Fire use and wildfire occurrence in the cultural landscapes of the region shaped by agriculture, pastoralism and forestry, have considerable impacts both in negative and positive ways on landscape patterns, land productivity, biodiversity and the atmosphere with implications on air quality, human health and security, and climate change. The UNECE Convention on Long-Range Transboundary Air Pollution 23, the European Landscape Convention 24 or the Convention on the Conservation of European Wildlife and Natural Habitats (Bern Convention) 25 are examples of conventions that are addressing tangible and pressing regional issues and that are relevant although not yet explicitly referred to a potential regional wildland fire regime. Thus, the overall aim of the Forum will be a first step towards the development of an agreement on international cooperation to enhance fire management 20 Note: The objectives of the Forum are not intended to interfere or whatsoever with the Community Civil Protection Mechanism of the European Union (EU) and its efforts of coordination of international / multinational assistance in (wildfire) emergencies in Europe and outside the EU

37 June 7-8, 2011 capability in the region. Right from the beginning of the political and technical planning and consultation process in preparation of the Forum is considered to be connected to the outcomes of the 5 th International Wildland Fire Conference (May 2011) and a follow-up process. 2.6 Bilateral reciprocal agreements with multilateral character: Examples Looking back a decade the wildland fire season in the United States during the year 2000 at that time was the worst fire season in more than 50 years. Almost 100,000 fires burned more than 2.8 million hectares of forest and range lands. This was approximately twice the U.S. ten-year average. The season was long and difficult and firefighters faced dangerous burning conditions throughout the western U.S.A. 26 Faced with this unprecedented situation, and with a forecast for a continuing hot and dry weather pattern, fire managers realized they would need to reach beyond U.S. borders for assistance. During the remainder of the 2000 fire season, the U.S. received assistance from more than 1200 Canadian firefighters, 96 fire specialists from Australia and New Zealand and 20 Mexican firefighters. These additional resources performed important roles in the U.S. fire fighting efforts. Some international fire fighters provided much needed support to fire crews on the fireline while others performed as middle managers on incident management teams. International agreements with Canada and Mexico were in place prior to the 2000 fire season but none existed with Australia and New Zealand. Why Did the U.S. go to Australia and New Zealand? U.S. fire managers have had informal study group exchanges with Australia and New Zealand for more than 50 years. These exchanges provided opportunities to share information about each other s programs and experiences. Based on that exchange relationship and the knowledge of the Australian and New Zealand firefighting systems, when the U.S. fire situation reached a critical level in 2000, the U.S. approached Australia and New Zealand and asked for their help. The 96 fire experts that came to the U.S. were integrated into the U.S. fire fighting 26 The essential parts of this section are taken from USDA Forest Service (2003)

38 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 organization where they served in middle management positions on fires, freeing up U.S. fire managers to take on more critical tasks. The Relationship in 2000 The justification for seeking the assistance of Australia and New Zealand in 2000 was based on an assumption that a U.S. Public Law called the Wildfire Suppression Assistance Act, gave the authority to the U.S. Secretaries of the Interior and Agriculture to enter into assistance agreements, especially because of the critical situation facing fire managers on the fire lines. U.S. fire managers, who coordinated this activity during 2000, did not understand the clearance and vetting processes required by the U.S. Department of State prior to putting Australia and New Zealand fire fighters in harms way on the fire lines. Fortunately, no serious incidents occurred, and all of the Australian and New Zealand firefighters returned home safely and from all accounts, they performed ably and were a credit to their countries. When the 2000 fire season was over, U.S. fire managers understood that the assumed authority used to bring Australia and New Zealand fire fighters over would not be viable in the future. And the biggest issue that arose was a concern about tort claim liability. Solicitors and risk managers from Australia and New Zealand, upon review of U.S. law did not feel that their firefighters were provided sufficient legal coverage by the U.S., if an Australian or New Zealand firefighter, even properly performing his duties, was involved in unforeseen fire line incident which caused injury or damage to other parties. New arrangements needed to be developed and signed by all parties before the U.S. could once again request and receive help from Australia and New Zealand. The 2001 Christmas Fires in Southeastern Australia The devastating fires around Sydney and other areas of southeast Australia in 2001, like the U.S. fires in 2000, drew worldwide attention. The U.S., trying to return the favor of help that it received in 2000, offered to provide some assistance. Thanks to a break in the weather, the professionalism of the their fire fighters, and to the incorporation of some lessons learned from their experiences in the U.S. in 2000, Australian fire managers were able to control their fires. However, had U.S. help been needed, the ability and timeliness of the U.S. to provide assistance would have been greatly hampered by the lack of signed arrangements and

39 June 7-8, 2011 operating plans. This once again highlighted the need for the completion of more formal arrangements and protocols between the countries which would meet the concerns of all parties. The 2002 U.S. Fire Season After a quiet wildland fire season in 2001 in the U.S., the 2002 wildland fire season was just as challenging to fire managers as As conditions in the forests and rangelands of the Western U.S. worsened and record setting hectares burned in several states, the U.S. again prepared to call for international assistance. However, Canada which provided valuable and lengthy support to the U.S. in 2000 faced its own fire problems stretching from Quebec to British Columbia. And there were still unresolved issues with arrangements and protocols with Australia and New Zealand. Throughout 2001 and up to August of 2002 U.S., Australian, and New Zealand fire managers, risk managers and solicitors had been proposing and reviewing potential options to solve the liability concerns raised after the 2000 fire season. One alternative that was explored was purchasing sufficient liability insurance to meet risk managers requirements, but the cost was prohibitive, and the policies would have been too complex. The best possible solution was to change U.S. law that would give any international firefighter brought to the U.S. under the Wildfire Suppression Assistance Act, tort liability coverage equivalent to that provided to U.S. Government fire fighters. In early August the bill was passed and signed by the President of the U.S. The language in the bill provided the assurance required by Australian and New Zealand and U.S. fire managers were once again allowed to request international assistance from Down Under. Signatures of the Secretaries of Agriculture and the Interior were quickly inked on the official Arrangement papers and posted overnight to Australia and New Zealand. The Australian States of Victoria, New South Wales, Tasmania, Western Australia, and South Australia and New Zealand signed the Arrangements. Within a week of the passage of the legislation, 50 Australian and New Zealand fires specialists were on U.S. fire lines filling, as they did in 2000, critical mid-level management fire positions in operations and aviation. The Bush Fire Season in Australia The bush fire season in Australia saw some of the worst fires in

40 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 over 50 years. The hardest hit states were Victoria and New South Wales. In later January of 2003, the State of Victoria requested U.S. assistance. Thirty six U.S. fire fighters went to Australia for a month. A twenty person hand crew worked shoulder to shoulder with Australian fire fighters on the fire line. An infrared scanning aircraft with crew and infrared photo interpreters was also sent. And two small incident management teams took on management responsibilities for portions of a massive fire in the Alpine Region of Victoria. The U.S. fire fighters shared their knowledge and experience with their Australian counterparts and also gained a great deal of respect for the challenges facing Australian fire fighters in the management and suppression of bush fires. This request for U.S. assistance tested and solidified the two-way nature of the Arrangements signed in The 2003 U.S. Wildland Fire Season The 2003 wildland fire season initially did not appear that it would reach the proportions of needs that were required in 2000 and However, by mid- July the affects of a multi-year drought, record high temperatures and low humidity and the right weather conditions caused numerous large fires to rage out of control in the Northern Rockies. For the third time in four years U.S. fire managers were facing the need for more international assistance, and Canada, as in much of 2002, was busy suppressing devastating fires of its own in several provinces. And once again, based on the 2002 Arrangements, 40 Australian and New Zealand fire fighters came to the aid of the U.S. The 2003 request reaffirmed the true value of being able to call upon Australian and New Zealand firefighters during extreme wildland seasons. It also began to seem routine. Other Values of the Arrangements Instituting these Arrangements not only provided for a clear mechanism to share fire fighting resources during critical times, it also established formal procedures for cooperative exchanges outside of critical fire season time periods. Knowledge of each other s capabilities and systems before the fires start can pay important dividends in the heat of battle. These cooperative exchanges have the following objectives:

41 June 7-8, To share information on training, qualification, and certification systems and requirements; - To share knowledge about operational procedures and systems including ground and air operations; - To observe each others prescribed fire programs; - To study how each country uses fire fighting equipment and apparatus; - To observe how each country uses relationships at national, state and local levels to coordinate and access firefighting resources such as volunteers; - To share information on fire prevention and education. The Future Through mobilizations of firefighters and numerous exchange activities, these arrangements have repeatedly proven the value of having effective, flexible, cooperative and formal relationships. These Arrangements are not static but must be periodically reviewed, adjusted, and re-approved by the signatories. The U.S. will continue to work with its partners in Australia and New Zealand to improve and expand on these valuable relationships in order to cooperatively address the common global challenges of wildland fire management. 3. Progress in developing guidelines, protocols and standards for increasing efficiency and effectiveness of transnational cooperation In addition to bi- and multilateral agreements the international community has developed during the last years a number of proposals, templates and models for improving governance, efficiency and effectiveness of international cooperation in wildfire disaster risk reduction, management and response. The tools include an international wildland fire terminology, methods for wildland fire risk identification at national, regional, and global levels, non-binding guidelines for fire management and smoke management, including dedicated eco-zonal fire management guidelines. The use of a standardized, commonly accepted wildland fire incident management system for international cooperation in a disaster situation has been proposed. The global wildland fire network has also developed a template for international wildland fire management cooperation agreements for the use by countries interested in entering into formal relationships and agreements on reciprocal assistance with other countries facing similar issues. Training in fire disaster management through development of internationally compatible standards

42 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 and competency, as well as certification of international fire responders, are important elements of improving international cooperation in wildland fire management. In the following some key activities are described. 3.1 International Wildland Fire Terminology The very basic prerequisite for international cooperation in fire management is a commonly agreed terminology, a language that is understood by all partners intending to develop cooperation in fire management. In a number of countries very useful terminologies have been developed. This includes English-speaking countries in which fire terminologies are becoming increasingly compatible at international level. However, terminologies show some differences in the use and meaning of terms. In some countries specific terms have been developed that are not in use in other countries. As the English language is becoming the major language used for international cooperation in fire management it has been proven useful to develop a basic English glossary with English explanations of the terms, which would then be useful for translation. The Global Wildland Fire Management Terminology, first published by FAO (1986) was updated by the Global Fire Monitoring Center (GFMC) on behalf of FAO in The glossary has not been printed as it is considered a dynamic document, open for any changes considered necessary. The glossary is available as an interactive search engine on the web. 27 In the 1999 version, the only non-english language updated was German. FAO added French and Spanish. 28 In 2010 the GFMC published the Russian and Mongolian version (together with English and German) International Statistical Wildland Fire Data Collection Internationally agreed methodologies and procedures for the establishment of fire databases and formats for national fire reporting are not in place. Both databases and national fire reports (assessments) are important for decision support at national, regional and international levels, as well as for assessment of needs and impacts and for targeted cooperation in fire management. The FAO Global Forest Fire Assessment (a special report of

43 June 7-8, 2011 the Global Forest Resources Assessment 2000 [FRA-2000]) 30 and the reports from 12 Regional Wildland Fire Networks, summarized and evaluated in the Fire Management Global Assessment , revealed the lack of current, consistent and coherent statistical data sets. The concept of the detailed Global Wildland Fire Assessment 2004, initiated by the GFMC, was used for a number of national reports submitted to the Regional Wildland Fire Networks. 32 However, the assessment covered only a marginal fraction of countries. Information flow from national and regional level to an international clearing house for receiving, processing and disseminating fire data as well as other fire information back to countries and other users, connected with a network of national fire management agencies, must be ensured. The earlier recommendation by the UNECE/FAO/ILO in 1996 to establish a Task Force to produce a proposal for a harmonized and coordinated data collection and reporting system that will meet the demands of various user communities is therefore repeated with urgency. 33 The next step forward to overcome uncertainties and inconsistencies of fire inventories is the development of a global satellite-based vegetation fire inventory. The Global Observations of Forest and Land Cover Dynamics (GOFC/GOLD) project, an element of the Global Terrestrial Observing System (GTOS) programme, sponsored by the Integrated Global Observing Strategy (IGOS), provides a forum for international information exchange, observation and data coordination, including calibration and validation of sensors and algorithms, and a framework for establishing the necessary long-term monitoring systems. The GOFC/GOLD Fire Mapping and Monitoring Theme is aiming at refining and articulating the international observation requirements and making the best possible use of fire products from the existing and future satellite observing systems, for fire management, policy decision-making and global change research Template for International Wildland Fire Management Cooperation The International Wildland Fire Summit of 8 October 2003 provided an 30 Global Forest Fire Assessment : 31 Fire Management Global Assessment 2006: 32 Global Wildland Fire Assessment 2004: 33 Initial proposal for a global fire dataset by the ECE/FAO International Conference Forest, Fire, and Global Change : 34 GOFC/GOLD Fire Implementation Team: see also section 6 of this paper

44 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 important forums for discussions of how to manage the future of international wildland fire management and share solutions to global problems. One of the outcomes of the Summit was a paper that offered a template and information on cooperation in wildland fire management to countries interested in entering into formal relationships and agreements with other countries facing similar issues. 35 The paper is intended to enhance current international coordination and cooperation by providing information on the following: - A template outlining areas to consider when developing international cooperative agreements; - Listing of the types of cooperation and assistance that may occur between countries; - The responsibilities of countries sending assistance and of those receiving assistance; - Websites containing information and examples of existing cooperative agreements and arrangements. 4. Internationally Compatible Training, Standards and Competency; Certification of International Fire Responders Capacity building of human resources is a key prerequisite for efficient planning and implementation of sustainable fire management. Many countries that are in need of developing or reviewing fire policies or upgrading existing fire management methods and / or technologies, however, do not have own resources or expertise in capacity building in fire management. International cooperation in fire management is critical to support those countries. Priority for international cooperation should include advanced capacity building of target groups influential or responsible for developing fire policies, fire management planning and implementation. Multi-stakeholder, inter-sectoral and inter-agency approaches will be a key consideration. Non-government organizations and the private sector are important target groups to be included. Capacity building of instructors (training for trainers) is a key prerequisite for the success of building capacities at local to national levels. Several fire management handbooks are available that are tailored for use in countries that need to build fire management capacity by applying advanced knowledge in fire ecology and fire management, including participatory approaches in fire management (Community-Based Fire Management) Published in International Forest Fire News (IFFN) No. 29, p : iffn_29/content29.htm

45 June 7-8, 2011 Advanced international training courses for fire management specialists working at high-level positions in their home country and in the private sector will support the development of a culture of trans-national cooperation. Experience has been gained by several UN interagency training courses conducted by the United Nation University (UNU) and GFMC in Africa. A vision is to work through a decentralized worldwide network of training institutions in which donor organizations would collaborate. The development of training materials for international use is desirable. The recently evolving Partnership for Environment and Disaster Risk Reduction (PEDRR) is currently developing training materials on Ecosystembased Disaster Risk Reduction for Sustainable Development, in which a module Integrated Fire Management has been included. 37 In the case of fire suppression first steps have been taken by developing competency standards that will ensure the smooth cooperation between firefighting units of different nations, i.e. their inter-operability in international missions. The EuroFire project, financed by the EU Leonardo da Vinci programme and implemented jointly by the Global Fire Monitoring Center (GFMC), the International Association of Fire and Rescue Services (CTIF) and Rural Development Initiatives Ltd. Between 2006 and 2008, reviewed competencybased wildfire training systems to identify best practice examples from Europe and around the world. This research was the basis for the production of competency-based basic training materials specifically for use in European countries. The key target end-user groups for the EuroFire project included: firefighters, the rural and land-based sector, sectoral organizations and education and training institutions. 38 Meanwhile EuroFire competency standards and training materials have been translated to Russian and tested in Europe and neighboring countries in East Europe / Caucasus in the frame of the Environment and Security (ENVSEC) programme. 39 In future competency-based standards could serve for certification of firefighters to be deployed in fire response international missions EuroFire project website with competency standards and training materials for download:

46 International Forest Fire Symposium on Commemorating the International Year of Forests Fire Management Guidelines Fire Management Guidelines are needed for the various user levels, ranging from practical guidelines for local fire managers to guidelines for land-use planning and policy development. Guidelines must consider the specific natural (ecological) conditions of vegetation fire, as well as the social, cultural, economic and political environment. Valuable guidelines exist for local to global use. However, in many countries these guidelines are not known or not applied, or need to be adapted for the specific conditions, or need to be translated. Fire management guidelines for international use have been developed by international organizations since the 1990s and are available on the Internet: 40 - International Tropical Timber Organization (ITTO) Guidelines on Fire Management in Tropical Forests (1997) - The WHO/UNEP/WMO Health Guidelines for Vegetation Fire Events (1999) - The FAO Guidelines on Fire Management in Temperate and Boreal Forests (2002) - The Fire Management Voluntary Guidelines (2006) 41, with its implementation mechanisms, the Fire Management Actions Alliance (2007) 42 - FAO Legislative Study Forest fires and the law. A guide for national drafters based on the Fire Management Voluntary Guidelines (Morgera and Cirelli, 2009). While guidelines have been developed primarily to serve countries to develop sound, sustainable fire management capacities, including fire management policies and implementation strategies, they are also providing guidance on standard approaches or standards in fire management that have been proven internationally and which will facilitate international cooperation in fire management. 6. International Systems to be Shared: Global Wildland Fire Monitoring and Early Warning There are a number of fire management support tools that are based on international Earth Observation Systems (EOS). These systems include spaceborne sensors for fire detection and monitoring, and terrestrial networks of 40 Overview / portal:

47 June 7-8, 2011 hydrometeorological services for recording and forecasting of fire weather. The Global Observations of Forest and Land Cover Dynamics (GOFC/ GOLD) project, an element of the Global Terrestrial Observing System (GTOS) programme, sponsored by the Integrated Global Observing Strategy (IGOS), provides a forum for international information exchange, observation and data coordination, including calibration and validation of sensors and algorithms, and a framework for establishing the necessary long-term monitoring systems. The GOFC/GOLD Fire Mapping and Monitoring Theme is aimed at refining and articulating the international observation requirements and making the best possible use of fire products from the existing and future satellite observing systems, for fire management, policy decision-making and global change research. 43 GOFC/GOLD and GFMC are closely interacting with the United Nations Office for Outer Space Affairs (UNOOSA), UNOSAT (Operational Satellite Applications Programme of the United Nations Institute for Training and Research UNITAR), the International Charter Space and Major Disasters, and the Group on Earth Observations (GEO) with its Global Earth Observing System of Systems (GEOSS). In 2005 a global multi-hazard early warning system has been proposed in the Hyogo Framework for Action. Subsequently a concept for the development for a Global Early Warning System for Wildland Fires has been endorsed by the United Nations and presented at the Third International Conference on Early Warning (EWC-III), March The Global Early Warning System for Wildland Fires, which is also a Task of the Group on Earth Observations (GEO) 45 aims at developing: - Early warning of fire danger, on a global basis, that will provide international agencies, governments and local communities with an opportunity to mitigate fire damage by assessing threat likelihood and possibility of extreme behaviour enabling implementation of appropriate fire prevention, detection, preparedness, and fire response plans before wildfire problems begin. - A robust global operational early warning framework with an applied system that will provide the foundation with which to build resource- 43 GOFC/GOLD Fire Implementation Team: 44 Website of the Global Early Warning System for Wildland Fires: EWS.htm 45 GEOS Task DI-09-03B Implementation of a Fire Warning System at Global Level -

48 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 sharing agreements between nations during times of extreme fire danger. - Development of local expertise and capacity building in fire management for system sustainability through technology transfer and training. 7. International Wildfire Incident Management System: A Proposal As a result of severe fires over a number of years, national leaders have demanded a more coordinated approach to the management of wildfires, including receiving from or sending firefighting assistance to other countries. However, the ability to effectively cooperate between countries is still limited by organization and communication barriers. In the USA, State and Federal legislators, concerned at the lack of uniform emergency management protocols, directed federal, state, and local government to develop common incident management systems that provide a framework to enable wildland fire protection agencies to effectively facilitate clear response authority, acquire and mobilize resources, coordinate interagency actions, and provide effective management during incident response. A fundamental element of incident management was the creation of the Incident Command System (ICS), which provides standardization through consistent terminology and established organizational structures to enable effective, efficient incident management. Australia and New Zealand, faced with similar emergency response issues, evaluated incident management systems around the world, elected to adopt the ICS and modify it to meet their specific needs. The complexity of incident management, coupled with the growing need for multi-agency and multi-functional involvement at incidents has increased the need for standard inter-agency incident management systems not only within a country/state but increasing internationally. Many countries have adopted similar or common systems of addressing emergencies. In addition a number have developed firefighting agreements based on a common system enabling interoperability when lending support to other countries. In the past this is usually to support adjoining States or Countries within the same geographical region. Since 2000 we have seen examples of this being broadened by support provision occurring from different hemispheres. In 2000 and 2002, Australia and New Zealand sent critically needed incident managers to the USA. Similarly early in 2003 the USA reciprocated sending fire specialists to Australia. Canada and the USA frequently exchange firefighting forces, especially along their borders. New

49 June 7-8, 2011 Zealand sent firefighting forces to Australia in 2002 and used during the wildland fire emergency in Ethiopia in ICS was also The Incident Command System may need to be adapted to suit a particular country s existing political, administrative or cultural systems, customs and values. Where the primary purpose is to enhance emergency management within a country, such adaptations are not only beneficial, but may be essential to have the ICS system adopted. If the purpose of adopting ICS is to enhance cooperation between countries, through the sharing of resources such as fire management teams, it is highly recommended that the sending country and the receiving country both use the same emergency management system. A strategic paper produced by the International Wildland Fire Summit in Sydney, Australia, 8 October 2003, suggested that such a system should be the ICS. Given that ICS is a proven model in many countries and given that training materials for ICS are freely available, there is considerable benefit to be gained by a country adopting this system. It is proposed to introduce an International Wildfire Incident Management System (IWFMS) based on the incident management components discussed previously, including the principles of the ICS. This system would not necessarily require that specific components, such as ICS, be the incident management system of the country receiving or providing firefighting assistance. However, IWFMS components would need to be previously agreed upon, ideally in a formal Arrangement, and utilized by all countries cooperating in the wildfire emergencies. IWFMS would also be considered as a candidate system to be introduced in the UN-driven process to strengthen the international system of responding to environmental emergencies. The UNEP and OCHA have established the international Advisory Group on Environmental Emergencies (AGEE) as their most important cooperation and support mechanism for the response to environmental disasters. The AGEE is an international forum that brings together environmental experts from around the world to share information, expertise and lessons learned for improved response to environmental emergencies worldwide, and in particular in 46 See: International Arrangements on the Sharing of Wildland Fire Suppression Resources between the United States of America and Australia and New Zealand: USA-Australia-NZ-Int-Arrangements.pdf 47 See: The Ethiopia Fire Emergency between February and April 2000: country/et/et_1.htm

50 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 developing countries. In 2007 AGEE founded the Rosersberg Initiative, which aims at strengthening the global regime that governs environmental emergency response and preparedness. 48 The international firefighting assistance during the wildfire emergencies in Greece (2007, Russian Federation (2010) and Israel (2010) have revealed the need for introducing a unified incident management system, especially for the international deployment of aerial firefighting assets. Following the International Wildland Fire Summit (2003) an interest group was formed at the 4 th International Wildland Fire Conference and recommended internationally concerted action. 49 In a series of International Aerial Firefighting Conferences ( ) this idea consolidated. 50 In 2010 the International Fire Aviation Working Group (IFAWG) was founded and officially launched at the meeting of the UNISDR Global Wildland Fire Network / Wildland Fire Advisory Group at GFMC The terms of reference have been laid down in the IFAWG Charter: The International Fire Aviation Working Group (IFAWG) will work under the framework of the UNISDR Wildland Fire Advisory Group (WFAG) / UNISDR Global Wildland Fire Network (GWFN) as an advisory committee with the following principal objectives: - Sharing of relevant information, especially information that will support the promotion and improvement of safety in the sector; - Providing a conduit or facilitation mechanism for the sharing of resources between jurisdictions; - Identifying of opportunities for harmonisation of operating practices and establishment of consistent standards, where appropriate; and recommend or initiate suitable harmonisation action, including the development of voluntary guidelines; - Providing advice and guidance to nations and the United Nations regarding fire aviation through the UNISDR Wildland Fire Advisory Group / Global Wildland Fire Network Initiative/tabid/2647/language/en-US/Default.aspx See website of the last AFF Conference in Spain (December 2010), which includes the reports of all AFF conferences between 2008 and 2010: IFAWG website:

51 June 7-8, Conclusions The United Nations International Strategy for Disaster Reduction (UN- ISDR) and its Wildland Fire Advisory Group are working to assist and strengthen the efforts of United Nations bodies, other international organizations, and nongovernmental organizations, and a large number of national agencies responsible for fire managements, to reduce the negative impacts of wildland fires and to promote safe, ecologically benign and environmentally safe concept of fire use in ecosystem management. Similarly, the UN-ISDR Global Wildland Fire Network (GWFN), the Global Fire Monitoring Center (GFMC) and the FAO are working systematically to increase the intra- and inter- regional cooperation in wildland fire management in the world. Consultations of the Regional Wildland Fire Networks in , expressed by the outcomes of the International Wildland Fire Summit of 2003 and the 4th International Wildland Fire Conference in 2007, revealed that the majority of countries worldwide is ready to establish and strengthen regional and international dialogues on cooperation and exchange of information, research and wildland fire management, including through agreements. It is hoped that the recommendations of this 5 th International Wildland Fire Conference in South Africa will further contribute to this endeavor

52 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Acknowledgements The author is indebted to Dale K. Dague, U.S. Forest Service, Fire & Aviation Management, Washington, D:C., for a review of this paper. Annex I Figure 1. Overview of international organizations sharing responsibilities and tasks in global wildland fire management Goldammer, J.G., Fire Management. Review of International Cooperation. FAO Fire Management Working Paper FM18, 45 p

53 June 7-8, 2011 Annex II At the 4 th International Wildland Fire Conference (Sevilla, Spain, May 2007) the representatives of the Regional Wildland Fire Networks and the participants of the joint regional sessions agreed to the need to develop synergies through coordinated and collective action to address the most pressing problems related to fire management globally. 54 Further the conference participants recommended that: - The international wildland fire community pursue the development of a global-scale international resource sharing strategy to assist countries with fire management planning activities (including prescribed fire for ecological purposes and fuels management), and active support during periods of wildland fire; - The FAO promote the global adoption of Incident Command System (ICS) including the publishing of an annual list of countries which have implemented ICS; - Regional strategies for fire management be developed and designed to the specific needs of regions; - An international framework for fire management standards be developed and regional wildland fire training be supported, especially to meet the needs for capacity building in developing countries; - Scientific research programmes addressing the consequences of changes of climate, land use and land cover, and socio-economic changes on fire regimes, environment and society must be supported at all levels; - The Strategy to Enhance International Cooperation in Wildland Fire Management and the implementation of the Fire Management Voluntary Guidelines be encouraged and endorsed; - Agencies and groups be encouraged to participate in the Fire Management Actions Alliance in support of their adoption of the Voluntary Guidelines; - The UNISDR Global Wildland Fire Network, the Regional Wildland Fire Networks and the Secretariat of the global network, the Global Fire Monitoring Center (GFMC), be supported by national agencies and international donors aimed at fostering international cooperation in fire management, including collecting and disseminating fire information, arranging and enhancing international policy dialogue, and supporting projects; - A series of Regional Consultations tentatively addressing Global Change and Wildland Fire: Regional Solutions for Fire Management be held 54 The results of the 4th International Wildland Fire Conference are documented at:

54 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 globally, within the next 1-2 years, to progress the global issues that are impacting people, resources and livelihoods; - The 2nd International Wildland Fire Summit tentatively addressing Global Change and Wildland Fire: Fire Management Solutions for Mitigation and Adaptation be held within the next 2 to 4 years under the auspices of the United Nations and partners

55 June 7-8, 2011 References Note: Numerous references have been provided as footnotes or embedded in the text of this paper in order to facilitate online reading. Further search for documents on international cooperation in wildland fire management is facilitated by the search engine on the GFMC homepage ( For searching documents only a relevant term (without adding fire etc.) needs to be entered. Other references cited in the text: Goldammer, J.G., and C. de Ronde (eds.) Wildland Fire Management Handbook for Sub-Sahara Africa. Global Fire Monitoring Center and Oneworldbooks, Freiburg Cape Town, 432 p. (ISBN X). Heikkilä, T.V., R.Grönqvist, and M. Jurvélius Wildland Fire Management: Handbook for Trainers. Ministry for Foreign Affairs of Finland, Development pdf (PDF, 12 MB) Khee-Jin Tan, A The ASEAN Agreement on Transboundary Haze Pollution: Prospects for Compliance and Effectiveness in Post-Suharto Indonesia. New York University Environmental Law Journal 13, Morgera, E., and M. T. Cirelli Forest fires and the law. A guide for national drafters based on the Fire Management Voluntary Guidelines. FAO Legislative Study 99, FAO, Rome, 161 p. Nguitragool, P Environmental cooperation in Southeast Asia: ASEAN s regime for transboundary haze pollution. Routledge, Abingdon and New York, 191 p. USDA Forest Service International Arrangements on the Sharing of Wildland Fire Suppression Resources between the United States of America and Australia and New Zealand. International Forest Fire News No. 29, Arrangements.pdf

56 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Wildland Fires and Human Security: Challenges for Fire Management in the 21 st Century Johann Georg Goldammer* Global Fire Monitoring Center, Secretariat of UNISDR Global Wildland Fire Network / Wildland Fire Advisory Group, Georges-Koehler-Allee 75, D Freiburg, Germany info@gfmc.org Abstract Changes of fire regimes and an increasing vulnerability of humans to direct and secondary consequences of wildland fire are observed in many regions globally. Both increasing and decreasing intensity of land use are associated with set fires and often uncontrollable wildfire episodes. This was revealed by the fire emergencies in Bolivia and Russia in July-August Wildfires burning at the interface between wildlands, the fringes of suburbs, metropolitan agglomerations and rural settlements are increasingly impacting large populations, notably by extreme air pollution affecting human health and mortality. Other critical issues include wildfires burning on terrain contaminated by various types of hazardous chemicals (e.g. by mercury) and radioactivity. Remnants of military activities and armed conflicts, e.g. unexploded ordnance (UXO), depleted uranium ammunition and landmines are posing additional threats. Wildfires occurring in such contaminated terrain are resulting in secondary damages, such as chemical and radioactive air pollution and explosion of UXO and landmines on active or abandoned mined areas. Fires occurring during armed conflicts and causing collateral damages and fire purposely set as means of conflicts have been noted increasingly over the past decade. The Global Fire Monitoring Center (GFMC) has launched an initiative in 2008 to address challenges of fire management on contaminated terrain. This endeavour is supported by the Council of Europe and the Environment and Security Initiative (ENVSEC), an activity of UNDP, NATO, UNEP, OSCE and UNECE. Since 2008 the GFMC is publishing annual Global Wildland Fatalities reports. Keywords: Wildfire, wildland fire, human security, radioactive contamination, chemical contamination, unexploded ordnance, land mines 1. Introduction Observations and projections of zonal fire regimes reveal several major

57 June 7-8, 2011 trends which imply that relationships between fire and the environment are going to change. There are some generally accepted major trends: Continuing excessive application of fire in land use and land-use change in the former perhumid equatorial rainforest belt, associated with extreme droughts that are attributed to regional climate change and seeming to be unprecedented since the beginning of recording of weather data in the late 19 th Century, continue to contribute to extended wildfires, notably in impoverished / degraded vegetation; Wildfires are a major agent synergistic to and exacerbating the impacts of climate change (increase of severity and duration of dry spells, increasing aridity) and impacts of human activities (e.g., non-sustainable use of forest and non-forest lands). This is particularly the case in ecosystems subjected to savannization or steppe formation at the expense of forest / tree cover (throughout the paleo- and neotropical floral kingdoms and the Holarctic); Increasing occurrence of fire in global wetland and peat bog biomes due to cultivation / conversion involving drainage and fire use, and an increasing readiness of desiccated organic layers to burn as a consequence of extending vegetation periods, droughts and fire seasons; In the boreal zone interactions are expected between climate change, change of fire regimes and permafrost regimes, with consequences on species composition, habitability of permafrost lands for humans, and release of currently still trapped radiatively active paleogases. The following section focuses on the cultural landscapes of the temperateboreal Eurasian region where major trends are primarily influenced by human and cultural activities that are rather different from other regions of the world Fire in the Cultural Landscapes of Temperate-Boreal Eurasia In preparation of the first Regional Forum on Cross-boundary Fire Management in the region of the United Nations Economic Commission for Europe (UNECE) the Global Fire Monitoring Center (GFMC) provides rationale and arguments for developing a collective transnational view at recent socioeconomic, political and land-use changes affecting fire regimes in temperateboreal Eurasia Note: An earlier version of this paper has been presented at the 5 th International Wildland Fire Conference, South Africa, 9-13 May In preparation by the UNECE/FAO Teams of Specialists on Forest Fire and supported by the Council of Europe as a contribution to the UN International Year of Forests in Draft agenda: fire.uni-freiburg.de/intro/unece-fire-forum-2011-draft-agenda.pdf

58 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Within the last decade temperate-boreal Eurasia has experienced a number wildfire episodes that have resulted in severe environmental damages, high economic losses and considerable humanitarian problems. This has prompted the United Nations Economic Commission for Europe (UNECE) to address recent socio-economic, political and land-use changes, as well as climate change, that are affecting fire regimes of temperate-boreal Eurasia. 57 The UNECE region, which is extending from Western Europe to Asia s Far East and spanning over more than 180 degrees longitude, is bearing a large variety of ecosystems and land-use systems. While specific causes and underlying reasons of extreme wildfires may be quite different throughout the region there are many fire-related issues in common. For instance, in the Western Eurasian part of the region many neighboring countries are sharing similar natural, cultural and social conditions that are determining the occurrence and impacts of wildfires. Besides changes that are evidently understood by the public there are hazards and risks that have been perceived only recently due to a better scientific understanding of the multiple impacts of fires, notably in the context of climate change and human health, but also with regards to an increasing vulnerability of society to direct and secondary effects of fire, e.g. in the context of fires burning at the interface of residential and industrial areas, or in terrain contaminated by industrial deposits or armed conflicts. Land-use change and heritages of former land use constitute a major driver of changing fire regimes in cultural landscapes of Eurasia. The UNECE Forum therefore aims at analyzing the following recent developments in the region: Consequences of rural abandonment and urbanization on agriculture, pastoralism, forestry and wildfire hazard; Rural exodus affecting availability of rural work force, including availability of rural firefighters; Re-privatization of formerly nationalized forests resulting in vacuums of forest management in smallholder forest estates; Weakened governance over forestry and decreased fire management capabilities in many Eastern European and Central Asian countries as a consequence of 57 Regional Forum on Cross-boundary Fire Management (United Nations Economic Commission for Europe UNECE region), planned to be held at the United Nations in Geneva in early 2012, coordinated by the Global Fire Monitoring Center (GFMC) and the UNECE/FAO Team of Specialists on Forest Fire, and supported by the Council of Europe

59 June 7-8, 2011 the transition of national economies, resulting in uncontrolled or illegal forest use and increase of related wildfires; Increasing problems of wildfires affecting the perimeters of metropolitan areas, settlements and developments dispersedly located in wildlands; New standards for reducing gaseous and particle emissions from fossil fuel, biofuel and other open burning that are affecting human health; Concerns about the impacts of radiatively active trace gases and particle emissions from vegetation fires affecting the functioning of the atmosphere and of ecosystem, e.g. the consequences of the transport of fire-emitted black carbon to the arctic environment; Vulnerability of society to fire-generated air pollution, new scientific insights in the impact of fire emissions on human health; Resulting new conflicts in fire management, e.g., controversial views on the acceptance of prescribed burning; Secondary problems associated with wildfires, notable on territories contaminated by radioactivity and remnants from armed conflicts (e.g., unexploded ordnance, land mines, uranium-depleted ammunition); Consequences of climate change on cultural fire regimes and ecosystem vulnerability, e.g. transformation of former fire-excluded or -protected ecosystems or land-use systems such as peat bogs and other organic terrain, or high-altitude mountain ecosystems, such as in the European Alps. On the other hand there is an increasing perception by the public and by policy makers to review and adopt innovative integrated fire management solutions and to exploit of the results of successfully completed international fire research projects in Eurasia for developing adequate public policies affecting fire management and operational fire management. Indeed, the recently published White Paper on Use of Prescribed Fire in Land Management, Nature Conservation and Forestry in Temperate-Boreal Eurasia (Goldammer, 2010a) reveals that the use of fire including historic swidden (shifting) agriculture, and other disturbances by land cultivation have contributed to shape landscape patterns of high-value ecological and cultural diversity in large parts of temperate-boreal Eurasia, e.g. heathlands, open grasslands and meadows. In the eastern Euro-Siberian biota, e.g. in the light taiga, natural fires have shaped open and stress-resilient forest ecosystems (Sannikov and Goldammer, 1993). The rapid socio-economic changes in the past four decades and the recently increasing trend of rural exodus all over Eurasia, however, have resulted in

60 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 abandonment of traditional land-use methods (Dimitrakopoulos and Mitsopoulos, 2005). With the elimination of these disturbances by cultivation, including traditional burning practices, large areas of Europe are converting to fallow lands, a process that is associated with ecological succession towards brush cover and forest, and an overall loss of open habitats. Besides the loss of valuable biodiversity the abandoned lands constitute an increase of wildfire hazard a trend that is revealed by a growing number of extremely severe fire disasters. Similarly, the exclusion of fire in natural ecosystems such as northern boreal and sub-boreal coniferous forests in Eurasia has resulted in changing vegetation composition and an increase of wildfire hazard, notably in Central-Eastern Eurasia. Changing paradigms in ecology and nature conservation recently have led to reconsidering fire-exclusion policies in certain sectors of land / landscape management, nature conservation and forestry. However, the use of prescribed fire in ecosystem management in Europe may not exclusively target those vegetation types that have been shaped by fire over historic time scales, but to introduce fire as a tool to substitute abandoned cultivation practices (Goldammer, 2010a). A sound understanding of the pros and cons of prescribed fire application is necessary as well as the consideration of side effects of fire use. Large areas threatened by land abandonment are embedded in industrialized regions in which society is becoming increasingly unreceptive to fire emissions the fire and smoke episode in Western Russia in 2010 is a striking example for this increased vulnerability (Goldammer, 2010b). Legal restrictions for open burning must be understood in the context of clean-air rules and the need for reducing gaseous and particle emissions that are threatening human health (cf. chapter 18 of this volume). The concerns of those considering prescribed fire emissions a contribution to the increase of the anthropogenic greenhouse effect and thus global warming is mixing in this debate. Here is where new approaches and terminologies are developing, e.g. the critical review of necessary and unnecessary burning in the agricultural domain in northern Eurasia. 58 On the other side it is noted that nature conservation agencies, nongovernment actors and the general public meanwhile have a rather sound and better understanding and perception of the nature of fire as compared to the situation two to three decades ago. International (regional) dedicated networks 58 The International Conference on Open Burning and the Arctic (Russia, November 2010), which explored the impacts of emissions from open fires on Arctic climate, particularly black carbon emissions from set fires in Northern Eurasia, provide a critical view of the escalating fire use in the agricultural sector (cf. Clean Air Task Force (CATF) [2009])

61 June 7-8, 2011 and research projects such as the Eurasian Fire in Nature Conservation Network (EFNCN) 59, within which the White Paper on Use of Prescribed Fire was developed, and particularly the European Integrated Project Fire Paradox, significantly contributed to the acceptance of fire use in wildfire hazard reduction and fire suppression (Sande Silva et al., 2010; see also Birot, 2009). 3. Recently Emerging Fire Threats in Eurasia The Eurasian region is experiencing a range of fire problems that are an expression of heritages and recent changes in the cultural landscapes. Some examples are given in the following. Armed Conflicts Fires occurring during or after armed conflicts or during political unrest constitute a major humanitarian and security issue. The history of fire use as a weapon during wars is as long as the history of armed conflicts of humankind (Pyne, 1995). Most recent occurrence of wildfire during armed conflicts, however, cannot clearly be assigned to intentional tactical or strategic purposes. Most of the fires highlighted in the following are rather collateral damages during conflicts, or fires that were otherwise started at times of armed conflict. Conflict parties often used wildfires as opportunity and reason for mutual accusations. The most recent conflict-related wildfires occurred mainly in the South Caucasus, Near East and the Central Asian Hindu Kush regions. Nagorno-Karabakh (2006) During the period of June to September 2006, extended wildfires affected territories situated close to the Line of Contact (LoC) in and around the Nagorno- Karabakh region. Countries involved in the unresolved conflict around Nagorno- Karabakh accused each other to have started the fires intentionally. The fires affected large areas of abandoned lands around the LoC and adjoining agricultural and forest lands. Impediments to control the fires included the threat of landmines and unexploded ordnance in the area affected by fires, as well as the general tensions between armed forces along the LoC. Concerns over the fires in the affected territories resulted in the UN General Assembly Resolution A/RES/60/

62 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 The Situation in the Occupied Territories of Azerbaijan (15 September 2006). 60 The Organization for Security and Cooperation in Europe (OSCE), with the technical assistance of the United Nations Environment Programme (UNEP) and the Global Fire Monitoring Center (GFMC), assessed the short-term and longterm impacts of the fires on the environment. In his report to the UNGA the OSCE Chairman-in-Office recommended a number of short- to long-term measures aimed at improving fire management capability in the countries concerned and to contribute to peace building in the region (UN General Assembly, 2007). 61 Between 2007 and 2011 the recommendations were implemented by the GFMC using funding of the Environment and Security (ENVSEC) programme and the Euro-Mediterranean Major Hazards Agreement (EUR-OPA) of the Council of Europe. Afghanistan / Pakistan (2006) In 2006 the armed conflict in the border region between Afghanistan and Pakistan escalated. Afghanistan-based NATO forces entered Pakistan's airspace from the neighboring Nooristan province. The air raids on two border villages Daroshot and Azo (Arandu) involved dropping of bombs, which ignited wildfires in the surrounding forests of the area. 62 Israel-Lebanon (2006) During the armed conflict shelling, air raids and rocket attacks started numerous fires on the territories of Israel and Lebanon at a time of drought and extreme wildfire risk (Achiron-Frumkin and Frumkin, 2006). In Israel 800 forest fires were induced by rockets (400 of these wildfires required response) affecting 1,200 ha of forests (mainly coniferous). In addition about 6,600 ha of nature reserves and national parks, and of open landscapes proposed for conservation as nature reserves, as well as ca. 7,000 ha pasture lands were burned. In Lebanon the total area of burned forests was ha and that of burned productive trees was ha UNGA Resolution A/RES/60/285 The Situation in the Occupied Territories of Azerbaijan : st UN General Assembly Session, Agenda item 17 The situation in the occupied territories of Azerbaijan, Letter dated 20 December 2006 from the Permanent Representative of Belgium to the United Nations addressed to the Secretary-General: OSCE-UNGA-ENG.pdf Report of Association for Forests, Development and Conservation (AFDC) made available to and archived at GFMC

63 June 7-8, 2011 Georgia (2008) During the armed conflict in Georgia in August 2008 a number of forest fires occurred as a consequence of military activities in several sites of the country. According to reports by government authorities and non-government organizations the fires burned between 13 August and end of August 2008 (Ateni Gorge and Borjomi Gorge). Several fires affected also two national parks and one nature reserve. Three fires burned within the Borjomi-Kharaguali National Park. A joint mission of OSCE and UNEP to assess the environmental impacts of the conflict in Georgia confirmed the collateral damages caused by fires in Borjomi Gorge and noted that additional areas had burned along the main corridors of combat activities (roads between South Ossetia and Gori Region). 64 Wildfires and Land Mines The countries in the region most affected by land mines are Bosnia and Herzegovina, Croatia, Serbia, Macedonia, Georgia, Ukraine, and Armenia. The origin of the land mines and unexploded ordnance (UXO) in Bosnia and Herzegovina, Croatia and Serbia is from the civil war from the last decade of 20th century. It is estimated that about 300,000 ha are contaminated by land mines and UXO (mostly along to the line of conflict during the civil war). According to a report from Bosnia and Herzegovina about 127,000 ha of forests or ca. 10% of the total forest cover are contaminated by UXO and land mines (Pešković, 2008). Croatia reports 95,000 ha mined forests and other lands with total of ca. 100,000 land mines left (Jungwirth, 2009). This is a significant problem and challenge for forest fire management since wildfires burning on mined lands cannot be fought on the ground using conventional equipment. Wildfires triggering explosions of land mines have caused casualties in several cases and resulted also in reluctance to attack wildfires, or in orders fire services to stay out of the red zones. Similar reports from the Line of Contact (LoC) between India-Pakistan in Jammu and Kashmir reveal that landmines are the main hindrance in controlling forest fires, e.g. the incidences in July and December 2009 when a number of landmines laid along the LoC exploded during the wildfire. 65 In Israel wildfires in 64 Report of the Joint OSCE / UNEP Environmental Assessment mission to Georgia (29 September to 3 October 2008): Assessment-Georgia-Oct-2008-OSCE-34577_en.pdf 65 Landmines explode, hamper efforts to put out spreading LoC forest fires, published on 9 July 2009 by and Bushfire sets off 6 landmines in Poonch, published on 13 December 2009 by Rising Kashmir: risingkashmir.com/index.php?option=com_content&task=view&id=19095&itemid=1-43 -

64 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 minefields on the Golan Heights threatening a UN peacekeeping battalion in August Unexploded Ordnance (UXO) Unexploded Ordnance (UXO) is found on several hundred thousand hectares of forests and other lands throughout Western, Eastern and Southeastern Europe. Remnants of World War I battles along the frontlines of 1917 in Southern Macedonia have repeatedly created problems, e.g. during the fire season of 2007 when more than 70 incidents of explosions of ammunition triggered by forest fires were noted (Goldammer and Nikolov, 2007). In Germany, the battlegrounds of the final phase of World War II in Brandenburg State around Berlin are still highly contaminated by unexploded artillery grenades and bombs. In addition, former and active military exercise areas and shooting ranges, with some of them dating back to the early 1900s, some established after World War II, are posing high risk to civilian populations and especially firefighters (Goldammer, 2010; Goldammer et al., 2009a). A recent estimate reveals that ca. 250,000 ha of former and active military training and shooting ranges in Germany are contaminated by UXO. 67 Besides the above-mentioned land mine contamination in the Balkans and the South Caucasus the combat grounds in and around the Nagorno-Karabakh territory represent one of the major UXO-polluted terrains worldwide. 68 In the Near East the aftermath of the Israel-Lebanon War of 2006 brought problems with unexploded cluster bombs. 69 Threats arising from Wildfires Affecting other Military Assets Wildfires affecting other military assets include ammunition depots, nuclear and conventional research facilities and storage facilities, and active military shooting and exercise ranges. Forest fires entering ammunition storage facilities in the territories of the former Soviet Union in recent years. In 2008 artillery shells and other ammunition at a storage facility in Ukraine, exploded when a forest fire swept into the depot. Details of the causes of some of other reported incidents remain unclear, e.g. a fire and explosions at a munitions depot in southern Ukraine in 2004 in which five people were killed, and a fire at a Soviet-era military base in 66 Feueralarm: Grasbrände in der Zone, published on 3 August 2009 by the Austrian Army 67 This assessment of January 2011 has been extracted from the databank Nature Conservation and Military, David Foundation, Germany (unpublished data on file at GFMC). 68 c.f. footnote no c.f. footnote no

65 June 7-8, 2011 Kagan, Uzbekistan, which spread to an ammunitions depot in July 2008, igniting a series of explosions that killed three people and injured 21 others; or the fire burning an arsenal near Ulyanovsk in November During the fire and smoke pollution episode in Western Russia in July/ August 2010 several military depots and nuclear facilities were threatened by fire. In the first week of August wildfires overrun a weapons storage facility near Moscow (the Central Air and Technical Naval Base 2512), with an estimated loss of 200 airplanes and half of the buildings of the base destroyed. At the same time the Russian military garrison Naro-Fominsk near Moscow moved all its artillery ammunition and rockets to a safer location as wildfires advanced in the region. In the same week wildfires threatened a factory in Kolomna that produced guided missiles, the Novovoronez nuclear power station near Voronez and the Tryokhgorny nuclear closed city in the Urals (Soviet code name: Zlatoust-36) where nuclear warheads are assembled and dismantled. A critical situation developed in the closed nuclear city of Sarov (Arzamas-16) where wildfires advanced towards nuclear arms producing facilities Emissions from Fires Burning on Contaminated Terrain In some countries forests and other lands are contaminated by various types of hazardous chemical and radioactive pollution. Wildfires occurring in such contaminated terrain may result in secondary air pollution. The territories most affected by radioactive pollution have been contaminated by the release of radionuclides during the failure of the Reactor Number Four of the Chernobyl Nuclear Power Plant in Among the total are of 6 million ha of radioactively contaminated terrain in Ukraine, Belarus and Russia the most polluted forest area covers over 2 million ha in the Gomel and Mogilev regions of Belarus, the Kiev region of Ukraine, and the Bryansk region of the Russian Federation. The main contaminator was found to be caesium-137 ( 137 Cs); in the core zones of contamination, strontium-90 ( 90 Sr) and plutonium-239 ( 239 Pu) were found in high concentrations. Generally, under average dry conditions, the surface fuels contaminated by radionuclides the grass layer and the surface layer of peatlands are consumed by fire. Most critical is the situation in peat layers, where the radionuclides are deposited. The long-range transport of radionuclides lifted in the smoke plumes of wildfires and their fallout on large areas were investigated in 70 Data of this section on file at GFMC: For a detailed situation description and a map of Sarov and surroundings, with active fires, see:

66 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 detail in 1992 (c.f. review by Goldammer et al., 2009b). A recent study presented at the conference Twenty-Five Years after Chernobyl Accident: Safety for the Future (Kiev, Ukraine, April 2011) concluded that radioactive fallout from a large forest fire occurring in the Chernobyl Exclusion Zone could affect the food chain and thus be considered threat to human health and security (Oliver et al., 2011; Zibtsev et al. 2011). Recent research reveals that, as a consequence of climate change, mercury deposits once protected in cold northern forests and wetlands will increasingly become exposed to burning. Mercury is released to the atmosphere with fire smoke. Turetsky et al. (2006) quantified organic soil mercury stocks and burned areas across western boreal Canada; it was assumed that, based on ongoing and projected increases in boreal wildfire activity due to climate change, atmospheric mercury emissions will increase and contribute to the anthropogenic alteration of the global mercury cycle and to the exacerbating mercury toxicities for northern food chains. According to a study by Genualdi (2008) vegetation fires have the potential to release toxic industrial and agricultural pollutants previously deposited in ecosystems over the decades, lifted by fire and transported with smoke particles. 71 In the case of pesticides and polychlorinated biphenyls, or PCBs, these persistent organic pollutants can land in regions where the compounds are now banned or even in the Arctic, where they were never used. One of the studies looked at the long-range transport from Siberia by using satellite imaging of smoke plumes and modeling of air mass trajectories which allowed to track the source of pollutants emitted by Siberian wildland fires in 2003 and transported to the Pacific Northwest of the U.S., e.g. dieldrin and alpha-hexachlorocyclohexane (alpha-hch). 5. Conclusions and Outlook This brief overview paper has highlighted selected wildfire threats in the Eurasian space that are affecting human security. The aspects of vegetation fire smoke impacts on human health been covered elsewhere in detail (e.g., Goldammer et al., 2009b). Besides extended fire smoke episodes in South East Asia and South America, especially in Bolivia and Brazil in 2010 and previous 71 See also summary report Forest fires could spread pollutants, released by on 3 December 2009, available at GFMC repository:

67 June 7-8, 2011 years, it should be noted that the heat wave and wildfire smoke episode in Western Russia in July and August 2010 resulted in about 55,800 additional (above long-years average) deaths. These fatalities are attributed to premature deaths as a consequence of both the extreme heat and extended fire smoke pollution. 72 Furthermore the annual global wildland fire fatalities reports should be noted that are published by GFMC since The reports reveal that 345, 374 and about 386 people were killed directly by wildfires in 2008, 2009 and 2010 respectively (2010 number without the above-mentioned 55,800 premature deaths in Russia attributed to heat wave and fire smoke pollution). In conclusion it is stated that wildfires in the cultural and industrial landscapes of Eurasia are a significant threat to human security. Society in this region of the world is becoming increasingly vulnerable to wildfire a trend that needs to be taken into account when addressing the underlying causes for an increased fire hazard and fire risk in this part of the world. References Achiron-Frumkin, T, and R. Frumkin (2006) Preliminary assessment of environmental damages induced by the fighting in northern Israel-Lebanon, summer Based on preliminary reports issued by the Israel Nature and Parks Authority (INPA), the Jewish National Fund (JNF) - Forest Department, the Ministry of Environmental Defence (MOE) and the Ministry of Agriculture (MOA) and prepared for Israel s National Committee of the World Conservation Union (IUCN). Birot, Y. (ed.) (2009) Living with Wildfires: What science can tell us. A contribution to the science-policy dialogue. European Forestry Institute (EFI) Discussion Paper 15, 82 p. Dimitrakopoulos, A.P., and I.D. Mitsopoulos (2005) Global Forest Resources Assessment 2005 Thematic report on forest fires in the Mediterranean Region. Working Paper FFM/8/E, FAO, Rome, Italy. Susan Genualdi, S. (2008) Semi-volatile organic compounds as molecular 72 Report of the Economic Development Ministry of Russia, published in The Moscow Times, 27 October 2010, on file at the GFMC repository:

68 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 markers for atmospheric and ecosystem transport. Dissertation submitted to Oregon State University in partial fulfillment of the requirements for the degree of Doctor of Philosophy, Presented 18 September oregonstate.edu/xmlui/bitstream/handle/1957/9463/genualdi_final_thesis_p df.pdf?sequence=1 Goldammer, J.G., and N.N. Nikolov (2007) Ecological Damage Assessment of the Wildfires in the Former Yugoslav Republic of Macedonia in Joint Mission by the UNEP-OCHA Joint Environment Unit, UNEP, UNDP and GFMC. Goldammer, J.G., E. Brunn, G. Hoffmann, T. Keienburg, R. Mause, H. Page, J. Prüter, E. Remke and M. Spielmann (2009a) Einsatz des Kontrollierten Feuers in Naturschutz, Landschaftspflege und Forstwirtschaft Erfahrungen und Perspektiven für Deutschland. Naturschutz und Biologische Vielfalt 73, Goldammer, J.G., Statheropoulos, M., Andreae, M.O. (2009b) Impacts of vegetation fire emissions on the environment, human health and security A global perspective. In: Wildland Fires and Air Pollution (A. Bytnerowicz, M. Arbaugh, A. Riebau, and C. Andersen, eds.), Elsevier B.V., Developments in Environmental Science, Vol. 8. DOI: /S (08) Goldammer, J.G. (ed.) (2010a) White Paper on Use of Prescribed Fire in Land Management, Nature Conservation and Forestry in Temperate-Boreal Eurasia. Results and recommendations of the Symposium on Fire Management in Cultural and Natural Landscapes, Nature Conservation and Forestry in Temperate-Boreal Eurasia and members of the Eurasian Fire in Nature Conservation Network (EFNCN), Freiburg, Germany, January Fire Ecology Research Group/ Global Fire Monitoring Center, 28 p. fire.uni-freiburg.de/programmes/natcon/efncn- White-Paper-2010.pdf Goldammer, J.G. (2010b) Preliminary Assessment of the Fire Situation in Western Russia. Analysis of 15 August 2010, presented at the State Duma, Moscow, 23 September International Forest Fire News No. 39 (in prep.). Online publication: Dateien/ GFMC-RUS-State-DUMA-18-September-2010-Fire-Report.pdf Jungwirth, O. (2009) Mine Action Organization The Croatian Experience. Paper and data presented in: Advanced Seminar Wildfires and Human Security: Fire Management on Terrain Contaminated by Radioactivity, Unexploded Ordnance (UXO) and Land Mines, Kyiv / Chornobyl, Ukraine,

69 June 7-8, 2011 October Abstract Volume (J.G. Goldammer and S. Zibtsev, eds.). OSCE-Seminar-Ukraine-Brochure-Final-06-Oct-2009.pdf Oliver, C.D., Zibtsev S.V., Hohl A.M., Goldammer J.G., McCarter J., Petrenko M.M., and Borsuk O.A. (2011) Needs For Development Of Wildfire Management In The Chornobyl Exclusion Zone. Paper presented at the conference Twenty-Five Years after Chernobyl Accident: Safety for the Future (Kiev, Ukraine, April 2011). Pešković, B. (2008) Country Report Bosnia and Herzegovina International Forest Fire News No. 37 (in prep.) Sande Silva, J., F. Rego, P. Fernandes, and E. Rigolot (eds.) (2010) Towards Integrated Fire Management Outcomes of the European project Fire Paradox. European Forest Institute Research Report 23. Sannikov, S.N., and J.G. Goldammer (1996) Fire ecology of pine forests of Northern Eurasia. In: Fire in ecosystems of boreal Eurasia (J.G. Goldammer and V.V. Furyaev, eds.), Kluwer Academic Publ., Dordrecht, 528 pp. Turetsky, M.R., Harden, J.W., Friedli, H.R., Flannigan, M.D., Payne, N., Crock, J., Radke, L.F. (2006) Wildfires threaten mercury stocks in northern soils. Geophys. Res. Lett. 33, L16403, doi: /2005gl UN General Assembly (2007). Sixty-first session, Agenda item 17 The situation in the occupied territories of Azerbaijan. UNGA Document A/61/696, 20 p. Zibtsev, S., C.D. Oliver, J.G. Goldammer, A. Hohl, J. McCarter, A. Niccolai, M. Petrenko, and O. Borsuk (2011) Wildfires risk reduction from forests contaminated by radionuclides: A case study of the Chernobyl Nuclear Power Plant Exclusion Zone. Paper presented at the 5th International Wildland Fire Conference, Sun City, South Africa, 9-13 May

70

71 ORAL PRESENTATION I

72

73 June 7-8, 2011 Forest Fires and Management in China Xiaorui Tian* State Forestry Administration s Key Open Laboratory of Forest Protection Research Institute of Forest Ecology, Environment and Protection, CAF Beijing ,China. Tianxr@caf.ac.cn Abstract Since 1987, the government attaches great importance to fire management. But there are still many fires occurrence in China every year. For the period , the yearly average fires were 9,493, which caused burned areas 333,796 ha. The yearly cost of fighting fires was 171,220,000 RMB (about 26,341,538 US$). Human activities were mainly attributable to the forest fire, occupying more than 98%. Most forest fires occurred in spring and autumn. Most fires occurred in southern China, accounting for 52% of the total each year, 37% in the southwest, 6% in northwest, and 4% in northeast and inner Mongolia, and 1% in other areas. It is expected the average temperature in northeastern China could increase 2.22 C in scenario A2 and 2.55 C in the 2040s in scenario B2. That may cause the potential burned area will increase by an average 3% and 20% in 2040s under scenario A2 and B2, respectively. The fire season will prolong on average by five days and 18 days in the 2040s under scenario A2 and B2, respectively. Measures were made for reducing the fires, such as improving the organization system, fire equipments, strengthening fire monitoring and developing the fire research, ect. INTRODUCTION Fire plays an important role in forest ecosystem. After the great fires occurred in 1987 in northeastern China, the government pays more attention to the fire management. The fire agencies were founded in local government at all levels. The measures were made on fire prevention and extinguishment. From 1988 till now, statistics show that the average number of forest fire is 7,935, and the affected forest area is 92,000 hectares. Chinese government attaches great importance to forest cultivation and prevention, especially grasps the intensification of prevention of forest fire as an important measure of the construction of the ecological and forestry prevention. CAUSES OF FOREST FIRES Hunan s activities Most fires are due to man's fault while inappropriately using fires in

74 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 production and daily life. According to statistical analysis for , human activities were mainly attributable to the forest fire, occupying more than 98%. Take 2010 for instance, charcoaling, burning paper as sacrificial offerings, smoking in forest areas, or fires caused by children and the mentally handicapped account for 34.1%, 18.1%, 5.9%, 4.3%, and 3.5% respectively. In Daxing anling, northeastern China, the lightning fires accounted for about 30%. In southwestern and southern China, many fire caused from prescribed fires in farmland. In Northeast China, hunter's casual smoking and cooking fires by medical herb pickers or mushroom pickers often cause forest fires. FIRES DISTRIBUTION Fire peak seasons The occurrence of forest fires in China showed two peaks in a year (Fig. 1). In northern China, the fire season includes two periods, which are spring (from March to June) and autumn (from September to November). But in southern China, most fire occurred in dry season (from November to next May). In general, there are two peaks of fire occurrences. For example, in 2005 the most fires occurred in April (4,604 ha, 40%), and the most burned areas in October (128,662 ha, 40%) (Fig 1). Fires Fires Burned area Burned areas (ha) 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 0 Month Fig. 1. Fire occurrences by month in

75 June 7-8, 2011 Regional distribution Most fires occurred in Southern forest, accounting for 52% of the total each year, 37% in the southwest, 6% in northwest, and 4% in northeast and inner Mongolia, and 1% in other areas. For example, in 2005 the Hunan province had most fires (3,204 fires, accounted for 28% of the country) and Heilongjiang had most burned areas (132,248 ha, 46%) and burned forests (24,307 ha, 33%) (Fig. 2). Fires Fires Bruned areas Burned Forest Bruned areas (ha) Beijing Hebei Inner Mongolia Jilin Jiangsu Anhui Jiangxi Henan Hunan Province Guangxi Sichuan Yunnan Chongqing Gansu Ningxia Fig. 2. Fires distribution in provinces in 2005 RECENTLY FIRE SITUATION For the period , the yearly average fires was 9,493, in which fire alarm (< 1ha), average fire (1-100 ha), big fire ( ha) and conflagration (>1000 ha) were 5452 (57%), 4014 (42%), 23 (0.2%) and 4 (0.04%), respectively. The fires caused burned areas 333,796 ha in each year, in which 138,712 ha (42 %) was occupied by forests. In the period, the most burned areas occurred in 2003 (1,123,751 ha) and most fires (14,144) in 2008 (Fig 3). The yearly cost of fighting fires was 171,220,000 RMB (about 26,341,538 US$), and brought the other losses 131,257,500 RMB (20,193,460 US$). Meanwhile, the wildfires caused 69 people died and 68 wounded in each year

76 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Fires 16,000 14,000 12,000 10,000 8,000 6,000 4,000 2,000 Conflagration(>1000 ha) Big fire( ha) Average fire(1-100 ha) Fire alarm(< 1ha) 1,200,000 1,000, , , , , Burned areas (ha) Year Fig. 3. Fires occurences burned areas in China for EXPECTED FIRES IN THE FUTURE UNDER CLIMATE CHANGE Climate change in 2040s ( ) Temperature and precipitation are two important factors affecting climate, which inevitably become the factors of greatest concern in the analysis of future climate changes. In the baseline period ( ) the average annual temperature of the northeastern China was 1.7 C and annual precipitation mm. Precipitation is mainly concentrated in the period from June to September (accounting for 76.3%) and accounted for 14.9 and 4.6% of annual total precipitation in the spring and autumn fire seasons. During the baseline period the observed average temperature in the peak of the spring fire season (from March to May) was 3.2 C (ranging from C) and the average monthly precipitation was 23.7 mm (ranging from mm). The average temperature in northeastern China could increase 2.22 C (ranging from 1.13 to 3.51 C) in scenario A2 and 2.55 C (ranging from 1.35 to 3.39 C) in the 2040s in scenario B2, which are significantly larger than those in the rest of the country (Tian et al., 2011). The annual precipitation of northeastern China could increase 55.9 and 5.0 mm in the 2040s under scenarios A2 and B2 respectively. The mean relative humidity shows a small change and wind speed is expected to decrease slightly. The average relative humidity could increase 0.17% and 0.90% in the 2040s under both scenarios and wind speed could change by 0.04 and 0.09 m s

77 June 7-8, 2011 Changes in potential burned areas During fire seasons the average monthly temperature could significantly increase in the 2040s under scenarios A2 and B2. Under the two scenarios, precipitation in June and October could be significantly reduced by 4.2 mm (5%) and 5.0 mm (6%), suggesting that the drought trend occurring in early summer and autumn would lead to increased fire risk. The FWI system is correlated with the area burned and the ratio of FWI is a better index to reflect the relationship between fire danger and potential burned areas. Approximately a 1 to 1 relationship between FWI and area burned has been suggested, based on statistical data of several decades (Harrington et al., 1983; Wittrock & Wheaton, 1997). For example, if the FWI were to increase by 50% for a scenario simulation we could expect a similar increase in area burned (Flannigan et al., 2001). Under scenario A2, the average FWI ratio is 1.03 in the 2040s, which in March, April, May and October are expected to be 0.88, 1.09, 1.08 and 1.07 respectively. This suggest that potential burned areas are expected a decrease slightly in March and increase in the other months. During the spring fire season the central region of Heilongjiang shows a high FWI ratio, and burned areas with high FWI ratios are expected to expand significantly in October, in Jilin, southern Liaoning and northern Heilongjiang (Tian et al., 2011). Under scenario B2, the average FWI ratio is 1.21, which suggests that the potential burned area could increase by an average 20% in the 2040s. The FWI ratios in March, April, May and October are 1.21, 1.22, 1.05 and 1.36 respectively. October is the month with the most significant change in the FWI ratio. In March and April the FWI ratios are higher in the east and in May in the central regions of Heilongjiang. In October potential burned areas increased are mainly found in southeastern parts, i.e., Heilongjiang, eastern Jilin and southern Liaoning. Changes in length of fire season Compared with the baseline period, under scenarios A2 and B2, the number of days with high, very high and extreme fire danger ratings will increase, while the length of the high fire danger rating season increase is expected to increase significantly under scenario B2. Under scenario A2, the number of days with high, very high and extreme fire danger ratings will increase on average by five days in the 2040s, during which the number of days of high, very high and extreme fire danger rating will increase by 0, 2 and 3 days respectively (Tian et al., 2011). The number of days of high fire danger ratings will increase significantly in the east, while the length of the fire season will decrease in most of the west. Compared with the baseline period ( ), the fire season could be prolonged by

78 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 days in the 2040s under scenario B2, during which the number of days of high, very high and extreme fire danger ratings could increase by 3, 7 and 8 days, respectively. The length of the fire season could increase in most of northeast, but significantly in the center. EFFORTS TO REDUCE THE FIRES Effective measures are adopted to enhance the management of forest fire control, and minimize the number and losses of forest fires. Laws and regulations Since 2004, the Chinese central government has issued a series of documents, schemes, and regulations. Such as Notification From the State Council on Further Implementing the Forest Fire Prevention, State Emergency Preplan on Handling Particularly Serious Forest Fires, and Forest Fire Regulations. These drafts and regulations greatly enhanced the scientificity and standardization of forest fire prevention career in China. Organization system At present, China government sets up the National Forest Fire Prevention office, which is composed of 19 member organizations. State council also holds meeting to deploy forest fire management. Officials of governments at all levels take their own responsibilities. Till 2010, 3326 forest fire command headquarters with 63,000 member staff and 3544 administrative offices with 20,000 member staff have been built in China. There are fire agencies in the government at all levels. Fire sources management and education Fire sources were strictly controlled in key forest regions. Personals entering woodlands need to be checked and notified the fire sources management in the region. Billboards and posters are located in places. Broadcast, TVs, magazines, and networks were fully used to publicize the importance of forest fire prevention in key forest regions. So the public's awareness of the importance of fire prevention has been reinforced in past twenty years. Fire monitoring In China, the satellites and airplanes were used in fire detection except

79 June 7-8, 2011 monitoring on the ground. There are 18 air bases in northeastern and southwestern China used in fire season. Each year more than 70 aircraft are rented, including more than 40 helicopters. Many satellites were used in fire season for fire detection, such as FY-1C, FY-1D, NOAA Series (NOAA-12, NOAA-14, NOAA- 15, NOAA-16, NOAA-17), EOS/MODIS (terra, aqua). Fire equipments There are 16,000 forest fire brigades in the whole country, which include 510,000 fire fighters. There also exists a forest armed police army (more than 10,000 in total). Since 2006, the central government has invested 2.6 billion RMB for fire prevention construction work. Over 200 projects were developed in key forest region to reduce the fire danger by using the comprehensive measures. So the infrastructures and facilities in high-risk environments and key point s areas are greatly improved. Funds on forest fire prevention and fighting are also included in the government budgets. Those were built of 16,000 fire watchtowers, 1,310,000 km highways, 1,190,000 km fire belts or fuel breaks. 63,000 special vehicles and 6700,000 fire machinery are equipped. There are 3 state-owned warehouses and 11,000 regional warehouses. More than 100 air-bombers are used in distinguishing fires in 15 provinces, flying more than 10,000 hours annually. Fire danger rating system A national fire danger rating system are founding and a net of fire weather stations will finish for better monitoring the fire danger. Some research projects are progressing on basic fire sciences and technologies application. CONCLUSION Although the government made great efforts on fire management, the serious fire situation is still a problem faced by the government. Most fires are caused by human activities. It is expected the average temperature will increase in most areas of China, and the potential burned area also increase in the future under the climate change scenarios. Measures should be made on fire prevention, monitoring, and fire fighting

80 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 REFERENCES TIAN XIAO-RUI, SHU LI-FU, WANG MING-YU, ZHAO FENG-JUN Forest fire danger ratings in the 2040s for northeastern China. Forest Studies in China. (in Press) FLANNIGAN M D, WOTTON B M Climate, weather and area burned. In: Johnson E A, Miyanishi K, eds. Forest Fires: Behavior & Ecological Effects. Hardcover: Academic Press, HARRINGTON J B, FLANNIGAN M D, VAN WAGNER C E A study of the relation of components of the Fire Weather Index to monthly provincial area burned by wildfire in Canada Canadian Forestry Service Information Report PI-X-25. Petawawa National Forest Institute, Chalk River, Canada, 65 WITTROCK V, WHEATON E Climate variations, fire characteristics and budget implications: preliminary analysis of their relationships. Saskatchewan Research Council (SRC), Saskatoon, Saskatchewan. SRC Publication No. R E

81 June 7-8, 2011 The Report of Forest Fire Organization, System and Skill in Korea Si-Young Lee 1*, Houng Sek Park 2 1 The Professional Graduate School of Disaster Prevention, Kangwon National University, Joongang-ro1 Samcheok-si, Gangwon-do, , Republic of Korea LSY925@Kangwon.ac.kr 2 Dongkuk University, Pil-Dong 3Ga Chung-ku, Seoul, Republic of Korea Abstract In this study, we studied and analyzed the change of forest fire policies in order for helping to improve a established forest fire policies. We divided into 7 periods for easier analysis. The first period was 'the period of pre-establishing the Korea forest service after independence ( ). And the second was 'the period of pre-basic plan of forest' after establishing the Korea forest service( ). The third was 'the period of the first basic plan of forest'( ). The forth was 'the period of the second basic plan of forest'( ). The fifth was 'the period of the third basic plan of forest'( ). The sixth was 'the period of the forth basic plan of forest'( ). And the seventh was 'the period of the fifth basic plan of forest'(2008-th present). In the result, the fire fighting organization, system and techniques was made rapid progress after Goseong forest fire in Kangwon region in April For future large scale forest fire attacking, the establishment of more systematical regional forest fire action plan and exercise from this would be needed. Introduction This study was surveyed in 2010 when it was the 50 th anniversary of forest fire statistics starting. Therefore, we analyzed forest fire statistics during 50 years and policy for making an alternative proposal and for preparing the new policy under future climate change condition. In this process, we classified the change of organization, system and skill relating forest fire by several periods

82 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Methodology For analyzing Korea forest fire policy from 1945 to 2010 and proposing an alternative plan, we researched the diversity literature on forest fire like 1996 Goseong forest fire white paper, 2000 eastern coastal region forest fire white paper, 2005 Yang Yang forest fire white paper 50 years of Korean forestry policy, The comprehensive countermeasures for forest fire prevention, The forest fire statistics, The footprint of forest fire administration during 20 years and etc. Results and Discussion 1. The period of pre-establishing the Korea forest service after independence ( ) In this period, the service of forest fire prevention and attack was mainly controlled by the fire department, the bureau of public order in the Ministry of home affairs The temporary special law for forest protection was legislated for decrease in forest fire in The skill for forest fire prevention in this period was the organization of the forest fire guard which was composed up officers of local government and members of a forestry association. The guard attacked forest fires using agricultural implements. 2. The period of pre-basic plan of forest after establishing the Korea forest service ( ) The Koreat forest service was established in Jan The main policy in this period was the know-how of forest fire attack written in 1968 and the handbook for forest fire guarding and prevention was proclaimed in And the forest aviation headquarters was inaugurated in this period. This meant that the air attack was started. For improving of attack efficiency, the forest fire attack equipment like a forest fire rake and a forest fire cleaner was distributed. The surveillance and attack was performed according to handbook

83 June 7-8, 2011 <A banner about forest protection in 1960s> <The watchtower for forest fire observation>

84 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 <The forest fire attack by local residents> 3. The period of the first basic plan of forest ( ) The forest fire warning system was reinforced in this period. The measure for forest fire prevention was classified by 3 stages (prevention, attack and restoration). The restrict policy of entering mountain was implemented in 1973 and the system of forest protection liabilities was modified and improved in In this period, the helicopter was introduced for attack, but, the public were mobilized in general forest fire attack. 4. The period of the second basic plan of forest ( ) In this period, the early warning system according to weather condition was introduced. The law for the restricted area designation of a mountain entrance was legislated in The term for forest fire prevention was designated legally. The air attack using helicopter was operated for the first time and the efficiency of attack was improved by changing to mechanical equipment in this period

85 June 7-8, 2011 <The first air attack using helicopter (Hughes 500D, 1981)> 5. The period of the third basic plan of forest ( ) In this period, the largest forest fire which was ignited on Goseong in Kangwon province was occurred in April This fire had many impact on various area related to disaster in Korea. Therefore, the government was newly organized The controller of forest fire in Korea forest service which supervised the task related to forest protection and prevention and attack of forest fire in The department of forest fire prevention training was newly organized in forest human resources department institute. The air unit and ground attack unit were organized and large-sized helicopter like KA-32T was introduced. Various equipments for surveillance and ground attack was introduced and skills were improved. <The opening ceremony for the organization of forest fire ground units on Seoul in 1996>

86 International Forest Fire Symposium on Commemorating the International Year of Forests The period of the forth basic plan of forest ( ) In this period, the largest forest fire to today named eastern coastal forest fire was occurred in 2000 and Yang Yang forest fire, which destroy Temple of Naksan was occurred in For these large damages of forest fire, the forest fire was regulated as national disaster. for surveillance and ground attack was introduced and skills were improved. The national interest like this made Korea forest service to build a forest fire action plan quickly The prescribe for task and role of forest fire institution was legislated as Korea forest service regulation 588. This determined the task and role of institution when forest fire occurred clearly. The combined guideline of field command for forest fire attack and field manual were enacted following large forest fire case. The forest fire management center for eastern coastal region opened in this period. <The eastern coastal forest fire on Samcheok in 2000> <The YangYang Forest fire in 2005>

87 June 7-8, 2011 <The joint training of forest fire ground units in 2004> <The air attack by large sized helicopter(s-64e) in 2002> <The forest fire damage on the temple of Naksan in 2005>

88 International Forest Fire Symposium on Commemorating the International Year of Forests The period of the fifth basic plan of forest (2008-present) In this period, The main theme of plan was scientific forest fire disaster prevention and action as tactical task of forest protection which was an important resource. The detail of this plan was composed three parts. First was the establishment of scientific surveillance system and backfire countermeasure. This meant the development of prior action system. Second was the specialization of forest fire management organization. And third was the development of advanced technique for forest fire prevention and attack action. The last was development of skill for rapid restoration and re-forestation. Especially, the totally government measure was reinforced. The special forest fire measure of great drought in 2009 and the legislation of the law of forest conservancy (2010) which integrated various regulations related forest fire divided various laws, and time-, reason-specific measure was a good example in this period. The improvement of fore fire surveillance technique and attack skill was implemented. The conditions control systems and forest fire mechanized system and equipment using IT technique was a example of advancement in this period. References Goseong County Goseong forest fire white paper. YangYang County Yang Yang forest fire white paper. Korea Forest Service years of Korean forestry policy. Korea Forest Service years of Korean forestry policy. Korea Forest Service eastern coastal region forest fire white paper. Wotton, B.M. and M.D. Flannigan Length of the fire season in a changing climate. Flannigan et al Future Area Burned In CANADA. Wlliam J. Geroot Adapting to Future Wildland Fire Regimes. Kripalani et al Response of the East Asian summer monsoon to doubled atmospheric CO2: Coupled climate model simulations and projections under IPCC AR4. Im et al Regional Climate Model Simulations of the Middle East. Tom swetnam Climate, People and Fire in the Western US: Long-term Perspectives from Tree-Rings. Goldama Impact of Climate Change, Socio-economic and Land-use Changes on Fire Regimes in the Asian Region

89 June 7-8, 2011 WRF-fire Simulation of the 2005 Yangyang, Gangwon Province Forest Fire Sang-Ok Han 1*, Yonghee Lee 1, Chonghoon Jin 1, Eunhee Kim 1, Jongcheol Ha 1, Heechoon Lee 1, Byungdoo Lee 2 Forecast Research Lab./NIMR 1 Korea Forest Research Institute 2 Abstract In dry spring season, mountainous Gangwon province is one of the most vulnerable area to forest fire in Korea. Yangyang Forest Fire in 2005 was a disaster due not only to its huge burnt area but also to the loss of historic temple building, Naksan temple. This study tried to simulate the Yangyang Forest Fire using the WRF-fire modeling system. Even though there are no such high resolution topographic and fuel data available as is needed to simulate the fire adequately, the simulated fireline was fairly similar to the actual fire spread. From the simulation result it was found that the WRF-fire system could be a useful tool to predict fire-lines for extinguishing activities in the near future when such high resolution topographic and fuel data as few meters become available in Korea. Introduction As the fire-spread model became coupled in the WRF meteorological modeling system in recent years, the meteorological research communities such as NIMR/KMA got interested in the fire-spread prediction. The so-called WRF-fire modeling system works in such manners that the atmospheric part provides wind fields with the fire modeling part while the fire modeling feeds the heat and moisture flux back to atmosphere in the lowest model layer. The coupled fire model computes the fuel burning in the subgrids within atmospheric land-surface grids as a function of wind, fuel, and topographic height. The subgrids are in general less than 1/10 of atmospheric grids. Thus, fire modeling works in far finer grids thant those of atmosphere requesting very high resolution fuel and topo data. In general, those of 1/3 arc seconds are required for reasonable simulations of fire-spread from the WRF-fire modeling system

90 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 In dry spring season, mountainous Gangwon province is one of the most vulnerable area to forest fire in Korea. Yangyang Forest Fire in 2005 was a disaster due not only to its huge burnt area but also to the loss of historic temple building, Naksan temple. In order to simulate the Yangyang Forest Fire, the authors made use of the land cover data obtained from Korean Ministry of Environment in geographic information system format in order to generate 4-meter resolution fuel data. Thus, it was made possible to run the WRF-fire with reasonable results. Methodology The WRF-fire modeling system was used to simulate the Yangyang Forest Fire of Gangwon province in The atmospheric initial and boundary conditions for the fire simulation were provided by 5-km resolution short-range modeling system in KMA which is called KLAPS. The grid spacing for the fire model subgrids is 50 meters which is 1/20 of 1- km atmospheric grid. The 4-meter resolution fuel data set of Anderson 13 categories were prepared using land cover data set from Korea Ministry of Environment. Results and Discussion In Figure 1, the example image of simulated fire-burning area of the Yangyang Forest Fire is suggested in 3-dimensional view image. Compared to the actual burning area the simulation seems to be quite useful, because we expect the result will be much better when the fuel data is more realistic in terms of resolution and fuel category. Thus, further efforts to obtain the realistic fuel data are required to have more realistic fire-spread forecasts

91 June 7-8, 2011 Figure 1. The 3-dimensional image of the simulated Yangyang Forest Fire

92 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Soap-based Firefighting Foam for Forest Fire, Current Forest Fire Situation in Japan Kazuya Uezu*, Yuki Ishizaki, Isamu Akiba, Tomonori Kawano, BartDewanker, HidenoriYasui and Toru Matsumoto R&D Center of Fire and Environmental Safety, The University of Kitakyushu, 1-1 Hibikino, Kitakyushu, , Japan. Abstract Class A firefighting foam concentrates, containing surfactants in major proportions, reduce the usage of water for firefighting. Due to the characteristic properties of surfactants, the agents reduce surface tension of water, and create a superior foam blanket when mixed with air. Thus, it allows firefighters to extinguish a fire much faster than water alone. In Japan, however, the firefighters hesitate to use these agents because they show significant toxicity against aquatic organisms. In other countries, some worry about the environmental impact from firefighting activities using the firefighting foam. We have been developing a novel firefighting foam concentrate with significantly lower environmental risk, consisting of soaps, chelating agent, and diluents. Soaps are the key major components in our firefighting agent, and possess very high biodegradability and very low toxicity particularly for aquatic organisms such as a fish. In this presentation, I will talk about the novel firefighting foam for forest fire and current forest fire situation in Japan. INTRODUCTION We have worked on "Environmentally Friendly Soap-based Firefighting Foam" with the Kitakyushu City Fire and Disaster Management Bureau, Shabondama Soap Co.,Ltd, and the firefighting enterprises since 2003(Mizuki et al., 2007; Goto et al., 2007;Mizuki et al., 2010). The novel firefighting foam for structure fire was a commercial reality in This activity is consistent with policy of Kiakyushu city that has a magnificent heritage of techniques and human resources as a "manufacturing city" and aims to become the "World Capital of Sustainable Development". The collaboration among industry, academia, and government is useful and powerful, and the teamwork is will be further advanced through the

93 June 7-8, 2011 mediation of the new research center. The Research and Development center of Fire and Environmental Safety (RDFES) was established in April 2008 as a research institute within the Faculty of Environmental Engineering, the University of Kitakyushu. The RDFES is the first academic institute in Japan to contribute to environmental engineering and firefighting technology for social safety, and focuses on the environmental researches to overcome the worldwide serious firefighting problem, for example huge forest fires, and consequently contributes to create the epoch-making products for the environmental conservation and the safety of citizens. In this project, we have been developing a novel firefighting foam concentrate with significantly lower environmental risk, consisting of soaps, chelating agent, and diluents. Soaps are the key major components in our firefighting agent, and possess very high biodegradability and very low toxicity particularly for aquatic organisms such as a fish. Furthermore, the ecotoxicological evaluation and the environmental risk assessment will be performed. This project has been promoted by the collaboration among Shabondama Soap Co.,Ltd, Morita Holdings Corporation, the University of Kitakyushu, and Kitakyushu City Fire and Disaster Management Bureau under Japan Science and Technology Agency (JST) Project, Adaptable and Seamless Technology Transfer Program through Target-driven R&D ( ). The research in this program is based on fundamental findings from universities, etc., and aimed at developing toward commercialization via regional industry-academia-government collaborative research. EXPERIMENTAL Constituents of Soap-based firefighting Foam Concentrate The long chain fatty acid salts, oleate andlaurate, were themajor components. The highly biodegradable chelating agent,methylglycinediacetic acid (MGDA) saltor N,N-Bis (carboxymethyl) glutamate tetrasodium (GLDA) saltwas added to the soap-based firefighting foam concentarate. Propylene glycol (PG) andhexylene glycol (HG) were also added to maintain fluidity. Foaming Test 4L of fire-fighting agent solution at 293K were sealed into the fire pump (capacity 8 L). Next, nitrogen gas was poured in the fire pump up to 0.85 MPa. The solution was drained off from the firepump to a bubble collection tank (made of stainless steel, 200 mm h 1200 mm). The steady-state foam height was measured as the distance between the bottom of the tank andthe top of the foam, andthe bubble rate was calculated

94 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Pour Point Test This test was based on Japanese Industrial StandardsK 2269 (the pour point test method). Ice, sodium chloride, and dry ice were used to lower the cooling bath temperature. 45 ml ofthe sample (firefighting foam concentrate) was poured up to the test tube s marker height and was sealed. The test tubewas installed in the outer tube and warmed up to 315 K. The sample was retained for 5 s at the preset temperature. The pour point is defined as the temperature at which the sample lost the fluidity. Toxicity Assay For toxicity assays, paramecia in the stationary phase were washed once with the EBIOS medium made up with the ultra-pure water and then washed twice with the EBIOS media made up with different waters to be used. The tests with P. bursaria were carried out on 12-well microplates. Each well on theplates was filled with 0.9 ml of EBIOS media harboring100 paramecium cells plus 0.1 ml of detergent solutions.then the cells were incubated for 12 h at 296 K under continuous dark condition, and the numberof living cells was counted at the end of incubation nunder a stereomicroscope (SMZ645; Nikon, Tokyo). Medaka fish was also used for toxicity assays of soap-based firefighting test according to the previous paper (Lin et al., 2006). Biodegradability Test Activated sludge from a wastewater treatment plant was used as the medium exposed to soap-based firefighting foam concentrate. Batchrespirometric tests are commonly used to identify organic fractions and kinetic parameters in aerobic biological processes (Yasui et al., 2008). Since oxygen consumption isassociated with biological substrate removal from the liquor and bacterial growth/decay in the system, the respirograms allow construction of a reliable modelstructure. Oxygen uptake rate were obtained using a batch respirometer (Challenging SystemsInc., USA (AER-8)). The temperature of the incubation vessel and the sensing device were maintained at 297 K in a temperature-controlled incubator. For collecting oxygen uptake rete, a small scrubber consisting of caustic material was set between the incubation vessel and the sensing device to absorb CO 2 produced by bacterial respiration from the headspace gas. RESULTS AND DISCUSSION When water is discharged to burning materials, most water flows out and contributes little inextinguishing a fire because of its high surface tension

95 June 7-8, 2011 Firefighting foam concentrate containing surfactants reduces the water surface tension and induces foam formation. The water can stay around theburning materials for a long time and fulfill its role as an extinguisher. Therefore, the wettability, foaming property, and foam stability are required for firefighting foam. From the results of preliminary experiments, the foaming property is found to correlate strongly with the firefighting performance. The fluidity of firefighting foam is also important so that it is available even in coldseason. The bubble rate and the pour point are useful indicators for the foaming property and thefluidity. Sodium oleate, potassium laurate, and potassium palmitate were employed as soap components since they are well known to enhance the wettability, the foaming property, and the foam stability of aqueous solution. Soap combines easily with minerals such as calcium and magnesium ions in tap water. It is also able to lose the interfacial activity immediately. Chelating agents should be added in a soap-based firefighting foam to inhibit the soap-mineral complex formation in tap water. MGDA and GLDA possessing an amino acid backbone were selected as a chelating agent because of its better biodegradation behavior. Diluents containing PG and pure water were added to maintain the fluidity of the agent. From the results of the foaming test and the pour point test, the percentage of fatty acid salt, chelating agent, and diluents are determined to be 14.0 wt. %,33.4 wt. %, and 52.6 wt. %, respectively. The acute toxicity of oleate, laurate, and palmitatewas evaluated using green paramecia (Paramecium bursaria) and Paramecium caudatum under various water conditions. In the low mineral culture medium prepared with distilled water, the median lethal concentration (LC 50 ) for each fatty acid ranged from 5.8 to 144 ppm (w/v). The toxic levels of fatty acid salts differed in the following order: laurate oleate, palmitate. The toxic levels of oleate and palmitate salts were ca. 10-fold lower than those of laurate salt. When river water and local tap water were used for culturing instead of ultra-pure water, the toxic levels of all fatty acid salts were drastically lowered compared to the low mineral condition by 30- to 100- fold ( ppm w/v). Similar detoxification effects were observed when Ca or Mg was added to the low mineral culture media, indicating that the toxicity of fatty acid salts can be notably lowered as the mineral content increases. The toxicity of GLDA is equal to the toxic levels of oleate and palmitate salts. Surprisingly, the toxicity was decreased when GLDA and laurate salt coexist. Other constituents have also low toxicity to green paramecia. Three biodegradable fractions in the soap-based firefighting foam were observed from the aerobic respirometrictests. The three fractions are considered to be fatty acid salts, the chelating agent GLDA, and the glycols (PG and HG). All fractions were found to degrade within the first day of incubation. Kinetic modeling of the degradation of the soap-based firefighting foam was studied by

96 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 aerobic respirometric analysis. The model for aerobic degradation of the soapbased firefighting foam was based on ASM1 (Yasui et al., 2008). The three fractions showed different degradation kinetics, and the biological degradation rate is almost comparable as the typical organic matters in sewage. REFERENCES Goto, K., Lin, C., Kadono, T., Hirono, M., Uezu, K. and Kawano, T. (2007) Ecotoxicity of a soap component (sodium oleate) and a synthetic detergent cocktail using green paramecia assayed in natural water samples from East Asia. Journal of Environmental Engineering and Management. 17 (6): Lin, C., Kadono, T., Yoshizuka, K., Uezu, K. and Kawano, T. (2006) Assessing the eco-toxicity of novel soap-based fire-fighting foam using medaka fish (Oryziaslatipes, Red-orange variety) adopted to river and sea water conditions. ITE Letters 7 (5): Mizuki, H., Uezu, K., Kawano, T., Kadono, T., Kobayashi, M., Hatae, S., Oba, Y., Iwamoto, S., Mitsumune, S., Owari, M., Nagatomo, Y., Umeki, H. and Yamaga, K. (2007) Novel environmental friendly soap-based fire-fighting agent. Journal of Environmental Engineering and Management 17 (6): Mizuki, H., Toyomura, M., Uezu, K., Yasui, H., Kawano, T., Akiba, I., Kawahara, T., Hatae, S., Sakamoto, N., Akiyama, M., Mizota, C., Umeki, H. and Yamaga, K. (2010) Microbial Biological Degradation of Soap-Based Fire- Fighting Agent in Aquatic Condition. Journal of Environmental Engineering and Management 20 (2): Yasui, H., Sugimoto, M., Komatsu, K., Goel, R., Li, Y.Y. and Noike, T. (2008) An approach for substratemapping between ASM and ADM1 for sludge digestion.water Scienceand Technolgy54 (4):

97 June 7-8, 2011 Spatial pattern of human caused forest fire in Korea Hanbin Kwak 1, Woo-Kyun Lee 1*, Saborowski Joachim 2, Si-Young Lee 3, Myoung-Soo Won 4, Kyo-Sang Koo 4, Myung-Bo Lee 4 1 Department of Environmental Science and Ecological Engineering, Korea University, Korea *leewk@korea.ac.kr 2 Chair of Ecoinformatics, Biometrics and Forest Growth, University of Göttingen, Germany 3 Department of Disaster Prevention and Safety Engineering, Kangwon National University, Korea 4 Division of Forest Disaster Management, Korea Forest Research Institute Abstract Most forest fires in Korea are concentrated in certain areas spatially and they are also highly related to human activities. This site-specific characteristic of forest fire can be analyzed by spatial analysis, such as generalized linear mixed model (GLMM) with spatial structure which considers spatial autocorrelation. In this article, we used GLMM approach to quantitative effect of topology, human accessibility and forest cover with spatial autocorrelation. Under assumption these factors, which are slope, elevation, aspect, population density, distance from road, forest cover, are related to forest fire occurrence, the explanatory variables of each factors are prepared using GIS based process. The explanatory variables are transformed for linearity. We firstly tried to test the fixed effect of forest fires and to explain which factors influence in occurrences of forest fire by using generalized linear model (GLM) with Poisson distribution in the meanwhile over-dispersion of response data also detected and variogram analysis was performed by using standardized residuals of GLM. Secondly, GLMM was applied to optimize the fixed effect and random effect. Fitted results were validated with correlation test and RMSE. The result of this study shows that the slope, elevation, squared aspect index, population density and distance from road were determined as significant factors which can explain forest fire occurrence. The spatial parameters were estimated to kilometers as range and as nugget. The results of GLMM were more stable without outlier than those of GLM. Finally forest fire risk was mapped. The risk near the region with high population density was relatively high. Elevation shows the suppress effect on fire. Keywords: forest fire, wildfire, mixed model, spatial autocorrelation, variogram

98 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 INTRODUCTION Forest fire is one of the major disasters which damage both forest ecosystem and human society. And it is very strong threat in Korea because about 65 percent of Korean peninsula is covered by forest. If we take a look into the reason why forest fires are occurred, all the forest fires in Korea are caused by human. There is no natural fire by lightening on condition of Korean meteorology. Because most forest fires are closely related tohuman activities, the fires have a tendency to be occurred on spatially concentrated sitesrepeatedly in Korea. The most frequent reason is accidental fire by human. The second is fires by field incineration (Kwak et al. 2009). To prevent forest fire caused by human activities, it is important to know which factors are related and how much the influences is to the forest fire occurrence. Spatial count data arise in many situations in epidemiology, ecology, and agriculture (Zhang 2002). Typical methods of statistics have limits to detecting specific relationship including spatial autocorrelations. So it is needed more additional techniques and handling count data with spatial autocorrelation is being important for ecology and forestry. To analyze spatial autocorrelation, geostatistical methods are well known for point based data and recently hybrid method so called regression-kriging was adopted which is kinds of mixed model with general regression and ordinary kriging (Hengl et al. 2007; Hengl et al. 2004; Oliver 1990; Stein 1999). In statistics, the effort to consider spatial effect in regression model is also proceeding. The researches for stabilization of mixed model, many GLMM models based on different assumption were prepared (Christensen 2002; Diggle 1998; Venables and Ripley 2002; Zhang 2002). These methods are used in variety field of researches not only forest fire but also epidemiology, social geography, remote sensing and so on. Spatial distribution is very important in forest fire, specially a hundred percent of fires are occurred by human, many spatial analyses have been adopted

99 June 7-8, 2011 STUDY SITE The study site is entirety of South Korea (Figure 1). Figure 1. Study area with the points of forest fire occurrence

100 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 MATERIALS Forest Fire Data In Korea, almost 74.2 percent of forest fires are occurred in spring season (Figure 2). Because there are public holiday and traditional celebrate day which are usually done in or near the forest area in this season, people who have activities in the mountain and forest area are dramatically increased. From this we supposed that human accessibility was highly related to forest fires in Korea. Forest fire occurrence history was collected by Korea Forest Service (KFA). This daily fire point contains informationn of occurred time and address. The address of fire occurrence was geo-coded to let data have spatial coordinate by using a land registry map from Korea Cadastral Survey Corporation. Figure 2. Seasonal variation of forest fires count which is summed from 1991 to The point data was converted to density data with 5km quadrats for quantitativeness (Figure 3). The reason we fixed the size of quadrat to 5km is to minimize an error on converting point to quadrat count because the average of polygon area of land registry map which was used for geo-coding was within 25 square kilometer. After that, mean value of all external variables which consist of topographical factors, human accessibility and tree species factors within quadrats also were input. Totally 4474 grids cover whole study site, South Korea. For model estimation, half of grids which are 2237 were used. The others were used for mapping of results and validation

101 June 7-8, 2011 Figure 3. The distribution of forest fire by 5 km quadrat count. Explanatory Variables In this study, we focus on the spatial characteristics of forest fire occurrence, not temporal characteristics. So we supposed that the important factors which influence forest fire occurrence are three major reasons related to spatial distribution. That is topographical factors, tree species and human accessibility which arestatic spatial factors. Spatial Autocorrelation Spatial data have a property of spatial autocorrelation which is hard to be explained by ordinary statistical method because observations are more related to nearby locations than would be expected (Anselin 1988; Legendre 1993). The first law of geography which related to this phenomena is known as follows: Everything is related to everything else, but near things are more related than

102 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 distant things (Tobler 1979). So, when it comes to spatial characteristics, a spatial-considered tool should be employed to analyze the spatial dependency. MODEL PREDICTION Figure 4. Process for predicting forest fire occurrence with GIS based quadrat count data and GLMM in this study. RESULTS AND DISCUSSION Model Estimation Coefficients of GLM and GLMM were estimated with general linear model shown as Table 1. The population density was the best explanatory variable which has lowest standard error and p-value. It shows that population density factor can explain the linear relationship to forest fire. The distance from road was also highly significant. These results show that human accessibility is considerable factors in forest fire. Topographic factors also showed high relationship except the aspect index in GLMM. In GLMMM the significance was decreased on aspect index. Distance from the road and elevation have negative effect on forest fire occurrence. If the values of these factors are increase, people have limit accessibility. Quite large differences of coefficients between GLM and GLMM were existed in intercept and aspect index factors. The ranking of forest cover

103 June 7-8, 2011 class in terms of fire occurrence probability was Etc> Grassland > Needle leaf forest > Mixed forest> Broad leaf forest. Etc class of forest cover means forest which is not used for tree growth and this class is distributedd near city and army community. Grassland includes cultivated land, pasture and fruit garden. In winter, field incineration whichh have done illegally make large number of fires in Korea. Table 1. Estimated Coefficients, standard error and significant level Variables GLM GLMM Coefficient Std. error P-value Coefficient Std. error (Intercept) Population density(kde) exp(distance from the road) Elevation Slope Aspect Index Aspect Index^2 Needle leaf forest (N/A) < 2e < 2e E (N/A) Broad leaf forest Mixed forest Grassland Etc P-value < 2e < 2e-16 < 2e Figure 5. Variogram of standardized Pearson residuals of GLM

104 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Variogram of standardized Pearson residuals are shown as Figure 5. They were fitted with the spherical models which have the coefficient value as Table 2. In variogram analysis the sill is the value at which the variogram became flat. It means the variance of the two separated point of spatial data. The nugget relates to the variance between pairs of points separated by very small distances (Western et al. 1998). Table 2. Estimated coefficients of the random effects (Spherical spatial structure). Primitive estimation by Estimated GLMM spatial structure Model Standardized residuals of GLM Parameter Value Parameter Value Spherical Range Range Sill Sill - Nugget Nugget Probability Maps Prediction map of forest fire occurrence was derived from estimated risk. The spatial characteristics of forest fire occurrence which are expressed in the prediction map show that strong relationship between forest fire and the population factor. The megacities such as Seoul, Daejeon, Daegu and Busan have high forest fire risk. And the forest area also well reflected on analysis. The western part of peninsula which has low forest area shows relatively low fire danger. And the high elevation area through Taebaek Mountains and Sobaek Mountatins which is located on Eastern part has low level of fire risk. It seems to reflect the topographical factors well so it shows the high similarity to the real fire distribution. And in terms of the distribution of fires risk, the map from GLM shows that a high risk region is scattered irregularly (Figure 6). On the other hand, estimated risk by GLMM shows more clustered pattern which is similar with real distribution of forest fire occurrence (Figure 1). The tendency to be concentrated to hot spot was increased on GLMM. It means GLMM reflects the spatial correlation well. In risk prediction, we summed up kriged residuals to estimated risk by fixed part of regression (Kleinschmidt et al. 2001). For this step, kriged standardized residuals were converted to raw residuals for mapping. When risk map was predicted, GLMM can consider ordinary kriging for interpolation and it can consider different spatial structure, such as spherical, Gaussian, exponential. But it is known that extrapolation is more difficult problem in spatial data and temporal data (Dormann 2007). The further objective of fire risk model should be focused on spato-temporal prediction. However, if meteorological factors are

105 June 7-8, 2011 added in mixed model for temporal variation, prediction will be more difficult problem than spatial aspect. (a) Observed fire count (b) GLM (c) GLMM Figure 6. Prediction map of GLM and GLMM CONCLUSION The aim of this study is to predict forest fire occurrence using spatial data. Furthermore to improve the accuracy of the GLMM model was applied for the enough reflection of spatial dependency. In this model geostatistical spatial structure was added to the ordinary regression. Because a quadrat count of forest fire is known that it follows the Poisson distribution, a Poisson regression was employed to estimate the external influences of parameters. But over dispersion was detected on data, it was adjusted by scale factor. In the result of GLM, the population density was highly significant factor on fire occurrence. Elevation and slope were also significant. There was spatial autocorrelation in the standardized Pearson residuals when variogram analysis was performed. Spatial parameters were initial value for spatial structure in GLMM. GLMM showed more stable and clustered result than GLM and the predictions are more related each other spatially. Spatial autocorrelation is important factor for estimation in statistical model. Also, spatial autocorrelation can be useful factor in management of forest fire. Since most fires are caused by human, spatial autocorrelation may be understood as determination factor which is related to distance. For example, decision making of fire guard network can be done. High density network should be in range of around 30kilometers from high risk region according to spatial autocorrelation. Especially, we focused on spatial characteristics so the temporal consideration

106 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 was insufficient such as climate factors. But considering spatial part is cornerstone of spatio-temporal analysis. In the next research, the climate condition which has temporal information should be incorporated to analyze the seasonality of forest fire. REFERENCES Anselin, L Spatial econometrics: methods and models. Springer. Christensen, O.F Bayesian prediction of spatial count data using generalized linear mixed models. Biometrics 58(2): 280. Diggle, P.J Model-based geostatistics. Journal of the Royal Statistical Society. Series A, Statistics in society 47(3): 299. Dormann, C.F Methods to account for spatial autocorrelation in the analysis of species distributional data: a review. Ecography 30(5): 609. Hengl, T., Heuvelink, G., and Stein, A A generic framework for spatial prediction of soil variables based on regression-kriging. Geoderma 120(1-2): Kleinschmidt, I., Sharp, B., Clarke, G., Curtis, B., and Fraser, C Use of generalized linear mixed models in the spatial analysis of small-area malaria incidence rates in Kwazulu Natal, South Africa. American Journal of Epidemiology 153(12): Kwak, H., Lee, W.K., Won, M.S., Koo, K.S., Lee, M.B., and Lee, S.C Spatial and temporal pattern of the human-caused forest fire occurrences in Korea July Proceeding of the ESRI User Conference, San Diego, California, Unpagenated CD-ROM. Legendre, P Spatial autocorrelation: trouble or new paradigm? Ecology 74(6): Oliver, M.A Kriging: a method of interpolation for geographical information systems. International Journal of Geographical Information Science 4(3): 313. Stein, M Interpolation of spatial data: some theory for kriging. Springer Verlag. Tobler, W Smooth pycnophylactic interpolation for geographical regions. Journal of the American Statistical Association: Venables, W., and Ripley, B Modern applied statistics with S. Springer verlag, New York, NY. Western, A., Blöschl, G., and Grayson, R Geostatistical characterisation of soil moisture patterns in the Tarrawarra catchment. Journal of Hydrology 205(1-2): Zhang, H On estimation and prediction for spatial generalized linear mixed models. Biometrics 58(1):

107 June 7-8, 2011 INDONESIAN FOREST FIRES MORATORIUM AND GREENHOUSE GAS EMISSION REDUCTION Abstract Bambang Hero Saharjo* Faculty of Forestry, Bogor Agricultural University, Bogor 16001, West Java, Indonesia. Indonesian goverment promised to reduce GHG emission by 2020 as much as 26 % or 41 % if supported by International cummunity. It could be realized if the main contributor of GHG emission seriously managed and solved. Forest and land fires is one of the main contributor for Indonesian greenhouse gas emission which produce during burning through land preparation using fire done by the local people or commercial activities, which lead to the global climate change impact direct or indirectly. Avoiding the forest and land fire will be one of the best solution. It could be reach not only by fighting the fires but also through goverment regulation and sustainable forest management such as moratorium. Keywords: moratorium, fire, emission, reduction, GHG INTRODUCTION Forest fires occur either because of anthropogenic or natural causes. The majority of fires around the globe are caused by human activity. It has been estimated that annually fires been across up to 500 million ha of woodland, open forests, tropical and sub-tropical savannas, million ha of boreal and temperate forests and million ha of tropical forests (Goldammer, 1998). Fire is one of the oldest tools known to humans. It has been used as a management technique in land clearance for centuries. For the thousands of farmers, ranches and plantation owners on the edge of the agriculture frontier pushing into forest, fire is the obvious mechanism. It is normally the least expensive and most effective way of clearing vegetation and of fertilizing nutrient poor soils (Rowell and Moore, 2000). Human probably had a role in starting forest fires in recent millennia, and may have deliberately burned forest to improve hunting for

108 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 thousands of years (Qadri, 2001). As prehistoric human settlers of the Indonesian archipelago began to switch from hunting and gathering to growing crops, they use fire to clear agricultural plots in the forest. The cycle of forest clearing, cultivation and abandonment is known as Swidden, kaingin or shifting cultivation, an agricultural system adopted throughout most of the region over a period of thousands of years. INDONESIAN FOREST FIRE SITUATION With the stepping up of economic activities in Indonesia s outer islands, forest fires have become common place, occurring every year. During pronounced El Nino years, when conditions are usually dry, fire and smoke problems tend to be much more serious. Serious fire occurred in , 1987, , , and Less serious fires occurred in Again significant fires have been reported in March 2000, 2002, 2004, Based on data made by Department of Forestry (2007), it had been known that until 2006 the increasing of hotspot was very significant as it can be seen that from 8 (eight) fire risk provinces in Indonesia only North Sumatra which produce less hotspot (-6.50%), meanwhile other provinces produce more hotspot that varies from % until 1, %. The total forest and land burned during 2006 predicted about 6 million ha. Number of hotspot during according to the data taken by Department of Forestry (MoE, 2008) shown that hotspot detected decreased significantly compared to hotspot detected in 2007 at average of %. Most of hotspot detected in the period of January-December 2007 totally about 10,280 or about % was in the community (MoE, 2008), followed by hotspot detected in the estate crops was 2,644 (16.48%) in the estate crop, in the forest concession was 1,691 (10.54 %) and in the forest plantation was 1,430 (8.91 %). Hotspot detected in peat land during January-December 2007 totally about 3,127 both in Sumatra (2,036) and Kalimantan (912) or about % compared to the all hotspot detected, 16,045 (MoE, 2008). Hotspot detected in mineral soil totally about 12,828 both in Sumatra (5,866) and Kalimantan (6,962) or about % compared to the all hotspot detected. Based on hotspot data made by Minstry of Forestry, it had been knwon that the total hotspot detected during the year 2008 was 30,704 and in the year 2009 was 37,659 hotspots (until 16 November 2009) that was separated in the forest area about 22.6 % (8,493 hotspot) and non forest area 77.4 % (29,081 hotspot) means hotspot in the year 2009 increased than in the year Hotspot detected in the year 2010 was also significantly decrease compared to the previous year

109 June 7-8, , 2008 and Total Hotspot Year Total Hutan Figure 1. The hotspot detected during (Bappenas, 2009) INDONESIAN GREENHOUSE GAS EMISSION An assessment of Indonesia s peat land GHG emissions from fire, peat oxidation and loss of AGB, completed according to IPCC Tier 2 standards, shows average annual net emissions of 903 Mt CO 2 yr -1 between 2000 and 2006 (Bappenas, 2009). This estimate is based on (a) team estimates of emissions from oxidation of 220 Mt CO 2 /yr using land use and land cover data from and previously published emissions factors, (b) loss of AGB of 210 Mt CO 2 /yr based on past rates of deforestation and carbon stock in peat swamp forests and (c) a fire emissions estimate of 470 Mt CO 2 /yr from van der Werf et al. (2008) that has been disaggregated into controlled and uncontrolled burning. Controlled burning is defined as fires occurring in land converted into peat and agricultural land. This estimate is broadly similar to the National Climate Change Council estimate and the data presented in the Second National Communication (SNC) to the UNFCCC. Following on from the peat CO 2 study, van der Werf et al. (2008) used several approaches to estimate annual average fire emissions from peat and forest fires. Their mean annual estimate from of 470 Mt CO 2 /yr is now widely accepted, and this study has been used for both the Indonesian National Climate Change Council (DNPI) assessment of the national GHG cost abatement curve and the Government of Indonesia s Second National communication

110 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 (SNC) to the UNFCCC. EMISSION REDUCTION The majority of the peat emissions during period were estimated to be a result of uncontrolled burning (defined as fires occurring outside of licensed areas and contributing 46% of total emissions), peat oxidation (25%) and biomass removal (24%) with the main source regions being Sumatra (44%) and Kalimantan (40%) (Bappenas, 2009). Emissions show a strong inter-annual variation due to factors that influence dry season rainfall such as El Nino and there has also been a reduction in loss of peat swamp forest in the period Sumatra and Kalimantan dominate the national peat emissions profile with fire-related emissions being greater in Kalimantan than Sumatra, while oxidation emissions are greater in Sumatra than Kalimantan. This pattern probably reflects the fact that development peat land in Sumatra preceded that in Kalimantan. These facts shown that fire emission which estimate around 470 Mt CO 2 /yr is the significant greenhouse gas emission that should be gradually reduced. There are several activities could be done regarding the reducing of greenhouse gas emission in term of forest fire such as forest fire prevention and suppression activities. Protection of forests from fires accroding to Indonesian goverment (Bappenas, 2009) would be taken through the following activities: enhancing coordination among stakeholders and roles of local institution, improving the infrastructure and equipment of fire control, strengthening the fire brigade Manggala Agni, avoiding and extinguishing forest fires in areas that are prone to fires, handling post forest fires and developing early warning system through hotspots monitoring with NOAA and MODIS. In order to make sure every activity well done, several target indicator established such as reduction of hotspots in Kalimantan, Sumatera and Sulawesi -20% per year, area of forest burned reduced 50% compared to the condition of year 2008 and increase the capacity of local government and community in controlling forest fire hazard in 30 operational Areas (Bappenas, 2009). Others activities would be done also to reduce fire emissions in term of fire prevention activities such as law enforcement, land preparation without fire, demonstration plot and incentive to the local people who do not use the fire for land preparation. Reducing deforestation alone has according to some scientist, the potential to conserve 1.5 Gt of carbon per year. It will also preserve species which we may need under elevated CO2 regime. Proper management in forestry and agriculture is a prerequisite for a stable global climate. It is much cheaper in the long run and

111 June 7-8, 2011 makes more sense to manage and conserve present land resources-which includes other value added benefits-than to continue with present unsustainable practices (Sorensen, 1993). Research done in the burnt peat swamp forest in Central Kalimantan, Indonesia, shown that moratorium fire and logging will give better environment to vegetation to grow up which increased above ground carbon stock due to the increasing of above ground biomass. Repeated burned site above ground carbon stock only 6.49 ton/ha, while 15-years undisturbed secondary peat swamp forest burned carbon stock 132 ton/ha. According to the Bappenas (2009) emission reduction target (Table 1), the main activities related to forestry and peatland is avoiding forest fire and combating illegal logging. Table 1. Indonesian Emission Reduction Target (Bappenas, 2009) Emission Reduction Sector Plan (Giga ton CO2e) Action Plan 26 % 41% Forestry and Peatland Waste Agriculture Industry Energy and transportation Controlling forest fire and peat fire, Water resource management, Forest and land rehabilitation, HTI, HR. Reducing Illegal Logging, Avoiding deforestation, community engagement. Building Landfill, wasting management based on 3R and integrated water waste management in urban area Introduction low emission rice, water irrigation efficiency, applying organic fertilizer Energy efficiency, applying renewable energy Applying bio fuel, engine efficiency Agency Ministry of Forestry, Ministry of Environment, Ministry of Public Work, Ministry of Agriculture Ministry of Public Work, Ministry of Environment Ministry of Agriculture, Ministry of Environment Ministry of Industry Ministry of Transportation, Ministry of Energy, Ministry of Public Work

112 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 CONCLUSION Avoiding forest and land fire as the main contributor of GHG emission in Indonesia will be one of the best solution to reach the emission reduction target. The activities done to reach the target actually not only fighting the fire during the fire season but it could be reach also using the legal formal approach such as moratorium that initiated by the government. It will be better if the field implementation realized through sustainable forest management manner such as do not let repeated burning occur and keep burnt forest area un-touch for certain period of time. REFERENCES BAPPENAS Reducing carbon emissions from Indonesia s peatlands. Goldammer, J.G., Early warning systems for the prediction of an appropriate response to wildfires and related environmental hazards. IDNDR early warning conference 1998, Postdan, Berlin-Heidelberg-New York: Springer-Verlag. MoE Status Lingkungan Hidup Indonesia (SLHI). Qadri, S.T Fire, Smoke and Haze. ASEAN-ADB. 246 pp. Sorensen, K.W Indonesian peat swamp forest and their roles as a carbon sink. Chemosphere 27(6):

113 KEYNOTE SPEECH II

114 International Forest Fire Symposium on Commemorating the International Year of Forests

115 June 7-8, 2011 Climate Change Impacts on Forest Fire Characteristics and Restoration Strategy in Korea 기후변화에따른한국의산불전망과피해지복원전략 Myungbo Lee* Korea Forest Research Institute, Seoul, , Korea Abstract Global change resulted from climate, population, and land use change, is altering the fire regimes. The decadal variability of the occurrence of forest fire for the recent two decades from 1991 to 2008 was analyzed. Across the Peninsula, forest fire frequency is negatively correlated with monthly mean temperature, relative humidity, and precipitation amount, while positively correlated with the monthly days without precipitation and the average duration of no precipitation for the total data period. To adapt to global change, Advanced and integrated forest fire strategies such as targeted social program based on accurate fire causes, communities-based fire management, fuel management and international cooperation program were needed. Forest ecosystem suitability and economic benefit of local residents should be considered in restoration of burnt area by forest fire. 요약 기후변화, 인구변화, 토지이용변화등으로초래된지구변화는산불패턴에변화를야기하고있다 년부터 2008 년까지산불발생과기상영향인자를분석한결과, 산불기간은일찍시작되어점점길어지고있으며, 산불빈도는평균온도, 상대습도, 강우와는부의관계가, 강우후경과일수와건조일기간은양의관계가있었다. 기후변화에따른산불관리전략으로는정확한산불원인및면적통계에따른대상별산불정책개발및집행, 지역사회기반산불관리정책개발, 산불근본대책으로서의숲가꾸기, 국제공조등을들수있다. 산불피해지복원은지역주민의경제적수익과더불어산림생태계의안정성을도모하여야한다

116 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 서론 (Introduction) 기후변화, 인구변화, 토지이용변화등으로초래된지구변화는과거산불패턴에변화를야기하고있다. 전지구적으로대규모의산불이더강하고빈번하게발생하여지구촌산림생태계와인류의건강을위협하고있다 년세계산불컨퍼런스에서는지구변화, 지구온난화가현실이고지구적산불건수와강도가증가하며사회에미치는영향이점차커지고있다고하였다. 이에따라, 국가적, 지구적차원의보호 (protection), 적응 (adaption), 완화 (mitigation) 할수있는적절한산불정책과전략을개발하고, 경관수준에서의토지관리에산불을통합해야한다고하였다. 아울러각국은산불관리평가, 법적체계와전략구비, 지속가능한산불관리능력과조직구성, 산불관리계획수립과인적자원개발을지원해야한다고권고하였다. 지구온난화에따른기상이변으로인한산림재해의대형화등에따라산불등자연재해에대한예측기술의개발이필요하고이에대비, 대응할수있는사회적요구에부응하기위한새로운예방체계의개발이필요하다. 기후변화와산불과의연관성에관한최근연구동향은온도상승에따른 CO 2 증가 온실가스의증가로인한지구의기후변화에영향 산불발생과피해규모증가예측 산불로인한지구온난화가능성 자연생태계의영향등의 Feedback 시스템에대한다방면의연구가시도되고있다. 이논문에서는한국에서기후변화가산불형의변화에미치는영향을고찰해보고, 우리나라의앞으로의이에따른산불관리전략을제시하고자한다. 또한산불피해지의복원전략을논의하기위해장기간에걸친산불피해지모니터링결과를제시하고, 이에따른복원전략을제시하고자한다.. 1. 기후변화에따른산불전망 IPCC-WGⅡ 에서는기후변화로인한영향에대해극한날씨의강도와빈도가증가 ( 제 4 차보고서 ) 함을밝혔다. 우리나라또한최근이상고온, 건조등기후변화영향으로건조일수, 산림내지피물등연소물질증가등으로 2000 년동해안 (23,794ha), 2002 년청양ㆍ예산 (3,095ha), 2005 년양양 (1,141ha) 등대형산불이빈발하고있다. 특히겨울철고온건조현상으로인한가뭄현상이나타나고농경활동의조기실시와겨울철야외활동증가등산불발생요인의증가로산불발생위험성은더욱높아질것으로전망됨에따라이에대한대책연구가필요하다. 산불발생에미치는기후조건을파악하기위해, 년동안의산

117 June 7-8, 2011 불발생과기상인자사이의상관관계를분석하였다. 산불발생에미치는기상인자로기후변화를직접적으로나타내는변수인기온, 산불발생을억제하는수분관련변수들인상대습도, 강수량, 무강수일수, 건조일간격평균, 모두 5 개의인자를분석하였다. 그결과, 건조계절산불발생은 1990 년대에는증가하고, 2000 년대에는감소하는것으로나타났고 1990 년대의경우, 기온, 무강수일수, 평균건조일간격이증가하고, 상대습도, 강수량은감소하였다 년의건조계절내산불은 11 월부터증가하다가 2-4 월에급격한증가를보이고 5 월에급감하는형태였다. 산불발생의계절내변동은 4 월까지강수량증가, 무강수일수감소, 그리고건조일간격평균감소가산불발생에작용하지만, 기온증가로인간활동이많아지고, 상대습도가감소가호조건으로작용하여산불발생이급증하였다 년대와 2000 년대계절내변동성을비교해본결과, 2000 년대산불발생이 2 월부터시작하여 3 월부터빈번하게나타나 4 월까지나타났다. 이는이시기봄철건조기가평년보다일찍찾아온점에서비롯되었을가능성이있으며, 이는산불기간이길어짐을의미한다. 서울 경기권역, 강원권역, 충청 호남권역, 영남권역으로구분하여기상인자와의관계를분석한결과, 모든권역에서상대습도와강수량은줄었으며, 기온, 무강수일수와건조일간격평균은권역에따라다소차이가있으나전반적으로늘어났다. 2. 산불발생특성변화에따른산불관리전략 효과적인산불전략을수립하기위해서는무엇보다도정책토대가되는산불통계의정확성이요구된다. 10 년평균연 478 건의산불중사실상원인불명으로분류되는입산자실화로인한산불이 43% 을차지하고있다. 따라서효과적인산불정책을개발하고집행하기위해서는이 43% 에대한원인조사가분명하게이루어져야한다. 입산자실화에대해실제그원인을조사한결과, 약절반가량이영농활동으로인한소각에기인하였다. 이러한측면에서최근산림청산불방지과에산불조사계를설치하고, 산불전문조사반을운영하여발생원인을정확하게구명하고자하는노력은시의적절했다. 그러나, 예방차원에서시군구까지강화활필요성이있다. 피해상황통계와관련되어유엔 FAO 산하의 GOFC-GOLD (Global Observation of Forest and Land Cover Dynamics) 기구내에산불팀이신설되어위성영상을활용하여산불정보를구축하고있다. 향후탄소배출과관련된피해면적산정시각국이제출한통계자료를검증하는단계까지계획하고있어, 국내에서도각지자체의피해면적통계를검증하여대응해야한다

118 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 지역사회의참여없이는성공적인산불관리는어렵다. 정확한원인조사를통해도출된통계를바탕으로교육대상별로목적과방법이분명한산불사회교육프로그램이개발되어야한다. 한편 2011 년남아프리카공화국에서개최된산불컨퍼런스에서는지역사회기반산불관리정책이주요화두였다. 이는즉지역사회의이해와참여를기반으로한산불정책을수립하고, 집행되어야함을의미한다. 컨퍼런스성명서에서도성공적으로산불위험을줄이고, 토지및환경의생산성과안전성을증진시키기위해서참여형접근법 (participatory approaches), 지역사회기반산불관리 (communities based fire management) 을통해산불관리에시민사회를포함시켜야함을포함하고있다. 이는지역주민이산불정책에있어서대상자가아니라, 정책의개발자, 집행자로의변화를의미한다. 이를위해서는사회적공감과사회교육이필요한데, 지역노인회를위한산불관리프로그램을개발하는것은소각행위자가산불관리자가된다는측면에서좋은예이다. 초대형산불로인한산불의강도가강해지고, 산림인접지에대한개발이증가하면서주택과인명피해가점점늘어나는추세이다 년호주검은토요일산불과 2002 년미국콜로라도, 오리건, 뉴멕시코, 애리조나일대에서동시다발로발생한산불은산불에강한숲가꾸기의중요성을부각시켰다. 불의확산속도와강도등에미치는인자는일반적으로지형과연료 ( 산림에위치하고있는나무와풀등유기물질 ) 와마지막으로기상인자를들수있다. 이세가지인자중우리가산불에대비하여준비할수있는인자는연료뿐이다. 즉산불이천천히타도록하기위해사람의힘으로산의급한경사를편평하게할수도없을뿐더러, 바람의세기와방향또한우리의힘으로는바꿀수없다. 따라서우리가할수있는최선의방향은산림을산불로부터피해를줄일수있도록가꾸는것이다. 우리나라는전세계에서부러워하는조림성공국가이다. 이제숲가꾸기사업을통해산림이건강하고, 동시에산불에도강한숲으로재탄생할수있도록노력해야한다. 대규모산불의빈번한발생은산불이더이상한나라만의문제가아니라, 국제적인이슈임을말해준다. 이는자원공유를통한공동진화, 전문화된경험과지식의공유등국제협력의강화가필요하며, 그수단이개발되어야함을의미한다. 이를위해서는산불관리와산불비상상황에대한상호협조와관련된다자간, 양자간합의가이루어져야하며, 산불관리에있어서경계를뛰어넘는협력에대한지구적차원의합의 (Global agreement) 가도출되어야함을의미한다. 한편, 국제공조와진화자원공유를위한표준절차및매뉴얼작성되고있으므로, 적극적으로참여해야할것이다. 결론적으로 2009 년호주산불에서도출된결과와비슷하게, 향후의산불관리전략은진화자원등의하드웨어확충도좋지만, 지역사회기반산불관리

119 June 7-8, 2011 등의사회교육프로그램개발과확충된진화자원을효율적으로이용하는방안, 좀더근원적인해결책인숲가꾸기등소프트웨어적인측면을보강해야할것이다. 3. 산불피해지복원과정모니터링결과 1996 년고성산불 (3,762ha 소실 ) 을계기로산불피해로인한산림생태계영향및회복과정을구명하고친환경적인산불피해지복구관리기법을개발하고자고성, 강릉, 삼척, 울진등지의산불피해지에서생태계변화모니터링을실시하고있다. 산불피해지의생태계변화과정을종합하면다음과같다. 가 ) 산불후대량토사유출은 3 4년간지속되며이를억제하기위하여동해안산불피해지에대해서는산불피해를받아벌채한통나무를산지사면에수평쌓기를실시토록하였다. 그효과는편책공을설치하는것과비슷한토사유출억제효과를나타냈으며통나무를쌓음으로서야생동물의서식처로도이용되었고결국참나무류의맹아가자라지력향상을촉진시키는효과를가질것으로추정된다. 나 ) 산불피해지는전소된곳과지표화피해를받은임분등이섞여모자이크상의모양을나타낸다. 전체적인식생변화는초기화본과식물에의한초지화가이루어지며 3~4 년이경과하면싸리등의관목과참나무류맹아가섞여있는관목림상태로바뀌고 5~6 년이경과하면아교목층과관목층의층분화가이루어진다. 자연복원시킨경우남사면은굴참나무가북사면은신갈나무가우점하는경향을보였다. 다 ) 곤충, 야생동물 ( 포유류, 조류 ) 등의변화는식생의회복정도에따라초지성종에서점차산림성종으로변화하는모습을보였다. 나비의경우산불당해년도에는초지성나비가많아졌다가점차감소하며임연부서식나비는점차증가하였다. 메뚜기군집의경우수목류가증가함에따라메뚜기아목은줄어들고여치아목은증가할것으로예상된다. 조류의경우산불발생후초기에는개활지선호종이대부분이며, 일부생육하고있는교목에서박새류가관찰되었다. 라 ) 이동성이강한어류는산불후 3 년간모니터링한결과산불의영향이없는것으로나타났으나저서성대형무척추동물의경우에는재, 낙엽, 토사퇴적등계류상태에따라다양한변화과정을겪지만연수가경과함에따

120 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 라꾸준히증가하는추세를보였다. 마 ) 수목병원균상을조사한결과, 참나무류에는녹병, 튜바키아점무늬병, 흰가루병등다양한병해가계속발생되는것으로조사되었으며, 병해피해로인해수고나생장이다른지역참나무에비해적은것으로조사되었다. 송이균생존률은 48% 에달하였다. 강릉, 삼척산불피해지에미생물제제를처리한결과, 처리구와대조구를비교했을때대부분의토양특성이처리구가대조구에비해양호하게나타났다. 바 ) 경관생태기법을적용한복원전략수립을위하여피해지복원을위한경관지수 (CBI) 를선정하였으며피해지복원을위해인공위성사진 (RS, NBR) 를기본경계로설정하는방법을연구하였다. 이러한생태계변화과정에대한모니터링결과는산불피해지의식생변화조건에따라어떠한복원과정을적용할것인지를결정하는데중요한인자로서임분이정상적인임분의상태에도달하는단계까지지속적인관찰을실시할예정이다 4. 향후산불피해지복원관리전략 산불피해지복원은산주및지역주민의경제적수익을전제하지않을수없다. 따라서목재이외의단기소득자원의조성에대한고려를해야한다. 반면에산림생태계의안정성을도모하여야한다. 이는국가적차원에서해당피해지역에서유지되어야할생물자원량으로산정하는기준치 ( 현재는없음 ) 에조기도달할수있도록노력하는것이다. 한편대형산불의재발을억제하기위한내화수림대의조성, 내화성임분개량등의대한노력도경주하여야한다 년동해안산불피해지복구시적용한경관조림지는경관적가치창출뿐만아니라산불강도를완화시켜주요시설을보호할수있는수단도된다. 또한인공위성사진 (RS, NBR) 을활용함으로써보다신속하고정확한피해지구획과피해평가시스템구축이필요하다. 향후복원전략을수립할때에는산림의구조를계량적으로평가할수있으며, 산림의생태적기능과밀접한관계를보이는경관생태지수에기반한경관생태기법을적용해야할것이다. 즉산불전의경관구조상태를기준으로, 복원시이와유사한경관구조를갖도록조성또는유도해야할것이다. 훼손된산림생태계는자연회복력에의해현재환경에적합한수준으로적응하며점차천이과정이이루어질것이다. 따라서입지조건이양호한곳에

121 June 7-8, 2011 서는식생과여기에기반을둔생물, 무생물도순차적으로회복되어정상임분이될것이다. 하지만질적인면에서살펴보자면목재생산성의저하와심재부후등의문제가도출된다. 따라서이런지역에서는지위조건에따라목재생산림은우선맹아림유도후수종갱신으로건전임분유도가가능하다. 그러나척약지의경우에는자연력에만의존하기에는시간이오래소요되므로우선적으로식생이정착할수있도록토양안정화작업이우선적으로필요하다. 인용문헌 (References) IPCC Good Practivce Guidance for Land Use, Land-Use Change and Forestry (J. Penman, M. Gytarsky, T. Hiraishi, T. Krug, D., Kruger, R. Pipatti, L. Buendia, K. Miwa, T. Ngara, K. Tanabe, F. Wagner, eds.). Intergovernmental Panel on Climate Change (IPCC), IPCC/IGES, Hayama, Japan. Korea Forest Service Statistical Yearbook. Korea Forest Service

122 International Forest Fire Symposium on Commemorating the International Year of Forests

123 ORAL PRESENTATION II

124 International Forest Fire Symposium on Commemorating the International Year of Forests

125 June 7-8, 2011 Effects of disturbance intensity on arthropod communities in burned pine forests in Korea Tae-Sung Kwon 1*, Young-Kyu Park 2, Joo-Hoon Lim 1, Sae-Han Ryou 3, and Cheol-Min Lee 1* 1 Department of Forest Ecology Research, Korea Forest Research Institute, Seoul, , Republic of Korea. 2 Korea Beneficial Insects Laboratory, Jangam-ri 236, Anseong city, Gyounggi-do, , Republic of Korea. 3 Department of Science Education, Seowon University, 241 Musimseoro, Heungduk-gu, Cheongju, Chungbuk, , Republic of Korea. Abstract Forest fires are one of the most frequent and important causes of forest disturbances, and they are gradually increasing due to climate change. This study aimed to find their impact of fire and human activity on arthropod communities in burned forests. Twelve study sites in three burned areas were selected for this study. Intensities of disturbance in the study sites were characterized as follows: 0 (no fire), 1 (weak fire), 2 (strong fire), and 3 (strong fire followed by human disturbance). Arthropods were collected using pitfall traps. Fourteen arthropod taxa (families, orders or classes) which are relatively homogeneous in feeding habits and abundant, were analyzed. Depth of litter layer was selected as an environmental indicator for disturbance intensity because it decreased linearly as degree of disturbance increased. Change of arthropod abundance in response to disturbance differed among functional guilds. As disturbance intensity increased, abundance of detritivores decreased, but herbivores increased. However, abundance of predators varied between taxa. Formicidae and Araneae abundance increased in disturbed sites, whereas Carabidae and Staphylinidae did not change. Abundance of Thysanura and Diptera was most highly correlated with disturbance intensity and may be used as a bioindicator for forest disturbance. Arthropod communities were more diverse in forests of intermediate disturbance

126 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Introduction Forest damages have increased due to climate change. For example, climate change has been linked to increased wind damage (Peterson 2000; Peltola et al. 2010), landslides (Kim and Chae 2009), tree mortality (Allen et al. 2010), bark beetle outbreak (Walton et al. 2008), and mega forest fires (Beverly and Martell 2005). In Korea, various environmental disturbances, including forest fires, have steadily increased due to climate change (Lim et al. 2006). Due to successful reforestation in South Korea, combustibles, such as leaf litters, dead branches and wood have accumulated in forests, resulting in an increased risk of mega forest fires. In late April 1996, a large fire burned 3,672 ha of forests in Goseong in Gwangwon province (Korea Forest Research Institute 1997). In April 2000, the largest fire recorded in Korea burned 23,794 ha of forests across Goseong, Gangneung, Samcheok in Gwangwon province and Uljin in Gyoungbuk province. It caused serious disturbance of forest ecosystems and economic damage in the local communities (Ro et al. 2000). Various disturbances play important roles in the formation of the structure and in the function of ecosystems. Moderate disturbances promote heterogeneity of habitat and increase biodiversity through migration of organisms (Kwon and Park 2005; Noske et al. 2008). Gaps in forests created by fire or wind produce different microenvironments and vegetation structures from surrounding and support overall high species diversity (Lain et al. 2008). In addition, dead wood created by disturbance provide habitats and food for various organisms (Lachat et al. 2006; Gibb et al. 2006). After the mega fire in 2000, homogeneous landscapes composed of pine forests (Pinus densiflora) were replaced with diverse landscape mosaics (Choung et al. 2004). To increase biodiversity in forests, moderate disturbances, such as prescribed fire, clear cutting, and thinning, were artificially applied (Gondard et al. 2003; Glasgow and Matlack 2007; Kwon et al. 2010a; Maleque et al. 2010). In Korea, reforestation of burned forests has been usually applied for recovery of burned forests. After the mega fire in 2000, however, natural regeneration of vegetation was suggested for recovery of the burned areas (Choung et al. 2004). After debate, both types of regenerations were used for recovery of burned forests. Prior to reforestation, dead trees and branches were logged and small trees and shrubs were removed, leading to increased disturbance to burned forests. However, to our knowledge, no study has been conducted to investigate the impact of fire intensity and reforestation on arthropod communities. Arthropods are important components of forest ecosystems. They perform a variety of ecosystem functions, including nutrient cycling, pollination, litter decomposition, and density control (Kremen et al. 1993; Niemelä et al. 1996;

127 June 7-8, 2011 Triplehorn and Johnson 2005). They respond rapidly to ecosystem changes because of their short generation time, high mobility, and dependence on temperature (Kremen et al. 1993; Samways 1994; Schowalter et al. 2003; Maleque et al. 2009). Hence, arthropods are useful bioindicators to estimate the impact of various disturbances, such as climate change, forest thinning, and insecticides (Schowalter et al. 2003; Yi and Moldenke 2008; Kwon TS 2008; Kwon et al. 2010b). Ground fires (hereafter weak fire) mainly burn understory vegetation, whereas crown fires (hereafter strong fire) burn all vegetation. Thus, the effect of fires on forest ecosystems depends greatly on their intensity (Whelan 1995). As noted above, it is expected that arthropods may be further affected by reforestation of burned forests. This study aims to evaluate the impact of disturbance intensity due to fire and reforestation on arthropod communities in burned forests. Arthropod communities were compared among four disturbance intensities (no fire, weak fire, strong fire, and strong fire followed by reforestation) in three burned areas. Materials and methods Study sites and disturbance degree This study was carried out in three burned areas (Goseong, Gangneung, and Samcheok) in the north-eastern coast region of South Korea (Fig. 1). Forests in the north-eastern coast region are composed mainly of pine forests and have most frequently experienced forest fires in South Korea due to a strong and dry northeasterly wind in early spring. Forests of 3,672 ha in Goseong were burned on 23 to 25 April in 1996 (Korea Forest Research Institute 1997)

128 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Fig. 1. Study sites. Circle: unburned site in pine forest (Disturbance Degree, DD = 0), triangle: site in pine forest burned by weak fire (DD = 1), open rectangle: site burned by strong fire (DD = 2), dark rectangle: site burned by strong fire and reforested afterward (DD = 3). Estimation logic for DD is shown in text. This fire was started accidentally during army training and the economic loss was about $1,800,000. In April 2000, a very large fire burned forests of 23,794 ha in four counties, including Goseong, Gangneung, Samcheok, and Uljin. It was the largest recorded fire in Korea and it burned 0.37% of the total forest in South Korea (Ro et al. 2000). In April 2004, a medium fire burned pine forest of 430 ha in Gangneung. A total of 12 study sites (4 study sites in each of 3 burned areas) were selected for this study (Fig. 1). Disturbance intensity was characterized as one of four disturbance degrees (hereafter DD). DD of unburned pine forest was 0, and DD after a weak fire was 1. DD after a strong fire was 2, and DD after a strong fire followed by reforestation was 3. The average annual temperatures in Goseung, Gangneung, and Samcheok in 2005 were 11.9 C, 12.9 C, and 12.1 C, respectively (Korea Meteorological Administration 2005). Annual rainfall in 2005 was 1,349 mm in Goseong, 1,650 mm in Gangneung, and 1,433 mm in Samcheok. Study sites are located 1 to 12 km from the coast, and at elevations of 20 to 192 m. Site aspects are north (8 sites), south (2 sites: Samcheok, DD 2 and 3), east (Donghae, DD 2), and west (Goseong, DD 1). In Goseong, DD 2 and 3 were located at N38 o 18', E128 o 29', DD 1 at N38 o 19', E128 o 27', and DD 0 at N38 o 20', E128 o 30'. In Gangneung, DD 1 to

129 June 7-8, 2011 were located at N37 o 36', E129 o 01', DD 0 at N37 o 35', E129 o 01'. In Samcheok, DD 1 to 3 were located at N37 o 14', E129 o 16-18', and DD 0 at N37 o 20', E129 o 14'. Site slopes were mostly 3 to 30 o and 2 sites (DD 2 and 3) in Samcheok were steep (ca. 50 o ). Soils at study sites originated from granite, and soil textures were silt loam, loam, or sand loam. Red pines (Pinus densiflora) of about years old were the main trees in DD 0 or 1 in 3 study areas. Shrubs and herbs were growing moderately under pine trees. No or a few pine trees were found in DD 2 or 3. Shrubs and herbs were growing well in understory vegetation there. Vegetation, litter and soil property Vegetation at the study sites was investigated by a botanist (Dr. Ryu) who recorded the plants growing in a plot of 450 m 2 (width 10 m, length 45 m) where pitfall traps were set up. To estimate vegetation structure, the coverage (%) of the tree, sub-tree, shrub, and herb layers were estimated. Depth of the litter layer was measured in 10 replicates per plot (450m 2 ), and the average of the values was used for analysis. Five soil samples (ca. 100 ml) per plot were randomly sampled from the surface to a depth of 5 cm depth using a small shovel after removing the litter layer and samples were pooled for each plot. Soil samples were analyzed in the soil analysis laboratory of the Korea Forest Research Institute for estimation of ph, soil texture (i.e., composition of sand, fine sand, and silt), organic matter, total nitrogen, and cation exchange capacity (cmol c /kg). Arthropods sampling Arthropods were collected in pitfall traps consisting of a plastic cup (depth 6.3 cm, mouth diameter 8 cm, bottom diameter 6 cm). Twenty pitfall traps were buried at each study site for 10 days in late May The survey period is considered best for sampling of arthropods since arthropods are active and abundant in this high temperature and low rainfall season from late May to early June in Korea (Kwon et al. 2005; 2010a). For 10 days, each of 20 traps was placed 5 m apart from adjacent traps along two parallel lines about 10 m apart. Each trap was filled about one-third with ethylene glycol as a preservative. All arthropod specimens were isolated from debris in the laboratory, stored in 80% ethyl alcohol, and identified to order or family level using taxonomic keys (Choi 1996; Triplehorn and Johnson 2005) under a stereomicroscope. All the specimens were deposited in the Insect Specimen Storage Room of the Korea Forest Research Institute

130 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Functional guilds of arthropods Microarthropods, such as springtails and mites, were excluded from analysis because of high variability of abundance (Kwon 2008). Arthropods that have relatively homogenous feeding functions at the coarse taxa (i.e. family, order, or even class) level were used for analysis. Crustacea, Thysanura, and Diplopoda are representative detritivores. Diplopoda were not used in analysis due to low abundance (Table 1). Although Diptera have several feeding functions, they are mainly detritivores. Orthopterans were assigned into two groups: detritivores (Othroptera-D) and Herbivores (Orthoptera-H). Othroptera-D included Rhaphidophoridae, Gryllidae, and Gryllotalpidae, whereas Othroptera-H included Acrididae, Tettigoniidae, and Tetrigidae. Sap-feeding Homoptera were assigned into two groups: Aphididae and other Homoptera. Aphididae are relatively abundant and different in life cycle and habit compared with other Homoptera. Larvae of Lepidoptera were classified as herbivores. Hemiptera were excluded from analysis because they include both predators and herbivores. Four coleopteran families, Carabidae, Staphylinidae, Curculionidae, and Chrysomelidae, were used. Carabidae and Staphylinidae were grouped as predators, while Curculionidae and Chrysomelidae were grouped as herbivores. Of the hymenopterans, only Formicidae (ants) as predators were used. Other Hymenoptera were not used due to their diverse feeding functions (i.e., herbivore, parasites or predators). Araneae (spiders) were used in analysis as predators. Chilopoda are predators but were not used because of low abundance. Opiliones were abundant (Table 1), but were not used because they include both predators and detritivores (Triplehorn and Johnson 2005). Blattaria may not be specific to forests so they were not used for analysis. Statistical analyses The number of arthropod individuals collected per trap was log transformed (ln N+1) to reduce variance. ANOVA was used to find differences in abundance of arthropods or in values of environmental factors according to scales of DD, and Neuwman-Keuls multiple comparison test was used afterwards. As litter depth decreased linearly with increasing scales of DD (see Results). it was used as a representative factor for intensities of disturbance. Regression analysis on each arthropod group was performed using litter depth as the independent variable. Multidimensional scaling (MDS) using Euclidean distance was used for ordination of arthropod communities. MDS ordination was conducted using PC- ORD (ver. 5.17) (McCune and Mefford 1999), and the other analyses were performed using STATISTICA (Statsoft 2004)

131 June 7-8, 2011 Results Abundance (i.e., number of individuals) of the arthropods collected in the three study areas is presented in Table 1. A total of 23,131 arthropods (except springtails and mites) belonging to five Classes were collected. The numbers of arthropods collected were 8,742 in Goseong, 7,503 in Gangneung, and 6,666 in Samcheok. Abundance decreased from north to south (Fig. 1). Table 1. Number of individuals of arthropods collected in pitfall traps Microarthropods such as springtails and mites were not included in table. Class Order Family Study area Proportion Total (Function, et al.) Goseong Gangneung Samcheok (%) Function Crustaceae Detritivore Arachnida Araneae 2,048 1, , Predator Opiliones Insecta Microcoryphia Detritivore Blattaria Isoptera Orthoptera (Herbivore) Herbivore (Detritivore) Detritivore Hemiptera Homoptera Aphididae Herbivore (Other) Herbivore Hymenoptera Formicidae 3,651 1,888 2,990 8, Predator (Other) Lepidoptera (Larvae) Herbivore Coleoptera Carabidae Predator Staphylinidae Predator Insecta Coleoptera Chrysomeridae Herbivore Curculionidae Herbivore (Other) , Diptera 1,670 1,342 1,661 4, Detritivore (Other) Diplopoda Chilopoda Total 8,742 7,503 6,886 23, Formicidae was most abundant, representing 37% of total. Diptera (20%), Araneae (19%), and Coleoptera (11%) were also abundant. Opliones, Blattaria, and Hymenoptera (except Formicidae) were not used in this analysis, but were relatively abundant (1-4%). In Table 1, groups with feeding functions were used

132 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 for analysis and accounted for 87% of total arthropods. When abundance of arthropods analyzed were compared among the four DDs, Thysanura, Orthoptera-H, and Homoptera (other) abundance were significantly different (p < 0.05), whereas Diptera was marginally significantly different (p = 0.054) (Table 2). As DD increased, detritivorous Thysanura and Diptera decreased linearly, whereas the herbivorous Homoptera increased. Abundance of the herbivorous Orthoptera-H was not linearly related with DD, but was higher in high DD (2 and 3) than in low DD (0 and 1). Table 2. Number (ln n+1) of arthropods according to disturbance degree (DD) in the study sites. Details on DD are shown in text and Table 1. The different letters behind SE indicate significant difference (p < 0.05) between groups according to Newman-Keuls multiple comparison test. Disturbance Degree ANOVA-test Function Taxa (df = 3, 8) Mean SE Mean SE Mean SE Mean SE F p DetritivoreCrustacea Thysanura a ab ab b Diptera Orthoptera-D Herbivore Orthoptera-H ac a b bc Homoptera (other) a ab ab b Aphididae Lepidoptera (larvae) Chrysomeridae Curculionidae Predator Araneae Carabidae Staphylinidae Formicidae When vegetation and soil variables were compared among DDs, tree coverage and litter depth were significantly different (p < 0.05), and shrub coverage, organic matter, and CEC were marginally significantly different (p < 0.1). Tree coverage and litter depth decreased with increasing DD. Organic matter in DD 0 and 1 was 3.3 to 3.5, whereas in DD 2 and 3 it was 1.5 to 1.7. CEC was higher in DD 0 and 1 (10 to 11.4) than in DD 2 and 3 (7 to 8). Although soil texture (i.e. composition of soils) was not significantly different among DDs, its change may be related to the loss of fine soils at burned sites. Sand increased with increasing DD, but fine sand and silt decreased. These results show that severe disturbance led to loss of soil, litter layer, organic matter and CEC

133 June 7-8, 2011 Because litter depth decreased linearly along with increasing DD, it may be an environmental indicator to intensities of disturbance (Table 3). The abundance of the four taxa which were significantly different in Table 1 was arranged along litter depth (Fig. 2). Thysanura, Diptera, Lepidoptera (larvae), and Formicidae were significantly correlated with litter depth (p < 0.05). In regression analysis using litter depth as an explaining variable, it explained about 33 to 70% of total variations (R 2 of Thysanura = 0.70, Lepidoptera = 0.60, Diptera = 0.49, and Formicidae = 0.33). However, abundance changed differently among feeding guilds. Abundance of the detritivorous Thysanura and Diptera increased as litter depth increased, whereas the herbivorous Lepidoptera and the predatory Formicidae decreased. Table 3. Variables of vegetations and soils according to disturbance degree in the study sites. Details on degree of disturbance are shown in text and Table 1. The different letters behind SE indicate significant difference (p < 0.05) between groups according to Newman-Keuls multiple comparison test.. Disturbance Factor ANOVA (df = 3, 8) Mean SE Mean SE Mean SE Mean SE F p Vegetation Tree coverage (%) a a b b Sub-tree coverage (%) Shrub coverage (%) Herb coverage (%) No. of species Soil Depth of Litter depth (cm) a ab ab b Sand (%) Fine sand (%) Silt (%) ph Organic Matter (%) Total nitrogen (%) CEC (cmol+/kg)

134 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Fig. 2. Log (ln n+1) transformed abundance (individuals collected at a pitfall trap) of arthropods according to depth of litter. The selected arthropods show significant (p < 0.05) or marginally significant (p < 0.1) differences among scales of disturbance degree (DD). When beta values of all taxa in the regression analysis were compared, general relationships between feeding guilds and litter depth appeared (Fig. 3). The herbivorous taxa had negative beta values ranging from to -0.21, whereas the detritivorous taxa had positive beta values ranging from to The predatory Fomicidae (-0.31) and Araneae (-0.22) had negative beta values, but the predatory Staphylinidae (0.005) and Carabidae (0.02) had beta values near 0. Beta values were significantly different among three functional guilds (Kruskal- Wallis nonparametric ANOVA, H = 6.5, p < 0.05). Such results were also confirmed when abundance of 14 taxa were pooled into three feeding guilds (Fig. 4). Abundance of all three feeding guilds was significantly correlated with litter depth (p < 0.01). The beta value of detritivores was 0.37 with R 2 = 0.54, of herbivores was with R 2 = 0.51, and of predators was with R 2 = Predators might have a negative beta value due to the negative value of abundant Formicidae and Araneae

135 June 7-8, 2011 Fig. 3. Beta value of 14 taxa in regression models using litter depth as independent variable. Significance of regression models (i.e., beta value of regression model is not zero) are represented as follows; * p < 0.01, ** p < 0.05, and *** p < Fig. 4. Log (ln n+1) transformed abundance (individuals collected at a pitfall trap) of three feeding guilds according to depth of litter. Feeding guilds of 14 taxa are shown in Table

136 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 The above results can be comprehensively summarized by MDS ordination (Fig. 5). In MDS ordination, Axis I (55%) and Axis II (40%) explained 95% of total variation. Angle of original ordination was moved slightly for visualization. Communities of arthropods were sequentially arranged according to scale of DD along Axis II. Arthropod communities in the least disturbed sites (DD 0) were closely located in the uppermost part of Axis II, whereas largest disturbed sites (DD 3) were closely located in the lowest. Moderately disturbed sites (DD 1 and 2) were located in the middle part of Axis II. Distances between sites are longer in moderately disturbed sites than in un-disturbed or highly-disturbed sites. This result indicates that moderate environmental disturbance contributes to greater diversity of community structure. Fig. 5. MDS ordination of arthropod communities. Two axes explain 95% of total variation. Angles of two axes were moved slightly for visualization of arthropod communities according to disturbance degree (DD). Circle: unburned site in pine forest (DD = 0), open triangle: site in pine forest burned by weak fire (DD = 1), dark triangle: site burned by strong fire (DD = 2), and reversed dark triangle: site burned by strong fire and reforested afterward (DD = 3)

137 June 7-8, 2011 Directions and scales of vectors of environmental variables almost coincided with the results shown in Table 3. The vectors of tree coverage, litter depth, and organic matter showed the same direction as arthropod communities of DD 0, showing their negative relationships with DD. TN and CEC also displayed the same direction. The vector of sand had the same direction as DD 3, whereas fine sand and silt had the reverse direction. This shows that coarse soils are positively correlated with scales of DD but fine soils are negatively correlated with DD. The arrangement of 14 taxa in Axis II was significantly correlated with the beta values in Fig. 3 (Spearman rank correlation, r = 0.73, p < 0.05). Homoptera, Lepidoptera, Formicidae, and Araneae, which have negative beta values, were located in the lower part of Axis II, whereas Thysanura, Diptera, and Crustacea, which have positive beta values, were located in upper part of Axis II. Discussion The expected patterns, including decrease of litter layer, loss of fine soils and nutrients supporting growth of plants, are recognized by present study. However, the content of nutrients in soils temporally increased in early stage of forest fire. Immediately after forest fire in Goseong in 1996, Ca, K, and Mg concentrations were higher in burned forests, whereas organic matter and nitrogen concentrations were not changed (Mun and Choung 1996). Oh et al. (2001) reported that organic matter and nutrient concentrations were higher in one of two burned sites for 10 months compared to the control site, but was lower after 11 months. However, one burned site had continually higher concentrations. In our study sites, nutrients were more abundant in unburned site than in burned sites at all times. Loss of soil and nutrients in burned forests may differ due to various environmental factors such as vegetation and topography (Choung and Kim 1987; Edmonds et al. 2000; Oh et al. 2001). Large amounts of ash produced by the 2000 mega fire polluted streams and the sea. Accordingly, the outflow of soils and nutrients caused by forest fires may significantly influence the surrounding ecosystems. In forest ecosystems, leaf litters plays key roles in material recycling, control of surface microenvironment, outflow restraint of sediment and soil nutrient, and influencing inhabitants such as microbes and arthropods (Lee et al. 2004). The decline of litter layer was related linearly with intensity of disturbance caused by forest fire and artificial disturbance. Abundance of detritivores linearly decreased as depth of litter layer decreased. It is likely that detritivores feeding on litters can be useful bioindicators for determining forest disturbance. By 10 years after a forest fire in eastern Australia, mouse density increased logarithmically

138 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 with increasing litter (Fox and Matzner 1987). Lee et al. (2006) found that density and weight of rodents were higher in natural recovery sites (DD 2) than in reforestation sites (DD 3) because natural recovery sites may provide a better habitat (coarse woody debris and understory vegetation) for rodents. More litter dwelling arthropods inhabited in natural restoration sites versus reforested sites which may explain the success of rodents. Thinning of Pinus koraiensis planatation reduced litter layer, resulting in a decrease of the detritovorous Diptera (Kwon et al. 2010a). In Argentina, the abundance of Diptera was much lower in logged shrub forest compared with dry forest (Molina et al. 1999). Such results show the similar impact of environmental disturbance caused by thinning and forest fire. However, Diptera belong to other feeding guilds in addition to detritivore guid. In Korea, dipterans collected by pitfall trapping at 60 sites in 12 high mountains were classified based on feeding guilds in family level. Among 50 feeding guild defined families of 61 total families, 25 families belonged to the detritivores (50%), followed by herbivores (18%), predators (6%), parasites (5%), and aquatic inhabitants (5%). However, at the individual level, 91 % were detritivores, followed by herbivores (6%), predators (0.8%), aquatic inhabitants (0.8%), and parasites (0.7%) (Kwon unpublished). Thus, grouping of Diptera as detritivores may not be far from a reality. Although Crustacea, Thysanura, and Orthoptera-D have homogenous feeding habits that are dependent on leaf litter, their abundance changed differently depending of various environmental disturbances caused by thinning or forest fire (Yi and Moldenke 2008; Kwon et al. 2010a). These arthropods may not be abundant enough to reflect a stable pattern related with disturbances. Many dipterans inhabit in leaf litter and soil in temperate forests of the Eurasian continent (Hovemeyer 2000), and the litter layer was the most important factor affecting the development of soil dwelling dipteran larvae (Frouz 1997). Therefore, Diptera has a high possibility for use as a bioindicator for various environmental disturbances related with the loss of litter layer (Kwon et al. 2010a). In this study, herbivores were positively correlated with intensity of disturbance (i.e., reciprocal to litter depth) because herbivores obtain food and live in herbs or shrubs which grow after the destruction of trees. However, Bae et al. (2011) reported that species richness and abundance of moths collected using light traps in Samcheok from 2006 to 2009 were highest in unburned site (DD 0), followed by surface fire site (DD 1), and lowest in crown fire site (DD 2). This pattern is opposite of ours. The reason for this difference is not clear but it may be due to different study methods. The present study was carried out in three areas for one season, whereas that of Bae et al. (2011) was carried out in one area for a longer period. In butterfly fauna surveyed from 2005 to 2010 in Samcheok, species richness and abundance were higher in burned forests than in unburned

139 June 7-8, 2011 well-conserved forests (Kwon unpublished). This agrees with our results but disagrees with those of Bae et al (2011). However, in butterfly data in Uljin for 3 years, abundance and richness did not differ between burned and unburned forests (Kwon unpublished). It may not be correct to conclude general patterns using data collected from restricted areas due to the high spatial variation of arthropod communities. Herbivores may respond differently to changes in the vegetation structure caused by environmental disturbances. Herbivores feeding on herbs and shrubs may increase after forest destruction, whereas herbivores feeding on trees may decrease. However, the latter may increase with recovery of vegetation. Bae et al. (2011) reported that moths feeding on trees increased continuously after forest fire. Predator groups may be more vulnerable to disturbance than any other group because they are more abundant in stable environments (Choi et al. 2010). However, in the present study, the abundance of Formicidae and Araneae increased in response to disturbance, whereas Carabidae and Staphylidae did not change. In studies on feeding guilds of arthropods, Formicidae is usually grouped as a predator, but its food niches are diverse at the genus or species level (Hölldobler and Wilson 1990; Lobry de Bruyn 1999). Ponerinae ants are usually predators, whereas several species of developed Formicinae and Myrmicinae (e.g., Lasius japonicus, Paratrechina flavipes, and Myrmica carinata in present study) have herbivorous roles, using aphid honeydew or plant sap in addition to their main roles as predators. This may be one of the reason that the change of abundance of Formicidae is similar to that of herbivores. Although all arthropods in Araneae are predators, their habitats and predation strategies vary (Lee and Lee 1990; Uetz et al. 1999). Of the spiders (i.e. Araneae) collected in the present study, 81 % were ground foragers and 19% were web spinners. Web spinners decreased as intensity of disturbance increased, while ground-foraging spiders increased (Kwon unpublished). Abundance of ground-foraging spiders mainly determined the pattern of Araneae. Carabidae and Staphylinidae were associated with habitats provided by fire residuals (Gandhi et al. 2001). Most species of Carabidae and Staphylinidae have functional hind wings and, therefore, have strong dispersal ability. It is likely that these beetles from unburned forests may actively migrate to burned forests after fires (Kwon and Park 2005). Various taxa and functional guilds respond differently to forest disturbances. Thus, it is more appropriate to use diverse arthropods rather than one taxa or one functional guild to understand the impact of disturbances on forests (Lawton et al. 1998; Buddle et al. 2006; Maleque et al. 2009). It is generally accepted that data obtained at fine taxonomic levels such as genus or species may have higher quality information than that of coarse levels such as order or family. Hence, a single taxon or group (i.e., butterflies, moths, carabid beetles, dung beetles, ants,

140 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 spiders, syrphid flies, etc.) is more frequently used to assess effects of environmental change on ecosystems (Lee and Lee 1990; Kwon et al 2005; Masís and Marquis 2009; Kwon et al. 2010b; Lee and Ishii 2010; Bea et al. 2011). However, identification of arthropods at fine levels requires more time, money, and expertise. At coarse levels such as family, order or class, a variety of information from various fine levels would be pooled, resulting in the reduction of stochastic variation occurring at fine levels. In addition, the summation of information may lead to emergence of properties of the group that are not shown in each of its components. Coarse filter approaches to biodiversity monitoring may have some benefit for arthropods (Oliver and Beattie 1993). Identification of arthropods at coarse levels is easy, and errors during the identification may occur less than at fine levels. Hence, information obtained from coarse taxonomic levels may be more important than generally expected. The different abundance changes among three functional guilds, detritivores, herbivores, and predators, show complexity in the response of arthropods due to environmental disturbance. This complexity may be theorized by simple arguments. If disturbance in forests is restrictively defined as tree kill, change of vegetation structure from various environmental disturbances such as fire, strong winds, clear-cutting, and pest damage may be similar in terms of disappearance of trees followed by growth of shrubs and herbs, and loss of litters (Choung et al. 2004; Walton et al. 2008; Kwon et al. 2010a; Peltola et al. 2010). In moderate disturbance, ground fire and thinning are similar in their patterns of selective tree kills. In this study, disturbance from fire and human activity acted additively on arthropod communities as evidenced by high correlation of DD and litter depth, and of litter depth and abundance of arthropods. Loss of litter caused by disturbance led to a decline of detritivores. However, death of trees led to growth of herbs and shrubs, resulting in increase of herbivores. In contrast, as predators have more flexible food niches than other feeding guilds, their changes are expected to be different from the simple pattern of herbivores or detritivores. When their favorite prey items decrease because of disturbance, they may alternatively feed on new increasing prey item. More diverse patterns were found among predators compared herbivores and detritivores. Diversification of arthropod communities in intermediate disturbance was found as well as other studies (Blair and Launer 1997; Wang and Chen 2010). Our results suggest that natural recovery of burned forest led to less impact to and more diversity of arthropod communities compared with reforestation of burned forests

141 June 7-8, 2011 Acknowledgements This study was carried out in the research project of Korea Forest Research Institute, Ecosystem monitoring and development of ecosystem recovery technology in burned forest (Project No. FE ). References Allen CD, Macalady AK, Chenchouni H, Bachelet D, Mcdowell N, Vennetier M, Kitzberber T, Rigling A, Breshears DD, Hogg EH, Gonzalez P, Fensham R, Zhang Z, Castro J, Demidova N, Lim JH, Allard G, Running SW, Semerci A, Cobb N (2010) A global overview of drought and heat-induced tree mortality reveals emerging climate change risks for forests. For Ecol Manage 259: Bae YS, Chae DY, Ju YD, Bae JH, Kim JM, Ahn NH, Lee CM (2011) Changes of species diversity on moth communities at forest fire region in Samcheok, Korea. Korean J Appl Entomol (in Korean with English summary) (in press) Beverly JL, Martell DL (2005) Characterizing extreme fire and weather events in the Boreal shield ecozone of Ontario. Agric For Meteorol 133:5-16 Blair RB, Launer AE (1997) Butterfly diversity and human land use: Species assemblages along an urban greadient. Biol conserv 80: Buddle CM, Langor DW, Pohl GR, Spence JR (2006) Arthropod responses to harvesting and wildfire: Implications for emulation of natural disturbance in forest management. Biol conserv 128: Choi SS (1996) Soil Zoology. Wonkwang University Pub., Jeonju, p 488 (in Korean) Choi WI, Choi KS, Lyu DP, Lee JS, Lim JL, Lee SH, Shin SC, Chung YJ, Park YS (2010) Seasonal changes of functional groups in coleopteran communities in pine forest. Biodivers conserve 19: Choung YS, Kim JH (1987) Effects of fire on chemical properties of soil and runoff, and phytomass in Pinus densiflora forest. Korean J Ecol 10(3): Choung YS, Lee BC, Cho JH, Lee KS, Sang IS, Kim SH, Hong SK, Jung HC, Choung HL (2004) Forest responses to the large-scale east coast fires in Korea. Ecol Res 19: Edmonds RL, Agee JK, Gara RI (2000) Forest Health and Protection. The McGraw-Hill Companies, p 630 Fox AS, Matzner E (1987) The role of fire in the scleomorphic forests and shrublands of eastern Australia. In L. Trabaud (ed.), The role of fire in

142 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 ecological systems. SPB Academic Publishing, Hague, pp Frouz J (1997) Changes in communities of soil dwelling dipteran larvae during secondary succession in abandoned fields. Eur J Soil Bio 33(3): Gandhi KJK, Spence JR, Langor DW, Morgantini LE (2001) Fire residuals as habitat reserves for epigaeic beetles (Coleoptera: Carabidae and Staphylinidae). Biol Conserv 102: Gibb H, Rettersson RB, Hjältén J, Hilszczański J, Ball JP, Johansson J, Alegrim O, Danell K (2006) Conservation-oriented forestry and early successional saproxylic beetles: Responses of functional groups to manipulated dead wood substrates. Biol Conserv 129: Glasgow LS, Matlack GR (2007) Prescribed burning and understory composition in a temperate deciduous forest, Ohio, USA. For Ecol Manage 238: Gondard H, Romane F, Aronson J, Shater Z (2003) Impact of soil surface disturbances on functional group diversity after clear-cutting in Aleppo pine (Pinus halepensis) forests in southern France. For Ecol Manage 180: Hölldobler B, Wilson EO (1990) The Ants. Harvard University Press, Massachusetts, p 756 Hovemeyer K (2000) Ecology of Diptera. In Manual of palaearctic Diptea. Eds. Papp, L. and B. Darvas pp science Herald, Budapest Kim WY, Chae BG (2009) Characteristics of rainfall, geology and failure geometry of the landslide areas on Natural Terrains, Korea. J Eng Geol 19(3): (in Korean with English summary) Korea Forest Research Institute (1997) Second year report on ecological study on forest-fired area in Goseong (in Korean), p 155 Korea Meteorological Administration (2005) Weather database. [Online] Kremen C, Colwell RK, Erwin TL, Murphy DD, Noss RF, Sanjayan MA (1993) Terrestrial arthropod assemblages: their use in conservation planning. Conserv Biol 7(4): Kwon TS (2008) Change of Abundance of Arthropods in Pine Forests Caused by Aerial Insecticide Spray. Arch Environ Contam Toxicol 54: Kwon TS, Kim SS, Chun JH, Byun BK, Lim JH, Shin JH (2010b) Changes in butterfly abundance in response to global warming and reforestation. Environ Entomol 39(2): Kwon TS, Park JK (2005) Comparative study on beetle fauna between burned and unburnd forest. Jour Korean For Soc 94(4): (in Korean with English summary) Kwon TS, Song MY, Shin SC, Park YS (2005) Effects of aerial insecticide sprays on ant communities to control pine wilt disease in Korean pine forest. Appl

143 June 7-8, 2011 Entomol Zool 40(4): Kwon TS, Yang HM, Shin JH, Kim SK, Yi HB (2010a) Effects of thinning on abundance and community structure of arthropods in a Pinus koraiensis plantation. Kor J Appl Entomol 49(3): (in Korean with English summary) Lachat T, Nagel P, Cakpo Y, Attignon S, Goergen G, Sinsin B, Peveling R (2006) Dead wood and saproxylic beetle assemblages in a semi-deciduous forest in Southern Benin. For Ecol Manage 225: Lain EJ, Haney A, Burris JM, Burton J (2008) Response of vegetation and birds to severe wind disturbance and salvage logging in a southern boreal forest. For Ecol Manage 256: Lawton JH, Bignell DE, Bolton B, Bloemers GF, Eggleton P, Hammond PM, Hodda M, Holt RD, Larsen TB, Mawdsley NA, Stork NE, Srivastava DS, Watt AD (1998) Biodiversity inventories, indicator taxa and effects of habitat modification in tropical forest. Nature 391: Lee CM, Ishii M (2010) Changes in species diversity of ground beetle assemblages in the riverbed of the Yamato River. Jpn J Environ Entomol Zool 21(1): Lee EJ, Lee WS, Rhim SJ (2006) Differences in density and body weight of rodents in different restored forest area after forest fire. Jour Korean For Soc 95(3): Lee GH, Lee HP (1990) Guild structure and seasonal occurrences of spider communities in pine plantation habitat. Korean J Ecol 13(2): Lee KS, Choung Y, Kim SC, Shin SS, Ro CH, Park SD (2004) Development of vegetation structure after forest fire in the east coastal region, Korea. Korean J Ecol 27(2): Lim JH, Shin JH, Lee DK, Suh SJ (2006) Climate change impacts on forest ecosystems: research status and challenges in Korea. Korean Journal of Agricultural and Forest Meterology 8(3): Lobry de Bruyn LA (1999) Ants as bioindicators of soil function in rural environments. Agric Ecosyst Environ 74: Maleque MA, Ishii H, Maeto K, Taniguchi S (2010) Seasonal prevalence of arthropods after line thinning of overstocked Japanese cedar (cryptomeria japonica D. Don) plantations in central Japan. Landscape Ecol Eng 6: Maleque MA, Maeto K, Ishii HT (2009) Arthropod as bioindicators of sustainable forest management, with a focus on plantation forests. Appl Entomol Zool 44(1): 1-11 Masís A, Marquis RJ (2009) Effects of even-aged and uneven-aged timber management on dung beetle community attributes in a Missouri Ozark forest. For Ecol Manage 221:

144 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 McCune B, Mefford MJ (1999) PC-ORD. Multivariate analysis of ecological data, Version 4.0. MjM Software Design, Gleneden Beach Molina SI, Valladares GR, Gadner S, Cabido MR (1999) The effects of logging and grazing on the insect community associated with a semi-arid chaco forest in central. Argentina J Arid Environ 42: Mun HT, Choung YS (1996) Effects of forest fire on soil nutrients in pine forests in Kosong, Kangwon Province. Korean J Ecol 19(5): Niemelä J, Langor D, Spence JR (1996) The importance of small-scale heterogeneity in boreal forests: variation in diversity in forest-floor invertebrates across the succession gradient. Ecography 19: Noske NM, Hilt N, Werner FA, Brehm G, Fiedler K, Sipman HJM, Gradstein SR (2008) Disturbance effects on diversity of epiphytes and moths in a montane forest in Ecuador. Basic Appl Ecol 9: 4-12 Oh KC, Kim JK, Jung WO, Min JK (2001) The changes of forest vegetation and soil environmental after forest fire. J Korean Env Res and Reveg Tech 4(3): Oliver I, Beattie AJ (1993) A possible method for the rapid assessment of biodiversity. Conserv Biol 7: Peltola H, Ikonen VP, Gregow H, Strandman H, Kilpelainen A, Venalainen A, Kellomaki S (2010) Impacts of climate change on timber production and regional risks of wind-induced damage to forests in Finland. For Ecol Manage 260: Peterson JC (2000) Catastrophic wind damage to North American forests and the potential impact of climate change. Sci Total Environ 262: Ro DK, Gong JS, Lee SH, Kim CM, Kim CJ, Seo SA, Paek JH, Kim SH, Sim WB, Son YM, Lee WK (2000) Damange of forests in the east coast fires. In: Report of the east coast fires in 2000 (I), pp The joint association for the investigation of the east coast fires, Seoul. (in Korean) Samways MJ (1994) Insect Conservation Biology. Chapman and Hall, London, p 358 Schowalter TD, Zhang YL, Rykken JJ (2003) Litter invertebrate responses to variable density thinning in western Washington forest. Ecol Appl 13: StatSoft, Inc. (2004) Statistica for Windows. Version 6. Tulsa Triplehorn CA, Johnson NF (2005) Borror and Delong s introduction to the study of insects. Tomson brooks/cole, p 863 Uetz GW, Halaj J, Cady AB (1999) Guild structure of spiders in major crops. J Arachnol 27: Walton A, Hughes J, Eng M, Fall A, Shore T, Riel B, Hall P (2008) BC Ministry of Forest and Range, provincial-level projection of the current mountain

145 June 7-8, 2011 pine beetle Outbreak, Victoria. BCMPB.v5. BeetleProjection.Update.pdf Wang S, Chen HYH (2010) Diversity of northern plantations peaks at intermediate management intensity. For Ecol Manage 259: Whelan RJ (1995) The ecology of fire. Cambridge University Press, Cambridge, p 343 Yi HB, Moldenke A (2008) Responses of litter-dwelling arthropods to four different thinning intensities in Douglas-fir forests of the Pacific Northwest, USA. Ann Zool Fennici 45:

146 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Impacts of Forest Fires on Ecosystems Functioning including Biodiversity and Climate Change: a perspective from Hindu-Kush Himalayan (HKH) Region Sundar P. Sharma 1 * and Johann G. Goldammer 2 1 Member, UN-ISDR Wildland Fire Advisory Group and Global Wildland Fire Network, Coordinator, UNISDR-Regional South Asia Wildland Fire Network, Afaldole, Lalitpur, Nepal. Tel , Cell: , sharmasp1966@yahoo.com, Web: 2 Global Fire Monitoring Center, Secretariat of UNISDR Global Wildland Fire Network / Wildland Fire Advisory Group, Georges-Koehler-Allee 75, D Freiburg, Germany info@gfmc.org Abstract Wildfires in high altitude Hindu Kush-Himalayas (HKH) ecosystems are a major driver for destruction of pristine biodiversity, including the habitats of many rare species. During the long and intense dry seasons occurring annually in the region, wildfires are a regular phenomenon, many of them having a potential to cause major damages; e.g., serious degradation of forests, changes of ecosystem properties, and deterioration of social and economic conditions in some land-use systems and natural vegetation types. Fires occurring in the highlands of Tibet, Sikkim, Bhutan and the northern part of Nepal at altitudes from 2,700 to 3,800m above sea level often cross national borders, especially during the dry winter fire season (November to January) (Schmidt-Vogt1989). Whereas, the fires are more common in lowlands in the Hindu Kush-Himalayan region during the hot dry and windy summer season (February to May) (also associated with agricultural burning). Observations indicate that the occurrence of wildfires is increasing as a consequence of regional warming and extended dry spells. The southern slopes of the mountains are primarily affected, since they are generally warmer and drier compared to northern slopes and are therefore exposed to high human pressure. Atmospheric Brown Clouds (ABC) are a hot topic and burning issue among scientists, politicians and scholars in the region in recent years. ABC are consequences of socio-economic changes, an expression of escalating air pollution and a major driver of regional climate change. It s net positive feedback

147 June 7-8, 2011 to the regional climate shape a vicious circle with increasing occurrence of wildfires in the region. This paper reviews the impacts of wildfires on ecosystems particularly to the biodiversity and regional climate. INTRODUCTION Wildfires in high altitude ecosystems of Hindu Kush-Himalayas (HKH) region in recent years have become a major driver for destruction of pristine biodiversity, including the habitats of many rare species. They are also affecting cultural heritage sites and land-use systems that provide the basis for livelihoods to a population of around 150 million people living in the mountain region. Most importantly, the secondary consequences of wildfires include the destruction of soil protecting vegetation cover, affecting water regimes for a population of 1.4 billion (UNEP 2007). These fires are often border crossing in nature and initiated political discourse in transboundary smoke haze pollution. Fires occurring in the highlands of Tibet, Sikkim, Bhutan and the northern part of Nepal at altitudes from 2,700 to 3,800m above sea level often cross national borders, especially during the dry winter fire season (November to January). Observations indicate that the occurrence of wildfires is increasing as a consequence of regional warming and extended dry spells. The southern slopes of the mountains are primarily affected, since they are generally warmer and drier compared to northern slopes and are therefore exposed to high human pressure. The wildfires in the recent past in the southern stretch of the HKH region not only contributing to regional and the overall global problem but also pose a higher risk to the communities if looked at from the point of view of the fragile Himalayan ecology (SAARC 2009). At the same time, the Atmospheric Brown Cloud (ABC) has come into discourse among scientists, researchers, politicians and academia after a preliminary assessment report published by UNEP in 2002 ( Now, it is a hot topic and burning issues particularly in south Asia regarding regional climate change and in impacts on air quality, agricultural production, human health, changing glacier dynamics and wildfire regimes (Ramanathan et al. 2008, Thompson 2006, Gustafsson et al. 2009). Despite there is a general consensus on one of the regional climate driver, ABC, the source of it is still under debate (Ramanathan et al. 2008, Thompson 2006, Gustafsson et al. 2009, Economic Times 2008). Transboundary wildfires and haze pollution is an emerging issue in south Asia. Moreover, fire is the most important disturbance agent in global vegetation cover worldwide, affecting between 3 and 4 million square kilometers annually

148 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 (GFMC 2009). While there is clear evidence of the historic role and timescale of fire in many ecosystems, along which many fire-dependent ecosystems evolved, the current trend provides evidence of increasing use of fire in land use and landuse change as well as an increase of destructive wildfires (uncontrolled and unwanted fires). Burning of forests and other vegetation is a major driver of transferring carbon from the terrestrial sphere to the atmosphere. It constitutes a significant source of radiatively active (greenhouse) gases and aerosols. Fires globally consume about 5% of net annual terrestrial primary production per annum, and release about 2-4 billion metric tons of carbon (C) per year. Approximately 0.6 billion tons of carbon emitted to the atmosphere come from tropical deforestation and peat fires, the global figure is equivalent to about 20-30% of global emissions from fossil fuels (GFMC 2009). A recent study (by the National Center for Ecological Analysis and Synthesis, California) reveals that the emissions of carbon dioxide, a greenhouse gas adding to global warming, are about equal to half the output from burning fossil fuels such as coal, Balch and colleagues wrote in a study published 23 April 2009 in the journal Science on the role of fire in the climate system. Further, one of the key messages of the UNEP report (UNEP 2007) is that the Earth s surface is warming. This is now evident from observations of increases in global average air and ocean temperatures, widespread melting of snow and ice, and rising global average sea level. Other major impacts include changes in water availability, land degradation, food security, and loss of biodiversity. The Conference of the Parties to the UNFCCC has identified fire management as one of the options for reducing emissions from deforestation and forest degradation in developing countries (REDD concept) as fire is one of the defined 44 Essential Climate Variables (ECVs). While in the past century the global average temperature increased by 0.74 C, the best estimate of the Intergovernmental Panel on Climate Change (IPCC) for additional warming over the current century is projected to be from 1.8 to 4.0 C. Climate change may further exacerbate the loss of biodiversity and degradation of land, soil, forest, freshwater and oceans. The projected increase in frequency and intensity of heat waves, storms, floods and droughts would dramatically affect many millions of people including those living in Hindu-Kush Himalayan region. INCREASING TREND OF WILDFIRE INCIDENTS During the long and intense dry seasons occurring annually in the region,

149 June 7-8, 2011 wildfires are a regular phenomenon, many of them having a potential to cause major damages; e.g., serious degradation of forests, changes of ecosystem properties, and deterioration of social and economic conditions in some land-use systems and natural vegetation types. The ecosystems and society are very vulnerable to wildfires, in general, and to the secondary disasters, such as landslides and flash floods, that follow disastrous wildfires. A Case Study: Nepal Active fire incidents and total burning days are increasing in Nepal from the recent past (Figure 1) impacting huge loss of lives, properties and natural resources. Forest fire is considered one of the climate induced disasters in the National Adaptation Programs of Action (Nepal NAPA 2010). In 2009 alone forest fires claimed 49 lives injured 9 people; about 146,742 hectare of forest destroyed and caused the loss of about Rs 134,415,000 (GFMC 2010, MoFSC, 2009, RSAWFN 2009). A total of 9 people were reported dead, 3 people seriously injured, 431 houses were completely destroyed, 92 animals killed and more than 82,000 hectares forests were burnt in 2010 fire season (RSAWFN 2009). A proper damage assessment of a fire is not being practice in Nepal. Moreover, there is no any record of impacts of fire on wildlife, medicinal plants, secondary disasters (health and water induced disasters) and, regional climate (atmospheric brown cloud). Figure 1: Daily fire incidence in Nepal from 27 March 2008 to 21 May (From 2009 to 2010 Burning days increased by 22% and Active fires increased by 30%, Peak fire generally occurring in last week of April) Data source: FIRMS-MODIS/FAO/UMD/NASA. On 12 March 2009, the Moderate Resolution Imaging Spectroradiometer (MODIS) on NASA s Aqua satellite caught a glimpse of a relatively rare event: large scale forest fires in the Himalaya Mountains of Nepal (Image 1). Places where the sensor detected active fires are outlined in red. The image is centered on Nepal and shows the towering Himalaya Mountains arcing through the small country. Many national parks and conservation areas are located along the northern border of the country, and the fires appear to be burning in or very near some of them

150 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Image 1: Huge plumes of smokes coming from mega-fires in Makalu-Barun National Park, Langtang National Park and Annapurna Conservation Area of Nepal which cross the India border. Source: Earth Observatory, 12 March South Asia Regional Fatality Record In south Asia 109 people died and injured in wildland fires, notably in Nepal (58) and India (51) in 2009 ( 2010/GFMC-Bulletin pdf). Annual Regional Wildland Fire Fatalities Report covering the year 2009 is based on the evaluation of media (newspaper, radio broadcasts and TV) published on the internet, as well as agency reports (Note: This report includes all death and injuries of firefighters and civilians related one way or the other to wildfires, i.e. accidents during firefighting, accidents on the way to respond to a fire, wildfires affecting people in wildlands or structures ignited by wildfires, or other indirect accidents, e.g., by smokereduced visibility, or people admitted to hospitals, medical treatment or deceased due to wildland fire smoke inhalation. Since only a fraction of wildfire-related accidents may be reported through media and subsequently published on the web ( and data stored by the UNISDR-Regional South Asia Wildland Fire Network, it has been presume that the attempt to compile a regional statistical database is still very incomplete and may scratch the surface only

151 June 7-8, 2011 The global report reveals that 374 deadly wildland fire fatalities and 160 injuries were reported worldwide in Risk to downstream population from secondary disasters Observation indicates a major portion of the forest areas in the mountains of the HKH region burnt annually. Already large areas of the Himalayan forests have cleared indiscriminately for agriculture making them vulnerable to soil erosion and landslide. Large amount of sediment production in upstream areas that is transported through rivers causing river bed rise and consequently resulting in flooding in the downstream of the HKH. An estimate from Nepal in 2009 fires revealed that about 40 percent forest cover has been lost in upstream areas from the wildfires. The situation might create a risk of soil erosion, mud-flow, debris flow and flooding in the downstream areas imposing a high risk to people leaving in downstream areas during the rainy seasons. Damage assessment, rehabilitation works and warning system development are the urgent needs. WILDLAND FIRES AND BIODIVERSITY CONSERVATION Forest fire was once viewed as synonymous with forest destruction and degradation (Botkin 1990). Certain forms of disturbance, however, are now held by ecologists and conservation biologists to play a fundamental and creative role in maintaining the natural heterogeneity in environmental conditions that organisms experience through space, time, or both. The importance of disturbance to the ecology of species and conservation of biodiversity has gained widespread recognition (Brawn et al. 2001, cited by Connell 1978, Sousa 1984, Pickett & White 1985, Petraitis et al. 1989, DeGraaf & Miller 1996a, Askins 2000). In restoration ecology, landscape ecology, and the concept of ecosystem management, natural disturbance is now generally recognized as essential for maintaining biodiversity (Brawn et al. 2001, cited by Alverson et al. 1994, Askins 2000). Some of the models that related disturbance to enhanced species diversity are: density-independent mortality of organisms changes in habitat and resource levels natural disturbance on species diversity at the landscape or regional scale (spatially structured populations) Conservation strategies involving the management of disturbance through

152 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 some combination of flooding, application of fire, or the expression of wildfire, and use of certain types of silviculture have the potential to maintaining biodiversity at the local, landscape, and regional scale (Brawn et al. 2001). However, uncontrolled wildland fires have a greater degree of threat to biological diversity at ecosystem level (NBS 2002) but it can be improved and/or maintained when controlled burns are applied as fires vary more in terms of type and intensity resulting in highly variable fire effects (Trollope et al. 2008). Society tends to have various competing goals, many of which depend on biodiversity. When humans modify an ecosystem to improve a service it provides, this generally also results in changes to other ecosystem services. As a result of such trade-offs, many services have been degraded. The Millennium Ecosystem Assessment shows that human actions often lead to irreversible losses in terms of diversity of life on Earth and these losses have been more rapid in the past 50 years than ever before in human history (Green Facts). Biodiversity loss, due to unwise use of fire, has direct or indirect negative effects on several aspects of human well-being at local and national level, such as food security, vulnerability to natural disasters, energy security, and access to clean water and raw materials. It also affects human health, social relations, and freedom of choice. Impacts of disturbances in floral and faunal diversity The hypothesized ultimate agent of decline acting on a population is the unsuitable fire management programme implemented within the protected areas (Pfab and Witkowski 1999). Many aspects of the disturbance ecology of birds require further research. Important questions involve associations between the intensity and frequency of disturbance and the viability of bird populations, the scale of disturbance with respect to the spatial structure of populations, and the role of natural vs. anthropogenic disturbance (Brawn et al. 2001). Reilly et Al. (2006) observed that there was no significant effects of burning on total bird abundance or species richness in savanna. Patchy fires have been an important natural disturbance in tropical ecosystems for millennia (Reilly et Al. 2006). Fire is an important component in the maintenance of grassland diversity (Uys et al. 2004, cited by Freeman 1998, Lunt and Morgan 2002). It has long been used to manage grasslands for livestock production in many part of the world. Though the dominant grasses were strongly influenced by season and frequency of fire, forb diversity showed no consistent trends. Most forb species tolerate a wider range of season and frequency of fires than the dominant grasses. However, to accommodate those species with low tolerance of frequent fires, parts of the

153 June 7-8, 2011 landscape will require less frequent fires (Uys et al. 2004). Understanding and applying ecological disturbance offers opportunities to conserve large and diverse group species many of which are declining as a result of habitat loss and successional changes in habitat structure (Brawn et al. 2001). Uncontrolled forest fires have devastating effects on animals with limited mobility. But, Kiss and Magnin (2006) found that Mediterranean land snail communities are particularly resilient to fires. Although abundance is drastically reduced after fire, species richness and diversity of communities are preserved, whatever the fire regime, since the time lapse between two successive fires is longer than the time required for malacofauna recovery (i.e. around 5 years). Further, a study on bird species by Slik and Balen (2006) found that three years after fire the number of birds and bird species were similar for undisturbed and burned forests, but species diversity and turnover were significantly lower in the burned forests These differences are probably related to the fact that undisturbed forests provide more spatial variation in vegetation structure (niche space) than burned forests. The shift in bird species composition between undisturbed and burned forests was significantly related to changes in the forest understorey (appearance of pioneer herbs and woody plants after fire) and the middle- and upper storey (strong decrease in foliage cover after fire). The species composition in tropical forests is strongly related to (vertical) vegetation structure and plant species composition. When fires become too frequent for the vegetation to recover, both the flora and avifauna are likely to become seriously negatively affected to the point that recovery becomes impossible (Slik and Balen 2006). Theory suggests that species richness should be highest at intermediate levels of disturbance (Connell 1978, Huston 1979). However, this is not always the case. In the Scottish Highlands, plant diversity was found to increase with fire frequency (Hobbs et al. 1984) while, conversely, in the North American prairies, plant diversity decreased with increasing fire frequency (Uys et al. 2004, cited by Collins et al. 1995). Therefore, researchers should always be careful while applying disturbances in biodiversity conservation considering species-habitat interaction in particular geographical location. Fire and ecosystem diversity Under natural conditions, fire is the major driver that governs the development and function of most deciduous, coniferous, and mixed forest types at temperate latitudes (Brawn et al. 2001, cited by Wright and Bailey 1982, Attiwill 1994, Frost 1998). Fire, along with other factors, is key in maintaining entire ecosystems such as grasslands and savannas at temperate and tropical latitudes, and Mediterranean-type shrublands (Brawn et al. 2001, cited by Moreno

154 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 and Oechel 1994, McPherson 1997). Reductions in plant diversity and the strong dominance of the forest understorey by only a few species of pioneer herbs, shrubs and trees (Slik and Eichhorn 2003) also strongly reduce habitat heterogeneity in burned forests. (Slik and Balen 2006). Many studies (Brawn et al. 2001, Connell 1978, Sousa 1984, Pickett and White 1985, Petraitis et al. 1989, DeGraaf and Miller 1996a, Askins 2000, Wright and Bailey 1982, Attiwill 1994, Frost 1998, Moreno and Oechel 1994, McPherson 1997, Marynowski and Jacobson 1999.) revealed that the frequency and intensity of wildland fire is a determinant of the presence of many terrestrial habitat types (or ecosystems) (Figure 2) in many part of the world that support significant components of species and ecosystem diversity. Figure 2 (Left). Some terrestrial habitats qualit ordinate according to frequency and intens disturbance. Even-aged forest can result from fire throw, floods, or even-aged silviculture. Uneve forest results from small-scale wind throw or u aged silviculture. Woodlands and savanna resul moderate to light intensity frequent fires. Gra generally result from frequent and intense fire, g or both (Adopted from Brawn et al. 2001) Despite the importance of fire in shaping many forest ecosystems, most fire studies have focused on the effects of fire on vegetation; relatively less attention has been paid to the effects of fire on genetic and faunal diversity (Reilly et.al. 2006, cited by Parr and Chown 2003). Fire and alien species The most significant effects of alien grasses on ecosystems result from interactions between grass invasion and fire. Alteration of fire regimes clearly represents an ecosystem-level change caused by invasion (D Antonio et. al. 1992). The effects of alien grasses on ecosystem function (fire, nutrient loss, altered local microclimate, prevention of succession) are significant on the local scale and are becoming increasingly important on regional and global scales. Moreover, the interaction of competition with alien grasses, fire, and the prevention of succession now represents a substantial global threat to biological diversity on the genetic, population, and species levels (D Antonio et al. 1992)

155 June 7-8, 2011 Management of fire is destined to become an increasingly significant challenge due to increased fuel loads arising from the invasion of exotic grasses, either on the rainforest boundary or within patches where the canopy has been opened by fire (Liddle et al. 2006). The regional population is at risk by more frequent and more intense fire due to the invasion of exotic grass species and land use changes in the catchment which result hydrologic changes lead to increased drying of the rainforest habitat (Liddle et al. 2006). Similar situation has already been observed in Chitawan National Park of Nepal. Fire and nutrient cycling Fires themselves alter nutrient budgets profoundly; they volatilize some elements (notably carbon and nitrogen) while converting others into biologically more available, mobile forms for at least a short time. The selective loss of nitrogen in particular drives ecosystems toward nitrogen limitation. Nutrient losses to stream-water, ground-water, and the atmosphere are also enhanced following fire, and these can have significant effects on the chemistry of the atmosphere regionally and globally (D Antonio et al. 1992). Socio-economics of pyrodiversity Archaeology provides a long-term perspective on human actions and their environmental consequences that can contribute to conservation and restoration efforts (Hayashida 2005). In one hand, ecologists have increasingly turned to sociology of environment to explain and manage modern ecosystems and landscapes. And the other hand, archaeologists become more involved in research directed at contemporary environmental issues. But, they need to consider the potential uses and abuses of their findings in management and policy debates (Hayashida 2005). The use of prescribed burning is a much debated issue in ecology and forest management, particularly in areas prone to major wildfires that threaten people, property, and forests. In the United States, the likelihood of catastrophic fires increased with the practice of fire suppression that interrupted the natural and cultural fire regimes of the past and resulted in large fuel accumulations (Hayashida 2005). In the southern Appalachian mountains, frequent burning was used by native Americans to improve conditions for travel and game, and later by European settlers to improve grazing for livestock (Cathryn et al. 2006, cited by

156 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Brose et al. 2001). In Nepal major cause of fire is link with livelihood of people (e.g. to boost-up new flush of grass for livestock, to facilitate collection of nontimber forest products, hunting wild animals, etc.) (Sharma 1996). A study conducted in in Senegambia (Stoate et al. 2001) reveals that more sustainable use of savanna farmland could have both agronomic and wider conservation benefits, and the provision of information that accommodates farmers cultural and economic incentives could benefit both farmers and wildlife. Farmers in this study reported a continuing decline in the area of trees, shrubs and fallow land in the subsequent 30 years, suggesting an increase in the area of open land lacking shrubs and trees and representing unsuitable habitat for Whitethroats. Vaux et al. (1984) conducted the first study on the influence of fire on the economic value of forest recreation. The authors state, Willingness-to-pay is an appropriate measure for valuing the effects of fire on forest recreation. Using a random effects probit model (Loomis and Caban 1998) to account for the panel nature of the data, the average willingness to pay to reduce catastrophic fire on 2570 acres was $56 per household. Since acreage of habitat protected is a statistically significant variable in the willingness-to-pay function, this function can be used by managers to evaluate the incremental benefits of different fire management plans that reduce additional acres burned. These benefits can serve as justification for funding of prescribed fire and fuel reduction programs to protect critical habitat of an endangered species (e.g. Northern and California Spotted Owl). There is growing recognition that protection of other environmental values beyond traditional multiple uses needs to be incorporated into fire decision making (González-Cabán and Chase 1992, González-Cabán 1993). These values often reflect public desire to know that rare and distinctive ecosystems exist (e.g. existence value, proposed by Krutilla (1967) and will be protected for future generations (bequest value) as well as being available for visits at future times (option value). Along with recreation, these three values are sometimes referred to as Total Economic Value (Randall and Stoll 1983). These values may be particularly important for preservation of old growth forests for threatened species (Loomis and Caban 1998). Management options for biodiversity conservation Buildup of forest fuels such as thick shrub cover or woody debris contributes to the potential for wildfire in many ecosystems. Recreating natural fire regimes and actively managing fuels have often been advocated to maintain biological diversity and reduce the risk of catastrophic wildfire in forests (Wales et al. 2006, Hann Allen et al. 2002, Marston et al. 2001, Stuart 1998). Thinning

157 June 7-8, 2011 and prescribed fire are being used extensively across the interior western United States to reduce the risk of large, severe wildfires. However, the full ecological consequences of implementing these management practices on the landscape have not been completely evaluated. (Wales et al. 2006). Prescribed fire as a means of restoring or maintaining biodiversity is becoming common place throughout North America and elsewhere (Askins 2000), but long-term studies of periodic fire and particular species are few (Brawn et al. 2001). Therefore, researchers/land managers need to know how different fuel reduction practices affect wild animal (e.g. small mammal) populations to better integrate wildlife management with forest management (Cathryn et al. 2006). Now, there is a common understanding that fire suppression, timber harvest/thinning, and ungulate grazing have reduced wildfire frequency, increased conifer establishment or shrubby vegetation, and changed the overall pattern and structure of forests (Wales et al. 2006, Hann et al. 1997, Hessburg et al. 2000). A study (Barbero et al. 1987) reveals that when frequent fires occur with fire interval of under 10 years, vegetation structure tends to be very simple, with only low shrubs and grass, and fire regime may also induce change in fauna biodiversity (Kiss and Magnin 2006, Gill and McCarthy 1998). However, very less studies have been done so far in impacts of wildland fires on biodiversity at genetic, species and ecosystem levels in south Asian region. Our current understanding of plant and animal community responses to tropical forest fires is limited and strongly geographically biased towards South Asia. TRANSBOUNDARY FIRES, HAZE POLLUTION AND REGIONAL CLIMATE Atmospheric Brown Cloud (ABC): implications to regional climate change, glacier dynamics, human health, food security and wildfire regimes The Atmospheric Brown Cloud Project Report of UNEP released in November 2008 is the latest and most detailed regional assessment of the phenomenon. Wildland fires, the burning of agricultural wastes, fossil fuels in vehicles, industries and power stations and emissions from burning wood stoves, cow dung and other bio fuels form the haze - a mass of ash, acids, aerosols and other particles, where black carbon could now be responsible for between 10 to 40 per cent of current climate change (Ramanathan et al. 2008, Thompson 2006, Gustafsson et al. 2009). The brown cloud in the atmosphere consists of pollutant particles (primary

158 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 aerosols) and gases such as nitrogen oxides (NOx), carbon monoxide (CO), sulphur dioxide (SO2), ammonia (NH3), and many organic gases and acids. The brown cloud plumes are formed by the result of the combustion of biofuels in houses and industries, biomass burning, and fossil fuels in densely populated region (Ramanathan et al. 2008). Some major cities in Asia may be up to 25 per cent dimmer or darker than they were half a century ago. Reductions in visible light may also be harming agriculture, again with implications for poverty and for combating hunger under the Millennium Development Goals (MDGs). Emissions of black carbon may also be accelerating melting rates of glaciers in mountain ranges such as the Himalayas, with the dark particles absorbing sunlight and raising ice temperatures. In addition, black carbon a key component of brown clouds in some parts of the world is contributing to dimming and reducing the amount of sunlight hitting the ground in polluted parts of the globe (Ramanathan et al. 2008, Gustafsson et al. 2009). The implications of the ABC has been becoming a hot topic among scientists, politicians and scholars in the region in recent years. The atmospheric concentrations of brown clouds are large mainly during the summer season since precipitation removes the aerosols efficiently during other seasons. Satellite data have revealed that ABC plumes, measuring 1-3 km thick, surround the Hindu Kush-Himalayan region during November to March (Ramanathan et al., 2008). Wildfires in the countries in the region are contributing to a tick blanket of smoke (Brown Cloud) hanging over the south of Hindu-Kush Himalaya (mainly south of Nepal covering major cities e.g. Butawal, Birgunj, Jankpur, Biratnagar etc. and Uttaranchal and Himanchal Pradesh in India) on 25 April 2009 (compare Image 2 with Image 3). The brown cloud is blocked by the Himalaya range (Image 3 and Image 6) and confined in the south of it. These so called brown clouds have a major impact on air quality (Image 4), human health (lung, heart, skin and eye diseases), regional climate, and natural resources. Image 2: 420 fires were detected in past 24 hours in Nepal. (Source: FIRMS-MODIS/UMD/NASA, Date: 25 April 2009, 08:58:23 UTC) Image 3: Haze primarily coming from vegetation fires in the high altitude Himalayan region. The smokes is blocked by the range of Himalaya and confined to the south of the range. (Source: MODIS Rapid Response System (opened with KMZ file for Google Earth), 25 April 2009)

159 June 7-8, 2011 Transboundary wildfires and haze pollution which are one of the component of ABC and poorly researched, are an emerging issue in south Asia which needs to be addressed collectively. The cloud is considered mainly due to the vegetation fires, urban and industrial pollutions, agricultural burning, and bush clearing. This cloud is similar to the Atmospheric Brown Cloud (UNEP 2007) but appeared also in this summer which needs further validation. The brown cloud along the Himalaya hover over northern India, Pakistan and Nepal during the beginning of winter season (Image 5) forming a thick blanket. It is considered that the wildfires or agricultural fires in the region contribute to it, though they are not solely responsible. The haze generally blown from the west probably results from a combination of smoke, urban pollution and dust from neighboring Pakistan might play a role. The images (Left) clearly shows that this haze, which is border crossing in nature, mainly originated from Pakistan and India forming a thick cloud of haze along the border between Pakistan, India and Nepal. Image 4: Faded Sun in the Morning. Photo taken from the Thapathali Bridge, Kathmandu at 07:15, 29 April Image 5: MODIS on NASA s Aqua satellite image, acquired 16 November Thick smoke hides most of the areas of northern India (Amritsar, New Delhi, Jaipur), east Pakistan (Islamabad, Rawalpindi, Lahore, Okara) and south-west Nepal (Mahendranagar). (source: (NASA s Earth Observatory)

160 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Wildfires in the region contributed to a thick blanket of smoke hanging over the south of Hindu-Kush Himalayan covering major cities of Nepal (Viz. Dhangadi, Bhairahawa, Nepalganj, Pokhara, Kathmandu, Biratnagar etc.), the north and eastern region of India (Viz. Sitapur, Gorakhpur, Patana, Bhagalpur, Kolkota etc.) and most of the western part of Bangladesh on 16 March 2009, when the MODIS on NASA s Terra satellite passed over head and captured this image (Image 6). Image 6: Smoke haze over South Asia (source: Earth Observatory, 16 March 2009). Professor Lonnie Thompson of Ohio State University and a team of researchers found (2006) that high-altitude glaciers, despite residing in colder temperatures, are more sensitive to climate change. As more heat is trapped in the atmosphere, he said, it holds more water vapour. And when the water vapour rises to high altitudes it condenses, releasing the heat into the upper atmosphere, where high mountain landscapes feel the brunt of warming. He further stated that water the Himalayas have is dwindling fast. The glaciers in the Himalayas collect water from the monsoon in the wet season, and release it in the dry season. But how effective they are depends on how much water is in the glaciers (ENN 2008). Professor Lonnie Thompson of Ohio State University and a team of researchers found (2006) that high-altitude glaciers, despite residing in colder temperatures, are more sensitive to climate change. As more heat is trapped in the atmosphere, he said, it holds more water vapour. And when the water vapour rises to high altitudes it condenses, releasing the heat into the upper atmosphere, where high mountain landscapes feel the brunt of warming. He further stated that water the Himalayas have is dwindling fast. The glaciers in the Himalayas collect water from the monsoon in the wet season, and release it in the dry season. But how effective they are depends on how much water is in the glaciers (ENN 2008). UNEP (2007) reported annual losses, resulting from ground level ozone, of wheat, rice, corn and soya bean crops in China, Japan and the Republic of Korea alone may amount to around $5 billion a year. It has been warned that if the current rate of retreat continues unabated,

161 June 7-8, 2011 these glaciers and snow packs are expected to shrink by as much as 75 percent before the year 2050 (Ramanathan et al. 2008). Atmospheric Brown Cloud (ABC) Radiative Forcing The absorption of solar radiation by the surface and the atmosphere is the fundamental driver for the physical climate system, the biogeochemical cycles, and for all life on the planet. ABCs have significantly altered this radiative forcing over Asia. Brown clouds cause dimming (by at least 6 per cent in south Asia compared with the pre-industrial values) at the surface and soot in it increase solar heating of the atmosphere (absorbed solar radiation at the surface by +15 Watt per square meter compared with the pre-industrial values) (Ramanathan et al. 2008). Between 1950 and 2002, soot emissions increased three-fold, while sulphur emissions have increased seven-fold in India. Black carbon (BC) in the brown cloud has increased the vertically averaged annual mean solar absorption in the troposphere (from the surface up to 14 km in altitude) by about 15 per cent (about 14 W m -2 ) and the solar heating at elevated levels (1-4 km) over India by as much as per cent (6-20 W m -2 ) (Ramanathan et al. 2008). BC is formed through the incomplete combustion of fossil fuels, biofuel, and biomass, and is emitted in both anthropogenic and naturally occurring soot. It warms the Earth by absorbing heat in the atmosphere and by reducing albedo, the ability to reflect sunlight, when deposited on snow and ice. Black carbon stays in the atmosphere for only several days to weeks, whereas CO2 has an atmospheric lifetime of more than 100 years. Black carbon is a potent climate forcing agent, estimated to be the second largest contributor to global warming after carbon dioxide (CO2) (Ramanathan and Carmichael 2008). Not only vehicles and industries are the sources of air pollution and greenhouses gases but also fires contribute to them. Due to the haze, the amount of sunlight falling on the ground has decreased by six per cent in India (UNEP 2007) and that has a direct bearing on the plant growth. In the region, during the dry season, thousands of fires burn each year as people clear cropland and pasture in anticipation of the upcoming wet (growing) season. Intentional fires also escape people s control and burn into adjacent forest. The smoke from these fires crosses border affecting climate far away. A discourse among politicians and the scientist are going on the topic of the regional climate change. A United Nations Environment Programme (UNEP) study has warned that Asian cities from New Delhi to Beijing are getting darker, glaciers on the Himalayas are melting faster and weather system is getting more extreme because of high pollution levels (UNEP 2007). However, Sibal (Science

162 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 and Technology Minister of India) argued that the Asian countries are not responsible for brown cloud formation contending that the per capita emissions of India are 1.2 tonnes as against 23 tonnes in the US and 10 tonnes in European countries (The Economic Times 2008). But scientists say the effects of climate change are only getting worse. Climate change and global warming are significant challenges that the world, particularly the developing and least developed countries, face this century. Nado Rinchhen, Deputy Minister, Royal Government of Bhutan argued that although Least Developed Countries (LDCs) like Bhutan contribute the least to global warming, they will nonetheless be seriously affected by the impacts of climate change. It is important to realize that climate change is not just an environmental problem but a serious challenge to sustainable development and the livelihood of the Bhutanese people (Bhutan NAPA 2007). Hindu-Kush Himalayan region is critically vulnerableto poor people in rural and remote areas tend to be the most directly affected by the decline of biodiversity and deterioration or loss of ecosystem services due to the climate change (UNEP 2007). Brown cloud is primarily responsible for total cooling effect by 40%. Summer fires could contribute significantly to the melting of glaciers and outburst of the glacier lakes (GLOF) (UNEP 2007). Impact of brown cloud on monsoonal climate, glaciations and socio-economy over south Asia The South Asian region is characterized by diverse ecosystems and socioeconomic and cultural settings resulting from a wide range of land-use systems and climatic conditions. Consequently these ecosystems have diverse fire regimes and vulnerabilities. These so called brown clouds have a major impact on air quality, human health (lung, heart and eye diseases), regional climate, and natural resources. Due to this haze many domestic flights in Nepal have been cancelled and some international air flights have been delayed in Nepal these days. Brown cloud-induced dimming is considered as the major causal factor for the change of monsoonal rainfall pattern (intensity, duration and frequency). For instance, rainfall over the northern half of India has decreased. The number of rainy days for all India is also decreasing, although the frequency of intense rainfall is increasing, leading to more frequent floods in downstream and soil erosions, landslides and debris-flows in upstream areas. The densely populated areas in the region are highly vulnerable from water-induced disasters. Rainfall over the Indo-Gangetic Plain has decreased by about 20 per cent since the 1980s

163 June 7-8, 2011 (Ramanathan et al. 2008). According to the UNEP report, the toxic material could kill 340,000 people in China and India every year. The toxic clouds also threaten the massive Hindu Kush-Himalaya-Tibetan glaciers. CONCLUSIONS AND RECOMMENDATIONS Wildfires in high altitude Hindu Kush-Himalayas (HKH) ecosystems are a major driver for destruction of pristine biodiversity, including the habitats of many rare species. However, specific study on impacts of wildfires at genetic, species and ecosystem levels in HKH region is poorly understood. Occurrences and extents of wildfires in the countries in South Asia are increasingly impacting socio-economy and environment and are contributing to regional and global climate change. Fires affecting sensitive mountain ecosystems have considerable consequences on secondary disasters such as landslides, mudslides, erosion, increased water runoff, and flash floods. It has been recognized that the involvement of local communities, which are suffering most by the consequences of inappropriate burning practices and wildfires, is crucial to reduce the adverse impacts of fire. But, a lack of fire research and management capability exists in the region, including monitoring, early warning, early detection and ecological and socioeconomic impact assessment. Therefore, wildland fire management involving local communities (CBFiM approach) at Forest User s Group level could be a best strategy to: increase resilience of community to wildfire disaster and its secondary consequences like secondary disasters (soil erosion, landslides, flooding, regional climate change due to the Atmospheric Brown Cloud, etc.) establish linkages of ecosystem services with wildland fires management link wildfire risk reduction with sustainable livelihoods and development With much of the high altitude Himalayan region currently in the grip of a long and intense dry seasons occurring annually, the role of under-lying moisture deficits in the fluctuating nature of south Asia s experience with wildfire is notable, many of them having a potential to cause major damages; e.g., serious degradation of forests, changes of ecosystem properties, and deterioration of social and economic conditions in some land-use systems and natural vegetation types. Equally perhaps, international scientific study has confirmed that

164 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 significant climate change is under way. Atmospheric Brown Clouds (ABC) could be the major driver to the regional climate change and pollution though not well understood (Ramanathan et al. 2008, Gustafsson et al. 2009). Consequently, rainfall over the Indo-Gangetic plain has decreased by about 20 per cent since the 1980s negatively affecting to the agricultural production, changing glacier dynamics and wildfire regimes (GFMC 2009, Gustafsson et al. 2009, Ramanathan et al. 2008,). Climate change threats mainly to the largely agrarian population that still depends on subsistence agriculture for their daily livelihood in the region. It s net positive feedback to the regional climate shape a vicious circle with increasing occurrence of wildfires in the region. Observations indicate that the occurrence of wildfires is increasing as a consequence of regional warming and extended dry spells. Besides the role of climate extremes (droughts) as an aggravating condition for destructive fires, the majority of fires that are resulting in degradation of forests and other vegetation, are caused by human activities in the rural area. Since local communities will benefit first from reduced occurrence and severity of fires participatory approaches in fire management (community-based fire management) are imperative. Vegetation destruction by fire is a cross-sectoral issue and disaster risk affecting human health, security and livelihood in many countries the region, and contribute to destabilization of land cover and to climate change, policies addressing the pressing fire problems are needed at national to international levels. Therefore, it has been recommended: to enhance regional cooperation and collaboration in wildland fire research and development, particularly, among SAARC, ICIMOD and UNISDR-Regional Wildland Fire Network. to conduct further assessments of atmospheric brown cloud (ABC) including haze from wildland fires, as well as options for action to enhance and strengthen bilateral/multilateral and international cooperation in wildland fire management for creating synergies and sharing knowledge and technical and human resources among countries in the region by accepting and promoting principles, norms, rules, and decision making procedures within a guiding framework that individual countries agree on; to strengthen local communities coping with wildfires and aiding them in addressing the consequences of climate change and wildfires and the effects on their livelihoods; to assist countries in fire management planning, research and development,

165 June 7-8, 2011 enhancing institutional and technological capabilities and developing synergies through coordinated and collective action both within the region and internationally; to emphasize the improvement of participatory/community-based fire management approaches and institutional and technological capabilities at all levels; and to promote education and awareness-raising programmes on wildfire prevention. REFERENCES Bhutan NAPA (2007). Bhutan National Adaptation Programme of Action (NAPA), National Environment Commission, Royal Government of Bhutan/ United Nations Development Programme (UNDP)/ Global Environment Facility (GEF) (2007). Brawn, J.D., Robinson, S.K., and Thompson, F.R. (2001). The Role of Disturbance in the Ecology and Conservation of Birds, Annual Review of Ecology and Systematics 32: Cathryn, H., Greenberg, C.H, Otis, D.L., and Waldrop, T.A. (2006). Response of white-footed mice (Peromyscus leucopus) to fire and fire surrogate fuel reduction treatments in a southern Appalachian hardwood forest, Forest Ecology and Management 234: D Antonio, C.M., and Vitousek, P.M. (1992). Biological invasions by exotic grasses, the grass/fire cycle, and global change, Annu. Rev. Ecol. Syst. 23: Economic Times (2008). Asian countries not responsible for haze formation: Sibal: _not_responsible_for_haze_formation_sibal/articleshow/ cms ENN (2008). Tibetan glaciers rapidly melting, published by in 25 November GFMC (2009). Draft UN White Paper Vegetation Fires and Global Change presented to the United Nations Global Platform for Disaster Risk Reduction in Geneva, The Global Fire Monitoring Center (GFMC), Freiburg, Germany _gfmc.htm GFMC (2010). Global Wildland Fire Fatality Report 2009, Global Fire Monitoring Center (GFMC), Freiburg, Germany. Goldammer, J.G. (2007). Wildland Fires: Global and South Asian Perspectives on International Cooperation. In: National Round Table (RT) for Development

166 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 of Strategy for Wildland Fire Management in Nepal - Proceedings of Round Table meeting, Kanel, K.R., Shrestha, K.R., Acharya, K.P., Pachhai, I.B. and Sharma, S.P. (eds), 16 December 2007, Kathmandu, Nepal (in press). Goldammer, J.G. (2006). History of Equatorial vegetation fires and fire research in southeast Asia before the episode: a reconstruction of creeping environmental changes. Mitig. Adapt. Strat. Glob. Change 12: Goldammer, J.G. (1999). Forests on Fire. Science 284: Goldammer, J.G. (1993). Historical Biogeography of Fire: Tropical and Subtropical. In: Fire in the Environment: The Ecological, Atmospheric, and Climatic Importance of Vegetation Fires, P.J. Crutzen and J.G. Goldammer (eds.), John Wiley and Sons Ltd, England, ES 13: Goldammer, J.G. (1988). Rural land-use and wildland fires in the tropics, Agroforestry Systems, 6: Goudsblom, J. (2008). Fire: A Socio-Ecological and Historical Survey. A lecture note on Hot Topics and Burning Issues: Fire as a Driver of System Processes - Past, Present, and Future, 30 March 5 April 2008, Wageningen, The Netherlands. Green Facts (accessed in 1 May 2011.) Hayashida, F.M. (2005). Archaeology, Ecological History, and Conservation, The Annual Review of Anthropology 34: 43-65, Department of Anthropology, Pennsylvania State University, University Park, Pennsylvania. Gustafsson, Ö., M. Krusa, Z. Zencak, R.J. Sheesley, L. Granat, E. Engström, P.S. Praveen, P.S.P. Rao, C. Leck, and H. Rodhe (2009). Brown Clouds over South Asia: Biomass or Fossil Fuel Combustion? Science 23: DOI: /science ICIMOD web portal on Atmospheric Brown Cloud : org/?page=152, accessed in 23 September Kiss, L. and Magnin, F (2006). High resilience of Mediterranean land snail communities to wildfires, Biodiversity and Conservation 15: Kraus, P.D. and Goldammer, J.G. (2007). Fire Regimes and Ecosystems: An Overview of Fire Ecology in Tropical Ecosystems. In: Forest Fires in India Workshop Proceedings, Ashoka Trust for Research in Ecology and Environment (ATREE), (India), Institute of Silviculture, Freiburg University (Germany), Madurai, India, February. Liddle, D.T., Brook, B.W., Matthews, J., Taylor, S.M., and Caley, P. (2006). Threat and response: A decade of decline in a regionally endangered rainforest palm affected by fire and introduced animals, Science Direct, Biological Conservation 123: Monadjem, A. (1999). Population dynamics of Mus minutoides and Steatomys

167 June 7-8, 2011 pratensis (Muridae: Rodentia) in a subtropical grassland in Swaziland, Afr.J. Ecol. 37: NASA s Earth Observatory Web site: Hazards/view.php?id=35888, accessed in 14 November NASA s Earth Observatory Web site: Hazards/view.php?id=35917, accessed in 25 November 2008 NBS (2002). Nepal Biodiversity Strategy, Government of Nepal, Ministry of Forests and Soil Conservation. Nepal NAPA (2010). Nepal National Adaptation Programme of Action (NAPA), Climate Induced Disaster Thematic Working Group Report (Draft), For National Adaptation Programs of Action, 2010 News post online (2008). Thick clouds of pollution block sunlight in world s major cities, published by in 26 November NRBWCDR (2005). National Report on Bhutan for World Conference on Disaster Reduction (NRBWCDR), Kobe, Japan, January 2005, Department of Local Governance, Ministry of Home and Cultural Affairs, Bhutan (2005). Pfab, M.F., and Witkowski, T.F. (1999). Fire survival of the Critically Endangered succulent, Euphorbia clivicola R.A. Dyer fire-avoider or fire-tolerant?, Afr. J. Ecol. 37: Pyne, S.J. (1993). Keeper of the Flame: A Survey of Anthropogenic Fire. In: Fire in the Environment: The Ecological, Atmospheric, and Climatic Importance of Vegetation Fires, Ed: P.J. Crutzen and J.G. Goldammer, John Wiley and Sons Ltd. Reilly, L.O., Ogada, D., Palmer, T.M., and Keesing, F. (2006). Effects of fire on bird diversity and abundance in an East African savanna, Journal Compilation, East African Wild Life Society, Afr. J. Ecol. 44: Salvatori, V, Egunyu, F., Skidmore, A.K., Leeuw, J., and Gils, H.A.M. (2001). The effects of fire and grazing pressure on vegetation cover and small mammal populations in the Maasai Mara National Reserve, East African Wild Life Society, Afr. J. Ecol. 39: Slik, J.W.F. and Balen, S. V. (2006). Bird community changes in response to single and repeated fires in a lowland tropical rainforest of eastern Borneo, Springer, Biodiversity and Conservation 15: Loomis, J.B. and Caban, A.Z (1998). A willingness-to-pay function for protecting acres of spotted owl habitat from fire, Ecological Economics 25: Ramanathan, V., M. Agrawal, H. Akimoto, M. Aufhammer, S. Devotta, L. Emberson, S.I. Hasnain, M. Iyngararasan, A. Jayaraman, M. Lawrance, T. Nakajima, T. Oki, H. Rodhe, M. Ruchirawat, S.K. Tan, J. Vincent, J.Y. Wang, D. Yang, Y.H. Zhang, H. Autrup, L. Barregard, P. Bonasoni, M. Brauer, B. Brunekreef, G. Carmichael, C.E. Chung, J. Dahe, Y. Feng, S

168 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Fuzzi, T. Gordon, A.K. Gosain, N. Htun, J. Kim, S. Mourato, L. Naeher, P. Navasumrit, B. Ostro, T. Panwar, M.R. Rahman, M.V. Ramana, M. Rupakheti, D. Settachan, A. K. Singh, G. St. Helen, P. V. Tan, P.H. Viet, J. Yinlong, S.C. Yoon, W.-C. Chang, X. Wang, J. Zelikoff, and A. Zhu (2008). Atmospheric Brown Clouds: Regional Assessment Report with Focus on Asia, the United Nations Environment Programme (UNEP), Nairobi, Kenya RSAWFN (2010). Nepal Wildland Fire Damage Records of , UNISDR- Regional South Asia Wildland Fire Network /Global Wildland Fire Network/ Global Fire Monitoring Center (GFMC) SAARC web portal on Hazards and Disasters: Fire report.mht, accessed in 7 January SAARC web portal on Fire Disasters: accessed in 29 April Schmidt-Vogt, D. (1989). Fire in high altitude forest of the Nepal Himalaya. In: Fire in Ecosystem Dynamics, Proceedings of the third international symposium on fire ecology, Freiburg, Germany, Ed: J.G. Goldammer and M.J. Jenkins, pp Sharma, S.P. (2009). Climate Change and Wildland Fire Regimes: South Asia Regional Perspectives, International Symposium on Regional/National Impact of Climate Change on Fire Regimes (Proceedings), Korea Forest Service (KFS)/Global Fire Monitoring Center (GFMC)/Korea Forest Research Institute (KFRI)/Inje University, South Korea. Stoate, C., Morris, R.M., and Wilson, J.D (2001). Cultural ecology of Whitethroat (Sylvia communis) habitat management by farmers: winter in farmland trees and shrubs in Senegambia, Journal of Environmental Management 62: The Economic Times (2008). Nation/Asian_countries_not_responsible_for_haze_formation_Sibal/articles how/ cms, accessed in 25 November Trollope, W. Govender, N., Chris de B Austin, C.B.,Lynne Trollope, L., Kirkman, K., and Alex Held, A. (2008). Promoting Savanna Heterogeneity for African wildlife through greater Pyrodiversity, Kruger National Park, South Africa. Uys, R.G., William J. Bond, W.J., and Everson, T.M. (2004). The effect of different fire regimes on plant diversity in southern African grasslands, Science Direct, Biological Conservation 118: UNEP (2007). Global Environment Outlook: environment for development (GEO-4), Summary for Decision Makers, United Nations Environment Programme (UNEP). UN News Center and The Global Fire Monitoring Center (GFMC) s Recent

169 June 7-8, 2011 Media Highlights on Fire, Policies, and Politics: A thick haze over Hindu-Kush Himalaya region: Asian Brown Cloud (ABC) over Hindu-Kush Himalaya region: Biomass-burning 'behind Asian brown clouds': Brown Clouds over South Asia: Biomass or Fossil Fuel Combustion?: Climate change fan Nepal s fires : Fires in Nepal: Fires in Nepal: Fires in Nepal: Forest fire a major problem: Planting Trees and Managing Soils to Sequester Carbon: Third-World Stove Soot Is Target in Climate Fight: UN agency urges measures to slash non-carbon dioxide greenhouse gas pollutants: UN-backed clean stove initiative to save lives and heal environment: Wood And Dung Fires Feed Asia's Brown Cloud: B.C., Suring, L.H., and Hemstrom, M.A. (2006). Modeling potential outcomes of fire and fuel management scenarios on the structure of forested habitats in northeast Oregon. Landscape Urban Planning, doi: /j.landurbplan Yeaton; R.I. and Bond, W.J. (1991). Competition between Two Shrub Species: Dispersal Differences and Fire Promote Coexistence, The American Naturalist 138: %29138%3A2%3C328%3ACBTSSD%3E2.0.CO%3B2-N

170 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Assessment of the Crown Fire Hazard of Pinus densiflora based on the Crown Fuel Characteristics in Korea Sung Yong Kim¹, Yeon Ok Seo¹, Mi Na Jang¹, Gun Ik Lee¹, Roscinto Ian C. Lumbres¹, Byungdoo Lee², Kyongha Kim², Young Jin Lee¹ * ¹ Kongju National University, Yesan , Korea ² Korea Forest Research Institute, Seoul , Korea Abstract The objective of this study was to assess the crown fire hazard of Pinus densiflora forests in Korea based on its crown fuel characteristics (moisture content, crown base height and crown bulk density). The study sites were located in five areas, namely Youngju, Bonghwa, Yangyang, Daegu and Goryeong. Ten representative sample trees were destructively felled in each area to analyze the crown fuel characteristics. One focus of this study was to compare the crown fuel characteristics of the stands with forest tending works (Youngju) and the control stand (Bonghwa). Furthermore, comparison between the crown fuel characteristics of Pinus densiflora found in Gangwon province (Gangwon type) and Central region (Jungbu type) was conducted. Crown fuel characteristics based on the results in the five study areas were analyzed and allometric equations for crown fuel load estimation were also developed. These information will be used in the development of crown fire hazard map using Geographic Information System (GIS). Introduction Pure stand forests in Korea are widely distributed. These stands are very dense because thinning operations are not being done in these forests. Therefore, forest fire hazard is very high and abundant. As a result, burned and damaged properties, human casualties and forest ecosystem degradation caused by serious wild fire are gradually increasing in Korea. The distributions of Pinus densiflora stands in Korea are approximately 25% of the total forested area. However, these forests are threatened to be extinct (Korea Forest Service, 2009). One of the major causes is the abundance of forest

171 June 7-8, 2011 fire. The needles and branches of this type of forest are characterized of being flammable. Studies on the crown fuel characteristics of Pinus densiflora are limited and insufficient. Therefore, study on crown fuel characteristics is urgently needed to suppress the crown fires that are frequently occurring in Pinus densiflora stands. Thus, the objective of this study was to assess the crown fire hazard of Pinus densiflora forests in Korea based on the crown fuel characteristics (moisture content, crown base height and crown bulk density). Methodology The sample trees were cut from 0.2m above the ground for each sampled tree. Other measurements that were collected were tree total height, crown base height, crown width, crown length, diameter at breast height (DBH) and age. The biomass were categorized into stems, needles, branches (<0.5cm, 0.5-1cm, 1-2cm, 2-4cm, and >4cm diameter) and the green weight of these fuel components were also measured. This method was used by the different scientist on their studies (Koo et al., 2010). The moisture content of each component was analyzed using oven drying method. Samples of stems, needles and branches were dried at a temperature of 95 C for 240h and were weighed after. In this study, the available crown fuel loads were composed of needles and fine branches less than 1cm roundwood diameter (Call and Albini, 1997; Koo et al., 2010). Crown bulk density, moisture content and crown base height are important crown fuel characteristics that are needed in order to predict crown fire spread (Reinhardt et al., 2000; Koo et al., 2010). The crown bulk density (CBD) was calculated as follows: CBD kg/m CFL kg CV m CFL is crown fuel load(kg) and CV is crown volume(m³) The allometric equations for each crown fuel load component were analyzed using the measurements of fuel in the five regions

172 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Results and Discussion 1) Effects of Forest Tending Works on the Crown Fuel Characteristics The crown fuel load distribution in Youngju region (with forest tending works) were 50.3% for crown fuel(needles and >1cm branches), whereas in Bonghwa region (control), it were 62%. The average canopy bulk density observed in Youngju region was 0.19kg/m³, whereas in Bonghwa region, it was 0.30kg/m³. On the other hand, for the average crown base height, it was 5.0m in Youngju region and 3.1m in Bonghwa region. 2) Comparison of the crown fuel characteristics of Pinus densiflora in Gangwon type and Jungbu type It was observed that the crown fuel load of the Pinus densiflora in Gangwon type and Jungbu type were almost similar for each DBH class. However, the average CBD of Pinus densiflora in Gangwon type was higher than the average CBD of Jungbu type because the CV of the Gangwon type trees were lower. 3) Overall crown fuel characteristics of Pinus densiflora in Korea. Crown fuel load distributions of available crown fuel load of Pinus densiflora in Korea were 54%. Furthermore, the crown bulk density of this species was 0.37kg/m³. On the other hand, the crown bulk density of available crown fuel load was 0.2 kg/m³. 4) Allometric equations of the crown fuel load of Pinus densiflora in Korea. Allometric equations were developed based from the 50 sample trees to estimate crown fuel load components of Pinus densiflora in Korea. It was observed that the R values of the developed allometric equations were very high. Acknowledgements This study was carried out with the support of the (Estimation of crown fuel load on forest fire damage in Pinus densiflora stand) project provided by the Korea Forest Research Institute

173 June 7-8, 2011 References Koo, K.S., Lee, B.D., Won, M.S. and Lee, M.B Crown fuel characteristics of Japanese Red pine (Pinus densiflora) in Mt Palgong, Daegu. Journal of Korean Forest Society. 99(1): Call, P.T. and Albini, F.A Aerial and surface fuel consumption in crown fires. International Journal of Wildland Fire 7(3): Reinhardt, E.D., Keane, R.E., Scott, J.H. and Brown, J.K Quantification of canopy fuels in conifer forests: assessing crown fuel characteristics using destructive and nondestructive methods. Study plan on file at: U.S. Department of Agriculture, Forest Sevice, Rocky Mountain Research Station, Fire Sciences Lab, Missoula, MT. Korea forest service, Forest statistics annals 49:

174 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Bush Fire Management & Cooperation 산불관리와협력 ( 체제 ) Jae, Jin-joo* University of Seoul, Jeonnong-dong 90, Dongdaemun-gu, Seoul Korea Abstract Forest fires in addition to economic losses of forest resources, degradation of ecosystems, hwangyeongohyeomdeung huge collateral damage of the importance of preventing the cursor not to mention the two times. However, forest fire prevention activities, no matter how thoroughly even accidentally fires can occur naturally or artificially not only in the event of fire quickly and efficiently to minimize the damage evolution in action must be taken to cardiovascular. Therefore, a wildfire that the Forest Service, including the other organization s activities of the faithful and active suppression are required. Had already occurred in 2000 East of the wildfire problem, derived by analyzing the case and looking for ways to improve its fire suppression and efficient to the following suggestions to 1. Faithful self-role integrated command structure, even within the respective work on the actual mutual recognition of each institution by Each agency in performing its duties for the leadership of directors fulfill any integration 2. Fire suppression system, improved organizational Fire suppression equipment permanent installations, such as Australia RFS (Rural Fire Service) Take a morale of the firefighter's institution(the stability of employmentrelated) 3. For fire suppression is the sole cause notice tries to "Forest management" 1. 개요 일일대응능력을초과하는대규모산불, 24 시간이상되는산불또는동시다발의산불발생시산림청을포함한산불진압유관기관간유기적인협력체제

175 June 7-8, 2011 구축으로산불로인한인적 물적피해를최소화하는방안을제시하기위하여, 먼저 2000 년 4 월동해안산불사례로부터문제점을도출하고, 효율적인방안을강구해보고자한다. 2. 산불사례 2-1 발생개요 :30 고성군현내면 3GP 북방한계선산불을시작으로고성토성, 강릉사천, 삼척근덕으로이어지는동시다발산불이이어서, 삼척원덕및미로, 고성거진, 동해시일원, 그리고삼척도계지역에서추가발생되면서강원도동부지역전체가대형산불로휩싸였고, 결국대형재난지역으로선포됨. 2-2 산불피해 < 피해액 : 총 1,072 억원 > 산불발생면적 : 23,138ha (4. 24 항공촬영 ) - 고성 : 2,696 (12%), 강릉 : 1,447 (6%), 동해 : 2,244 (10%), 삼척 : 16,751 (72%) 인명 : 17 명 ( 사망 2, 부상 15) - 사망 : 강릉, 삼척각 1 명, - 부상 : 강릉 4 명, 동해 2 명, 삼척 8 명, 고성 1 명 이재민 : 299 세대 850 명 - 고성 : 70 세대 206 명, 강릉 : 159 세대 465 명, 동해 : 16 세대 47 명, 삼척 : 54 세대 132 명 재산피해 - 건축물 : 814 동 ( 주택 390, 부속사 283, 축사 129, 공장 6, 기타 6) - 가축 : 6,601 마리 ( 소 31, 돼지 259, 닭 1,869 염소 166, 개 133, 꿀벌 3,540, 사슴 80, 기타 523) - 영농자재 : 농기계 (1,328 대 ), 종자, 비닐하우스등 6 종 - 기타 : 송이 (31 톤 ), 장뇌 (134 천본 ), 공공시설등 2-3 복구비용 < 총복구비 589 억원 > 직접지원 : 30,069 백만원 - 보조 20,849(69%), 의연금 2,818(10%), 융자 5,515(18%), 자부담 887(3%) 이재민구호 : 3,384 백만원 ( 보조 566, 의연금 2,818) 주택복구 : 15,335 백만원 (321 동 ) - 보조 9,495, 융자 5,006, 자부담 834 농업시설 : 4,774 백만원 - 보조 4,720, 융자 451, 자부담 53 수산시설 : 584 백만원 ( 보조 526, 융자 58)

176 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 산림응급복구 : 5,992 백만원 ( 전액보조 ) 산림항구복구비 ( 별도 ) : 188,287 백만원 ( 국비 90% 지방비 10%) 간접지원 : 28,885 백만원 표고시설 : 8,300 ( 보조 2,400, 융자 2,800, 자부담 3,100) 영농자금 ( 융자 ) 10,000, 작목개발 ( 융자 ) 2,000 긴급구조경비 ( 보조 ) 477 공장지원 : 7,601 백만원 ( 융자 5,088, 자부담 2,513) 기타시설 : 507 백만원 ( 보조 249, 융자 258) 2-4 교훈및개선방향 구분문제점개선사항 지휘체계 상호협조 정보교환및소통 각기관의입장차이로통합지위체계가존재함에도지휘체계가일원화되지못함 산불전문가가아닌산불통합지휘자원활한지휘및작전수행미흡 소방의최초각지 ( 覺知 ) 산불발생정보가지방자치단체등수보 ( 受報 ) 기능미흡으로신속전파곤란 산불신고초기부터정보공유미흡으로원활한상호협조가되지못함. 산불진압범위에따른상호불신 ex) 소방은집불진압에만, 산림관련자는산불진압에만관심이있다 산불진화의특성상필요한헬기가각기관장들에의한다른명령으로인하여원활한소통이이루이지못하고있음.( 지휘권과의문제도존재함.) 통제단에유관기관의연락관이상주하지만, 현장간의연락은채널이틀리기때문에교신이힘듬. 헬기간무선주파수상이로통신미흡 명확한산불진화통합지휘체계확립 1) 각기관에대한적극적인홍보 산불에대한전문지식자의양성 산불진압전문가를참모로활용 소방상황실에서각지 ( 覺知 ) 한산불발생정보의신속한전파, 보고등정보공유에대한매뉴얼및시스템개선 정보공유에대한매뉴얼및시스템구축의필요성 협조체제구축외진압책임범위상호인정. 산불진화통합지휘자의지휘통제하에헬기운용 재난시공통주파수활용 1) 산불진압지휘체계의법적근거는산립보호법 ( 법률제 9763 호 )( 제정 , 시행 ) 로정립되었음

177 June 7-8, 산불진압책임등 3-1 관련법령 산림보호법 ( 제 33 조 2), 제 37 조 3), 제 38 조 4) 및제 39 조 5) 재난및안전관리기본법 ( 제 2 조 6), 제 3 조 7), 제 49 조 8), 제 50 조 9), 제 52 조 10) 2) 산림청장은산불의효율적인예방 진화체계를마련하여야하며, 지방자치단체의장과지방산림청장은이에따라산불을예방하고산불이발생하면진화하여야하며, 산불에대비하여산불예방과진화에필요한인력, 장비및예산을확보하는등의조치를하여야한다. 3) 산림청장및국유림관리소장또는지방자치단체의장은산불의규모에따라산불진화를위한통합지휘권을갖고진화하여야한다. 다만, 산불이국유림 공유림또는사유림에걸쳐발생하면특별자치도지사 시장 군수 구청장이통합지휘하여야한다.( 산림보호법제 37 조 ) 4) 산불진화를통합지휘하는지방자치단체장및국유림관리소장은산불이발생한현장에산불현장통합지휘본부를설치하여산불진화를지휘하고진화대원에게진화에필요한명령을할수있고, 산불유관기관의관계관을소집하여산불현장대책회의를개최하고기관별임무를부여하여야한다.( 산림보호법제 37 조 ) 5) 산불현장통합지휘본부장은산불진화와관련하여필요하다고인정되면소방관서, 경찰관서, 군부대및대통령령으로정하는산림관련기관및단체에의장에게산불진화, 현장통제등에필요한장비및인력의협조를요청할수있으며, 요청을받은기관및단체의장은특별한사유가없으면이에적극협조하여야한다.( 산림보호법제 39 조 ) 6) 재난을예방하고재난이발생한경우그피해를최소화하는것이국가와지방자치단체의기본적의무임을확인하고, 모든국민과국가ㆍ지방자치단체가국민의생명및신체의안전과재산보호에관련된행위를할때에는안전을우선적으로고려함으로써국민이재난으로부터안전한사회에서생활할수있도록함을기본이념으로한다. 7) 산림에서의화재도재난의하나로서관리되고있다.( 재난및안전관리기본법제 3 조 ) 8) 긴급구조에관한사항의총괄ㆍ조정, 긴급구조기관및긴급구조지원기관이하는긴급구조활동의역할분담과지휘ㆍ통제를위하여소방방재청에중앙긴급구조통제단 ( 이하 " 중앙통제단 " 이라한다 ) 을두고, 소방방재청장이단장이되며, 중앙통제단장은긴급구조를위하여필요하면긴급구조지원기관간의공조체제를유지하기위하여관계기관ㆍ단체의장에게소속직원의파견을요청할수있다 ( 요청을받은기관ㆍ단체의장은특별한사유가없으면요청에따라야한다.) 그리고중앙통제단의구성ㆍ기능및운영에필요한사항은대통령령으로정한다.( 재난및안전관리기본법제 49 조 ) 9) 지역별긴급구조에관한사항의총괄ㆍ조정, 해당지역에소재하는긴급구조기관및긴급구조지원기관간의역할분담과재난현장에서의지휘ㆍ통제를위하여지역통제단장을두며지역통제단에는각각단장 1 명을두되, 지역통제단장은긴급구조를위하여필요하면긴급구조지원기관간의공조체제를유지하기위하여관계기관ㆍ단체의장에게소속직원의파견을요청할수있고 ( 이경우요청을받은기관ㆍ단체의장은특별한사유가없으면요청에따라야한다.) 지역통제단의기능과운영에관한사항은대통령령으로정한다.( 재난및안전관리기본법제 50 조 ) 10) 재난현장에서는지역긴급구조통제단장이긴급구조활동을지휘 ( 치안활동과관련된사항은관할경찰관서의장과협의 ) 하며재난현장에서인명의탐색ㆍ구조, 긴급구조기관및긴급구조지원기관의인력ㆍ장비의배치와운용, 추가재난의방지를위한응급조치, 긴급구조지원기관및자원봉사자등에대한임무의부여, 사상자의응급처치및의료기관으로의이송, 긴

178 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 소방기본법 11) ( 제 2 조 ) 3-2. 기타규칙, 예규등 긴급구조대응활동및현장지휘에관한규칙 ( 소방방재청 ) 현장통합지휘본부장임무매뉴얼 - 산림청 산불현장통합지휘지침 - 산림청 산불진화기관의임무와역할에관한규정 ( 공동예규 ) - 산림청 3-3. 산불진압전담및유관기관 산림청및지방산림청 지방자치단체 ( 시 도및시 군 구 ) 긴급구조기관 ( 소방방재청, 소방본부및소방서 ) 긴급구조지원기관 ( 첨부 2) 3-4. 통합지휘본부설치산불형태 대형산불이나일일대응능력을초과하는동시다발산불로인한재난성산불 12) 발생시산불진화및이재민발생, 고립지역구호등재난상황 4. 산불현장지휘 4-1. 지휘절차 지휘권은산림보호법및관련규정에서정한절차에따른다. 산불현장의현장통합지휘는산림보호법제 37 조및같은법시행령제 25 조 급구조에필요한물자의관리, 현장접근통제, 현장주변의교통정리, 그밖에긴급구조활동을효율적으로하기위하여필요한사항에관하여여지휘한다. 그리고중앙통제단장은대통령령으로정하는대규모재난이발생하거나그밖에필요하다고인정하거나시ㆍ도긴급구조통제단장은필요하다고인정하면직접현장지휘를할수있으며재난현장에서긴급구조활동을하는긴급구조요원은현장지휘를하는각급통제단장의지휘ㆍ통제에따라야하며통제단장은재난현장의긴급구조등현장지휘를효과적으로하기위하여재난현장에현장지휘소를설치ㆍ운영할수있으며, 긴급구조활동에참여하는긴급구조지원기관의통합지휘본부장은현장지휘소에대통령령으로정하는바에따라연락관을파견하여야한다.( 재난및안전관리기본법제52조 ) 11) 산림은소방대상물의하나로서화재를예방, 경계, 진압하여야하는행정대상물이다.( 소방기본법제2조 ) 12) 재난성산불이란산불로인하여다수의인명사고와가옥피해가발생하고산림생태계에심각한영향을주는산불을말한다. ( 산불재난 위기관리표준매뉴얼. 행정안전부 산림청 )

179 June 7-8, 2011 의규정에따른다. - 산불현장에출동한유관기관 ( 소방, 경찰, 군부대등 ) 의선임지휘자는통합지휘본부장의지휘에따라야하며통합지휘본부에연락관을상주시켜야한다. 지휘권은공식적인절차를통해서이양한다. 지휘권이양시변경내용은현장대응요원전체와관계기관에게전파 통합지휘권은자치단체장및국유림관리소장으로일원화하며산불의규모 복잡성등을고려하여임무 역할을위임할수있다. - 위임받은임무 역할은단일한계통으로행사되도록한다. - 임무 역할을위임받은자는소관사항에대해서만임무를수행하여야하고, 다른임무를부여받은자의소관사항에서오류가발견되면통합지휘본부장또는다른임무를부여받은자를통해서시정하도록해야한다. 현장지휘는공식지휘계통인산림과장 - 시장 군수 - 시 도지사에의해이행되고, 이외의간부는도착순서와담당업무를고려하여보좌또는부여받은임무를수행 지원하여야하고부여된임무외다른임무에관여할수없다. 현장도착간부는현장통합지휘본부장에의해임무지정을받을때까지는산불현장통합지휘본부에대기하여야하며, 승인없이이탈할수없다 산불규모별현장지휘권자 산불규모 설명 산불현장통합지휘본부장 ( 현장지휘관 ) 통합지휘본부장보좌 중소형산불 관할지역에발생한중소형산불 특별자치도지사또는시장ㆍ군수ㆍ구청장또는위임받은자 단, 국유림발생산불은국유림관리소장이담당 시 군산림부서장 보호담당주무 대형산불 대형산불 13) 또는둘이상의시 군 구에발생한산불또는동시다발산불 시 도지사또는위임받은자 시 도산림부서장 그외둘이상의시 도에발생한산불 산림청장또는위임받은자 지방산림청장 * 산림보호법제 37 조및동법시행령 25 조수정재구성 13) 산림피해면적이 100 만제곱미터이상또는 24 시간이상지속되는산불 ( 산림보호법제 25 조 )

180 International Forest Fire Symposium on Commemorating the International Year of Forests 기관간대응절차 5-1. 산불시전체대응업무및관련기관 기관 기관별상황실 ( 시도상황실 ) 산불현장통합지휘권자 ( 지방자치단체장및국유림관리소장 ) 지방산림청 / 지방자치단체 단계 상황각지 (119 수보 ) 중소형 대형 중소형 대형 야간산불진행 대응단계 임무 신고접수및유관기관전파, 상급기관보고 ( 관할소방서출동명령 ) 현장지휘권인계 ( 산림부서장또는산림담당직원 시장 군수 구청장 ) 산불현장통합지휘본부장설치 운영 기관별상황실및산림청보고 현장지휘권인계 ( 시장 군수 구청장 시 도지사 ) 기관별상황실및산림청보고 진화계획수립 - 산불진행상황, 기상 지형 임상등고려 진화조편성및현장배치 현장상황을수시로파악하여대응조치강구 필요시진화자원 ( 헬기, 인력 ) 추가지원요청 현장대책본부설치및대책회의개최 - 진화계획, 진화반편성, 보급방안등토의 추가동원자원현장배치 ( 현장상황수시파악, 대응조치강구 ) 보도자료배포및대언론브리핑 보급및유조차지원방안강구 진화자원추가요청및진화조직재편성 ( 대형산불확대우려시 ) 현장대책본부설치및대책회의개최 - 진화계획, 진화반편성, 보급방안등토의 진화단재편성및현장배치 ( 상황수시파악, 대응조치강구 ) 보도자료배포및대언론브리핑 야간진화계획수립 진화조재편성및현장배치 ( 진화곤란시감시인력배치 ) 보도자료배포및대언론브리핑 현장상황및진화대원안전수시확인 ( 진화대장, 조장활용 )

181 June 7-8, 2011 산항본부 중소형 대형 헬기출동준비및이륙요청 투입예정진화헬기의파악및투입조치 헬기급유지원체계확인및계류장선정 다기관헬기진화시에는공중지휘기운영 대형산불발생시에공조체계강화를위하여현장대책본부에공중진화반배속 소방현장지휘자 소방대 경찰 의료기관 전기 / 통신 군부대 중소형 현장지휘소설치 ( 가급적통합지휘본부내에설치 ) 시도상황실보고및전파 선착현장지휘자산하소방대원지휘권선언 ( 산불현장통합지휘본부장도착시지시에따른다.) 상황파악및소방대출동 ( 각지사항보고, 전파 ) 기상상태등을감안, 헬기지원요청검토 대형 소방업무현장지휘권인계 ( 소방서장 시도지사 ) 중소형 현장상황수시파악및안전한퇴로확보 진화작전및진화계획구상 현장집결지통보 ( 출발지가다른경우 ) 차량은접근 퇴출이용이한곳주차 ( 필요시주차요원배치 ) 간이현장지휘본부설치 ( 진화지휘차량 ) - 산불발생원인조사및가해자검거반운영 ( 경찰협조 ) 산불규모및진행상황보고 ( 대책본부및유관기관 ) 화재진압, 요구조자발생시구조, 구급 ( 응급처치 ), 이송및응급조치 현장도착후통합지휘본부장보고, 현장통제, 증거확보 출동대기 현장도착후통합지휘본부장보고, 전기 / 통신차단 산불진압대지원 상기나열된대응업무는순서와무관함

182 International Forest Fire Symposium on Commemorating the International Year of Forests 산불대응절차 가. 신고접수ㆍ지령전파단계 신고접수 : 119 상황실또는각행정기관 - 신고접수한 119 상황실또는각행정기관은시장 군수 구청장과지방산림청및소방청상황실에보고 전파 - 그밖의관계행정기관이수보했을경우는산림청 긴급구조기관 ( 시도상황실 ) 유관기관등에게통보. 14) 출동지령 - 산화발생상황을전파받은시장 군수 구청장과지방산림청장은산화진화대출동지령및유관기관의지원가능자원파악 - 시도상황실은관할소방서에소방대출동지령 유관기관등에상황전파 - 전파대상 : 군부대, 지방경찰청, 한국전력공사, 한국가스공사, 한국가스안전공사, 한국전력공사, 한국전기안전공사, 한국통신 (KT), 상수도사업본부, 응급의료센터, 응급의료정보센터 (1339) 등 나. 현장대응 산불현장에선착한공무원이최초의지휘관이되며, 통합지휘본부장은현장통합지휘본부를설치하고출동산불진화대원및관계기관에게알림. 시 군산불현장통합지휘본부설치 - 다수기관및지자체가참여시산불통합지휘권자지휘하에현장의참여한기관들이함께통합대응을위한산불현장대책회의를개최하여기관별임무를부여, 산불진화계획등을수립함. - 필요시해당지역주민이나그지역안에있는사람을대피시킴. 15) - 통합지휘권자는대응요원의급식, 휴식, 샤워시설등기본처우를마련해야하며, 이에필요한물품의지원등제반사항은해당지자체에서수행하여야함. - 현장참여유관기관이추가산불진화와관련하여진화자원이필요하다고인정되면통합지휘본부장은각소방관서, 경찰관서, 군부대등유관기관에산불진화, 현장통제등에필요한장비및인력의협조를요청할수있으며요청을받은기관및단체의장은특별한사유가없으면이에적극협조하여야함 ( 소속기관진화자원이투입한경우에는그사실을통합지휘본부장에게알려통합적자원관리가되도록해야하며, 통합지 14) 15) 재난및안전관리기본법제18조 2항재난및안전관리기본법제40조 1항

183 June 7-8, 2011 휘본부장은경찰에게동사실을전달하여신속히현장에자원이도착할수있도록협조요청하여야한다.) ( 소방 ) 현장지휘소설치 16) - 소방현장지휘관은가시적인지휘위치에차량을위치시켜최대한사고현장을볼수있도록한다. 예하인원에게위치를통보한다. - 소방현장지휘관은상황실에지시사항을보고하고필요시상호교신을한다. 지휘명령권은정식으로위임되거나위임된것으로가정되며그렇지않은경우혼란이야기된다. - 현장지휘소조명을설치하고통신장비들을설치하며가능하다면휴대용테이블과지도, 경방계획도를설치한다. - 소방지휘관 ( 소방연락관 ) 은산불현장통합지휘본부에고정근무를하며통합지휘본부장의지휘에참여하고소관사항에관한사항은수집보고 전달한다. 상급기관보고 ( 산불상황보고및전파 ) - 통합지휘본부장은산불진화상황을시 도, 산림청, 유관기관에보고 전파하여야한다. - 소방현장지휘관은산불이발생한경우에시도상황실 ( 통제단장 ) 에게보고하여야함. 17) - 지원이필요한때시ㆍ도긴급구조통제단장 ( 시도상황실장 ) 은중앙통제단장 ( 소방청상황실장 ) 에게각각보고하여산불진압이신속히이루어질수있도록하여야함. 18) - 지역통제단장 ( 시도소방본부장 ) 은응급조치, 대피, 통행제한, 응원요청, 응급부담등의응급대책을실시하였을때는즉시시ㆍ도지사에게보고하여야함. 19) 지자체산림부서는통합지휘본부장으로부터상황정보확인및적정산화진화단투입 - 현장출동시출동한다는것과도착즉시출동인력, 보유장비등을통합지휘본부장에게통보하고통합지휘본부장은산불통합지휘본부설치시까지현장도착자원을통합지휘함 - 소방현장지휘관으로부터산불현장의안전상태파악 현장접근통제실시 20) 16) 17) 18) 19) 20) 재난및안전관리기본법제52조 6항긴급구조대응활동및현장지휘에관한규칙제4조 1항긴급구조대응활동및현장지휘에관한규칙제4조 3항재난및안전관리기본법제48조 1항산불및안전관리기본법제43조 1항

184 International Forest Fire Symposium on Commemorating the International Year of Forests 소방은소방활동구역을위해제 1 통제선을설치하고, - 경찰은화재진압및현장구조활동관계자이외의접근을통제하기위해제 2 통제선을설치함. 21) 경찰은현장통제, 교통통제, 치안등에필요시인력추가투입 한전, KT 등긴급구조지원기관은전화, 통신장비, 전기조명시설, 전력등현장대응에필요한물자지원 다. 종료단계 뒷불감시 인원, 장비점검 ( 지방 ) 산림청, 지방자치단체및소방, 경찰은사고원인조사 지휘권이양 - 복구자료인계 - 지자체사고수습주무부서는산불피해결정및이재민지원방안강구 - 사고원인에따라보상 배상 구상권청구 6. 효율적산림보호를위한제언 6-1. 자기역할을충실히 통합지휘체계안에서라도각각개별법에의한각기관의업무상호인정 각기관자체임무수행시라도통합지휘자의지휘내용충실히이행 6-2. 산불진압조직체계개선 호주의 RFS 22) 같은상설산불진압기구설치. ( 과거의영림서부활또는가칭산불소방서개청 ) 산불진화단원 ( 산림보호법제41조 ) 의사기진작을위한제도강구 ( 고용의안정성관련 ) 6-3. 산불진압은영림 ( 營林 ), 육림 ( 育林 ) 의한가지라는대의명분공고 ( 鞏固 ) 23) 21) 22) 23) 긴급구조대응활동및현장지휘에관한규칙제17조 1항 Rural Fire Service는각주, 각시마다철치되어있음 ( 국내에서는주는산불소방청, 시는산불소방서라번역, 사용하고있음 ) 공고 ( 鞏固 ) : ( 의지따위가 ) 굳고흔들림없다

185 June 7-8, 2011 첨부 1. 산불진압을위한기관별역할과대응절차 산불진압가. 기능 산불우선진화순위결정및신속, 적합한산불진화대출동조치나. 관련기관 산림청, 지방자치단체, 소방, 군부대, 경찰, 기상청다. 기관간대응절차 < 산림청 ( 주관기관 ) 산불방지과 > 진화우선순위판단및조치 산불진화대책회의개최 산불현장대책지원반파견조치파견및지휘자문 산불중앙사고수습지원본부구성 운영 산불현장통합지휘본부운영및활동사항파악 산불상황접수및전파 산불진화상황파악및언론보도 ( 대변인 ) 및담화문작성준비 잔불진화실시및진화완료상황확인 < 지방자치단체 ( 녹지 산림부서 ) 및지방산림청 ( 운영과 )> 산불발생신고접수 확인및전파 초동조치및산불진화 - 산불현장출동및산불현장통합지휘본부설치 운영 - 현장대책회의개최및진화계획수립 - 헬기지원요청및투입조치 - 산불상황대언론보도 진화상황보고 / 전파 - 헬기의추가투입여부판단및요청 - 유관기관진화자원추가지원요청 - 산불확산에따른진화대책수립 / 실행 - 적정지역을선정하여진화선구축 < 산림항공관리소 > 산불발생시산림진화헬기지원 다기관헬기진화시에는공중지휘기운영 대형산불발생시에공조체계강화를위하여현장대책본부에공중진화반배속

186 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 < 행정안전부 ( 산불대책과 )> 산불예방 진화에대한지자체장의산불경각심고취및대응태세독려 산불방지를위한지자체의현장대응인력지원 필요시담화문발표와관련된업무협조 < 국방부 ( 산불관리지원과 ) 합참 ( 합동작전과 ) 및군부대 > 필요인력및장비 ( 헬기등 ) 지원 군비행장이용및산림헬기급유, 군급유차량을이용한항공유운송지원 DMZ 내산불진화지원 ( 합참합동작전과 ) 산림헬기와공조체계유지및주관기관요청시산불진화참여 대형산불발생시공조체계강화를위하여현장대책본부에연락관파견 < 소방방재청 ( 방호과 ) 및소방관서 > 산불안전대책본부운영, 상황보고및전파 지자체의산불진화조치지원 산불신고접수시신속전파및대응조치 소방헬기, 소방차및소방인력지원 긴급지원체계의가동 대형산불발생시공조체계강화를위하여현장대책본부에연락관파견 < 기상청 ( 예보정책과 )> 산불관련기상자료제공 기상전문가현장파견자문 < 경찰청 지방경찰청및경찰서 > 진화인력, 경찰헬기의산불진화협조 산불발생지역교통통제 산불가해자또는방화범의검거 대형산불시공조체계강화를위하여현장대책본부에연락관파견. < 국무총리실 ( 안전환경정책관실 )> 상황전파및내각의필요조치시행 담화문문안작성및부처협의와관련된업무협조 < 농림수산식품부 ( 비상계획관실 녹색미래전략과 농산경영과 )> 산불발생상황전파, 진화지원 무단소각행위금지교육 계도 필요시담화문발표와관련된업무협조

187 June 7-8, 2011 < 환경부 ( 국립공원관리공단 )> 국립공원지역에대한산불진화 국립공원인근지역의산불발생에따른헬기등진화자원의지원 < 산림조합 > 진화인력및장비 ( 차량 ) 등의산불진화자원을지원 < 법무부 ( 비상계획관실 )> 필요시담화문발표와관련된업무협조 산불관련사범의신속한법집행및결과통보 홍보 < 보건진료관서 > 부상진화대원응급처치및후송을위하여구급인력및차량을지원 이재민구조, 구호가. 기능 산불로출입이통제되거나불가한고립지역의주민안전을위한유관기관간조치나. 관련기관 산림청, 지방자치단체, 소방, 경찰, 기보유부서, 적십자사등민간단체다. 기관간대응절차 < 지방자치단체 > 이재민구호용주부식, 음료수구호물품등확보 구호물품공급을위한헬기보유부서협조요청 산불진압대원음식물준비및제공 < 헬기보유부서 > 보유부서 : 산림청, 산림항공관리소, 군부대, 소방및경찰항공대 헬기이용요구조주민구조및대피활동 구호물품공급을위한헬기운용 < 소방서 > 대피완료시까지구조대투입음용수공급 ( 헬기활용 ) < 경찰서 > 화재현장통제선설치, 현장및교통통제등 경찰 POLICE LINE 을설치, 기자단및일반인들의위험지역출입통제

188 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 < 적십자사 > 급식차 ( 炊事차량 ) 등사전확보, 유사시제공준비 < 민간자원봉사단체 > 이재민및산불진압등활동자를위한음식등제공활동 헬기및소방차량소화용수확보, 공급을위한기관간대응절차가. 기능 소방차및헬기의소화용수의원활한확보를위한유관기관간조치나. 관련기관 산림청, 지방자치단체, 소방다. 기관간대응절차 < 산림청, 지방자치단체 > 헬기공급용 : 소하천등물막이설치를위한장비및인력동원 소방차공급용 : 소하천등물막이조치및우물, 정호활용을위한가반식펌프등등확보, 협조 풀장에서직접흡수가곤란한헬기의급수장소확보 ( 학교운동장등 ) < 소방서 > 호수, 하천및해양등에접근로확보취수활동 취수차량접근곤란시등가반식펌프활용취수활동 헬기공급소화수용차량배치 < 경찰서 > 물탱크차량신속한통행을위한교통통제 도시지역산불또는시가지연소우려시기관간대응절차가. 기능 도시지역연소로주민안전위협및중요시설물연소방지대책수립나. 관련기관 지방자치단체, 소방, 경찰다. 기관간대응절차 < 소방서 > 도시지역은소방관서에서초동진화를적극지원 도시지역내로의연소저지선확보진화활동

189 June 7-8, 2011 연소우려범위내주민구조및피난명령 민가 시설물보호등대상지역에따른임무및역할분담및산불방지업무지원 < 지방자치단체 > 연소우려범위내주민피난방송및대피유도활동전개 < 경찰서 > 시가지연소우려시교통및출입통제 ( POLICE LINE 설치 ) 치안유지및주민대피령발령에따른주민의보호 첨부 2. 긴급구조지원기관 근거재난및안전관리기본법 ( 제 3 조 8 호 ) 및동법시행령 ( 제 4 조 ) 기관명 교육과학기술부, 국방부, 경찰청, 지식경제부, 보건복지부, 환경부, 국토해양부, 방송통신위원회, 기상청, 산림청, 해양경찰청, 탐색구조부대, 군부대, 대한적십자사, 응급의료기관, 응급의료정보센터, 구급차등의운용자, 전국재해구호협회, 법제 3 조제 7 호의규정에의한긴급구조기관과긴급구조활동에관한응원협정을체결한기관및단체, 긴급구조에필요한인력과장비를갖춘기관및단체로서행정안전부령이정하는기관및단체 근거재난및안전관리기시행령 ( 제 4 조 ) 및동법시행규칙 ( 제 2 조 ) 기관명 유역환경청또는지방환경청, 지방국토관리청, 지방항공청, 보건소, 지하철공사, 도시철도공사, 한국가스공사, 한국가스안전공사, 한국농촌공사, 한국전기안전공사, 한국전력공사, 대한석탄공사, 대한광업진흥공사, 한국수자원공사, 한국도로공사, 한국공항공사, 항만공사, 한국원자력안전기술원, 한국원자력의학원, 국립공원관리공단, 전기통신사업법 제 5 조에따른기간통신사업자로서소방방재청장이정하여고시하는기간통신사업자

190 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Projection of Climate Change over East Asia Based on Global Downscaling with High Resolution AGCM Abstract Jaiho Oh*, Yoo-Lim Chung, Sumin Woo, and Kyoung-Min Lee Dept. of Env. & Atmos. sci., Pukyong National University, Deayeon 3 dong, Nam-gu, Busan, Korea, jhoh@pknu.ac.kr For the regional projection of future climate the dynamic downscaling technique has been used with MM5 based on the MPI-ECHO simulation for the period from 2001 to 2100 simulated based on the SRES A2 CO2 emission scenario. The future projection of warming rate is about 1.65 C higher than the global warming rate and almost three times faster than that shown during the last century (0.61 C/decade) over the Korea Peninsula and East Asia. As a result, the length of the summer period is projected to increase by two weeks, while the winter period is projected to decrease by three weeks in comparison to the last 100 years over the Korea. The most important thing related to global warming is not how much it will warm up but how quickly it will happen. The precipitation shows an inter-decadal fluctuation instead of monotonic increase with carbon dioxide in the atmosphere with large variation because of heavy convective rainfalls and tropical storms. Based on the future regional climate simulation the annual precipitation increases as much as 0.07 mm/day by It means we may have about 2.6 % more precipitation annually in comparison with current annual precipitation (2.68 mm/day). However, the number of rainy day diminished by 19 days in It illustrates that we may have more precipitation and less rainy days. Consequently we may have more chance to get heavy rainfall associated with global warming. Global downscaling with high resolution atmospheric general circulation model (GCM) has been used for regional detail climate projections over East Asia to avoid the uncertainty from choice of domain in the dynamic regional downscaling. This method has used prescribed sea surface temperature (SST) obtained from observed data as well as projected data by 4 Coupled GCMs reported in IPCC AR4. For present-day integration, AMIP SST data has been used to force the high resolution AGCM to produce present climate. The simulated climate has been examined in comparison with observed climate data. For future climate change experiment, two sets of time-slice experiments have been performed for mid-21st century and end-21st century. To force AGCM for future climate monthly mean SST anomalies are superimposed which are derived from 4 different coupled GCMs simulation based on the A1B reference scenario, onto the present-day SST boundary conditions

191 June 7-8, 2011 INTRODUCTION It has been a long desire to better understand the processes that determine regional climate and to evaluate regional climate change information which can be useful in impact studies and policy planning. To date, a relatively high level of uncertainty has characterized regional climate change information. This is mainly due to the complexity of the processes that determine regional climate change, which span a wide range of spatial and temporal scales and to the difficulty in extracting fine-scale regional information from coarse resolution coupled Atmosphere-Ocean General Circulation Models (AOGCMs). Coupled AOGCMs are traditionally used for generating projections of climatic changes due to the anthropogenic forcing. Climate modeling groups around the world have been performing an unprecedented set of coordinated 20th and 21st century climate change experiments, in addition to commitment experiments extending to the 22nd century for the Inter-governmental Panel on Climate Change Fourth Assessment Report (IPCC 2007). One of the aims of these simulations is to assess the ability of the global coupled climate models to make projections of future climate change. Hence the mean climate state of a model needs to be compared against observations. In defining the mean climate state of a model, there are a number of control run choices available in the IPCC AR4 archive. These include the pre-industrial or the present day trace gases runs or the 20th century all forcing runs. The second and obvious aim of the IPCC AR4 simulations is the future projections. There are several scenarios under which projections are available. The increased emissions of CO2 have led to the enhanced heating of the earth (IPCC, 2001) and the largest simulated climate changes are due to the radiative influence of CO2 (Chen et al., 2004). This idealized scenario provides opportunity to compare responses of the different models. Although the coupled AOGCMs are extremely useful to project the future climate under increase in atmospheric CO2 concentration, however, the horizontal resolution of present day AOGCMs is still relatively coarse, of the order of hundred kilometers, due to the centennial to millennial time-scales associated with the ocean circulation and computing requirements that these imply. Giorgi and Mearns (1991) have reported that regional climate is often affected by forcing and circulations that occur at the sub-aogcm horizontal grid scale. Accordingly, AOGCMs are unable to capture the fine-scale structure that characterizes climatic variables. Features like fine-scale structure are useful in impact assessment studies. Therefore, a number of techniques have been developed to generate fine-scale regional climate information based on the simulation by coupled AOGCMs. According to IPCC (2001), these are referred as "regionalization" techniques and

192 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 it has been classified into three categories as follows: High resolution and variable resolution Atmosphere GCM (AGCM) experiments; Nested limited area (or regional) climate models (RCMs); Empirical/statistical and statistical/dynamical methods. IPCC (2001) also reports that the assessment is based on different modeling tools that are currently available to obtain regional climate information, and includes: (a) an evaluation of the performance, strengths and weaknesses of different techniques in reproducing present day climate characteristics and in simulating processes of importance for regional climate; and (b) an evaluation of simulations of climate change at the regional scale and associated uncertainties. PROJECTION OF FUTURE CLIMATE CHANGE WITH REGIONAL DYNAMIC DOWN-SCALING TECHNIQUE MM5 has been integrated for 100 years ( ) to simulate the future climate over the East Asia (Oh et al., 2004; METRI, 2003b). The initial and boundary conditions have been derived from ECHO-G AOGCM based on IPCC SRES A2 scenario (METRI, 2003a; Oh et al., 2004). In this section the future surface air temperature and precipitation over Korean peninsula and East Asia have been discussed based on results of METRI (2003b). Temperature is one of the most sensitive factors to human beings among many climate variables. The long-term variation of temperature has significant impact on our daily life. It also affects long-term projection on the production of seasonal goods, energy consumption for cooling and heating, determination of seeding time and blooming season, etc. Furthermore, it is one of the important key factors to determine persistence of any species in the earth's ecosystem. According to third IPCC assessment report (IPCC, 2001), the global mean surface air temperature has increased 0.6 C during the last century. Furthermore, the warming over Korean peninsula is more dominant than global average. It is recorded 1.5 C warming during the 20th century. To understand the projection of future climate in Korean peninsula, it is necessary to review the characteristics of past temperature variation over the Korean peninsula and East Asia using observed data. Accordingly, the CRU 0.5 x0.5 grid data (New et al., 1999) for surface air temperature has been analyzed to illustrate past climate changes during the last 95 years for the period of over the Korean peninsula and East Asia region (24~49 N, 109~149 E)

193 June 7-8, 2011 The annual mean surface air temperature ( C) over the Korean peninsula and East Asia for 95 years has been presented in Fig. 1(a) and monthly surface air temperature anomaly for recent 10 years ( ) has been presented in Fig. 1(b). Temperature anomaly is calculated using climatological surface air temperature for the period of 30 years ( ). Average temperature during these 95 years is noted as 7.5 C, which is slightly lower than that of other region in the same latitude zone. It might have been caused from the fact that the cold and dry northwesterly is dominant during winter season. It brings a dry and chilly air mass to this region during winter season so that it may contribute to slightly cool mean temperature compared with other region where the Siberian high is not dominant. As shown in Fig. 1(a), the average surface air temperature of recent 10 years ( ) is 7.9 C, which is 0.7 C higher than that of 1900s (1901~1910). This represents another example of the global warming trend, which is observed in the most parts of the world (Parker et al., 1994; Easterling et al., 1997; Heino et al., 1999; Plummer et al., 1999). Researchers have pointed out that the rapid warming trend since 1970s is not occurring at particular region but it is a global phenomenon (Hansen and Lebedeff, 1987), it is more significant over the Korean peninsula and East Asia as high as 0.3 C/decade. In particular, the most significant high warming is recorded in 1990s as much as 0.6 C/decade. Fig. 1. (a) Annual mean surface air temperature ( C) over the Korean peninsula and East Asia for 95 years, (b) Monthly surface air temperature anomaly for recent 10 years ( ). Temperature anomaly is calculated using climatological surface air temperature for the period of 30 years ( ). In the upper panel the solid line represents 11 months running mean

194 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Fig 1(b) illustrates the different aspects of regional warming phenomena by presenting monthly mean surface air temperature anomalies from the climatological mean of 30 years ( ).The largest temperature anomaly is found in Based on these results we may conclude that the Korean peninsula and East Asia is not free from the global warming trend. Recent warming trend is much more significant than previous period. It can be a result of recent Industrial Revolution. Projection of future climate changes over Korean Peninsula and East Asia in the context of temperature changes is presented here. Fig. 2 shows the annual mean temperature changes in Korean peninsula and East Asia for the future 100 years ( ) the regional climate simulation with the SRES A2 CO 2 emission scenario. From these results, it is evident that the mean warming rate over Korean peninsula and East Asia is 0.61 C/decade for the next century. This warming rate is somewhat faster than the global temperature trend presented in Table 1. Based on the ECHO global climate model simulation the global mean temperature is C during the 2090s, which is about 3.7 C higher than 2000s. Meanwhile, the warming is 5.45 C over the East Asia during the next 100 year. It is about 1.65 C higher than the global warming. Fig. 2. Annual temperature changes in the Korean peninsula and East Asia for the next 100 years ( ) by the regional climate model based on the SRES A2 greenhouse emission scenario. Fig. 3 shows the relative changes in the daily mean, maximum, and minimum temperature for decades (selected every 20 years) in comparison with 2000s temperature. Until year 2100, there is gradual increase in these three variables. The magnitude of increase is 0.45 C, 0.34 C, and 0.56 C for the daily mean, maximum and minimum temperature respectively. It may not be significant. However, it is notable that the increase in the daily minimum temperature is somewhat larger than that of the maximum temperature. Then, we might be interested to know which season becomes significantly warm. Fig. 4 shows the mean temperature changes in autumn of 2090s, when the largest change has been recorded among four seasons in comparison with current climate. The increase in

195 June 7-8, 2011 the daily mean temperature is larger than 5 C at most regions. Note that the shaded region represents temperature changes larger than 6.5 C. However, we may expect similar warming in the daily minimum and maximum temperatures (Fig. not shown). Fig. 3. Relative changes in daily mean, maximum and minimum temperature. Changes in diurnal temperature are also presented. Fig. 4. Mean temperature changes in autumn of 2090s, when the largest change has been recorded among four seasons in comparison with current climate. The shaded region represents temperature changes larger than 6.5 C. The precipitation over East Asia for the domain (24~49 N, 109~149 E) has been investigated with the CRU data for the period of 1901 to In a comparison to other meteorological variables, precipitation shows large variation because of heavy convective rainfalls and tropical storms. The annual mean precipitation is mm and it increases as much as 3.4 mm/decade (Fig. 5a). The wettest year is 1954 with precipitation of mm, while the driest year is recorded in 1913 with precipitation of mm. The difference of monthly mean precipitation from the 30 years ( ) mean precipitation is shown in Fig. 5b for recent 10 years from 1986 to Unlike surface air temperature, it shows a large variability due to its characteristics of spatial and temporal variation

196 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Fig. 5. (a) Annual mean precipitation during 1901~1995, and (b) monthly mean precipitation anomalies of 1986~1995. Smoothening has been done using 11-year moving average. Precipitation Anomalies are based on climatological mean of 1961~1990 over East Asia. A projection of future precipitation over the East Asia has been presented in Fig. 6 based on the results of regional climate simulation SRES A2 scenario. It is represented in terms of annual mean precipitation (Fig. 6a) and the number of rainy days with precipitation intensity (Fig. 6b). Daily precipitation more than or at least 0.1 mm/day is considered as rainy day. Based on the future regional climate simulation the annual precipitation increases as much as 0.07 mm/day by It means we may have about 2.6 % more precipitation annually in comparison with current annual precipitation (2.68 mm/day). However, the number of rainy day diminished by 19 days in It illustrates that we may have more precipitation and less rainy days. Consequently precipitation intensity (PI) will increase, as it is ratio of annual precipitation amount and rainy days (Fig. 6b). It means, in other words, we may have more chance to get heavy rainfall associated with global warming

197 June 7-8, 2011 Fig. 6. (a) Annual mean precipitation (mm/day) and (b) number of rainy days (light line, right scale) and precipitation intensity(thick line, left scale) from 2001 to 2100 over East Asia. As shown in Table 1, the decadal mean of simulated global and regional precipitation has increased gradually, although the interdecadal oscillation can be observed in the regional precipitation clearly with its increasing trend. In Korean peninsula and East Asia the precipitation has either increased or decreased during the period of The precipitation has been estimated as much as 4.1 % increase during 2090s compared to that of 2000s. It is somewhat larger than the global mean precipitation increase which is estimated as 3.5 %. It has increased up to 1.9 % for 2010s compared to that of 2000s. However, it has decreased up to 1.1 % for 2030s. Subsequently it has increased up to 3.4 % for 2050s. It may reflect that there are two decadal oscillations in the simulated precipitation

198 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Table 1. Decadal mean and difference of simulated global and regional precipitation Precipitation Global East Asia (mm/day) Mean Difference Mean Difference 2000's 2010's 2020's 2030's 2040's 2050's 2060's 2070's 2080's 2090's % 0.4% 0.7% 1.1% 1.1% 1.8% 2.1% 3.8% 3.5% % 2.2% -1.1% -2.6% 3.4% 1.1% 1.1% 1.9% 4.1% The geographical distribution of precipitation changes during February (left panel) and autumn (right panel) in 2090s have been presented in Fig. 7. There is significant decrease in precipitation during February, while there is significant increase in autumn. During February the precipitation is decreased as much as 2 mm/day over the area of central China and over the ocean of East China Sea to south of Japan. During the autumn of 2090s, the precipitation increase is more significant over the ocean than over the land in comparison with 2000s precipitation (Fig. 7b). Fig. 7. Precipitation changes during (a) February and (b) Autumn in 2090s relative to 2000s. Shading indicates negative values in left panel and positive values in right panel

199 June 7-8, 2011 PROJECTION OF FUTURE CLIMATE CHANGE WITH HIGH RESOLUTION AGCM The weather and climate prediction using the global numerical model has been studied for a long time. Recently, the study has been performed improving the horizontal/vertical resolution in numerical modeling due to availability of enhanced computing resources. A review by Simmon and Hollingsworth (2002) showed that the increased resolution has had a positive effect on operational forecasting. This research with numerical global modeling below 30km resolution is in progress in several research groups in USA, England, Germany and so on. Especially, the Earth Simulator in Japan has performed a mesoscale resolving simulations of the global atmosphere with 10km resolution (Ohfuchi et al., 2004). So in this study, we have also examined the capability of Icosahedral-hexagonal Gridpoint Global Model (GME) in the projection of future climate change by time slice run based on the IPCC AR4 simulations by several coupled AOGCMs. The GME model is the operational mode of German Weather Service (Deutscher Wetterdienst). It is based on uniform icosahedral-hexagonal grid (Majewski et al., 2002). The GME gridpoint approach avoids the so-called pole problem. This model can adjust the resolution, so it is possible to simulate weather and climate change over both global and regional scale. So we have experimented GME model in different horizontal and vertical resolution (10 km~240 km/40~60 layers). We have used the KISTI IBM supercomputer to simulate the typhoon prediction at various resolutions. With high-resolution (10 km) and increase in vertical level (60 layer), the model captures even the very small scale local features as shown in Fig. 8. Fig. 8. Typhoon Simulations in different resolutions using GME model. (a) Precipitation and Mean Sea Level Pressure in 240 km and 40 layers, (b) 40 km and 40 layers, (c) 20 km and 60 layers and (d)10 km and 60 layers

200 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 For the current and future regional projection with time-slice method GME model with 40 km/ 40 layers has used. For present-day integration, the prescribed AMIP SST and sea ice data has been used to force the high resolution AGCM. The simulated climate has been examined in comparison with observed climate data. For future climate change experiment, two sets of time-slice experiments have been performed for mid-21st century and end-21st century. To force AGCM for future climate a superimposing monthly mean SST anomalies, which is derived from 4 different coupled AOGCMs simulation based on the IPCC SRES A1B emission reference scenario reported in the IPCC AR4. The AOGCMs provided SST and sea ice data are CGCM3.1(t63) (Flato 2005), CNRM_CM3 (Salas-Media et al. 2006), ECHAM5/MPI-OM (Jungclaus et al. 2006), and MIROC3.2(hires) (K-1 Model Developers 2004). The horizontal resolution of CGCM3.1(t63) and CNRM_CM3 is 2.8 degree by 2.8 degree, and those of ECHAM5/MPI-OM and MIROC3.2(hires) are 1.9 degree by 1.9 degree and 1.1 degree by 1.1 degree, respectively. For the evaluation of GME to simulate climate change the simulated climate (20C3M) for the period of by GME with the monthly AMIP SST and sea ice data (Taylor et al. 2000) has been compared with observational data as well as the results of the 20th Century Climate in Coupled Models (experiment acronym 20C3M ) runs as reported in IPCC AR4. As recommended by the IPCC AR4 panel, the models mean climate will be defined based on the last twenty years (corresponding to period) of the all forcing 20th century runs i.e. experiment designated as 20C3M. The proposed periods of middle and end of the 21st century climate are chosen the climate of and from the future climate change projection by those AOGCMs, respectively. The variability of AMIP observed SST during the period of has been introduced for middle and end of 21st century climate by superposing detrended AMIP SST to composite future SST data from 4 AOGCMs. Figs. 9 and 10 show the summer (June-August) and winter (December- February) surface temperature of current, middle and end of 21st century, respectively, together with the changes of future climate from current climate. A warming has occurred over the land and it is more significant during the high latitudes at the winter hemisphere. This warming trend becomes much clearer at the end of 21st century than during the middle of 21st century

201 June 7-8, 2011 Fig. 9. Summer surface temperature of (a) 20th century, (b) middle of 21st century, and (c) end of 21st century, respectively, together with the differences from 20th in (d) and (e). Differently from surface temperature the trend of summer precipitation does not show any organized development in future as shown in Fig. 14 except that the regional variation of precipitation during the summer in the Northern Hemisphere becomes more significant by the end of 21st century. Based on the analysis of the 22 AOGCMs in IPCC AR4 Kripalani et. al. (2007a, b) reported that the simulated spatial distribution of precipitation could be sensitive to model resolution. The high-resolution version produces more precipitation over the eastern side of the Tibetan Plateau, southern China up to Japan due to the high frequency of heavy precipitation simulated by this model

202 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Fig. 10. Winter surface temperature of (a) 20th century, (b) middle of 21st century, and (c) end of 21st century, respectively, together with the differences from 20th in (d) and (e). DISCUSSION AND CONCLUSION Through a series of GCM experiments it has been observed that there is increase in surface air temperature and precipitation over East Asia during the 21st century. Although the temperature increase is well reflected by various climate models, there is a large difference in the precipitation patterns. We may have clear understanding regarding the impact of carbon dioxide reduction in the atmosphere from the comparison of simulated climate based on the IPCC SRES A2 and A1B scenario. Striking feature like enhancement of summer precipitation is well simulated by climate models (Fig. 11). It may lead to enhancement of summer monsoon. However, it requires further studies for more clarification and confirmation

203 June 7-8, 2011 Fig. 11. Summer precipitation of (a) 20th century, (b) middle of 21st century, and (c) end of 21st century, respectively, together with the differences from 20th in (d) and (e). Several studies have been done on regional climate change based on statistical analysis over Asia. However limited studies have been done on the future climate projection with dynamic downscaling technique over Asia. Generally, all studies indicate warming climate trend over East Asia. However, there is a considerable difference in the future precipitation changes. Recently, Oh et al. (2004) made a projection of surface air temperature and precipitation in East Asia including Korean peninsula based on the dynamic downscaling with MM5 for the 100 years ( ) nested to the global projection with the MPI AOGCM based on IPCC SRES A2 scenario. This experiment results also show a significant increase in the surface air temperature in East Asia. However, the precipitation shows an inter-decadal fluctuation instead of monotonic increase as in temperature increase with carbon dioxide in the atmosphere. This regional climate change study is based on dynamic downscaling method. To get reasonable confidence on the regional climate change information, it is required to have various regional modeling studies including time-slice runs with high resolution AGCM to avoid the uncertainty from the domain of regional dynamic downscaling. Through these efforts we may have better understanding of

204 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 future climate projection and we may reduce or get rid of uncertainty which is involved in regional climate change information. Model inter-comparison studies can be helpful for this purpose. Recently there is a great concern on the extreme weather events, such as heavy rainfall and tropical cyclone. Also, many scientific challenges have been taken to reduce the uncertainty in future climate projection, such as systematic monitoring, enhancement in treatment of physical processes in the global as well as regional models, ensemble prediction, simulation with various scenarios including both direct and indirect effect of atmospheric aerosols, and consistent analysis of model projection for future climate. In future, we will have better projection for future climate changes and better understanding of the catastrophic disaster arising from greenhouse gases. Although none of these sets of time-slice experiments is able to capture accurately the response of regional climate change, the results may provide an essential data to estimate regionally detail temperature increase and changes in precipitation pattern over East Asia including the East Asian monsoon. The results might be utilized to set up both national and regional adaptation strategies for CO 2 induced global warming. ACKNOWLEDGEMENTS This study was supported by the Korea Meteorological Administration Research and Development Program under RACS And the computational resources for this study were provided under the PLSI program by Korea Institute of Science and Technology Information and PKNU super-compute center. REFERENCES Chen M., D. Pollard, E. J. Barron, 2004: Regional Climate change in East Asia simulated by an interactive atmosphere-soil-vegetation model. J. Climate, 14, Easterling, D., B. Horton, P. Jones, T. Peterson, T. Karl, D. Parker, G. Salinger, V. Razuvayev, N. Plummer, P. Jamson, and C. Folland, 1997: Maximum and minimum temperature trends for the globe. Science, 277, Flato. G. M., 2005: The Third Generation Coupled Global Climate Model (CGCM3) (and included links to the description of the AGCM3 atmospheric model). Giorgi, F., and L.O. Mearns, 1991: Approaches to the simulation of regional climate change: a review, Rev. Geophys., 29,

205 June 7-8, 2011 Hansen, J., and S. Lebedeff, 1987: Global trends of measured surface air temperature. J. Geophy. Res., 91, 13, Heino, R., R. Brazil, E. Forland, H. Tuomenvirta, H. Alexandersson, M. Beniston, C. Pfiser, M. Rebetez, G. Rosenhagen, S. Rosner, and J. Wibig, 1999: Progress in the study of climate extremes in northern and central Europe. Clim. Change, 42, IPCC, 2001: Climate Change 2001: The Scientific Basis. Contribution of Working Group I to the Third Assessment Report of the Intergovernmental Panel on Climate Change, Houghton, J. T., Y. Ding, D. J. Griggs, M. Noguer, P. J. van der Linden, X. Dai, K. Maskell, and C. A. Johnson, Eds., Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, pp IPCC, 2007: Climate Change 2007: The Physical Science basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, Solomon, S., D. Qin, M. Manning, Z. Chen, M. Marquis, K. B. Averyt, M. Tignor, and H. L. Miller, Cambridge Univ. Press, Cambridge, United Kingdom and New York, NY, USA, pp Jungclaus J. H., N. Keenlyside, M. Botzet, H. Haak, J. J. Luo, M. Latif, J. Marotzke, U. Mikolajewicz, E. Roeckner, 2006: Ocean circulation and tropical variability in the coupled model ECHAM5/MPI-OM. J. Climate, 19, K-1 Model Developers, 2004: K-1 Coupled GCM (MIROC) description. In: Hasumi H., S. Emori, K-1 Tech Report No. 1, Center for Climate System Research, University of Tokyo, National Institute for Environmental Studies, Frontier Research Center for Global Change, pp. 39 (Available at ccsr.u-tokyo.ac.jp/kyosei/hasumi/miroc/tech-repo.pdf). Kripalani, R. H., J.-H. Oh, H. S. Chaudhari, 2007a: Response of the East Asian summer monsoon to doubled atmospheric CO2: coupled climate model simulations and projections under IPCC AR4. Theor. Appl. Climatol. 87, Kripalani, R. H., J.-H. Oh, A. Kulkarni, S. S. Sabade, and H. S. Chaudhari, 2007b: South Asian summer monsoon precipitation variability: Coupled climate model simulations and projections under IPCC AR4. Theor. Appl. Climatol. 90, Majewski, D., and D. Liermann, P. Prohl, B. Ritter, M. Buchhold, T. Hanish, G. Paul, W. Wergen, and J. Baumgardner, 2002: The Operational Global Icosahedral Hexagonal Gridpoint Model GME: Description and High- Resolution Tests. Mon. Wea. Rev., 130, METRI, 2003a: 온실가스증가로인한한반도주변의기후변화탐지. 온실가

206 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 스저감기술개발사업최종보고서, pp METRI, 2003b: 한반도및동아시아지역기후변화시나리오기법개발 (II). 최종보고서, 기상연구소. Morlet J., 1983: Sampling theory and wave propagation. NATO ASI Series, FI, Springer, New, M., M. Hulme, and P. Jones, 1999: Representing twentieth-century spacetime climate variability. Part I: Development of a mean monthly terrestrial climatology. J. Climate, 12, Oh, J.-H., T. Kim, M.-K. Kim, S.-H.Lee, S.-K.Min, and W.-T. Kwon, 2004: Regional climate simulation for Korea using dynamic downscaling and statistical adjustment. J. Meteorol. Soc. Jpn., 82, Ohfuchi W, H. Nakamura, M. K. Yoshioka, T. Enomoto, K. Takayo, X. Peng, S. Yamane, T. Nishimura, Y. Kurihara and K. Ninomiya, 2004: 10-km Mesh Meso-scale Resolving Simulations of the Global Atmosphere on the Earth Simulator Preliminary Outcomes of AFES (AGCM for the Earth Simulator), J. of the Earth Simulator, 1, Parker, D. E., P. D. Jones, C. K. Folland, and A. Bevan, 1994: Interdecadal changes of surface temperature since the late nineteenth century. J. Geophys. Res., 99, 14,373-14,399. Plummer, N., M. J. Salinger, N. Nicholls, R. Suppiah, K. J. Hennessy, R. M. Leighton, B. Trewin, C. M. Page, and J. M. Lough, 1999: Change in climate extremes over the Australian region and New Zealand during the twentieth century. Clim. Change, 42, Salas-Mdlia D., F. Chauvin, M. Deque, H. Douville, J. F. Gueremy, P. Marquet, S. Planton, J. F. Royer, S. Tyteca, 2006: Description and validation of the CNRM-CM3 global coupled model. Clim. Dyn. Simmons A. J., and A. Hollingsworth, 2002: Some aspects of the improvement in skill of numerical weather prediction, Quart J. Roy. Meteor. Soc., 128, Taylor. K. E., D. Williamson, and F. Zwiers, 2000: The Sea Surface Temperature and Sea-Ice Concentration Boundary Conditions for AMIP II Simulations, PCMDI Report No. 60, pp

207 June 7-8, 2011 Vegetation Regeneration and Soil Erosion for Ten Years Following Forest Fires in Korea Yeonsook Choung* Department of Biological Sciences, Kangwon National University, Chuncheon , Republic of Korea Abstract The recovery processes in terms of vegetation and soil have been monitored in permanent plots and runoff plots for ten years. Also, the effectiveness of several soil conservation measures on runoff and soil erosion has been tested. The prefire Pinus densiflora forests have been successfully recovered into mainly Quercus-dominated forests though the speed and the pattern varies depending on various factors such as prefire species composition, burn severities and initial regeneration. Soil erosion was severe in low regeneration regions occupied by prefire young pine stands. The experiments on the soil conservation measures emphasize that fast regenerating vegetation significantly contributes to preventing soil erosion and proper hill treatment such as woodchip mulching is needed for the poorly regenerated regions immediately after fires. INTRODUCTION All forest fires in Korea are caused by humans (Korea Forest Service 2009). Logging and frequent forest fires from heavy forest use had been important factors causing disturbance in Korea. More than 90% of forests in South Korea are secondary forests that are less than 40-years-old (Korea Forest Service 2009). This strongly implies that until recently, there have been frequent and serious disturbances to forested landscapes (Hong 1998). In the early 1970s, however, the government implemented strong forest resource policies including changing fuel sources from wood to coal, prohibition of slash-and-burn farming, prevention and suppression of forest fires, and plantation of large areas. These policy changes have affected the recovery and successional patterns of forests throughout Korea (Northeast Asian Forest Forum 2000). Significantly, they have reduced forest losses caused by massive logging. Nonetheless, fire is still a major disturbance to forested landscapes. The consumption of wood by fires has strikingly increased

208 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 since the 1990s as forest stands have developed and accumulated biomass (Korea Forest Service 2009). The east coast forest fires, the largest recorded, occurred in April 2000 and burned a total of 23,794 ha of forested land, an area equivalent to 0.37% of the total forested area in South Korea (Ro et al. 2000). After the east coast fires, serious issues were raised regarding conventional forest recovery management and the capacity of natural regeneration capacity in the affected forests. The actual spatial patterns of burn severity and recovery, and their implications over largescale landscape processes have been little studied in Korea. Most studies concerning forest fires in South Korea have mainly concentrated on initial floristic changes after fires compared to unburned areas (e.g. Kim et al. 1981; Kim et al. 1999), and on soil chemical characteristics (e.g. Kim et al. 1999). The lack of information has led to the establishment of the Joint Association for the Investigation of the East Coast Fires with the support of the government. This association has subsequently investigated various aspects of disturbances and forest recovery over the range of affected forests. The considerable attention given was enough to spur various levels of studies for the impact of fires on forest dynamics in Korea. Since the investigation immediately after the fires, the recovery processes in terms of vegetation and soil have been monitored in permanent plots and runoff plots for ten years. In addition, the effectiveness of several soil conservation measures on runoff and soil erosion has been tested. VEGETATION IN THE FIRE YEAR AND TEN YEARS LATER Burn severity and initial regeneration immediately after the fires We analyzed the differential severity and post-fire recovery of pre-fire forest types of different stand age both at stand and species level. Analysis showed that pine stands were the most severely burned, while conversely pinehardwood and hardwood stands were less vulnerable. This implied that pine forests had fire-prone characteristics. Vegetation recovery went the opposite way; that is, the regenerating vegetation cover was 71% pre-fire hardwood stands, and 65% and 53% of pinehardwood and pine stands, respectively. However, these recovery rates were strikingly fast, considering that investigation took place about 3 months after the fires. Fire did not initiate successional processes, but tended to accelerate the predicted successional changes by releasing pre-fire understory species that

209 June 7-8, 2011 survived the fires and regenerated by sprouting. The dominant pre-fire tree species (P. densiflora) was susceptible to fire and not resilient enough to reestablish in competition with oak species. Contrary to pines, the abilities of oak species, mainly Quercus mongolica and Q. variabilis, to survive fires and to resprout vigorously made them dominant at most post-fire stands. These shifts in species abundance caused drastic changes to the landscape: from pine-dominated to oakdominated stands without any notable change in species composition. Forest regeneration for ten years Vegetation recovery in terms of stand development and species composition have been monitored for ten years following fires. To compare the recovery process according to differential burn severities and initial regeneration classes, permanent plots were established in prefire Pinus densiflora forest stands. Upper layers were developed faster at the severe-burn plots than those at light-burn plots. Among plots with different regeneration classes, 'high' plots showed fast recovery developing the sub-tree layer in the seventh year, while at 'low' plots vegetation developed the shrub stage. Unlike layer development, the number of species showed slight difference with time. Species from all burned plots tends to increase with time due to the recruitment of seed plants. Species composition is highly correlated with prefire composition as plant species have been regenerated from prefire existed underground organs. Quercus mongolica and Q. variabilis are mostly dominant resulting in the replacement of P. densiflora at most severe burn plots except for low regeneration plots where the abundance of P. densiflora is increasing fast in recent years. SOIL EROSION AND SOIL CONSERVATION MEASURES Runoff and soil erosion for ten years Most forest fires occur in spring in Korea. Following fires in spring, summer monsoon damage soil surface severely. Therefore, at a prefire Pinus densiflora forest where severely burnt by East Coast fire in Samcheok, Korea in 2000, we have compared the relative effects of vegetation recovery and rainfall on the sediment yields and runoffs from Experiments were performed from 21 runoff plots with various level of vegetation coverage. The runoff and sediment yield have significantly decreased with time following the fire. However, significant differences of the sediment yields and runoffs among different vegetation recovery maintained during the experimental period. From the seventh

210 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 year, the relative effect of vegetation recovery on both of sediment yields and runoffs exceeded the effect of the rainfall events and its effects maintained since then. This has great implications regarding the managements of the disturbed soils by severe forest fires. Soil conservation measures The effect of soil conservation measures, such as mulching with wood chips, seeding with native plant species and log erosion barriers (LEBs), on runoff and soil erosion were examined using runoff plots. Wood chip mulching greatly reduced runoff and sediment yields and these effects were consistent regardless of the volume of rainfall. Neither seeding nor LEBs reduced runoff and sediment yields. No positive or negative effects of mulching, seeding or LEBs on ground vegetation cover were observed. The ineffectiveness of seeding and LEBs may have been due to the steep slope, the failure of germination and establishment of seeded plants, and the small diameter of logs. Treating hill slopes with mulch should be considered where post-fire regeneration is slow and there is an absence of organic material such as litter. CONCLUSION AND IMPLEMENTION FOR POSTFIRE FOREST MANAGEMENT For ten years after fires, prefire Pinus densiflora forests have been successfully recovered into mainly Quercus-dominated forests though the speed and the pattern varies depending on various factors such as prefire species composition, burn severities and initial regeneration. Soil erosion was more severe in low regeneration regions mainly occupied by prefire young pine stands. The results regarding soil conservation measures emphasize that fast regenerating vegetation contributes to preventing soil erosion and to preserving the substantial pools of nutrients in the biomass. Therefore, the protection of topsoil should be the first consideration in fragile and unstable environments. To prevent topsoil loss, the natural regeneration of vegetion should be allowed to accumulate biomass. In addition, proper hill treatments in poorly vegetated regions should be introduced immediately after fires to control severe soil erosion

211 June 7-8, 2011 REFERENCES Choung, Y., B.-C. Lee, J.-H. Cho, K.-S. Lee, I.-S. Jang, S. Kim, S.-K. Hong, H.-C. Jung and H.-L Choung Forest Responses to the Large-Scale East Coast Fires in Korea. Ecological Research 19(1): Kim, C.-G., K. Shin, K. Y. Joo, K. S. Lee, S. S. Shin and Y. Choung Effects of soil conservation measures in a partially vegetated area after forest fires. Science of the Total Environment 399:158~164. RO, D.-K., GONG J.-S., LEE S.-H., KIM C.-M., KIM J.-C., SEO S.-A., PAEK J.-H., KIM S.-H., SIM W.-B., SON Y.-M. & LEE W.-K Damage of forests in the East Coast Fires. In: Report of the East Coast Fires in 2000 (I), pp The Joint Association for the Investigation of the East Coast Fires, Seoul

212 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 System for the Reduction of CO 2 Emitted from Forest Fire Jun-ichi Kudoh and Masanao Ito Tohoku University, Japan kudoh@cneas.tohoku.ac.jp An addendum 1) was prepared to facilitate negotiations among the parties in the tenth session of the Ad Hoc Working Group on Further Commitments for Annex I Parties under the Kyoto Protocol. The Copenhagen deal requires countries to submit the pledges to reduce the green house gases emissions by the end of January, The idea of putting a cap on the amount of CO 2 emission has though caused some developing and emerging nations like China, India, Brazil or South Africa disagree with the clauses of this addendum. Their argument is that CO 2 emission is natural when a country is industrially developing like theirs and by putting a cap in the amount of CO 2 emission; the developed nations are trying to threaten their capacity to expand. Even though all these emerging nations have been trying their best to promote the renewable sources of energy and decrease the amount of the green house gases, they do not want the development of their nation be hindered by some external force and thus are not agreeing with the attempts to make emissions cut legally binding or setting long term goals for reducing green house gases. This disagreement has caused various concerns for the whole global environment scenario. In this paper, we present two techniques by whose implementation all the nations can come under the Kyoto protocol and the second commitment period can be implemented and global environmental problem caused by CO 2 can be reduced by a big amount. 1. Improvement of Monitoring Technique of Forest Fires and fast transmission of the message to extinguish the fire The amount of CO 2 released every year from the forest fires is considerably higher than from other sources. In a study of 100 Mha of Canadian Boreal forest for the effects of climate, carbon dioxide concentration and fire disturbance on net biome production, net primary production and vegetation dominance by creating a model, it was obtained that the carbon balance of that region was driven by changes in fire distribution from 1948 to ). Even though the trees work as natural sink for carbon dioxide, there are large quantities of forest fires occurring

213 June 7-8, 2011 all over the world 3 which emit CO 2 in large quantities. From the IPCC s Fourth Assessment Report, the greenhouse gases emission has increased by 70% from 1970 to 2004 to a total amount of 49 billion CO 2 tons, and out of which, CO 2 comprises of around 76.7% which accounts to about 37.6 billion tons of the total emission 4. And from FAO s data, the average amount of CO 2 emitted from the year 1997 to 2006 from forest fires is around 9.2 billion tons, which comprises of 18.8% of the total CO 2 emission and it also states that there is an increment in the frequency of forest fires every year 3). While GFMC (2009) reports it to be anywhere from 7.3 billion CO 2 tons to 14.7 billion CO 2 tons 5). This variation in the data of the green house gas emission with such large error extent indicates that a proper measurement method is not implemented yet. So an establishment of a technology for the detection of forest fires and the starting the extinguishing process within one hour of the detection is a must. Currently there are various satellites such as MODIS (with spatial resolution of 1km but time spacing of 2 times in a day) monitoring for forest fires but there is a big time lag between the detection and the extinguishing of the fire and even when the fire is detected it is already too late. So, newer satellites with spatial resolution of about 5m and temporal resolution of about 5 minutes should be launched for full time monitoring of the forest fire and relay of the fire alarm in real time to the required authority to start the fire extinguishing procedure within one hour of the fire outbreak. If this can be achieved, a large quantity of forests can be protected and the CO 2 released from the forest fires can be reduced by a drastic amount by decreasing the amount of the total forest fires on the earth. Figure 1 is an example of the fire smoke, which is observed by Khabarovsk Fire Fighters. Fig.1. Fire started breaking out almost immediately 6)

214 International Forest Fire Symposium on Commemorating the International Year of Forests Construction of Measurement Model for the Calculation of the amount of CO 2 emitted in Forest Fires The construction of a measurement model to calculate the accurate amount of CO 2 emitted from the forest fire is very essential. Currently there are simulation models available for the calculation of the amount of CO 2 absorbed by a portion of forest according to the type of trees in the forest. Using those models to calculate how much carbon the forest would have absorbed had it been there. Furthermore, there is a satellite called the GOSAT 7) (Greenhouse Gases Observing SATellite) which is the world s first satellite dedicated to greenhouse-gasmonitoring. GOSAT, launched by Japan Aerospace Exploration Agency(JAXA), measures the density of CO 2 and methane from 56,000 locations on the earth s atmosphere. And, there is a need for the establishing a different model (Emission model) for the precise calculation of CO 2 emitted by forest fire. By using the mixture of above three techniques i.e. the model for calculation of CO 2 absorbed by the forest (Absorption Model), the GOSAT data and the new model for the calculation of CO 2 released because of the fire (Emission model), exact calculations of the CO 2 stored in forest and emitted in forest fires can be done. 3. Application of the above mentioned techniques 1. and 2. into CO 2 Emission Trading Scheme The addendum states that when a forest fire occurs as a force majeure, it does not fall in the category for the calculation of CO 2 emitted, and force majeure is not intended to excuse negligence and other malfeasances. For example, in Russia there is approximately 1 billion tons of CO 2 released from forest fires and around 90% every year resulting from human carelessness and malfeasance 8).By the application of the above two techniques, first the forest fires can be detected and extinguished early which results in the reduction of global CO 2 emitted every year. By using the monitoring technique of forest fires and the CO 2 level measurement model as tools much more forests of the developing and emerging countries can be taken by the developed countries under forest management under emission trading scheme. With the establishment of the two techniques, there is a lesser need to worry about forest fires and the CO 2 emitted as a result of those forest fires, and the CO 2 stored by the trees as natural sink can be used by the developed countries as their CO 2 count while the timber and the forest products can be used by the developing nation for further development

215 June 7-8, Implementation of the 2 nd Commitment Period of the Kyoto Protocol. With the establishment of the above mentioned two technologies, emission trading (cap and trade) between the countries will be promoted to a greater extent and this also would be a very big step in solving the problem that the world environment is facing now. Currently various researches are going on in the Siberian forests of Russia and a case study for the improvement of the monitoring technique of forest fires and the calculation of the CO 2 emitted from forest fires through the model can be conducted there and if proven successful can be applied by the other nations under the Kyoto Protocol. And finally by taking these steps, the second commitment period can be implemented by all the nations. Figure 2 is a proposal system from this paper. Tool 1 Tool for the reduction of CO 2 emitted from forest fires Development of sensors for monitoring of fire outbreak Development of method for fast relay with the message Start extinguishing procedure within one hour of outbreak Forest Fire Reduction Tool 2 Tool for the measurement of CO 2 emitted from forest fires Development of prototype algorithm from fire analyzing satellite data Embedding the volume of CO2 emitted in forest fire into GOSAT level 4 High precision of the prototype model Tool 3 Emission Trading Scheme of CO2 emitted from forest fires Global CO 2 Reduction Fulfillment of the 2 nd Commitment period of Kyoto Protocol Implementation in International Market System for Reduction of CO 2 Emitted from Forest Fire Fig. 2. Proposal system. 5. Case Study Russia constitutes of about 20% of the total forest area in the globe and emits around one billion tons of CO 2 every year, so the effect of reduction of forest fires in Russia is huge. For that purpose, case study with should be implemented in Siberia with coalition of Japanese and Russian government with the collaborative research with the Russian Academy of Science being the base. We are considering the project scheme for the case study as shown in Fig

216 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Scheme for Russia as Case Study Huge effect in the reduction of CO2 emission in a nation with about 1 billion tons of CO2 emission yearly Natural Resources and Environment Ministry/Ministry of Agriculture /Space Agency Gazprom/Oil Company/Electricity authorities/energia.. Russian Government CO 2 emission Ministry of Education, Culture, Sports, Science and Technology, M. of Environment, M. of Economy and Public Works and Services, M. of Foreign Affairs, M. of Forestry.. JAXA/Electricity Authority/Steel Manufacturing Company/Trading companies. Japanese Government Forest Dept., Fire Dept Timber union, Companies. Companies Far Eastern Branch of the Russian Academy of Sciences Database Fire Info (PJ Working Capital) Tohoku University Kudoh Group Database Fig. 3. Considering scheme of the project. 6. Conclusion This study, if implemented, is certain to reduce the total emitted global CO 2 and can work as a trump card to solve the environmental problem that the world is facing by a mutual understanding and involvement of all the developing, emerging and developed nations. References Ben Bond-Lamberty, Scott D. Peckham, Douglas E. Ahl & Stith T. Gower Fire as the dominant driver of central Canadian boreal forest carbon balance, Nature Vol. 450,(1 st Nov, 2007) p.28 GFMC, Vegetation Fires and Climate Change Interactions, FAO Committee on Forestry Alexander P. Lubyakin, Russian Federation Agency of Forestry Avialesookhrana the branch of Far-Eastern aviation forest fires fighting base

217 POSTER SESSION

218 International Forest Fire Symposium on Commemorating the International Year of Forests

219 June 7-8, 2011 No. Title Presentor Affiliation Page P-01 P-02 P-03 P-04 P-05 P-06 P-07 P-08 P-09 P-10 P-11 P-12 P-13 P-14 P-15 P-16 P-17 P-18 P-19 P-20 P-21 P-22 P-23 P-24 P-25 P-26 P-27 Error correction of watershed map for forest functions classification Abundance of butterflies in new grassland formed by forest fire and in forest Influences of disturbance intensity on diversity and abundance of arthropod predators in fired-pine forest, Korea Vegetation type of Pinus densiflora stands and damage characteristics of oak stands in post- forest fire area in Korea Comparison of difference between Gwangneung and Mt. Gumsan of soil respiration, litterfall production and decomposing ;itter in natural broad-leaved forest Community fluctuation and functional feeding groups of benthic macroinvertebrates in forest fire area of Samcheok Community stability of benthic macroinvertebrates in forest fired area of Uljin Forest fire risk mapping based on analytic hierarchy process Changes of species diversity on moth communities at forest fire region in Samcheok, Korea A study on the possibility of Melia azedarach L. of planting in sterile land Method of manufacturing soil-map of natural regeneration site after forest fire, Goseong. Effects of edge and burn severity on post-fire vegetation regeneration Assessing canopy fuel characteristics for major conifer species in Korea Crown characteristics and fuel load estimation of Pinus densiflora S. et Z. in Gyeongbuk province Estimations of biomass and carbon stocks in the different forests types of the Philippines Hot spot analysis on forest carbon stock using Getis- Ord Gi* spatial statistic Physiological responses of Pinus densiflora by seawater watering Effects of previous-fertilization treatment and fertilizer-induced stress on the growth and nutrient contents of Chamaecyparis obtusa seedlings Dynamics of advanced seedlings of Pinus koraiensis in three stand types Synersistic effects of a landslide risk in forest fire damaged area of Namwon city using aerial photographs Estimation on forest fire potential hazard by fuel load change of forest strata in forest stand Analysis of spatial distribution for forest fire damaged areas in East-coast of Gangwon province The spatial change of forest Fire occurrence during the spring season between 1990s and 2001s Chemical composition of precipitation and wet deposition in forest area, Gangwon Trends of timber products of the east coast forest fire area after 2000 in Korea Monitoring of vegetation recovery in the east coastal forest bunt area of Korea using multi-temporal landsat images Mapping and evaluation of forest carbon stock derived from the national forest inventory data and landsat imagery Hyejung Roh Cheolmin Lee Cheolmin Lee Kwangsoo Lee Imkyun Lee Korea Forest Research Institute Korea Forest Research Institute Korea Forest Research Institute Korea Forest Research Institute Korea Forest Research Institute Jongeun Lee Andong National University 226 Jongeun Lee Andong National University 230 Changjae Kim Yonsei University 234 Cheolmin Lee University of Incheon 239 Bonhag Koo Korea Forest Research Institute 245 Jooho Hwang Tae-Eun Research Institute 248 Sangwoo Lee Konkuk University 254 Sungyong Kim Kongju National University 258 Mina Jang Kongju National University 262 Youngjin Lee Kongju National University 266 Kyoungmin Kim Korea Forest Research Institute 271 Sunhee Kim Korea Forest Research Institute 275 Jiyoung An Korea Forest Research Institute 279 Heemoon Yang Korea Forest Research Institute 285 Choongshik Woo Korea Forest Research Institute 289 Myoungsoo Won Korea Forest Research Institute 294 Byungoh You Korea Forest Research Institute 302 Myoungsu Won Korea Forest Research Institute 306 Heejung Youn Korea Forest Research Institute 309 Seongyoun Lee Korea Forest Research Institute 313 Seungho Lee Korea Forest Research Institute 318 Jongsu Yim Korea Forest Research Institute

220 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 No. Title Presentor Affiliation Page P-28 P-29 P-30 P-31 P-32 P-33 P-34 P-35 P-36 P-37 Comparision of small aera estimation techniqes for providing forest growing stock volume from NFI6 field data Analysis on land-use and land-cover change using medium resolution satellite image Forest cover change detection for Mt. Bogdkhan and Tuul River Basin in Mongolia Idea of rifling application of water mist nozzlefor forest fire extinguishing Effects of forest fire on the water quality in forest catchments The effect of microbial compost on soil and vegetation in forest-fire areas Allocating initial attack resources in the republic of Korea with a scenario based optimization model Eco-toxicity of firefighting foams assessed in aquatic and semi-aquatic compact biotopes A combustion analysis of surface fuel burning experiment according to density variation Changes of fuel consumption in forest by thinning and pruning P-38 Scattering of tsunami wave from a forest Semyong Chang P-39 Development of algorithm for analyzing priority area of forest fire surveillance using viewshed P-40 Evaluating the feasibility of wood damaged by forest fire on solid biomass fuel P-41 Chemical properties of a soap-based class A foam for Takayoshi wildfire Kawahara Effectiveness of rehabilitation treatments for forest P-42 restoration in post-wildfire area: A case on a hillside Changwoo Lee slope in eastern coastal forest area, Republic of Korea P-43 P-44 P-45 P-46 P-47 P-48 P-49 P-50 P-51 P-52 P-53 P-54 Introduction and growth of Pinus densiflora at the naturally regenerated forest stands after forest fire Policy of forest fire prevention and analysis of forest fire during Chosun dynasty The analysis for cognition of forest fire danager of structure in wildland-urban interface The fertilization effect on the growth and foliage nutrients of containerized quercus acutissima at the forest fire damaged area Study on regional spatial autocorrelation of forest fire occurrence in Korea A research on the fire extinguishment resource utilization of the fire department specialized in forest fire Differences in bird communities by postfire restoration Development of U-ICT forest disaster management system based on LBS(Location based services) Liming effects on soil condition of a forest fire burned stand in southern Korea The flora of tree pathogens in forest-fired regions in Gangwon province in Korea Changes of photosynthetic characteristics of Japanese larch(larix kaempferi) seedlings raised under different CO 2 concentration Long term change of avifauna referring to the restoration method after the forest fire on Goseong, Korea Ilbin Jung Korea Forest Research Institute 327 Hyunkook Cho Korea Forest Research Institute 332 Junghwa Chun Korea Forest Research Institute 335 Choiseok Hwan Korea Forest Research Institute 338 Hyungtae Choi Korea Forest Research Institute 346 Jeongsoo Park Seoul National University 351 Byungdoo Lee Oregon State University 356 Tomonori Kawano The University of Kitakyushu 367 Janghwan Kim Hoseo University 371 Kyosang Koo Korea Forest Research Institute 375 Kunsan National University 379 Gyesun Ryu Korea Forest Research Institute 384 Soomin Lee Korea Forest 388 Joohoon Lim Donghyun Kim Houngsek Park Jaekyung Byun Research Institute Shabondama Soap Co., Ltd., Japan Korea Forest Research Institute Korea Forest Research Institute Korea Forest Research Institute Kangwon National University Korea Forest Research Institute Wookyun Lee Korea University 423 Junghun Kim Hoseo University 429 Eunjae Lee Donghyun Kim Seungwoo Lee Sanghyun Lee Donghun Ji Kisup Lee Seoul National University Korea Forest Research Institute Korea Forest Research Institute Korea Forest Research Institute Korea Forest Research Institute Korea Institute of Environmental Ecology

221 June 7-8, 2011 Error Correction of Watershed Map for Forest Functions Classification 유역단위산림기능구분을위한유역구분도오류보정 Abstract Hye-jung Roh 1*, Jun-heon Jeon 1, Young-kyu Park 1, Hyung-ho Kim 2, Jin-young Hwang 2 1 Korea Forest Research Institute, Seoul, , Korea, 2 Gyeongsang National University, Jinju, , Korea poulpican@sungshin.ac.kr Forests provide various socio-economic functions, based on the differentiated needs of the human. It is essential that the functions of forests should be identified and evaluated in order to meet social needs. The aim of this study was to correct error in watershed using two methods, and to compare the results. Methods were Simple Merge (SM) and Topology using Gaps (TG). The results showed that SM was simpler and more applicable, but there were stair shape watersheds. Although TG was showed smoother watersheds, total count of watershed was less than SM and it generated errors such as sliver polygon. Thus, we suggested SM was adequate for forest functions classification map. 요약 산림은목재, 임산물생산등의경제적기능과산지재해방지, 수원함양, 산림휴양등다양한사회생태적기능을제공하고있다. 산림이지니고있는다양한기능을객관적으로평가하고, 정확히파악하여산림의기능을극대화하고산림자원에대한사회적수요를충족시키기위해서는산림기능평가와같은객관적평가가필요하다. 이연구는유역단위의산림기능구분도의기초자료인유역구분도가내재하고있는오류들을단순병합방법과토폴로지방법을이용하여오류를보정하고자하였다. 연구결과, 단순병합방법은간단하고, 적용하기용이하였지만, 계단형유역이나타났다. 반면, 토폴로지방법은자연스러운곡선형태의유역형태가나타나고, 불필요한유역들이줄어듦에따라총유역개수도단순병합방법에비해적었지만, 슬리버폴리곤을만드는경우가나타났다. 따라서단순병합방법에의한유역구분도보정방법이적절한것으로판단된다

222 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 서론 (Introduction) 산림은목재, 임산물생산등의경제적기능과산지재해방지, 수원함양, 산림휴양등다양한사회생태적기능을제공하고있다. 산림이지니는다양한기능을정확히파악하여산림의기능을극대화하고산림자원에대한사회적수요를충족시키기위해서는산림기능평가와같은객관적평가가필요하며, 수많은공간자료가사용되므로 GIS 를기반으로분석하고있다. 국유림과공유림에대한산림기능구분이완료되었으므로사유림에대한산림기능구분도제작이요구되고있다. 우리나라의사유림은대부분소규모이므로유역단위의집단화된산림기능구분이적합하다. 그러나 DEM 자료를기반으로제작된유역구분도는직선형오류및슬리버폴리곤형태의오류를내재하고있으므로유역구분도의오류수정은산림기능구분도제작의필수적인과정이다. 이연구는사유림의산림기능구분도작성을위해유역구분도의오류를제거하기위한해결방안을모색하는데그목적이있다. 재료및방법 (Methodology) 경상남도의산림면적은 707,103ha 로써전국산림의약 11.1% 를차지하고있다. 유역구분도의오류보정을위한접근방법으로단순제거및병합과토폴로지방법을사용하였다. 단순제거및병합방법은일정면적이하의유역을주변유역에병합하는방법으로써주변에있는유역중면적이가장크거나경계가가장많이접하고있는유역에편입시킴으로써오류를제거하는방법이다. 반면토폴로지 (topology, 위상 ) 는지리적개체들사이의인접성, 연결성및포함에대한공간적관계를의미하고, 토폴로지정리는유효성검사를통해위배되는규칙을확인하고수정하는과정을가리킨다. 이연구에서토폴로지방법은일정면적이하의유역을인위적인 GAP( 규칙위배 ) 을생성하고, 토폴로지규칙중 must not have gaps 하여토폴로지를정리하는방법을사용하였다. 유역구분도를검토한결과, 1ha 미만의유역에서일반적으로오류가많이발생함에따라오류보정작업을수행하였다. 결과및고찰 (Results and discussion) 연구결과, 단순병합방법은적용하기쉽고, 유역구분도의변형이없다는장점이있다. 그러나유역구분도의정사각형유역이계단형으로변형되는것

223 June 7-8, 2011 으로나타났다. 반면, 토폴로지를이용한인위적인갭생성방법은단순제거및병합방법에비해자연스러운유역형태가나타나고, 정사각형유역이자연스러운직선형태로나타나는장점이있다. 그러나인위적인갭을제거하기위해서는적절한범위 (tolerance) 를설정해주어야하며, 설정된그범위때문에오히려슬리버폴리곤이생성하거나인위적인갭을제거하지못하여오히려자료의무결성을저하시키는것으로나타났다. 또한산림기능구분도의축적이 1:25,000 임을감안할때, 단순병합방법이적절할것으로판단된다. 그러나위의두방법에서시스템적으로규정할수없는오류에대해서는육안검수작업이필요하다. 위의방법의적용은산림기능구분뿐만아니라유역단위의분석시유역구분도오류를보정하기위한방안모색에의미가있다. (a) (b) (c) (d) (e) (f) Figure 1. Comparison by methods; (a), (d) : raw data, (b), (e) : simple merge, (c), (f) : topology using gaps

224 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 인용문헌 (References) Korea Forest Service Statistical yearbook of foresty. Süleyman S. M., Hakan H. M., Bahadır A., Erdem E. M. and Ferruh Y Topological error correction of GIS vector data. International Journal of the Physical Sciences Vol. 5(5), pp , May

225 June 7-8, 2011 Abundance of butterflies in new grassland formed by forest fire and in forest Kwon, Tae-Sung 1, Sung-Soo Kim 3, Joo-Hoon Lim 2, Cheol-Min Lee 1*, and Sei-Woong Choi 4 1 Department of Forest Ecology Research, Korea Forest Research Institute, Seoul, , Republic of Korea 2 Department of Forest Conservation, Korea Forest Research Institute, Seoul, , Republic of Korea 3 Research Institute for East Asian Environment and Biology, , Amsa 3 dong, Gangdong-gu, Seoul, Republic of Korea 4 Department of Environmental Education, Mokpo National University, Muan, Jeonnam , Republic of Korea Abstract kr-chulmin@daum.net Forest fire is an important variable in forest ecosystem succession and is a dominant large scale disturbance factor. We investigated butterfly in grassland formed by forest fire, secondary forest, and primary forest, to estimate the impacts of forest fire in ecosystem. Butterfly was counted using the line transect method once a month from April to September for three years 2007 to Butterfly habitats were determined based on their larval habitats and classified as either forest inside (FI), forest edge (FE), or grassland (GL). The survival strategy of butterfly based on the numbers of voltinism and food plants was classified into two groups: specialist and generalist. A total of 93 species and 1,833 butterflies were observed in three habitats. Grassland (G) with 56 species and 572 individuals, secondary forest (F) with 57 species and 585 individuals and primary forest (Fp) with 51 species and 676 individuals were observed, respectively. Number of species and individuals of butterflies (G, F, Fp) in the three habitats were different according to habitat types. Survival strategy was also different among habitats. The pattern between number of species and individuals was different. It is thought that species composition of butterfly is similar from early stage of succession to secondary forest due to good movement ability and is not greatly affected by habitat destruction caused by forest fire. In contrast, abundance was obviously different between three sites. It is likely that difference of abundance relate to habitat reduction caused by forest fire. In primary forest, specialists and tree-feeding species increased

226 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 요약 산불은산림천이과정에중요한역할을하고, 중요한교란요인이다. 우리는산불에의해형성되초지, 이차림, 일차림에서생태게에미치는나비의영향을파악하기위해조사하였다. 나비는선조사법으로조사하였고, 2007 년부터 2009 년 3 년동안월 1 회조사하였다. 유충의서식처에따라산림, 임연부, 초지의 3 가지로구분하였다. 나비의생존전략은발생회수, 먹이식물의수에의해 specialist 와 generalist 로구분하였다. 총 93 종 1,833 개체가관찰되었고, 초지에서 56 종 572 개체, 이차림에서 57 종 585 개체, 일차림에서 51 개체, 676 개체가발견되었다. 나비의종수와개체수는서식처에따라달랐으며, 생존전략역시달랐다. 초지와이차림사이에종구성이별로달라지지않았는데이는높은이동성때문으로사료된다. 일차림에서는 specialists 와나무를먹는종의비율이높았다. Introduction Forest damages caused by various disturbances have increased due to climate change. Since 2000, typhoons with strong winds and heavy rain caused frequent occurrence of landslide and induced increase of tree mortality. In 2009, hundreds of thousands of pine trees in southern region of Korea withered in the drought in winter season. The dry spring has been increasing occurrence of forest fire. The gap in forest created by disturbance develops herbs and shrubs instead of trees. Such change of forest structure and vegetation may change fauna in forest. Many studies on the diversity of butterflies have been conducted in various types of forest. However, little is known about how butterflies respond to change of forest structure and vegetation caused by forest fires. After forest fire, trees were destructed and herbs and shrubs grow afterwards in forest fire areas. In this process we expect that species that favor grassland increases while forest species decreases. However, as forest recovers, the situation will be again recovered with increasing of forest species. Similarly just after forest fire, the number of generalist that adopt in broad habitats and fast fliers will increase. However, as forest grows, we expect that the number of specialists that require rather limited habitat increases whereas the generalist decreases. We posed a hypothesis that the butterfly fauna will be changed after forest fire and succession. To test this, we surveyed the butterfly fauna in three different sites in grassland, secondary forest and primary forest

227 June 7-8, 2011 This study addresses the following questions. Did the herb and shrub feeding butterflies increase after forest fire? Did the generalist common in forest fire area and specialist recover as forest grows? Materials and Methods Study sites We selected three different types of habitats to compare butterfly fauna: grassland (G), secondary forest (F) and primary forest (Fp). Grassland sites, located in Mt. Hyunjongsan, Uljin, were changed from pine dominated forest into grassland by severe forest fires in 2000 and Secondary forest sites, Gilgokri, Uljin were covered with about 30 year-old pine (Pinus densiflora) trees. Primary forest sites, located in Gwangneung, were covered with year-old deciduous trees with Quercus serrata Thub. Ex. Murray and Carpinus laxiflora (Siebold and Zucc) Blume. Butterfly monitoring The abundances of butterfly species were counted using the line transect method described by Pollard and Yates (1993) or/and Yamamoto (1975). All surveys were conducted on cloudless days and began at 10:00-14:00. The survey was conducted by steadily walking along the study routes at a speed of approximately 2 km/hour, and then recording butterflies observed within a width of 10 m along the routes. When identification of the species by sight was difficult, the butterflies were caught by net, identified and then released. We conducted the survey once a month from April to September for three years 2007 to Classification of habitat and butterflies Butterfly habitats were determined based on their larval habitats and classified as either forest inside (FI), forest edge (FE), or grassland (GL). If a species occupied more than one habitat, their habitat was considered to be the one that was more frequently utilized. The habitats of each species were primarily determined using SS Kim s personal observations over approximately twenty years with the supportive data from Fukuda et al. ( ). The survival strategy of butterfly was followed by Kitahara and Fujii (1997). They classified the butterflies based on the numbers of voltinism and food plants:

228 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 specialist species are uni- or bivoltine and the number of food plant is less than 10 in one plant family, while generalist species are multivoltine and more than 10 food plant species in more than two families. However our limited information on food plants of butterfly species in Korea, we only considered the number of plant families for the classification excluding the number of food plant species. In the present study we categorized specialist species as uni- or bivoltine and limited in one families of food plant and generalist species as multivoltine and more than two families in food plants. Analysis We analyzed the relationship between variables using log-linear models for multidimensional contingency tables (Zar, 1997). The relationship between habitat and survival strategy was analyzed using number of species. The relationship between habitat (FI, FE, GL) and surveyed sites (G, F, Fp) were analyzed using number of species and individuals. Differences among surveyed sites, (1) G (grassland, Uljin) and F (secondary forest, Uljin), (2) G and Fp (primary forest, Gwangneung), (3) F and Fp, (4) G, F and Fp, respectively were analyzed with habitat types. Each analysis was carried out separately in three years and pooled three years. All statistical analyses were using Statistica ver 6.1. Results A total of 93 species and 1,833 butterflies were observed in three habitats in three years. Grassland (G) with 56 species and 572 individuals, secondary forest (F) with 57 species and 585 individuals and primary forest (Fp) with 51 species and 676 individuals were observed, respectively. In 2007 when forest fire broken out, the number of butterfly species in G with 29 species and 137 individuals was lower than that of F was 34 species and 184 individuals. However in 2008, the number of butterflies was increased in grassland (G) with 46 species and 276 individuals while secondary forest (F) had 45 species and 217 individuals. In 2008, the number of individuals was highest in all three surveyed sites (Fig. 1)

229 June 7-8, 2011 Figure 1. Occurrence of butterfly species according to the habitat types. (A) annual occurrence in number of species, (B) pooled occurrence in number of species, (C) annual occurrence in number of individuals, and (D) pooled occurrence in number of individuals. G: grassland after forest fire in Uljin, F: pine forest in Uljin, and Fp: primary forest in Gwangneung. Among 93 species identified in the present study, GL species (37 species) was the largest in number, followed by FI (28 species) and FE (28 species). According to the survival strategy, 10 species were classified into generalist and 31 species were into specialist. Here we excluded the species in the analysis that we cannot classify. It was shown that butterfly habitat and survival strategy were not independent. Eleven species of FI were specialist, while four species of FE were generalist and 10 species were specialist. Among GL, six species were generalist and 10 species were specialist. Survival strategy was also different among habitats. The pattern between number of species and individuals was different. In the number of species, specialist was larger than generalist in the three habitats. However, in the number of individuals large number of generalist and small number of specialist were observed in G and F (Fig. 2)

230 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Figure 2. Occurrence of butterfly species according to the life strategies. (A) annual occurrence in number of species, (B) pooled occurrence in number of species, (C) annual occurrence in number of individuals, and (D) pooled occurrence in number of individuals. G: grassland after forest fire in Uljin, F: pine forest in Uljin, and Fp: primary forest in Gwangneung. Discussion We expected that assemblage structure of butterfly was changed according to change of forest structure and vegetation caused by forest fire. However, species composition of habitat and survival strategy of butterfly between grassland formed by forest fire and secondary forest was not different. But, species composition of primary forest site differed from one of two sites. It is thought that species composition of butterfly is similar from early stage of succession to secondary forest due to good movement ability and is not greatly affected by habitat destruction caused by forest fire. In contrast, abundance was obviously different between three sites. It is likely that difference of abundance relate to habitat reduction caused by forest fire. Given difference of butterfly in primary forest, many specialists, which have strong habitat preference, using trees as feeding plant is due to inhabiting, compared to grassland and secondary forest. This study suggests that species composition of butterfly is not affected by forest fire, whereas abundance is affected

231 June 7-8, 2011 References Kitahara, M. and K., Fujii An island biogeographical approach to the analysis of butterfly community patterns in newly designed parks. Res. Popul. Ecol. 39: Yamamoto, Y Notes on the methods of belt transact census of butterflies. Jour. Fac. Sci. Hokkdaido Univ. Ser. VI. Zool. 20(1): Fukuda, H., E. Hana, T. Kuzuya, A. Takahashi, B. Tanaka, H. Tanaka, M. Wakabashi and Y. Watanabe The life histories of butterflies in Japan vol. I-IV. Hoikusha Co. Osaka. (in Japanese with English summary) Zar, J.H Biostatistical Analysis. Prentice Hall International, Inc., Englewood Cliffs, New Jersey. 663p

232 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Influences of disturbance intensity on diversity and abundance of arthropod predators (Araneae, Carabidae, Staphylinidae, and Formicidae) in fired-pine forest, Korea Abstract Kwon, Tae-Sung 1, Joo-Hoon Lim 2, Cheol-Min Lee 1* 1 Department of Forest Ecology Research, Korea Forest Research Institute, Seoul, , Republic of Korea 2 Department of Forest Conservation, Korea Forest Research Institute, Seoul, , Republic of Korea kr-chulmin@daum.net Forest fires are one of the most important natural disturbances in forest ecosystem. Recently, due to increase of mega forest fire caused by climate change, it needs to estimate impact of forest fire in ecosystem. Intensities of disturbance in the study sites were characterized as follows: 0 (no fire), 1 (weak fire), 2 (strong fire), 3 (strong fire, human disturbance). Arthropod predators (Araneae, Carabidae, Staphylinidae, and Formicidae) were collected using pitfall traps and compared. A total of 124,009 individuals belonging to 169 species were collected in 12 study sites. The species richness of arthropod predators collected were 124 species in Araneae, followed by Formicidae (18 species), Carabidae (16 species), and Staphlinidae (11 species). Formicidae was most abundant, representing 65% of total, followed by Araneae (28%), Staphlinidae (5%), and Carabidae (2%). For analysis, Arthropod predators were divided by guilds. Araneae was assigned into two groups: wandering spiders and web spiders. Carabidae and Formicidae were assigned into two groups: forest species and grassland species.. This study suggests that Formicidae is very suitable bioindicator for short term survey than Araneae, Staphlinidae, and Carabidae. 요약 산불은산림생태계에서가장중요한교란요인이다. 최근에기후변화로인한대형산불이증가함으로서, 산불이산림생태계에미치는영향을평가할필요가증가하고있다. 조사지의교란강도는 0( 산불미발생 ), 1( 약한산불 ), 3( 강한산불 ), 4( 강한산불, 인간의교란 ) 의 4 단계로구분하였다. 절족동물중포식자

233 June 7-8, 2011 인거미류, 개미류, 딱정벌레류, 반날개류를함정트랩으로채집하여비교하였다. 총 12 개조사지에서 169 종 개체가채집되었다. 종풍부도는개미류가 124 종, 개미류 18 종, 딱정벌레과 16 종, 반날개과 11 종이었다. 풍부도는개미류가가장높아전체개체수의 65% 를차지하였고, 거미류 28%, 반날개과 5%, 딱벙벌레과 2% 였다. 분석을위하여, 포식자를기능군으로구분하였다. 거미류는배회성과조망성, 딱정벌레과와개미류는산림성과초지성으로나누었다. 본조사결과개미류가단기성조사에서는거미류, 반날개과, 딱정벌레과에비해보다적합한환경지표생물이었다. Introduction Forest fire in forest ecosystem is one of the most important factors of natural disturbance. Fires cause serious disturbance of forest ecosystem and economic damage. Forest fire damages many species living in forest due to burning vegetation and litter, and changing soil component. For example, ground beetles were more abundant in unburned area than in burned area. Soil microarthropods were more affected by destruction of environment than direct damage caused by fire. However, change of species composition of the forest vegetation caused by fires increases biodiversity and habitat heterogeneity in forest. Some studies reported that forest fire positively affected species component. Moretti et al. (2004) found that species richness of ground beetles and spiders increased with fire frequency in the southern Alps. In addition, as disturbance intensity increased, abundance of detritivores decreased, but herbivores increased. Recently, forest fires tend to increase following global climate change. After the mid-1990s, mega forest fires are occurring more often in Korea. Several countries such as Australia, America, Greece, and Russia were greatly damaged by mega forest fire. The study on impacts of climate change on fire activity ongoing in Canada since the late 1980s predicted the increase of the fire weather and fire seasons in 21C (Podur and Wotton, 2010) Therefore, it is important to estimate disturbance caused by forest fire. Arthropod predators, such as Araneae, Carabidae, Staphylinidae, and Formicidae, have often been used as bioindicator for urbanization, fragmentation of forest, change of land use. Araneae depends on microclimate factors and physical structure of the habitat than species richness of vegetation. It was known that Araneae was affected by herb, litter, and soil layers after forest fire. Araneae assemblages showed significant difference according to vertical structure and spatial composition of a habitat because it was divided into wandering spider and

234 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 web spider according to predation method. Carabidae was known that wingless large species in disturbed area disappeared faster than small species. In addition, Carabidae has strong habitat preference and some opportunistic species quickly colonize the disturbed area. Staphylinidae has clear systematic character and responds sensitively to change of environment. Furthermore, specie component of Staphylinidae differs according to forest maturity. Formicidae has been used as indicator of various disturbances such as forest, clear cutting, and land use. Advantage of Formicidae is not affect by seasonal change and has strong habitat preference. However, there is not comparison study of 4 taxa as bioindicators, which is more useful. Therefore, to do effective environment assessment, 4 taxa need to be compared. Due to successful reforestation in Korea, combustibles such as litters, dead branches and woods have accumulated in forest. After mid-1990s, mega forest fires have been frequently occurring. In general, reforestation of burned forest has been applied for recovery of burned forests. However, after the mega fire in 2000, natural regeneration of vegetation was suggested for recovery of the burned areas. When traditional reforestation is carried out, dead trees and branches are logged and small trees and shrubs are removed prior to reforestation. From view of disturbance, the process of reforestation is additional disturbance. This study aims to evaluate which is better bioindicator according to disturbance intensity in forest. Arthropod predators (Araneae, Carabidae, Staphylinidae, and Formicidae) were compared among four disturbance intensities (no fire, weak fire, strong fire, and strong fire followed by reforestation) in three burned areas. Materials and Methods Study sites and disturbance degree This study was carried out in three burned areas (Goseong, Gangneung, and Samcheok) in the north-eastern coast region of South Korea. A total of 12 study sites (4 study sites in each of 3 burned areas) were selected for this study. Disturbance intensity was characterized as one of four disturbance degrees (hereafter DD). DD of unburned pine forest was 0, and DD after a weak fire was 1. DD after a strong fire was 2, and DD after a strong fire followed by reforestation was 3. Arthropods sampling Arthropods were collected by pitfall traps of plastic cup (depth 6.3 cm,

235 June 7-8, 2011 mouth diameter 8 cm, bottom diameter 6 cm). Twenty pitfall traps were buried at each study site for 10 d from late May The survey period is considered best for sampling of arthropods since arthropods are active and abundant in this high temperature and low rainfall season from late May to early June in Korea. For 10 days, each of 20 traps was placed 5 m apart from adjacent trap along two lines, which were parallel away about 10 m. Ethylene glycol was used as a preservative, and it filled one-third of each trap. Arthropod predators (Araneae, Carabidae, Staphylinidae, and Formicidae) were identified to species or morphospecies level using the taxonomic keys under a stereomicroscope. All the specimens were deposited in the Insect Specimen Storage Room of the Korea Forest Research Institute. Guilds of arthropod predators Arthropod predators, Araneae, Carabidae, Staphylinidae, and Formicidae, were used for analysis. Araneae was assigned into two groups: wandering spiders and web spiders. Carabidae and Formicidae were assigned into two groups: forest species and grassland species. However, Staphylinidae itself was used as one guild due to little information of lifecycle. Statistical analyses The number of arthropod individuals collected a trap was log transformed (ln N+1) to reduce variance and was used for analysis. Tukey-Kramer s HSD multiple comparison was used to find difference in species richness, abundance, and guild of arthropod predators according to DD. Correlation analysis between guilds of arthropod predators and environmental factors was performed. Multidimensional scaling (MDS) based on the Sorensen distance measure was used for two-dimensional ordination of arthropod predators, guild, Araneae, Carabidae, Staphylinidae, and Formicidae. Multi-response permutation procedures (MRPP) of a nonparametric protocol for testing the null hypothesis of no differences among arthropod predators, guild, Araneae, Carabidae, Staphylinidae, and Formicidae according to disturbance degree (DD) were used. MDS ordination and MRPP were conducted using PC-ORD (ver. 5.17) (McCune and Mefford, 1999), and the other analyses were performed using STATISTICA (Statsoft 2004). Results A total of 124,009 individuals belonging to 169 species were collected in 12 study sites. The species richness of arthropod predators collected were 124 species

236 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 in Araneae, followed by Formicidae (18 species), Carabidae (16 species), and Staphlinidae (11 species). Formicidae was most abundant, representing 65% of total, followed by Araneae (28%), Staphlinidae (5%), and Carabidae (2%). When arthropod predators analyzed were compared among the four disturbance degrees, specie richness and abundance of Araneae, Carabidae, Staphylinidae, and Formicidae did not show significant difference (P > 0.05, Fig 1). Fig. 2 showed MDS ordination of Araneae, Carabidae, Staphylinidae, and Formicidae. MDS ordination explained 86% of variation of Araneae and Formicidae, followed by Carabidae (78%), and Staphylinidae (33%). However, only MRPP of Formicidae was statistically significant (P = 0.038). Figure 1. Species richness and number (ln N+1) of individual of arthropod predators (Araneae, Carabidae, Staphylinidae, and Formicidae) collected in pitfall traps in 4 different disturbance degrees. The bars indicate the SE. Figure 2. MDS ordination of arthropod predators (Araneae, Carabidae, Staphylinidae, and Formicidae) according to disturbance degree (DD). Circle: unburned site in pine forest (DD = 0), open triangle: site in pine forest burned by weak fire (DD = 1), dark triangle: site burned by strong fire (DD = 2), and reversed dark triangle: site burned by strong fire and reforested afterward (DD = 3). MDS ordination of arthropod predators (Araneae, Carabidae, Staphylinidae, and Formicidae) used only species collected at more than 2 sites

237 June 7-8, 2011 Discussion MDS ordination of arthropod predators explained above 78% of variation except for Staphylinidae. However, in only Formicidae, sites were located according to four disturbance degrees. Araneae and Carabidae were recognized as useful bioindicators in many studies such as forest fire, urbanization, fragmentation. However, they need to do long term survey due to change of species component according to seasonal change. In contrast, Formicidae does not change species component according to seasonal change and has strong habitat preference. This study therefore suggests that Formicidae is very suitable bioindicator for short term survey than Araneae, Staphlinidae, and Carabidae. References Podur, J. and M. Wotton Will climate change overwhelm fire management capacity? Ecological Modelling 221: Moretti, M., M.K. Obrist and P. Duelli Arthropod biodiversity after forest fires: winners and losers in the winter fire regime of the sothern Alps. Ecography 27, McCune B. and M.J. Mefford PC-ORD. Multivariate analysis of ecological data, Version 4.0. MjM Software Design, Gleneden Beach

238 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Vegetation Type of Pinus densiflora stands and Damage Characteristics of Oak Stands in Post- forest fire Area in Korea 산불피해지소나무식생유형및참나무류피해특성 KwangSoo Lee 1*, SuYoung Jung 2, SukKwon Kim 1, Sangwon Bae 1 KyungJae Lee 1, HyunCheol Shin 1, NamChang Park 1, EunJi Bae 1, and JoonHyung Park 1 1 Korea Forest Research Institute, Seoul , Rep. of Korea 2 The Research Forests, Gyeongsang National University, Jinju , Rep. of Korea Beldel660@forest.go.kr Abstract This study was carried out to obtain basic database to manage damaged stands by forest fire by classifying vegetation type in pine forests and analyzing damage characteristics and sprout appearance in oak forests in Korea. Six vegetation units were surveyed in southern pine forest area of Gangwon where huge forest fires frequently occurred at. Vegetation communities were analyzed as five communities in pine stands such as Quercus variabilise community, Betula davurica community, Acer pseudosieboldianum community, Sasa borealis community, and typical (Pinus densiflora) community and in cork oak communities there were two groups such as Acer pseudosieboldianum group and pure cork oak forest group. This study found that there were significant differences of stands growth and their quality caused by community characteristics in tree layer of pine forests and oak tree stands of Quercus serrata Quercus mongolica, etc. were dominant species in subtree layer of damaged stands. Considering damage characteristics of oak tree stands caused by forest fire, there were four types of aboveground survivals, partial-aboveground survivals, soil surface residuals, and soil surface loss, and as their damage types there were significant differences in occurrence and growth of sprouting. As considering stand structure and species composition, post-fire treatment for damaged stands needs to properly control the pine forest in dominant position for up level and manage sprouting of oak trees for low level

239 June 7-8, 2011 요약 본연구는한국의산불피해지에서소나무임분의식생유형을구분하고참나무류의피해특성과맹아발생특성을분석하여산불피해지의임분관리를위한기초자료를확립하고자하였다. 대형산불피해가주로발생하는강원남부지역소나무류는 6 개의식생단위로나타났으며, 식생유형은굴참나무군락, 물박달나무군락, 당단풍나무군락, 조릿대군락, 전형군락등 5 개군락으로구분되었으며, 굴참나무군락은당단풍군과전형군으로구분되었다. 군락에따라임분의생장및형질도차이가발생하고있는것으로나타났으며, 아교목층이하에는졸참나무, 신갈나무등참나류가우점하고있는것으로나타났다. 산불피해지참나무류임분의피해특성은지상부생존형, 지상부일부잔존형, 지제부잔존형, 지제부손실형등 4 가지유형으로나타났으며, 피해특성에따라맹아발생과생장도차이가있는것으로나타났다. 따라서산불피해지에서임분관리는상층소나무와하층에서발생하는참나무류의맹아에대한지속적인관리가요구되었다. Introduction Gangwon-do area in Korea was continuously damaged by forest fire caused by site environmental characteristics. Once forest fire occurs at this area, it s easy to be an enormous disaster since this area consists of pine trees. As these pine forests are mostly consist of pure and artificial stands, they have high stand density and weak structure against forest fire. This study was investigated to obtain basic data for the purpose of post-fire survivor management by classifying vegetation type in pine forests and analyzing damage characteristics and sprout appearances in oak forests. Methodology Sample plots were selected at the eastern coastal area of Korea. Firstly, vegetation type in pine forests without forest fire damages was analyzed by using Braun-Blanquet method. Secondly, damage characteristics in post-fire pine forests and sprouts occurrence in post-fire oak stands were analyzed, respectively

240 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Results and Discussion Figure 1. Species importance value by vegetation units in Pinus densiflora forests (Pd : Pinus densiflora, Qm : Quercus mongolica, Qv : Quercus variabilis, Qs : Quercus serrata, Bd : Betula davurica, Ta : Tilia amurensis, Pl : Prunus leveilleana) Table 1. Description of sprout decay of oak trees in burned and unburned forest area Stand classification Post-fire tree survival stands Degree of sprout decay (%) Species Total Severe Moderate Sound Q. variabilis Q. mongolica Non-fire stands Q. mongolica References C.P. Weatherspoon and C.N. Skinner An Assessment of Facts Associated with Damage to Tree Crowns from the 1987 Wildfires in Northern California. Forest Science 41, No. 3, pp J. Englin, P.C. Boxall, K. Chakraborty, and D.O Valuing the Impacts of Forest Fires on Backcountry Forest Recreation Watson. Forest Science 42, No. 4, pp J. D. Stuart, M.C. Grifantini, and L. Fox Ⅲ Early Successional Pathways Following Wildfire and Subsequent Silvicultural Treatment in Douglas-Fir/ Hard- wood Forests, NW California. Forest Science 39, No. 3, pp

241 June 7-8, 2011 Comparison of Difference between Gwangneung and Mt. Gumsan of Soil Respiration, Litterfall Production and Decomposing Litter in Natural Broad-leaved Forest 천연활엽수림에서의토양호흡과낙엽생산량및분해력의지역간차이비교 Im-Kyun LEE 1*, Choonsig KIM 2 1 Department of Forest Conservation, Korea Forest Research Institute, 57 Hoegiro, Dong daemun-gu, Seoul, , Korea 2 Department of Forest Resources, Gyeongnam National University of Science and Technology, Jinju-si, Gyeongsangnam-do, , Korea iklee@forest.go.kr Abstract This study was carried out to estimate annual soil respiration and litterfall production and to measure litter decomposition at the two LTER sites installed in a deciduous broadleaf natural forest in Gwangneung (central cool temperate forest) and Mt. Gumsan (transitional zone from southern cool temperate to warm temperate forest) where belong to different biome, South Korea. We investigated the soil respiration for one year and measured litterfall production for two years, and measured decomposing litter for three years at the two LTER sites in Gwangneung and Mt. Gumsan. In the both two areas, soil respiration showed a clear seasonal variation, increased from April through July, peaked in August, and decreased thereafter. Also, seasonal trends of soil temperature were similar to soil respiration in the both two areas. These trends indicated a strong correlation between soil respiration and soil temperature. Annual soil respiration were 10.6 t CO 2 ha -1 for Gwangneung and 6.9 t CO 2 ha -1 for Mt. Gumsan, respectively. In the both two areas, litterfall production were significantly different among the sampling dates, whereas it was not significantly different between the years. The total annual mean litterfall production for two years were 5,934 kg/ha/yr for Gwangneung and 5,189 kg/ha/yr for Mt. Gumsan, respectively, and leaf litter accounted 63.2% for Gwangneung and 66.4% for Mt

242 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Gumsan, respectively. In the both areas, the leaf litter quantities were highest in Quercus serrata, followed by Carpinus laxiflora and C. cordata, etc., which are dominant tree species in the site. In the both two areas, the mass loss from the decomposition of leaf litter were fastest in C. cordata, followed by C. laxiflora and Q. serrata. 100% of litter for C. cordata, 98.7% for C. laxiflora, 85.3% for Q. serrata decomposed for 1,217 days in Gwangneung, while 84.1% of litter for C. cordata, 82.4% for C. laxiflora, 65.4% for Q. serrata decomposed for 1,070 days in Mt. Gumsan. Introduction Soils are the largest carbon pool in terrestrial ecosystems, containing m ore than two-thirds of total carbon in the terrestrial ecosystems. Soil respirati on (belowground respiration) is the major pathway of carbon transfer from so il to atmosphere, and a tiny amount of change in soil respiration rate may h ave profound impact on the atmospheric CO 2 budget, thus understanding soil respiration is crucial for the carbon balance of terrestrial ecosystems and for the global carbon balance (Ewel et al., 1987; Son et al., 2004). Litterfall inputs and litter decomposition represent a large and dynamic portion of the nutrient cycling in a forest ecosystem (You et al., 2000). In addition, the turnover of litter is a major pathway of the nutrient and carbon inputs to forest soils. Significant amounts of organic matter and nutrients in the soils can be transferred during litter decomposition processes (Lisanework and Michelsen, 1994). Natural hardwood stands in the temperate forest zone and warm temperate forest zone of the Korea are mixed with various kinds of deciduous tree species (Kim et al. 2003). Although several studies have reported soil respiration, litterfall inputs and litter decomposition in hardwood forest ecosystem in Korea (Kim et al., 1997; Kim et al., 2003; Koo, 2006), little is known about the direction and rates of change associated with mixed-hardwood forest ecosystem. The objectives of this study were 1) to estimate annual soil respiration, 2) to measure litterfall and nutrient quality; 3) to examine decomposition rates in Quercus serrata, Carpinus laxiflora and C. cordata litter at the LTER sites in Gwangneung and Mt. Gumsan deciduous broad-leaved forest in Korea

243 June 7-8, 2011 Methodology This study was conducted in the two broadleaf deciduous forest in Gwangneung, Gyeonggi-do, and in Mt. Gumsan, Gyeongnam-do, South Korea. These areas had been registered as an official LTER sites since 1998 and an official KoFlux site since 2002 (Kim et al. 2003). Gwangneung forest (37 44'45.7 N, '01.0 E) was located in the central part of the cool-temperate forest zone in Korea, and the soils were classified as brown forest soils developed on granite gneiss. Annual precipitation in the site averages 1,365mm and annual mean temperature is Major tree species are composed of Quercus serrata, Carpinus laxiflora, C. cordata, Sorbus alnifolia, Celtis jessoensis (Lim et al. 2003). Mt. Geumsan forest (34 45'41.4 N, '12.9 E) was located in the southern coastal area and belongs to warm temperate forest zone in Korea and the soils were classified as brown forest soils developed on granite gneiss. Annual mean temperature reaches 10.8 and annual precipitation is about 1394mm. Major tree species are composed of Quercus serrata, Chamaecyparis obtusa, Styrax japonica, Acer pseudo-sieboldianum, Carpinus tschonoskii, Stewertia pseudo-camellia. Soil respiration was measured using an infrared gas analyzer (EGM-4, PP System Inc., UK) connected to a SRC-1 chamber (100mm inside diameter) from April through October The chamber was gently inserted into a forest floor to a depth of less than 0.5cm, without the use of previously installed collars. Since insertion may cause a burst of CO 2 from the soil, measurements did not begin until a few minutes had elapsed from the placement of the chamber. Five randomly selected sample points were measured in each plot. Soil temperature was measured with a soil temperature probe (PP Systems Inc.) at 15cm depth, and soil moisture was measured gravimetrically adjacent to the chamber. Litterfall was collected in circular traps devised by Hughes et al. (1987) using 1.5mm nylon net. The collecting area was 0.25m 2. Twelve traps in three plots (20 20m) were installed 50cm above ground. Litter was collected at approximately monthly intervals from October 2005 to September Litter collected from each trap was transported to the laboratory and oven-dried at 65 for 48 hours. All dried samples were separated into leaf, bark, flowers, acorn, woody and miscellaneous components, and each portion was weighed. Mass loss and nutrient release in decomposing litter were estimated using the litterbag techniques employing 30 30cm nylon bags with a 1.5mm mesh size. Fresh leaf litter from the site was collected during the heavy litterfall season (late

244 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 November) in Collected litter samples were dried to constant mass at room temperature for 14 days and sorted into representative deciduous foliage in the stands. Ten grams of litter of air-dried three dominant four species (Q. serrata, C. laxiflora, and C. cordata) was weighed to the nearest 0.01g and places in numbered litterbags. Subsamples from each litter type were also taken to determine the oven-dried mass at 65 for 48 hours. The litterbags were randomly place on the forest floor on 24 December The thirty-six bags (3 plots 4 species 3 replication) in each sampling time were collected on six occasions over the study period. Collected bags were oven-dried at 65 for 48 hours. Litter in the bag was cleaned by gentle brushing with a soft paintbrush to remove mineral soil and weighed to determine litter mass loss rates. Litterfall and litter in the bag were ground in a Wiley mill to pass a 40-mesh stainless steel sieve. All nutrients (N, P, K, Ca, Mg) were analyzed by the standard method of the National Institute of Agriculture Science and Technology (1988). Results and Discussion Soil respiration Fig. 1. Seasonal soil temperature (left), and soil respiration rate (right) for two years in Gwangneung and Guemsan forest

245 June 7-8, 2011 Litterfall inputs Fig. 2. Seasonal patterns of litterfall production for three years in Gwangneung and Guemsan forest. Litterfall inputs References Fig. 3. Remaining mass of leaf litter for about three years in Gwangneung and Guemsan forest. Ewel, K.C., W.P. Cropper and H.L. Gholz Soil CO 2 evolution in Florida slash pine plantations: Ⅱ. Importance of root respiration. Canadian Journal of Forest Research 17: Kim, C., J.H. Lim and J.H. Shin Nutrient dynamics in litterfall and decomposing leaf litter at the Kwangneung deciduous broad-leaved natural forest. Kor. Jour. of Agri. and For. Meteolo. 5(2): Lim, J.H., J.H. Shin, G.Z. Jin, J.H. Chun and J.S. Oh Forest stand structure, site characteristics and carbon budget of the Kwangnung natural forest in Korea. Kor. Jour. of Agri. and For. Meteolo. 5(2): Son, Y., Y. C. Jun, Y. Y. Lee, R. H. Kim, and S. Y. Yang Soil carbon dioxide evolution, litter decomposition, and nitrogen availability four years after thinning in a Japanese larch plantation. Communications in Soil Science and Plant Analysis 35(7-8):

246 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Community fluctuation and Functional feeding groups of Benthic macroinvertebrates in Forest fire area of Samcheok 삼척산불지의저서성대형무척추동물군집변동과섭식기능군에관한연구 Myung-Eun Kim 1, Eul-Won Seo 1, Joo-Hoon Lim 2, Jong-Eun Lee 1* 1 Department of Biology, Andong National University, Andong, , Korea. 2 Department of Forest Conservation, Korea Forest Research Institute, Seoul, , Republic of Korea Abstract jelee@andong.ac.kr This study was conducted for searching the community structure, biological evaluation of water quality, community analysis and functional feeding groups (FFGs) of benthic macroinvertebrates in forest fire area of Samcheok. 요약 본연구는삼척산불지역계류에서식하는저서성대형무척추동물을이용하여산불후, 이들분류군의군집구조, 생물학적수질평가, 군집분석및섭식기능군을분석하였다. Introduction Benthic macroinvertebrates are common inhabitants of lakes and streams where they are important in moving energy through food webs. The term "benthic" means "bottom-living", so these organisms usually inhabit bottom substrates for at least part of their life cycle; the prefix "macro" indicates that these organisms are retained by mesh sizes of ~ mm (Rosenbergand Resh, 1993). Benthic macroinvertebrate species are differentially sensitive to many biotic and abiotic factors in their environment. Consequently, macroinvertebrate community structure has commonly been used as an indicator of the condition of

247 June 7-8, 2011 an aquatic system (Armitage et al., 1983; Rosenberg and Resh, 1993). Biotic index systems have been developed which give numerical scores to specific "indicator" organisms at a particular taxonomic level (Armitage et al., 1983). Such organisms have specific requirements in terms of physical and chemical conditions. Changes in presence/absence, numbers, morphology, physiology or behaviour of these organisms can indicate that the physical and/or chemical conditions are outside their preferred limits (Rosenberg and Resh, 1993). Presence of numerous families of highly tolerant organisms usually indicates poor water quality (Hynes, 1998). The most diverse group of freshwater benthic macroinvertebrates is the aquatic insects, which account for ~70% of known species of major groups of aquatic macroinvertebrates in Korea. Thus, as a highly diverse group, benthic macroinvertebrates are excellent candidates for studies of changes in biodiversity of forest fire area. Methodology 1. Collection and Identification Collecting method was done by qualitative and quantitative method. Larvae fixed with 70% Ethyl alcohol in the lab for two days and then preserved in 80% ethyl alcohol. All specimen which used in this study were preserved in animal taxonomic laboratory of Andong National University. 2. Community analysis 1) Dominance index(di) = (n 1 +n 2 )/N (N : Individual number, n1, n2 : Dominant and subdominant species individual number) 2) Diversity index(h`) = - (n i /N log2n i /N) (n i : i s individual number, N : Individual number) 3) Evenness index(e ) = H`/log 2 S (H` : Diversity index, S : Species number) 4) Richness index(ri) = (S-1)/Ln(N) (S : Species number, N : Individual number) 5) Functional feeding group(ffg) and Community stability Merritt-Cummins(1996) and Ro(2004) 6) Ecological score of benthic macroinvertebrate community(esb) = (S i Q i ) Q i : Environmental quality score of individual taxa S i : Species frequency to i environmental quality

248 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Results and Discussion 1. After the occurrence of forest fire, fluctuation in species and individuals investigated by year, 61 species, individuals in And in 2007, 71 species, individuals, in 2008, 58 species, 1,563 individuals. 2. After the occurrence of forest fire, dominance index showed the tendency to decreased every year. It showed that forest ecosystem restoration and valley ecosystem stabilization. 3. After the occurrence of forest fire, diversity index showed the tendency to increased and a high Qi score of species increased every year. And appearance ratio of biological indicator species was steadily increased. 4. As a result of functional feeding group by year, it showed a trend that individuals of GC type which is a functional group picking up and eating FPOM (fine particulate organic matter) from deposits in the bottom of water or benthic areas and performs an important function of material circulation in ecosystem decrease every year. Fig. 1. Species and individual number of functional feeding groups of benthic macroinvertebrates at the surveyed sites. References Margalef, R Information theory in ecology. Gen. Sys. 3, Merritt. R. W. and K. W. Cummins An introduction to the aquatic insects of North America, second Edition. Kendall/Hunt Publishing Co. Dubuque

249 June 7-8, 2011 Iowa. Edmunds G. F., S. L. Jensen and L. Berner, The mayflies of North and Central America. Univ. Minnesota Press, Minneapolis. 330 pp.. Pielou, E.C An introduction to mathematical ecology. New York, 165 pp. Ro, T. H. and D. J. Chun Functional feeding group categorization of Korean immature aquatic insects and community stability analysis. Korean J. Limnol. 37, Shannon, C. E. and W. Weaver The Mathematical Theory of Communication. University of Illinois Press, Urbana. Won, D. H., S. J. Kwon and Y. C. Jeon Aquatic insect of Korea. 415 pp, Korea ecosystem service press. Seoul

250 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Community stability of Benthic macroinvertebrates in Forest fired area of Uljin 울진산불지의저서성대형무척추동물의군집안정성분석 Hyeok-Young Kwon 1, Eul-Won Seo 1, Jin-Young Park 2, Joo-Hoon Lim 3, Jong-Eun Lee 1* 1 Department of Biology, Andong National University, Andong, , Korea. 2 Department of Applied Biology, Kyungpook National University, Sangju, , Korea 3 Department of Forest Conservation, Korea Forest Research Institute, Seoul, , Republic of Korea jelee@andong.ac.kr Abstract This study is to provide basic data on the effects of the forest fire on the ecology and recovery aspects by performing analysis on the comparison of community analysis, functional feeding group and community stability by understanding benthic macroinvertebrate s dispersion. 요약 본연구는울진산불지역의저서성대형무척추동물의군집분석, 섭식기능군및군집안정성분석을통해산불후, 계류생태계의복원양상에대해연구하였다. Introduction Forest Fire has had tremendous effects on the ecological succession processes, allochemical characteristics of soil and water quality of the nearby ecology and animals such as benthic macroinvertbrates. Benthic macroinvertbrates are river creatures that have variety of species. These creatures have a very small migratory pattern. It is easy to collect these specimens, and they are used in various ecological studies. Especially, aquatic

251 June 7-8, 2011 insects utilize various food resources in fresh water ecology. Therefore, functional feeding group s classification is suggested according to the food utilization characteristics. Water environment can be diagnosed based on the functional feeding group s make up. Therefore, to understand drastic changes in the mountainous species ecology due to the forest fires and to understand recovery processes, continuous and systematic investigation must be done on the recovery pattern and aspect of the ecosystem by monitoring composition of species, H, RI, Dominant species and community fluctuation of the benthic macroinvertbrates which has excellent adaptability according to the different habitat and of the special water environment. Methodology 1. Collection and Identification Collecting method was done by qualitative and quantitative method. Larvae fixed with 70% Ethyl alcohol in the lab for two days and then preserved in 80% ethyl alcohol. All specimen which used in this study were preserved in animal taxonomic laboratory of Andong National University. 2. Community analysis 1) Dominance index(di) = (n 1 +n 2 )/N (N : Individual number, n1, n2 : Dominant and subdominant species individual number) 2) Diversity index(h`) = - (n i /N log2n i /N) (n i : i s individual number, N : Individual number) 3) Evenness index(e ) = H`/log 2 S (H` : Diversity index, S : Species number) 4) Richness index(ri) = (S-1)/Ln(N) (S : Species number, N : Individual number) 5) Functional feeding group(ffg) and Community stability Merritt-Cummins(1996) and Ro(2004) 6) Ecological score of benthic macroinvertebrate community(esb) = (S i Q i ) Q i : Environmental quality score of individual taxa S i : Species frequency to i environmental quality

252 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Results and Discussion 1. As a result of examining an aspect of changes in species and individuals of E.P.T. taxa investigated in experimental area by year, Ephemeroptera was 21 species, inds./ m2, Plecoptera was 3 species, inds./ m2, and Trichoptera 14 species, inds./ m2 in And in 2009, Ephemeroptera was 9 species, 296 inds./ m2, Trichoptera was 4 species, 44.4 inds./ m2, and Plecoptera was 0 species, showing that species and individuals belonging to E.P.T. taxa decrease rapidly every year. Fig. 1. Fluctuation of E.P.T. taxa of species and individuals number by year. 2. As a result of community analysis by year, in 2007 just after the occurrence of forest fire, it showed that H' and RI relatively high in experimental area compared with control area, but in 2008 when The water system started to be influenced by the fire directly, it showed a trend that H' and RI decreased in the experimental area. 3. As a result of functional feeding groups (FFGs) by year, it showed a trend that species and individuals of GC type which is a functional group picking up and eating FPOM (fine particulate organic material) from deposits in the bottom of water or benthic areas and performs an important function of material circulation in riparian ecosystem decrease every year. 4. As a result of community stability by year, there was no significant difference in community stability between the experimental water system and control water

253 June 7-8, 2011 system in 2007, but as an environment of water system in forest fire area started to be somewhat destroyed, from 2008, it is shown that both species in I areaa which have great ability of resistance and recovery and species in II II area whichh live in relatively stable water system decreased a little. Fig. 2. Analysis of stability factors relative resistance and resilience in the control and experimental area (in each year). *Black : 2007, Red : 2008, Blue :2009 References Lee, J, E. and J. S. Kim Community fluctuation of benthic macroinvertebrates from forest fire area and non fores fire area. Korean J. Environmental Research. 1, Margalef, R Information theory in ecology. Gen. Sys. 3, Merritt. R. W. and K. W. Cummins An introduction to the aquatic insects of North America, second Edition. Kendall/ /Hunt Publishing Co. Dubuque. Iowa. McNaughton, S. J Relationship among functional properties of california grassland. Nature 216, Pielou, E. C Shannon's formula as a measure of specific diversity, its use and misuse. Amer. Nat. 100, Ro, T. H. and D. J. Chun Functional feeding group categorization of Korean immature aquatic insects and community stability analysis. Korean J. Limnol. 37, Shannon, C. E. and W. Weaver The Mathematical Theory of Communication. University of Illinois Press, Urbana. Won, D. H., S. J. Kwon and Y. C. Jeon Aquatic insect of Korea. 415 pp, Korea ecosystem service press. Seoul

254 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Forest Fire Risk Mapping Based on Analytic Hierarchy Process 분석적계층절차에기반한산불위험도지도화 Changjae Kim 1*, Shanmuganathan Jayakumar 2, Jaehoon Jung 1, Seongsam Kim 1, Soohee Han 1, Donghyun Kim 3, Joon Heo 1 1 Geomatics and Remote Sensing Lab, School of Civil and Environmental Engineering, College of Engineering, Yonsei University, Seoul, Korea 2 Department of Ecology and Environmental Sciences, School of Life Sciences, Pondicherry University, Puducherry, India 3 Division of Forest Disaster Management, Department of Forest Conservation, Korea Forest Research Institute, Seoul, Korea Abstract changjaekim@yonsei.ac.kr Forest fires have adverse ecological, economic, and social impacts. In this light, the present research aimed to construct a fire risk model using a GIS-based multi-criteria analysis. Analytic Hierarchy Process (AHP) was applied, and the experimental results were obtained from Indian expert group (IEG). Moreover, producer s accuracy, user s accuracy, and overall accuracy were introduced in order to compensate for the limitations of the evaluation method that most researchers use. Statistical evaluations revealed that the approach of the IEG together with the equal interval thresholding method provided the fire risk mapping result: 94.25% overall accuracy. Therefore, this paper proposes a robust forest fire risk mapping strategy. 요약 산불은생물적, 경제적, 그리고경제적으로상당한영향을미치고있다. 이런의미에서본연구는 GIS에기반한복수조건분석을이용하여산불위험도를지도화하는데그목적을두고있다. 분석적계층절차방법이적용되었으며인도산림전문가그룹의의견에기반한실험결과가획득되었다. 더구나, 기존

255 June 7-8, 2011 다수의연구에서사용되었던정확도평가방법의한계를극복하기위하여생산자정확도, 사용자정확도, 전체정확도가사용되었다. 인도산림전문가그룹의의견에기반한실험결과를평가한결과본연구는 94.25% 의높은전체정확도를나타내었다. 이에본연구는확고한산불위험도지도화전략을제공하였다. 서론 (Introduction) 산불은물리적, 생태적, 그리고환경적으로지대한영향을미치는재난이다. 지속적인화재의발생은숲의손실을발생시키는주요문제요소들중의하나이다. 산불을막기위해서기존에는화재감시타워나수작업의화재통제시스템을운영하였다. 그러나, 이는매우비효율적이며비용이많이소요된다. 이런의미에서산불위험도지도화는효율적이고실용적인화재모니터링및관리를위해서필수적인요소로등장하고있다. 화재위험모델링은화재에영향을미치는다양한요소들에기반한매우복잡한과정을거친다. 그러나, 원격탐사자료와지리정보시스템 (GIS) 을이용하게되면보다효과적으로이과정을수행할수있게된다. 본연구의목적은다음과같이정리할수있다. 첫째, 지리정보시스템에기반하여화재위험도모델링을수행한다. 둘째, 산림전문가의의견에기반하여산불에미치는요소들의점수화와가중치부여를수행한다. 셋째, 기존정확도평가방법의한계를극복하는다른평가방법을사용한다. 이를통해본연구에서제시한산불위험도지도화방법의정확도와신뢰도를확인한다. 본연구는우선아래의절에서는연구대상지에대한설명과화재에영향을미치는요소들, 그리고이들을이용하여화재위험도모델링을수행하는구체적인방법론에대해설명한다. 이후획득된결과를분석하여제시된화재위험도지도화방법이신뢰할만한정확도를보유하는지확인한다. 최종적으로본연구를통해획득한결과물들에대해결론에서정리한다

256 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 방법론 (Methodology) 대상지역은인도의 Namakkal district of Tamil Nadu에있는 Eastern Ghats 언덕이선택되었다. 이지역은해발고도가 180미터부터 1415미터에이르는지역으로그면적으로 503 제곱킬로미터에달한다. 지면기울기는평탄한지역부터 40도에이르는지역까지다양하다. 산불에영향을미치는주요다섯가지요소들은경사향, 경사도, 식생분포및밀도, 거주지및도로가있다. 이들요소들이만드는 GIS 레이어를이용하하여산림위험도를지도화하기위하여다음과같은식 (1) 을이용하였다. 5 FRV = W i S i (1) i= 1 여기서 FRV 는화재위험도, W i 는 i번째레이어의가중치, Si 는 i번째레 이어내의카테고리의화재위험지수이다. 레이어의가중치와카테고리의화재 위험지수를정량화하기위해서분석적계층절차방법 (AHP) 이적용되었다. 이 분석적계층절차방법은정량화문제에있어서발생할수있는주관성과불일 치성을해결하는데기여한다. 인도산림전문가 15 명을대상으로계층절차방 법을적용하였으며이를통해 3 명만이이방법을통해화재위험지수의정량화 에있어서객관성과일치성을보이는것으로분석되었다. 실험결과및분석 (Experimental Results and Analysis) 인도산림전문가의의견을바탕으로지도화한산불위험도는다음의그림 1( 좌 ) 에서보는바와같다. 그위험도는 Very-high, High, Medium, Low, Verylow의다섯단계로구분하였다. Very-high는전체면적의 5.87%, High는 52.31%, Medium은 30.13%, Low는 9.83%, 그리고 Very-low는 1.85% 를차지하였다

257 June 7-8, 2011 그림 1. 계산된산불위험도지도 ( 좌 ) 와실제산불발생지도 ( 우 ) 획득된산불위험도지도의정확도를평가하기위해실제산불발생지도 ( 그림 1( 우 )) 와정량작인비교를수행하였다. 산불에가장치명적인 very-high 위험지역만을기준으로아래의표 1에서보는바와같이생산자정확도, 사용자정확도, 그리고전체정확도를계산하였다. 전체정확도가 94.25% 에달하는높은정확도의결과를보여주고있다. 표 1. 생산자정확도, 사용자정확도, 전체정확도 생산자정확도 (%) 사용자정확도 (%) 전체정확도 (%) 결론 (Conclusions) 본연구는지리정보시스템을기반으로하여화재위험도를모델링하였다. 특히, 산불레이어들사이의상태적인중요도평가를통하여이를수행하였다. 이의평가를위해분석적계층절차방법이도입되었다. 인도타밀지역이연구

258 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 대상지로선정되었으며인도산림전문가들의의견을바탕으로산불위험도지도화를수행하였다. 정확도평가결과전체정확도가 94.25% 에이르렀으며, 이는본연구를통하여신뢰도높은산불위험도지도화방법을제시하였음을의미한다. 인용문헌 (References) Bahuguna, V.K., and Singh, S., (2002). Fire situation in India. International Forest Fire News (IFFN), 26: Brown, J.K. & Smith, J.K., eds. (2000). Wildland fire in ecosystems: effects of fire on flora. Gen. Tech. Rep. RMRS-GTR-42, Vol.2. Ogden, Utah: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station. 257 pp. Erten, E., Kurgun, V. and Musaoglu N., (2004) Forest fire risk zone mapping from satellite imagery and GIS a case study, ISPRS, Istanbul. Jaiswal, R.K., Mukherjee, S., Raju, K.D. and Sexena, R. (2002). Forest fire risk zone mapping from satellite imagery and GIS. International Journal of Applied Earth Observation and Geoinformation, 4: Leblon, B., (2005). Monitoring Forest Fire Danger with Remote Sensing, Natural Hazards, 35: Mbow, C., Goita, K. and Benie, G., (2004). Spectral indices and fire behavior simulation for fire risk assessment in savanna ecosystems. Remote Sensing of Environment, 91, Saxena, A. and Srivastava, P. (2007). Integrating biophysical characters, microclimate and human factors in forest fire risk modeling, In 4th International Wildland Fire Conference, Seville, Spain, May 2007 ( 2007/contributions/doc/SESIONES_TEMATICAS/ST1/Saxena_Srivastava_ 2_INDIA.pdf) Setiawan, I., Mahmud, A.R., Mansor, S., Mohamed Shariff A.R., and Nuruddin, A.A., (2004). GIS-grid-based and multi-criteria analysis for identifying and mapping peat swamp forest fire hazard in Pahang, Malaysia, Disaster Prevention and Management, 13:

259 June 7-8, 2011 Changes of Species Diversity on Moth Communities at Forest Fire Region in Samcheok, Korea Yang Seop Bae 1, Do Young Chae 1, Young Don Ju 1, Jung Hoon Bae 2, Jong Myung Kim 3, Neung Ho Ahn 4, Cheol Min Lee 5*, Tae Sung Kwon 5 1 Division of Life Sciences, College of Natural Sciences, University of Incheon, Korea 2 Institute of CHAM Ecology, Korea 3 Ecosystem Survey Team, National Institute of Environmental Research, Korea 4 Exhibition and Education Division, National Institute of Biological Resources, Korea 5 Division of Forest Ecology, Korea Forest Research Institute, Korea kr-chulmin@daum.net Abstract We investigated moth communities in Samcheok, Korea, to elucidate the influence of the East Coast Fire in Moths were collected with UV light traps 4 times a year from 2006 to 2009 at 3 sites: Unburned area (Site 1), Surface fire area (Site 2), and Crown fire area (Site 3) sites, respectively. A total of 3,804 individuals belonging to 727 species from 30 families were captured at the 3 study 1,727 individuals of 505 species (27 families) at Site 1, 1,193 individuals of 353 species (24 families) at Site 2, and 885 individuals of 340 species (25 families) at Site 3. Species richness and individuals of Noctuidae, Pyralidae, Geometridae, and Tortricidae were the highest at all 3 sites. All of species richness, individuals, and species diversity (H') were the highest in Site 1 for 4 years. The index of similarity (α-index) was higher between Site 2 and 3 in 2006, while higher between Site 1 and 2 in Moths were grouped into 16 categories by hosts of the larval stage. Individuals of moths of herbaceous feeding habit continuously decreased at Site 3, but proportion of individuals of tree feeding habits continuously increased. The present results showed that moth communities was relatively recovered according to vegetation change after forest fire and regeneration of moth communities in surface fire area was more faster than in crown fire area

260 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 요약 2000 년동해산불의영향을파악하기위해삼척에서나방군집을조사하였다. 나방은적외선등을이용하여 3 개조사지에서 2006 년부터 2009 년사이에연 4 회조사를하였다. 총 30 과 727 종 3804 개체가채집되었는데, 1 조사지에서 27 과 505 종 1727 개체, 2 조사지에서 24 과 353 종 1193 개체가채집되었다. 종수와개체수가많았던과는 Noctuidae, Pyralidae, Geometridae, Tortricidae 였다. 종수, 개체수, 종다양도는조사지 1 에서가장높았다. 유사도지수는 2006 년에조사지 2 와 3 조사지사이와 2009 년의조사지 1 과 2 사이에높았다. 나방류를유충의먹이식물에따라 16 개종류로구분하였다. 초본을먹은나방의개체수는조사지 3 에서지속적으로감소한데비해, 나무를먹는종은증가하였다. 산불후식생이회복됨에따라나방군집도회복되며, 지표화피해지에서회복속도가빠른것으로나타났다. Introduction Forest fire is one of major disturbance in forest ecosystem and determines landscape pattern due to creating various habitat in forest. However, forest fire causes physical, chemical, and biological change such as decrease of biomass, surface runoff according to soil erosion, and decline of water storage function. In Korea, mega forest fires occur intensively in the east coast region. In April 2000, mega forest fire burned 23,794 ha including Goseong, Gangneung, Samcheok, and Uljin, and was recorded by the largest fire in Korea. Vegetation in burned area in east coast region is recovering quickly. However, except for birds, little is known about how wildlife responds to vegetation recovery. Moths have abundant species richness and play important role such as herbivore, pollinator, and feed resource of small animals in forest ecosystem. Moths are useful bioindicator to estimate forest environment due to various feeding habit, such as evergreen tree, deciduous tree, lichen, moss, mushroom, etc., in larval stage. In this study, we investigated moth communities at unburned area, surface fire area, and crown fire area in Samcheok, Korea, to elucidate the influence of the East Coast Fire in In addition, change of moth communities was analyzed using feeding habit of larval stage of species collected

261 June 7-8, 2011 Materials and Methods In Samcheok damaged by East coast fire, moths were collected with UV light traps 4 times a year from 2006 to 2009 at 3 sites: Unburned area (Site 1), Surface fire area (Site 2), and Crown fire area (Site 3) sites, respectively. Moths were collected with UV light traps (Bioquip, U.S.A). Each traps consisted of a 22- W UV light of O type, powered by 12 V DC batteries. Moth sampling was conducted for 3 hours after dusk, avoiding the rainy day and the full moon day. Moths were identified at the species level when possible, based on available taxonomic keys and literature. When moths could not be identified on appearance, it was identified after dissection of genitalia. All material is now deposited in the Insect Specimen Storage Room of division of life sciences, college of natural sciences, university of Incheon, Korea. Analysis Community indices We calculated species diversity by the Shannon and Weaver s index of diversity, H (Shannon and Weaver, 1949) as follows: H = - ((n i / N) log 2 (n i / N)) where n i is the number of individuals of the ith species, N is the total number of individuals of all component species. To group the moth communities from three sites according to the similarity in species composition or communities structure, we conducted UPGMA cluster analyses based on Pianka s α index (Pianka, 1973) as follows: α = ( (n 1i n 2i )) / (( (n 1i / N 1 ) 2 (n 2i / N 2 ) 2 ) 1/2 N 1 N 2 ) where N 1 and N 2 represent the total number of individuals or species in landscape components 1 and 2, n 1i and n 2i represent the number of individuals or species of ith species or group in Site 1 and 2, respectively

262 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Feeding habit of larval stage Feeding habit of larval stage of moth collected was classified into 16 categories such as polypahgy, evergreen and deciduous trees, evergree tree, deciduous tree, grass plant, climbing plant, bamboo, fern, moss, lichen, mushroom, cereal, fruit, falling leaf, comb, and unknown (Kim et al., 1982; Inoue et al., 1982; Sugi et al., 1987). Results A total of 3,804 individuals belonging to 727 species from 30 families were captured at the 3 study 1,727 individuals of 505 species (27 families) at Site 1, 1,193 individuals of 353 species (24 families) at Site 2, and 885 individuals of 340 species (25 families) at Site 3. Species richness and individuals of Noctuidae, Pyralidae, Geometridae, and Tortricidae were the highest at all 3 sites. All of species richness, individuals, and species diversity (H') were the highest in Site 1 for 4 years. The index of similarity (α-index) was higher between Site 2 and 3 in 2006, while higher between Site 1 and 2 in 2009 (Fig. 1). Figure 1. Dendrograms showing the results of UPGMA cluster analysis based on Pianka s overlap index α among moth communities of 3 study sites in Samcheok, Korea in 2006 and

263 June 7-8, 2011 Figure 2. Species richness and individuals of moths of 1 different feeding habit at 3 study sites in Samcheok, Korea from 2006 to Moths were grouped into 16 categories by hosts of the larval stage. Individuals of moths of herbaceous feeding habit continuously decreased at Site 3, but proportion of individuals of tree feeding habits continuously increased (Fig. 2). Discussion The results revealed that species component, dominant species, and species diversity of moths showed relatively clear difference between unburned area, surface fire area, and crown fire area. However, the present results showed that moth communities was relatively recovered according to vegetation change after forest fire and regeneration of moth communities in surface fire area was more faster than in crown fire area. It is thought that moths are useful bioindicator to estimate change of forest environment due to having various feeding habit in larval stage. However, for a more detailed understanding of moths as indicator, feeding habit, lifecycle, and movement ability of each species need to be studied more

264 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 References Pianka, E.R The structure of lizard communities. A. Rev. ecol. Syst. 4: Shannon, C.E. and W. Weaver A Mathematical Model of Communication. Urbana, IL: University of Illinois Press. Kim, C.W., S.H. Nam and S.M. Lee Illustrated flora and fauna of Korea vol. 26 Insecta (VIII). Ministry of education republic of Korea. 919pp. Sugi, S., K. Nakatomi, R. Sato, H. Nakajima and M. Owada Larvae of larger moths in Japan. Kodansha. 453pp

265 June 7-8, 2011 A study on the possibility of Melia azedarach L. of planting in sterile land 멀구슬나무의척박지식재가능성에관한연구 Bon-Hag Koo 1 *, Choong-Hwa Lee 1, Chong-Min Park 2 1 KFRI, 57 hoegi-ro, Dongdaemun-gu, Seoul, , Korea 2 CBNU, 567 paekchede-ro, Deokjin-gu, Jeonju, , Korea darklina@hanmail.net Abstract This study was analyzed the possibility of planting Melia azedarach in sterile land. Seeds propagation made use of seed coat, non-seed coat and air drying. Soil made use of mixed perlite and vermiculite. Without gap of keeping and soil made germination over 75%. Growth was seen maximum height of 1.2m and double storied forests an annual. 요약 본연구는척박지에서멀구슬나무의식재가능성을분석하였다. 종자번식은종피, 종피제거, 양건법을이용하였으며, 토양은펄라이트와버미큘라이트를혼합하여사용하였다. 보관방법과토양의차이없이 75% 이상발아가이루어졌으며, 생장은 1 년에최대 1.2m 의수고와복층림을보였다. Introduction Melia azedarach was originated from Asia was introduced from Japan to Korea. Now, it grow in Jeju-do, Gyeongsangnam-do and Jeollanam-do. The tree is used as furniture tree, street tree, landscape tree and fruit is used as medicinal(tae-uk Kim, 2002; Seong-Cheon Hong, Su-Hyun Byun, Sam-Sik Kim, 1987). Study of Coast Salt Affected Environment(Do-Gyun Kim, 2010), it was relatively strong in salty wind and was distributed in inland area as Nepal(Yong

266 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 ha Kim, 2009). These conditions presume application to sterile environment. Methodology Seed was classified seed coat, non-seed coat and air drying. After storing in ground except air drying, the others carried sowing out in April. A group organized five pot and irrigated 500 ml without non-nutrient. Each group mixed 1:1 rate and made Soil : Perlite(Group A), Soil : Vermiculite(Group B). Results and Discussion It was different among groups and gradually decreased difference In July. After July, so all of group rapidly increased density that it was difficult to measurement. Table.1. Germination rate by processing method in July Group Germination(%) Seed coat Group A Soil + Perlite Non-seed coat Air drying Group B Soil + Vermiculite Non-seed Seed coat Air drying coat July Height measured average, minimum and maximum. Although it was different among groups, it shortly grew in summer. Especially, air drying highly showed average and maximum. Table.2. Growth by processing method and time Height (cm) Group Seed coat Group A Soil + Perlite Non-seed coat Air drying Group B Soil + Vermiculite Non-seed Seed coat Air drying coat July September Height( cm) : Average / Min Max

267 June 7-8, 2011 Germination rate and growth was not closely related. Regardless of keeping of seed, however, it came to rate and fast growth. Study concluded that Melia azedarach suited for double storied forests and fast greening. References Yong-ha Kim A Freld Guide to Trees & Shrubs. Korea National Arboretum Tae-Uk Kim. The woody plants of korea. Kyohak Publishing Seong-Cheon Hong, Su-Hyun Byun, Sam-Sik Kim Colored Illustrations of Trees & Shrubs in Korea Do-Gyun Kim Native Tree Species of Tolerance to Saline Soil and Salt Spray Drift at the Coastal Forests in the West-Sea, Korea. Korean journal of environment and ecology. 24(2)

268 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Method of manufacturing soil-map of natural regeneration site after forest fire, Goseong. 산불피해이후고성지역자연복원지에대한산림토양도작성 Jooho Hwang 1*, Taeho Jeon 1, Joohoon Lim 2, Kyung Won Seo 2 1 Tae-Eun Research Institute, Namyangju , Korea 2 Department of forest conservation, Korea Forest Research Institute, Seoul, , Korea hwangjooho@naver.com Abstract This study was carried out to manufacture soil map of compartment 309, sub-compartment 2, Yang-yang working plan area located in Injeong-ri, Jugwangmyeon, Goseong-gun, Gangwon-do, Korea where natural regeneration site after forest fire. First, soil information and profile survey were conducted in order to obtain information of reconnaissance soil map and soil type for the site. ArcGIS 9.3 program was used for the DEM(Digital Elevation Model) analysis and manufacturing soil map. After DEM was formed with digital map, an altitude, bearing and slope were analyzed. The result showed that the altitude was a little low and comparatively lots of an incline facing north topographies were observed. The slope of 0~20 showed distribution of 44.9ha accounting for 66.4%. Soil map showed that eroded soil was still distributed in 42.6ha, 63% of 67.6ha. The accumulated data after continuous research in the future will be used as preliminary data for the study of natural regeneration site after forest fire. 요약 본연구는산불발생이후자연복원을실시한강원도고성군간성읍인정리에위치한양양경영계획구 309 임반 2 소반의산림토양도를작성하기위해실시하였다. 우선대상지의입지구획과토양형을결정하기위해개황조사와토양단면조사를실시하였으며, 자료분석과산림토양도는 ArcGIS 9.3 프로그램을사용하여작성하였다. 수치지형도를이용하여 DEM 을생성하고표고, 방위, 경사를분석하였다. 분석결과표고는다소낮은것으로나타났으며, 비교적북사

269 June 7-8, 2011 면의지형이많았다. 또한경사분석결과 0~20 의비율이 66.4% 로 44.9ha 정도분포하는것으로나타났다. 산림토양도작성결과대상지면적의 63% 에달하는 42.6ha 가여전히침식토양이분포하는것으로나타났다. 앞으로조사가지속적으로진행되어자료가축적되면자연복원지의산림토양의회복과정을연구하는데필요한기초자료로활용될것이다. 서론 (Introduction) 1996 년고성산불이발생하기전까지산불피해지의복원정책은피해지전면적에조림을하는것이었다. 하지만 1996 년고성산불직후자연복원의필요성이처음제기되었고고성군죽왕면에자연복원연구지역이처음으로설정되었다. 또한 2000 년동해안산불이후복원방법에관해최초로본격적인사회적논의가이루어지게되었다 ( 환경부, 2006). 산불피해는단순히임목의피해뿐만이아니라산림토양에도직간접적으로영향을주어유실, 교란, 간섭등으로인해생산성이감소하게된다. 임목의생장에영향을미치는여러가지요인이있다. 이중에서지형및토양인자를포함하는입지환경인자는입목생장에결정적으로영향을주는것으로알려져있다 ( 김태훈등, 1988, 1991). 본연구는산불발생이후자연복원을실시한지 10 년이지난현재의산림토양분포를파악하기위해산림토양도를작성하는데의의가있다. 따라서자연복원지의산림토양의회복과정을규명하는데필요한기초자료로활용될것이다. 재료및방법 (Methodology) 1. 연구대상지. 본연구의대상지는산불발생이후자연복원을실시한강원도고성군죽왕면인정리에위치한양양경영계획구 309 임반 2 소반으로선정하였으며 ( 그림 1), 대상지의면적은 67.6ha 이고, 모암이조립질화강함으로이루어진지역이다

270 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 그림 1. 연구대상지 2. 수치지형도 본연구에서사용한수치지형도는국립지리원에서제작된 1:5000 축적의수치지도를사용하였으며, 수치지형도를이용하여 ArcGIS 9.3 프로그램으로수치표고모형 (Digital Elevation Model) 을생성하였다. 생성된 DEM 을이용하여대상지및표준지의표고, 방위, 경사값을분석하였다. 3. 현지조사자료 연구대상지의산림토양도를작성하기위해 2011 년 3~4 월에개황조사및산림토양조사를실시하였다. 개황조사자료로산림토양도의잠정구획을실시하였으며, 산림토양조사로 14 개의표준지를선정하여입지환경조사및토양단면조사를하였으며, 간략조사를실시하였다. 조사결과는 ArcGIS 9.3 프로그램으로산림토양도작성시 Point 자료로구축하였다. 4. 산림토양도작성 산림토양도작성시 ArcGIS 9.3 프로그램을사용하였다. 항공사진과개황조사자료그리고수치지형도를이용하여토양형별구획을실시하였으며, 각각의토양형별속성은현지조사자료를이용하여토양형을결정하였다

271 June 7-8, 2011 결과및고찰 (Results and Discussion) 1. 표고, 방위, 경사분석결과 대상지에대한표고, 방위경사를 ArcGIS 9.3 으로분석한결과는그림 2 와같다. 표고 100m 이하인면적이 45.0ha 이며, 100m 이상인면적이 22.6ha 로대상지전체면적에대한비율은 66.6% 와 33.4% 인것으로나타났다 ( 그림 2). 방위는북, 북동, 북서사면의비율이 48.1% 로 32.5ha 이며다음으로는남, 남동, 남서사면이 30.2% 로 20.4ha 에걸쳐분포하고있다 ( 그림 3). 경사분석결과완경사지와경사지의비율이 66.4% 로 44.9ha 를차지하고있다. 반면절험지는 1.8% 로국소적으로나타났다 ( 그림 4). 그림 2. 대상지표고분석결과 그림 3. 대상지방위분석결과 그림 4. 대상지경사분석결과 분석결과를이용하여조사표준지의표고, 방위, 경사값을획득하였다. 조사표준지의표고는 42~98m 까지분포하였다. 경사는 0~32.7 까지분포한다

272 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 방위는 13.4~357.5 까지분포하고, 무방위인표준지가 1 개소로나타났다. 2. 산림토양도작성 그림 5. 조사표준지 토양형에따른대상지의산림토양도작성결과는그림 3 과같이나타났으며, 토양형의분포비율은다음표 2 와같다. 침식토양의면적이 42.6ha 로전체면적 67.6ha 중에 63% 분포하고있으며갈색산림토양의비율은 34.3% 이다. 그림 6. 고성산불피해지산림입지토양도작성결과

273 June 7-8, 2011 표 2. 토양형별분포면적및비율 토양형 B₁ B₂ B₃ Er₁ Er₂ Er-c Im R 비율 (%) 면적 (ha) 고찰 본연구에서는산불이발생한후 10 년간자연복원을실시한지역의산림토양도를작성하였다. 산림토양도작성결과여전히침식토양이넓게분포하고있어앞으로도지속적인조사와관리방안이필요하다. 앞으로복구방법별로조사자료가축적되어, 복구방법에따른연도별토양분포의변화를비교한다면산불피해지의산림토양복구및관리방법과회복과정을연구하는데중요한기초자료가될것으로판단된다. 인용문헌 (References) 김태훈등 산림토양분류에관한연구. 입업연구원연구보고 37: 김태훈, 이충화, 구교상, 정진현 토양형별주요수종의생장. 입업연구원연구보고 42: 환경부 산불피해지식생회복및토양안정성증진기술개발

274 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Effects of edge and burn severity on post-fire vegetation regeneration 산불피해지식생재생에대한가장자리와피해강도효과 Joomee Lee 1, Myoungsoo Won 2, Hyunjoo Lee 1, Sangwoo Lee 3* 1 Graduate Program, Dept. of Environmental Science,Konkuk University 2 Korea Forest Research Institute, Seoul, , Korea 3 Dept. of Environmental Science, Konkuk University swl7311@konkuk.ac.kr Abstract This study aims to investigate the effects of edges and burn severity on the post-fire vegetation regeneration. Damaged areas of Samchiuck fire in 2000 was selected as the study areas, and we adopt 2 (high and low burn severity)Ⅹ2 (edge areas and interior areas) study design to compare the differences in post-fire vegetation regeneration. To measure burn severity and vegetation regeneration, delta Normalized Burn Ratio(ΔNBR) and delta Normalized Difference Vegetation Index (ΔNDVI) were adopted respectively. GRIDs (1kmⅩ1km), the unit of analysis, were generated and overlaid covering the entire study areas in GIS. Edge areas were identified as GRIDs touching boundary of damaged areas. The study results indicated that there was a statistically significant difference in vegetation regeneration due to difference in burn severity and edges. The results also suggested that GRIDs with high burn severity and edge areas showed the highest vegetation regeneration while GRIDs with low burn severity and interior areas. 요약 이연구는삼척산불피해지를대상으로피해강도 (burn severity) 와가장자리효과 (edge effects) 에따른식생의재생력에대한영향을분석하고자하였다. 피해강도와가장자리효과를분석하기위해 2( 저강도피해지역과고강도피해지역 )x 2( 피해지내부지역과가장자리내역 ) 을실험디자인으로채택하였다. 산불피해강도는인공위성영상을기초로하여정규탄화지수변화값

275 June 7-8, 2011 (dnbr) 과정규식생지수변화값 (dndv) 지수를이용하여평가하였다. 이결과에기초하여피해강도와가장자리효과가초기식생재생변화에주는영향과상관관계를비교분석하였다. 연구결과초기식생재생을나타내는 dndvi 값은강한피해강도가높은피해지역에서높으며, 피해지가장자리지역일수록더높은것으로나타났다. 즉가장자리지역이내부지역보다재생이활발하게이루지는것으로나타났다. 서론 (Introduction) 산불은산림생태계와인간의삶에중요한의미를갖는다 (Gustafson et al., 2004; Nunes et al., 2005). 따라서산림과학분야의연구자특히산림관리자, 산림정책입안자에의해중요한연구분야로인식되어왔다. 산불에의한식생에대한다양한변화는거의대부분이피해강도에의한것이다. 한편산불로생성된피해지가장자리지역은기온, 풍속, 상대습도와같은환경변화요소에많은영향을받는다 (Ranney, 1977; Chen et al.,1992) 년발생한삼척피해지를본연구는피해강도 (burn severity) 와가장자리효과 (edge effects) 에따른식생재생변화를분석하고, 연소강도와가장자리효과의복합적영향이식생의재생력에미치는영향을분석하였다. 재료및방법 (Methodology) 본연구에서는한국동해안에위치하고있는 2000 년발생한삼척피해지를연구대상지로선정하였다. 피해지산불강도를분석하기위해서 ERDAS 9.2 program 을사용하여다중분광밴드영상 (Landsat TM) 을기초로산불피해이전과이후의정규식생지수변화값 (dndvi) 및정규탄화지수변화값 (dnbr) 을측정하였다. 피해후초기식생재생을분석하기위해 년영상을이용하여정규식생지수변화값 (dndvi) 을계산하였다. 가장자리효과분석을위해 ArcviewGIS 3.3 를사용하였으며, 1x1km 격자를피해지전체를포함하도록생성하였다. 피해강도와가장자리효과를분석하기위해 2( 저강도피해지역과고강도피해지역 )x 2( 피해지내부지역과가장자리내역 ) 을실험디자인으로채택하였다. 피해강도에따른초기식생변화와피해강도와가장자리효과의복합적영향에따른초기식생재생변화를각각 T-test 와 Pearson 의상관분석을통해분석하였다

276 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 결과및고찰 (Results and Discussion) 산불강도 ( 저강도피해지역과고강도피해지역 ) 별식생재생차이 (Table 1) 를비교한결과식생재생변화량은피해강도가높은지역에서 로나타났다. 즉피해강도가낮은지역 (-0.037) 의비해식생재생변화량이상당히높은경향을보였다. 따라서고강도지역에서식생재생이빠르게일어난것으로나타났다. 두표본간의평균차이에대한통계량은 로유의수준 5% 에서통계적으로유의한차이가있는것으로나타났다. 이는높은피해강도에서식생의소멸로인공위성에낮아진식생반사율이감지된것이원인이며, 같은원리로인해낮은피해강도에서는상반된결과를보였다. 가장자리효과는 T-test 분석으로가장자리지역과내부지역의 dndvi 가치가통계적으로서로차이가나는것으로평가되었다. Table 2 에따르면, T-test 한결과가장자리지역의초기식생재생변화량은 0.01 으로산림내부지역의초기식생재생변화량보다 높은것으로나타났다. 그러나이는유의수준 5% 에서유의확률 0.18 을보여통계적으로유의수준은비교적낮은것으로나타났다 Table 3. T-test on dndviaccording to burn severity Variable Burn severity Mean t-value P High burn severity dndvi Low burn severity Total number of plots=23l, Number of plots in high burn severity= 117, Number of plots in low burn severity= 114 Table 4. T-test on dndviaccording to forest region Variable Forest region Mean t-value P Forest edge dndvi Forest interior Total number of plots=23l, Number of plots in forest edge= 165, Number of plots in forest interior= 66 Table 3 은피해강도와가장자리효과의복합적영향을분석하기위해 2x2 교차분석에대한일원배치분산분석결과이다. 결과에따르면고강도가장자리지역에속하는그룹 D 가 로재생변화량이가장높은것으로나타났다. 다음으로고강도내부지역인그룹 C(0.049), 그룹 B(-0.036), 그룹 A( ) 순으로나타났다. 즉초기식생재생변화량을나타내는 dndvi 값은피해강도가높은피해지역에서높게나타났으며, 피해지가장자리지역일수록더활성화되는것으로나타났다. 따라서식생의재생변화에피해강도의영향이크게작용하며, 이에가장자리효과가복합적으로작용하여초기식생재생변화량에영향을주는것

277 June 7-8, 2011 으로나타났다. Table 5. One-way ANOVA for four groups N Mean Std. Dev Min Max A B C D Total A: 저강도-내부지역, B: 저강도-가장자리지역, C: 고강도-내부지역, D: 고강도-가장자리지역 인용문헌 (References) Chen, J., and Franklin, J. F., Vegetation responses to edge environments in old-growth douglas-fire forests, Ecological Applications, Vol. 2(4): Gustafson, E. J., Zollner, P. A., Sturtevant, B. R., He, H. S., and Mladenoff, D. J., Influence of forest management alternatives and land type on susceptibility to fire in northern Wisconsin, U.S.A, Landscape Ecology, Vol. 19: Nunes, M. C. S., Vasconcelos, M. J., Pereira, J. M. C., Dasgupta, N., Alldredge, R. J., and Rego, F. C., Land cover type and fire in Portugal: Do fires burn land cover selectively?, Landscape Ecology, Vol. 20: Ranney, J. W., Forest island edges-their structure, development, and importance to regional forest ecosystem dynamics. Environmental Sciences Division Publication Number 1069.Oak Ridge National Laboratory, Oak Ridge, Tennessee, U.S.A

278 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Assessing Canopy Fuel Characteristics for Major Conifer Species in Korea 우리나라주요침엽수종에대한수관층연료특성평가 Sung Yong Kim¹ *, Yeon Ok Seo¹, Roscinto Ian C. Lumbres¹, Mi Na Jang¹, Gun Ik Lee¹, Byungdoo Lee², Kyongha Kim², Sung Ho Kim², Young Jin Lee 1 1 Kongju National University, Yesan, Chungnam , Korea. 2 Korea Forest Research Institute, Seoul , Korea. kitaco1@hanmail.net Abstract The objective of this study was to assess and analyze the hazard of crown fires based on the canopy fuel characteristics of major conifer stands in Korea. This study was also developed allometric equations for the canopy fuel load and canopy base height of the major conifer stands using the allomeric equations of biomass developed by the Korea Forest Research Institute (KFRI) and the data from the 5 th National Forest Inventory (NFI). The adjusted multiple coefficient of determination of the developed models ranged from 0.63 to 0.99 for canopy fuel load and 0.61 to 0.85 for canopy base height. The suggested models in this study could provide reasonable characterization of the canopy fuel load which is very significant in forest fire management in Korea 요약 본연구의목적은우리나라주요침엽수종에대한수관층연료특성을평가하고자하였다. 본연구에서는국립산림과학원 (KFRI) 에서제시한바이오매스상대생장식과제 5차국가산림자원조사 (NFI) 자료를이용하여우리나라에분포하는주요침엽수종에대한수관연료량과지하고추정모델을개발하였다. 본연구에서제시된수관연료량추정식의조정결정계수는 , 지하고추정식의조경결정계수는 로나타났다. 본연구에서제시된우리나라주요침엽수종에대한수관층연료특성의분석결과는산불관리에유용한정보를제시해줄것으로사료된다

279 June 7-8, 2011 Introduction Conifer stands have the widest distribution of forests in Korea. These stands are very dense because thinning operations are not completely being done in these forests. Therefore, fire hazard is very high and abundant in these stands(koo et al., 2010). Canopy fuels are the main fuel layer supporting canopy fire spread. Canopy structure largely determines combustion requirements and outputs, and consequently important fire behavior descriptors such as rate of fire spread and fire intensity(gragam et al., 1999). Therefore, study on canopy fuel characteristics is urgently needed to suppress the crown fires that are frequently occurring in conifer stands. Thus, the objectives of this study were to analyze canopy fuel characteristics (canopy fuel load, canopy base height, canopy bulk density) and develop allometric equations for the canopy fuel load estimation of major conifer stands in Korea. Methodology The present study was conducted to determine the forest fuel types that are subject to the incidence of crown fires. The definition of the fuel complexes in this category was based on historic fire regime characteristics and published fire case studies(rothermel, 1991). The variety of fuel complexes was further reduced given the available data in the National Forest Inventory (NFI) dataset covering certain forest types. The six major fuel types were categorized on the basis of the NFI forest cover type data and these were Central region Pinus densiflora, Gangwon region Pinus densiflora, Pinus thunbergii, Pinus rigida, Larix leptolepis and Pinus koraiensis. The stand and individual tree data source selected for this study was the National Forest Inventory (NFI). Canopy fuel load were analyzed using the database of the allomeric equations for biomass developed by the Korea Forest Research Institute (KFRI). Within the present study it was reasoned that, given the previous statistical analysis, linear regression analysis was used to develop models for predicting canopy fuel load (CFL) and canopy base height (CBH) from commonly measured forest stand descriptors(cruz et al., 2003). Given the heteroscedasticity present in the relationship between CFL, CBH and basal area, stand density, stand height and the consequent violation of required assumptions of linear regression analysis, these two variables were modeled using logarithmic transformation

280 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Results and Discussion Among the six broad fuel types, Pinus koraiensis had the highest mean CFL, 2.09kg/m², followed by Gangwon region Pinus densiflora 1.51kg/m², Pinus thunbergii 1.31kg/m², Pinus rigida 0.74kg/m², Central region Pinus densiflora 0.63kg/m² and Larix leptolepis 0.53kg/m². Among the species, Pinus thunbergii had the highest CBH. Furthermore, the CBH of six species did not show much variability between fuel types but in the CFL and CBD higher variations were observed. The CBD of Pinus koraiensis was the highest among the different trees while on the other hand, the CBD of Central region Pinus densiflora was the lowest among the different trees observed. Given the individual tree and stand structural differences among the fuel types considered in this study, the models to be developed were fuel type specific. From the linear relationships and nonhomogeneity of variance found in CFL and the explanatory variables under analysis, linear regression analysis of log transformed data was used to model CFL as a function of stand density and basal area. The adjusted multiple coefficient of determination of the developed models ranged from 0.63 to 0.99 for CFL. The adjusted multiple coefficient of determination of the developed models ranged from 0.61 to 0.85 for CBH. The study was conducted with an objective of assessing and analyzing the hazard of crown fire based on the canopy fuel characteristics of major conifer stands in Korea. Although not formally evaluated, the models seem to give a reasonable characterization of the canopy fuel which is very significant in forest fire management applications(cruz et al., 2003). Acknowledgements This study was carried out with the support of the Forest Science & Technology Projects (Project No. S120910L030120) provided by the Korea Forest Research Institute. References Gragam, R.T., Harver, A.E., Jain, T.B. and Tonn, J.R The effects of thinning and similar stand treatments on fire behavior in Western forests. U.S. Department of Agriculture, Forest Service. Pacific Northwest Research

281 June 7-8, 2011 Station. General Technical Report PNW-GTR-463. pp. 27 Cruz, M. G., Alexander, M. E. and Wakimoto, R. H Assessing canopy fuel stratum characteristic in crown fire prone fuel types of western North America. International Journal of Wildland Fire. 12: Koo, K.S., Lee, B.D., Won, M.S. and Lee, M.B Crown fuel characteristics of Japanese red pine (Pinus densiflora) in Mt Palgong, Daegu. Journal of Korean Forest Society.99(1): Rothermel, R.C Predicting behavior and size of crown fires in Northern Rocky Mountains. USDA Forest Service, Intermountain Research Station Research Paper INT-438. Ogden, UT. pp

282 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Crown Characteristics and Fuel Load Estimation of Pinus densiflora S. et Z. in Gyeongbuk Province 경북지역소나무림의수관특성과연료량추정 Mi Na Jang¹*, Yeon Ok Seo¹, Roscinto Ian C. Lumbres¹, Sung Yong Kim¹, Gun Ik Lee¹, Byungdoo Lee², Kyongha Kim², Young Jin Lee 1 1 Kongju National University, Yesan, Chungnam , Korea. 2 Korea Forest Research Institute, Seoul , Korea. kitaco1@hanmail.net Abstract The objective of this study was to develop allometric equations for predicting crown fuel load based on the crown characteristics of Pinus densiflora trees in Gyeongbuk province. The proportion of available crown fuel load in total crown fuel load was observed to be 48.7%. Diameter at breast height variable in the model explained 89%, 93%, 95% and 96% of the observed variation (P < 0.05) in the needles, available crown fuel load, total branch and total crown fuel load, respectively from regression models that were developed for estimating fuel load. On the other hand, crown length and crown width together explained 79%, 77%, 68% and 72% of the observed variation (P < 0.05) in the needles, available crown fuel load, total branch and total crown fuel load, respectively. The results showed that the crown length and crown width combine together as independent variables performed better and had higher R 2 on all fuel loads but had lower R 2 when used separately. The suggested allometric equation in this study could provide quantitative fuel load attributes for crown fire behavior modeling of Pinus densiflora stands in Gyeongbuk province. 요약 본연구의목적은경북지역에분포하는소나무림을대상으로수관특성에따른연료량을추정하기위한상대생장식을개발하고자하였다. 경북지역소나무림의수관연료량비율은잎 22.9%, 직경 1cm 이하가지 25.8% 로이

283 June 7-8, 2011 용가능한수관연료량은약 48.7% 인것으로나타났다. 연료량추정을위한회귀모델은흉고직경을독립변수로한식이모든연료량추정식에서결정계수 90% 이상으로매우높게나타났으며, 수관길이와수관폭은함께사용했을경우결정계수가 68%~79% 로각각독립변수로사용할경우보다비교적결정계수가높아지는것으로나타났다. 본연구는경북지역소나무림의다양한수관특성을이용한상대생장식개발을통해수관화모델에대한정량적인정보를제공해줄수있다. Introduction Forest fire brings many damages on ecological, economical and social aspects such as change in climate caused by increase of carbon dioxide emission, loss of lumber, forest products income, disturbance on industry and worst injuries and even casualties among human(korea Forest Service, 2011). In order to minimize and control forest fires, interest and efforts in predicting fire behavior in many countries are significantly increasing. Estimating the crown fuel load is very important in order to predict fire behavior such as crown fire intensity, flame length, fuel consumption during crown fire and crown fire severity(mitsopoulos and Dimitrakopoulos, 2007). Thus, this study was conducted to develop allometric equations for the crown fuel load estimation based on various crown characteristics of Pinus densiflora in Korea. Methodology 1. Data collection This study investigated Pinus densiflora stands in Youngju, Bonghwa and Daegu region in Gyeongbuk Province. Field measurements were done such as height of tree, diameter at breast height, crown width and crown length before the sample trees were cut 0.2m above the ground(kucuk et al., 2007). After the trees were cut, the stem of each tree was cut on a 1m interval. Crown fuel load estimate were classified into stem, needles, and branches(<0.5cm, 0.5~1cm, 1~2cm, 2~4cm, >4cm) and the weight of each crown fuel load were measured. Sampled fresh weights of each crown fuel load were also measured and then brought to the laboratory for drying. Samples were dried at 95 C for 240hrs and their dry weights were measured

284 International Forest Fire Symposium on Commemorating the International Year of Forests Statistical analysis Correlation and regression analyses were frequently performed in order to determine the relationship between two variable, Using these two analysis together will give very significant results(bilgili and Kucuk, 2009). Thus this study utilized both correlation and regression analysis. Correlation between different crown characteristics such as diameter at breast height, crown base height, height of the tree, crown length and crown width and needles fuel load, available crown fuel load(needles and <1cm branches), total branches fuel load, and total crown fuel load were analyzed. On the other hand, results of the correlation analysis on the crown characteristics was used as the independent variables while crown fuel loads was used as dependent variables in the regression analysis that were conducted. Lastly, logarithm regression model was used for predicting needles, branches and total biomass. Results and Discussion 1. Crown characteristics and distribution The observed mean diameter at breast of the sample trees in Gyeongbuk province was 16.83cm while the mean crown base height was 4.32m. On the other hand, the average height of the trees was 9.48m. The crown length of the sample trees ranged from 2.78m to 9.20m while the for the crown width, it ranged from 0.39m to 3.11m. The average needle fuel load and available crown fuel load (needles and <1cm branches) were 5.13kg and 11.13kg, respectively while total branches fuel load and total crown fuel load were 18.10kg and 23.23kg, respectively. The relative proportion of needles and branches(<0.5cm, 0.5~1cm, 1~2cm, 2~4cm, >4cm) in total crown fuel load were 22.9%, 13.25%, 12.59%, 15.55%, 26.19%, 10.34%, respectively. The proportion of available crown fuel load was 48.74%. 2. Crown fuel load Correlation analysis results showed that the diameter at breast height was closely related to all fuel loads. On the other hand, crown length and crown width were correlated well with the needles fuel load, available crown fuel load, total branches fuel load and total crown fuel load. Regression equations to predict crown fuels were based on the relationship between crown characteristics such as diameter at breast height, crown length and crown width to the crown fuel. The developed allometric equation for the crown fuel load estimation based on the

285 June 7-8, 2011 diameter at breast height, crown length and crown width were highly significant. The diameter at breast height variable alone explained 89%, 93%, 95% and 96% of the observed variation (P < 0.05) in the needles, available crown fuel load, total branch, and total crown fuel load, respectively. Crown length and crown width together explained 79%, 77%, 68% and 72% of the observed variation (P < 0.05) in the needles, available crown fuel load, total branch, and total crown fuel load, respectively. Acknowledgements This study was carried out with the support of the (Estimation of crown fuel load on forest fire damage in Pinus densiflora stand) provided by the Korea Forest Research Institute. References Mitsopoulos, I.D. and Dimitrakopoulos, A.P Allometric equations for crown fuel biomass of Aleppo pine(pinus halepensis MIll.) in Greece. International Journal of Wildland Fire. 16: Kucuk, O., Saglam, B. and Bilgili, E Canopy fuel characteristics and fuel load in young black pine trees. Biotechnology and Biotechnological Equipment. 21(2): Bilgili, E. and Kucuk, O Estimating above-ground fuel biomass in young Calabrian pine(pinus brutia Ten.). Energy & Fuels. 23:

286 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Estimations of Biomass and Carbon Stocks in the different Forests Types of the Philippines Roscinto Ian C. Lumbres 1, 2, Young Jin Lee 1* and Chung Weon Yun 1 1 Kongju National University, Yesan, Chungnam , Korea. 2 College of Forestry- Benguet State University, La Trinidad, Benguet, 2601, Philippines Abstract The most abundant Greenhouse gas in the earth s atmosphere is the carbon dioxide. The concentration of CO 2 has gradually increased by more than 30% since the pre-industrial times. Combustion of fossil fuels and deforestation are the main reasons why CO 2 is still increasing at unprecedented rate with an average of 0.4% per year. Thus, forest ecosystem can be considered as one of the major sources of C. In the Philippines, the C densities in natural forest diminished by approximately 50% after converting it into tree plantations or agricultural lands, whereas deforested and degraded lands with grasses has a C density of <15 Mg ha -1. On the other hand, this ecosystem is also considered as the major sinks for the C. The estimated C pool in the Philippine forests is approximately 1,100 Tg C. Introduction Scientists are saying that the earth is continuously warming. This phenomenon is called global warming. It is defined by Raga-as (2004) as a phenomenon of increasing global surface temperature and the potential climate change of global magnitude in response to the accumulation of atmospheric greenhouse gases predominantly CO 2 that is believed to have the greatest influence on global climate. According to Contreras (1997, as cited by Raga-as, 2004), the Intergovernmental Panel for Climate Change (IPCC) confirmed in 1990 during its First Scientific Assessment of Climate Change that there is a global warming. In the second assessment of IPCC in December 1995, greenhouse gases in the atmosphere were confirmed increasing (Raga-as, 2004). Strong evidence was concluded in the Third Assessment Report that human activities have affected the world s climate reported by the IPCC in 2001 (Lasco et al., 2004a). The United Nations Framework Convention on Climate Change (UNFCCC) defined climate change as a change of climate which is attributed directly or indirectly to human activity that alters the composition of the global atmosphere and which is in addition to natural climate variability observed over comparable

287 June 7-8, 2011 time periods (Lasco et al., 2004a). Greenhouse gases allow short-wave radiation to pass through the earth but prevent the release of long-wave radiation at the earth surface from escaping to outer space (Lofrangco, 2006). Carbon dioxide (CO 2 ), methane (CH 4 ), nitrous oxide (NO 2 ), chlorofluorocarbons (CFCs) and ozone (O 3 ) are considered Greenhouse gases (GHGs). Among GHGs, CO 2 is considered as the most abundant in the earth s atmosphere. According to Lasco et al. (2004a), the concentration of CO 2 has gradually increased by more than 30% since the pre-industrial times. He also noted that combustion of fossil fuels and deforestation are the main reasons why CO 2 is still increasing at unprecedented rate of an average of 0.4% per year. Its concentration has gradually increased over the years from 280 ppm during the pre-industrial period ( ) to 353 ppm in 1990 (IPCC, 1990 as cited by Lofrangco, 2006). From 1980 to 1990, C was the biggest green gas emitted into the atmosphere with 55% followed by CFC (17%), methane (15%), nitrous oxide (6%) and others (7%) (Raga-as, 2004). Approximately, 20.2 B tons of CO 2 were emitted in the atmosphere in In the Philippines, the forestry sector released C by means of deforestation, forest fires, and loss of C sinks by about 67% followed by the energy sector with 30% and industry with 3% (Raga-as, 2004). Murdiyarso (1996 as cited by Raga-as, 2004) stated that the C emission in the Philippines is M tons annually. Biomass and Carbon density in the different forest land covers in the Philippines Biomass represents the amount of C in the forest vegetation and it is measured to estimate the amount of C held by the forest (Raga-as, 2004). The trunk, branches and leaves of a tree is considered as aboveground biomass while roots are known as belowground biomass. Other C pools or a system which has the capacity to accumulate or release C in forest ecosystem are understory/ herbaceous, litters, necromass and soil. Using the aboveground biomass of the tree, the C stored for each tree can also be computed. According to Pulhin (2008), the typical biomass dry weight of a plant is ranging from 45% to 55% C and the recommended for the tropical trees is 47% (IPCC, 2006). The estimated C density in the Leyte Geothermal Reservation were Mg ha -1, Mg ha -1, Mg ha -1, Mg ha -1, Mg ha -1, Mg ha -1, and Mg ha -1 for the forest areas, shrubs/brush lands, Mahogany plantation, Mangium plantation, Gmelina plantation, Coconut, and Abaca/banana, respectively (Lasco et al. 2002). Raga-as (2004) conducted a study entitled Comparison of Carbon Benefits between Natural and Plantation Forest in Aurora Province, Philippines. The total biomass densities were Mg ha -1 and

288 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Mg ha -1 for the natural and plantation forest, respectively. The computed C stock densities were Mg ha -1 and Mg ha -1 for the natural and plantation forest, respectively. Lasco et al. (2004b) found out on their study that the average tree biomass density in the secondary forest of Mount Makiling Forest Reserve was 576 Mg ha - 1. On the other hand, the C stock density of this forest reserve was approximately 418 Mg ha -1 including the soil organic C. The multi-story agroforestry system in Mt. Makiling also has a biomass density accumulation of 236 Mg ha -1. The C stock density was estimated to Mg ha -1 (Sales, 1998, as cited by Palijon, 2009). In contrast, grassland areas such as dominated by Imperata cylindrica and Saccharum spontaneum had an 8.9 and 15.2 Mg ha -1 C density respectively which is much lower than forested areas. In the case of the La Mesa Watershed, the average tree biomass density for the 11 tree plantation species was Mg ha -1 (Lasco and Pulhin, 2006). On the other hand, the C density of these tree plantations ranged from 40 Mg ha -1 to 106 Mg ha -1 with an average of 66 Mg ha -1. The pure mahogany plantation had the lowest C density with 40 Mg ha -1, whereas the kaatoan bangkal plantation exhibited the highest with 106 Mg ha -1. In the mixed P. kesiya forest of La Trinidad, Benguet, the estimated averagebiomass of this forest was Mg ha -1. Using the IPCC default value of 47%, a total of Mg ha -1 were the carbon stocks present in the biomass of trees in this forest. The average biomass density of P. kesiya trees was estimated Mg ha -1 whereas for the other non species P. kesiya it was Mg ha -1 (Lumbres et al., 2011) In terms of mean annual increment (MAI) or the amount of C sequestered by a forest cover in a year, the secondary forest can sequester 0.9 Mg ha -1 yr -1 and 1.1 Mg ha -1 yr -1 in Leyte and in Mindanao, respectively. The average C MAI for this type of forest cover is 1.1 Mg ha -1 yr -1. In the case of tree plantation, the Benguet pine (Pinus keyisa) plantation can increase their biomass about 8.3 Mg ha -1 yr -1 while their C MAI is approximately 3.7 Mg ha -1 yr -1 using a 45% C content percentage. On the other hand, Benguet pine plantation with broadleaf species in Nueva Ecija has a biomass MAI of 6.4 Mg ha -1 yr -1 while its C MAI is 2.9 Mg ha -1 yr -1 (Sakurai et al. 1994, as cited by Lasco et al., 2004a). The Agoho (Casuarina equisitifolia) plantation in Iloilo has a biomass MAI of 3.9 Mg ha -1 yr - 1 while the C MAI was about 1.7 Mg ha -1 yr -1 (Lachica et al. 1994, as cited by Lasco et al., 2004a). The averages for the forest plantations land cover that were cited in the primer were 9.1 Mg ha -1 yr -1 and 4.2 Mg ha -1 yr -1 for the biomass and C MAI, respectively (Lasco et al., 2004a). The forest plantations can sequester more C annually than the secondary forest. This result is expected because most of the species planted in the forest plantation in the Philippines are fast growing as

289 June 7-8, 2011 compared to secondary forest. But old growth forest exhibit the most C stored in their biomass followed by the secondary forest, tree plantation, agroforestry and grassland, respectively (Lasco et al., 2004a). In the urban forest, Lofrangco (2006) conducted a C stock assessment in a golf course in Mandaluyong City. The estimated aboveground biomass was Mg ha -1 while the C stock density was approximately Mg ha -1. On the other hand, Florindo (2007) studied the west course of this golf course and found that the aboveground biomass was 1, Mg and C stored was Mg. In terms of C density, the estimated total aboveground biomass and C stocks were Mg ha -1 and Mg ha -1, respectively. Furthermore, Lumbres (2009) studied the biomass and C stocks available in the Burnham Park in Baguio city. In this study, the estimated aboveground biomass or tree biomass was approximately 4, Mg with 2, Mg of C while the belowground biomass or root biomass was approximately Mg with Mg of C that was stored in the Burnham Park. The total biomass density was Mg ha -1, whereas the carbon density was Mg ha -1. In an overall estimation, Lasco and Pulhin (2009) reported that the Philippine forest lands have approximately 1,100 Tg C. They also concluded that a reduction of approximately 50% of C density was observed after the conversion of a natural forest into tree plantation and perennial crop. Furthermore, 10 Mg ha -1 yr -1 and 3 Mg ha -1 yr -1 of C were sequestered by fast growing species and slow growing species, respectively through reforestation activities in the degraded forest lands (Lasco and Pulhin, 2009). Acknowledgement This study was carried out with the support of Forest Science & Technology Projects (Project No. S211010L010000) provided by Korea Forest Service. References Florindo, R. M Assessment of Carbon Storage Capacity of Urban Trees in West Course of Wack Wack Golf Course and Country Club, Madaluyong City, Manila. Unpublished Practium Report. University of the Philippines Los Baños. IPCC Guidelines for National Greenhouse Gas Inventories. Volume 4: Agriculture, Forestry and other land uses. Chapter 4: Forestland. Lasco, R.D. and F.B. Pulhin Forest land use change in the Philippines and climate change mitigation. Mitigation and Adapation to Climate Change Journal 5:

290 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Lasco, R.D. and F.B. Pulhin Assessing the Role of Watershed Areas in Mitigating Climate Change in the Philippines: The case study of the La Mesa Watershed. Journal of Environmental Science and Management 9 (1): Lasco, R.D. and F.B. Pulhin Carbon budgets of forest ecosystems in the Philippines. Journal of Environmental Science and Management 12(1): Lasco, R.D., J.S. Lales, M.T. Arnuevo, I.Q. Guillermo, A.C. De Jesus, R. Medrano, O.F. Bajar and C.V. Mendoza Carbon dioxide storage and sequestration of land cover in the Leyte Geothermal Reservation. Renewable Energy 25: Lasco, R.D., F.B. Pulhin, J.M. Roshetko and M.R.N. Banaticla. 2004a. Land use, land use change and forestry Climate Change Mitigation Projects: A Primer. World Agroforestry Centre. Southeast Asia Regional Research Programme. Lasco, R.D., I.Q. Guillermo, R.V.O. Cruz, N.C. Bantayan and F.B. Pulhin. 2004b. Carbon stocks assessment of a secondary forest in Mount Makiling forest reserve, Philippines. Journal of Tropical Forest Science 16(1): Lasco, R.D., F.B. Pulhin, R.V.O. Cruz, J.M. Pulhin and S.S.N. Roy Carbon budgets of terrestrial ecosystems in the Pantabangan-Carranglan watershed. Assessments of Impacts and Adaptations to Climate Change (AIACC) Working Paper No. 10: Lofrangco, I Determination of Carbon Sequestration Capacity of Trees and Palms in the East Course of Wack Wack Golf Course and Country Club, Madaluyong City, Manila. Unpublished Practium Report. University of the Philippines Los Baños. Lumbres, R.I.C Carbon Stock Assessment of Trees in Burnham Park, Baguio City, Philippines: A tool for Urban Environmental Management. Unpublished MS thesis. Benguet State University, Philippines. Lumbres, R.I.C., Lee, Y.J. and Seo Y.O Development of Height-DBH Growth Model and Biomass Estimation of Pinus kesiya Royle ex Gordon in La Trinidad, Benguet, Philippines. Asia Life Sciences. In Press Palijon, A.M Urban Forestry and Its Role in Climate Change Mitigation and Adaptation. Professorial Chair Lecture note. University of the Philippines Los Baños. Pulhin, F.B Methods in Assessing Carbon Stocks: Lecture notes. Training on Assessing Carbon in Forest Ecosystems. October 20-22, BPK Training Room, PCARRD. Raga-as, M.L Comparison of Carbon Benefits between Natural and Plantation Forests in Aurora Province, Philippines. Unpublished MS thesis. University of the Philippines Los Baños

291 June 7-8, 2011 Abstract Hot Spot Analysis on Forest Carbon Stock using Getis-Ord Gi * Spatial Statistic Getis-Ord Gi * 공간통계량을이용한산림탄소저장량핫스팟분석 Kyoungmin Kim*, Jungbin Lee, Eunsook Kim Korea Forest Research Institute, Seoul, , Korea greenann@forest.go.kr In order to acquire higher reliability, spatially explicit quantification of carbon stock is needed. Also an exploration of spatial cluster for high forest carbon is needed for effective forest management plan. In this study, therefore, we produced forest carbon map and explored carbon hot spot. For this, forest stand variables were extracted from NFI data and used to develop aboveground biomass(agb) regression models by species. Dominant/semi-dominant height and crown density were used as explanatory variables of AGB regression models. Spatial distribution of AGB could be estimated using AGB models, forest type map and the height map that was developed using forest type map and height regression models. Finally, mean carbon per hectare in Danyang county by this spatial modeling was tonc/ha and same statistic by existing forest statistical method was tocc/ha. These means were not significantly different at p=0.05 using paired t-test. Then we explored hot spot that was clustered with high forest carbon stock using Getis-Ord Gi * spatial statistic. Hot spot above 2.58 standard deviation was distributed in east of Danyang county and cold spot below standard deviation in west of Danyang county. Most of hot spot consists of natural forests(88.42%). Major forest type of hot spot was broad leaved forest(57.79%). Major age, DBH and density class of hot spot were IV & V class(69.47%), above medium DBH class(98.74%) and high density class(73.55%), respectively. 요약 온실가스통계의높은신뢰도를확보하기위해서는 IPCC 우수실행지침수준 3 에기반한공간적으로명확한산림탄소저장량의정량화가필요하다. 또

292 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 한효율적인산림경영계획수립을위해산림탄소저장량이높은지역과낮은지역이밀집하고있는공간클러스터를파악할필요가있다. 따라서본연구에서는충북단양군의지상부바이오매스를대상으로산림탄소지도를제작하고산림탄소저장량핫스팟을탐색하였다. 공간적으로명확한산림탄소저장량분포를추정하기위해국가산림자원조사표본점단위로계산된산림탄소저장량을임상도를이용하여공간규모를확장 (upscaling) 하였다. 그결과단양군산림탄소저장량은 3,329,451 tonc, 헥타당평균은 tonc/ha로추정되었다. 기존의 NFI 자료중심의비례배분법으로산출된평균 tocc/ha과 T-검정결과 t=0.177 (p>0.05) 로두가지방법간평균차이는통계적으로유의하지않다. 산림탄소지도를대상으로 Getis-Ord Gi * 공간통계량을이용하여높은탄소저장량이밀집한지역인산림탄소저장량핫스팟지역을탐색하였다. 분석결과표준편차 2.58 이상의높은탄소저장량이공간적으로집중된핫스팟은단양군동쪽에분포하고있으며표준편차 이하의낮은탄소저장량이공간적으로밀집한콜드스팟은단양군서쪽에분포하고있었다. 핫스팟의대부분은천연림 (88.42%) 으로구성되었으며활엽수가가장많은면적 (57.79%) 을차지하고있었다. 임분특성의경우 4, 5영급 (69.47%), 중경목과대경목 (98.74%) 이점유하고있으며소밀도가밀한지역으로 (73.55%) 으로분석되었다. Introduction In order to acquire higher reliability, spatially explicit quantification of carbon stock based on IPCC GPG tier 3 is needed. But existing forest statistics has limit for interpretation of spatial distribution of forest carbon because it is aggregated by administrative unit. Also an exploration of spatial cluster for high forest carbon is needed for effective forest management plan. In this study, to solve these problems we produced forest carbon map and explored carbon hot spot using Getis-Ord Gi* statistic. Methodology In order to upscale forest carbon from plot into landscape level, forest stand variables compatible with attributes of forest type map were extracted from NFI data and used to develop AGB regression models by species. Dominant/semidominant height and crown density were used as explanatory variables of AGB regression models

293 June 7-8, 2011 Spatial distribution of AGB was estimated by AGB regression models by species which were combined with stand height map that was developed using forest type map and height regression models, and layers of forest type map such as species, crwon density. Then AGB was converted to carbon stocks using CF(carbon fraction of dry matter) 0.5. Paired T-test for 117 plots was used to test mean difference of carbon between existing method based on National forest inventory data alone and spatial modeling method suggested by this study. Finally, we explored hot spot that was clustered with high forest carbon stocks using Getis-Ord Gi * spatial statistic. The standardized Gi * (d) is defined as Where wij(d) is a symmetric one/zero spatial weight matrix with ones for all links defined as being within distance d of a given i; W * i = Wi + wii, S * 1i = w (all j ). denote the usual sample mean and variance. Results and Discussion Mean carbon per hectare in Danyang county by this spatial modeling was tonc/ha and same statistic by existing forest statistical method was tocc/ha. These means were not significantly different at p=0.05 using paired t-test. Then we explored hot spot that was clustered with high forest carbon stock using Getis-Ord Gi * spatial statistic. Hot spot above 2.58 standard deviation was distributed in east of Danyang county and cold spot below standard deviation in west of Danyang county. Most of hot spot consists of natural forests(88.42%). Major forest type of hot spot was broad leaved forests(57.79%). Major age, DBH and density class of hot spot were 4 & 5 class(69.47%), above medium DBH class(98.74%) and high density class(73.55%)

294 International Forest Fire Symposium on Commemorating the International Year of Forests Figure 7. Forest carbon map (2009) Figure 8. Hot Spot of Forest carbon (2009) References Eun-Sook Kim, Kyoung-Min Kim, Jung-Bin Lee, Seung-Ho Lee, Chong-Chan Kim Upscaling Aboveground Biomass from Plot to Landscape Using National Forest Inventory Data and Forest Type Map. Korea Forest Society (in review). Ord, J. K., Getis, Arthur Local Spatial Autocorrelation Statistics: Distributional Issues and an Application. Geographical Analysis 27(4):

295 June 7-8, 2011 Abstract Physiological Responses of Pinus densiflora by seawater watering 바닷물살포에의한소나무의생리반응변화 Sunhee Kim* Korea Forest Research Institute, Seoul, , Korea The object of this study is to provide a method to minimize harmful influence on Pinus densiflora putting out forest fire using seawater, and then physiological response of Pinus densiflora is investigated after watering seawater. Five-year-old Pinus densiflora seedlings treated with seawater according the leafing stages (1 stage: bud, 2 stage: buds swell and new shoot growing, 3 stage: new needle sprouting). Seedlings were measured for shoot growth, photosynthetic responses and analyzed for chlorophyll contents. The shoot growth of all of the treated individuals shows almost no differences compared with those of controls, but some individuals in treatment of leafing stage 3 were withered and appeared browning in top of shoot. The photosynthetic rate in treatments of leafing stage 1 and 2 decreased after 14 days of treatment and has recovered after 28 days of treatment. The rate in treatment of leafing stage 3 drastically decreased and has not recovered completely. The net apparent quantum yield and carboxylation efficiency decreased in treatment of leafing stage 3 only. The chlorophyll contents did not change very much comparing with those of control. 요약 바닷물을이용한산불진화시에소나무에미치는악영향을최소화할수있는방법을마련하기위하여, 바닷물살포에의한소나무의생리적반응을조사하였다. 바닷물살포는개엽단계별 (1 단계 : 눈 ( 芽 ), 2 단계 : 눈이부풀고신초가자라는단계, 3 단계 : 신초에서잎이트는단계 ) 로처리하고신초생장, 광합성율변화, 광화학계와탄소고정계반응특성, 엽록소함량을측정하였다. 신초생장은모든처리구에서양호하였으나개엽 3 단계부터신초상단에가시적인피해가나타났다. 광합성율변화는개엽 1, 2 단계처리에서일시적으로감

296 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 소하나처리 28 일후회복하였고개엽 3 단계처리는완전히회복하지못하였다. 순양자수율과탄소고정효율은개엽 3 단계처리부터감소하였다. 엽록소함량은대조구대비유의적차이는나타나지않았다. 따라서개엽 3 단계이후바닷물을이용한산불진화를할경우, 소나무잎의생리적인감소를초래할수있을것으로사료된다. 서론 (Introduction) 최근극심한가뭄으로인한물부족의심각성이대두되면서산불발생시담수에의한진화에어려움이있으며, 있으며특히대형산불이나동절기산불의경우담수만으로진화하는데한계가있어바닷물을이용하여산불을진화하고자하는시도가있다. 국제적으로는이탈리아등지중해연안국가나유럽의프랑스에서도산불진화시에바닷물을사용하고있으며근년들어일본에서도도서지방이나해안가에서산불이발생한경우바닷물을살포하여산불을진화하고있다. 그러나국내에서는산불진화에바닷물을사용한사례가없고바닷물살포가산림생태계에미치는영향에대한연구자료가미흡하여현실적으로적용하기어려운실정이다. 따라서, 이연구는동해안산불위험지역산림의주수종인소나무를대상으로바닷물살포에대한영향을개엽단계별로파악하여바닷물을이용한산불진화시에산림생태계구성인자인소나무에미치는악영향을최소화할수있는방법을마련하는데그목적이있다. 재료및방법 (Methodology) 실험목은소나무 (Pinus densiflora) 5년생묘목을사용하였으며바닷물살포처리는개엽단계별 (1단계: 눈단계, 2 단계 : 눈이부풀고신초로자라는단계, 3단계 : 신초가다자라잎눈이트는단계 ) 로, 1단계는 4월18일, 2단계는 4월 27일, 3단계는 5월 2일에물뿌리개를사용하여 8L/ m2의바닷물을살포하였다. 바닷물살포후, 눈과신초의길이생장을모니터링하였으며 LED light source 와 CO 2 injector system이부착된휴대용광합성측정기 (Li-6400, Li Cor.) 를사용하여광도 PPFD 1000 umol m -2 s -1 에서의광합성율변화를모니터링하였다 (Figure 1). 또한광변화에따른광합성변화 (light response curve) 을측정하여 (Figure 2) 암호흡, 최대광합성능력, 순양자수율 (net apparent quantum yield) 을구하고 CO 2 농도변화에따른광합성변화 (A-Ci curve) 을측정하여광호흡과탄소고정효율 (carboxylation efficiency) 을구하였다. 엽록소함량은 DMSO법과

297 June 7-8, 2011 Arnon(1949) 의방법을이용하여엽록소 a, b, 엽록소 a+b 및엽록소 a/b 을산출하였다. 결과및고찰 (Results and Discussion) 바닷물살포처리에따른신초의길이생장을모니터링한결과, 처리구의일평균상대생장율은모두대조구와비교하여유의적인차이는없었으나개엽 3 단계에서처리한소나무의일부에서신초의상단이마르고갈변하여굽어지는현상이나타났다. 광합성율의변화는개엽 1 단계와개엽 2 단계에서처리한경우, 처리후 14 일경과시부터일시적으로감소하였으나 28 일경과하여광합성율을회복한반면, 개엽 3 단계처리구에서의광합성율은처리후 4 일경과시부터급격히감소하여처리후 7~14 일경과시대조구의 22~27% 로낮을값을나타내었고 21 일경과시에도대조구의 48% 로완전히회복하지못하였다 (Figure 1). 바닷물살포처리에따른광합성의광화학계및탄소고정계특성을개엽 1 단계처리구는처리후 35 일째, 개엽 2 단계처리구는 27 일째, 개엽 3 단계처리구는 21 일째측정한결과 (Figure 2), 개엽 1 단계와개엽 2 단계처리에서는대조구와유의적차이가나타나지않은반면, 개엽 3 단계처리에서는순양자수율과탄소고정효율이대조구의 60%, 39% 로감소하여바닷물살포로인하여광합성에서빛에너지를화학에너지로변환시키는광화학계의활성및 CO 2 고정계효소인 Rubisco 의활성이떨어진것으로판단된다. 모든처리구의엽록소함량은대조구와비교하여유의적차이가없어바닷물살포가엽록소의양에미치는영향보다는광을수확하는엽록소의질적능력에영향을준것으로사료된다. Figure 1. The change of photosynthetic rate at PPFD 1000 umol CO 2 m -2 s -1 in the leaves of Pinus densiflora for 35days after seawater treatment

298 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Figure 2. The light response curve of net photosynthetic rate in the leaves of Pinus densiflora at 35 days(leafing stage I), 28 days(leafing stage II), 21 days(leafing stage III) after seawater treatments. 인용문헌 (References) Arnon, D. I Copper enzymes in isolated chloroplasts, polyphenoloxidase in Betula vulgaris. Plant Physiology 24, Kim, Pan-Gi and Eun Ju Lee Ecophysiology of photosynthesis 1: Effects of light intensity and intercellular CO 2 pressure on photosynthesis. Korean Journal of Agricultural and Forest Meteorology 3(2),

299 June 7-8, 2011 Effects of previous-fertilization treatment and fertilizer-induced stress on the growth and nutrient contents of Chamaecyparis obtusa seedlings 전년도시비경험과당년시비스트레스가편백생장및양분함량에미치는영향 Jiyoung An*, Byungbae Park, Juhan Sung, Sukkwon Kim Korea Forest Research Institute, Seoul, , Korea Abstract It is required to produce healthy seedlings for restoration of destroyed forest by forest fire. The purpose of this study is to investigate early plant growth and nutrient uptake response of Chamaecyparis obtusa seedlings to previousfertilization treatment and fertilizer-induced stress during nursery production systems. In 2008, one year old of Chamaecyparis obtuse seedlings were supplied with a mixed nitrogen-phosphorus-potassium (N-P-K) fertilizer at the amount of 27.6 g m -2 N, 12.2 g m -2 P. After one year, showing similar growth of seedlings were selected and transplanted in the 30L pots. They were over-fertilized with Basacote 80 g pot -1. Plant growth of the none-fertilized plants in next year exhibited no differences as previous-fertilization treatment, whereas that of the over-fertilized plants which had experienced previous-fertilization treatment was better than the others. Carbohydrates distribution to belowground biomass was low at the over-fertilized seedlings regardless of previous-fertilization treatment. N and P concentration of leaf and root were irrelevant to previous-fertilization treatment. N and P content of tissues were high at the over-fertilized plants on the whole. This study proved that previous-fertilization treatment at nursery may increase tolerance to over-fertilization stress in the beginning of outplanting

300 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 요약 산불지초기복원성공을위해서는건전한묘목생산이반드시요구된다. 이연구의목적은묘목생산단계에서전년도다른시비경험과당년과시비스트레스처리가편백의초기생장및양분흡수량에미치는영향을구명하는것이다. 산불지초기복원성공을위해서는건전한묘목생산이반드시요구된다 년 1-0 묘편백을보성양묘사업소포지에심고무처리 (0 g m -2 N, 0 g m -2 P) 와 27.6 g m -2 N, 12.2 g m -2 P 비료를시비처리하였다 년 3 월말에생장이유사한묘목을선발한후 30 L 화분에식재하고, 대조인무처리와과시비스트레스 ( 지효성복합비료바사코트 80 g pot -1 ) 를처리한후생장과양분흡수량을조사하였다. 시비스트레스무처리에서는전년도포지시비경험에따른편백묘목생장에차이가없었지만, 시비스트레스처리에서는대조에비해전년도시비경험이수고및지상부생장을증가시켰다. 지하부로의탄수화물분배는전년도시비경험에관계없이시비스트레스처리에서낮았다. 잎과뿌리의질소와인농도는전년도시비경험에의한차이는없었고, 시비스트레스처리에서질소와인함량이대체적으로높게나타났다. 이연구를통하여양묘장에서의적절한시비경험이향후현장에서식재시수행하는시비로인한시비스트레스를줄일수있는방법임을알수있었다. 서론 (Introduction) 묘포장에서비료경험에대한묘목의생장반응및산지초기생장은 Salifu 와 Timmer(2003) 에의해제시된양분공급에따른식물생장, 조직내양분량, 양분농도의관계를나타내어주는개념적모델로그효과를알아볼수있다. 또한 Imo 와 Timmer(1992) 는 nutrient loading 을통해묘포장에서의묘목시비가다양한종류의침엽수종 (white spruce 등 ) 의체내에축적된양분을산지초기생장과정에서이용하여묘목의초기활착률을증가시켜옮겨심기성공률을높여준다는것을밝혔다. 편백은우리나라남부지방의대표적조림수종으로써산지식재시에시비처리를하는데이것이생장에도움을줄수도있지만과도한시비로인한스트레스로도작용을할수있다. 따라서본연구에서는전년도시비경험유무에따라당년과시비스트레스처리가편백의생장과양분에어떠한영향을미치는지알아보고자수행하였다. 이는산불지초기복원을위한건전묘목생산에있어서도, 포지에서의시비경험이건전묘목들이산지에옮겨심어졌을때과도한시비스트레스에어떤반응을보이는지알아보기위함이다

301 June 7-8, 2011 재료및방법 (Methodology) 실험수종인 1-0묘편백을 2008년 1년동안보성양묘사업소포지에심고무처리 (0mg -2 N, 0mg -2 P) 와 27.6g m -2 N, 12.2g m -2 P 비료를처리하였다. 건전하며균일한묘목을얻기위하여 2009년 3월말에생장이유사한묘목을각처리별로 12본을선발하였다. 선발된묘목을뿌리가상하지않도록굴취하여국립산림과학원산림생산기술연구소로옮긴후, 화분에심을때까지그늘진곳에가식한후충분히관수하였다. 실험에이용된화분은윗지름 36cm, 아래지름 26cm, 깊이 43cm 크기의화분으로화분당 30L의마사토를이용하였다. 화분토양에과시비스트레스환경을만들어주기위하여대조인무처리와과시비스트레스 ( 지효성복합비료바사코트 80g pot -1 ) 를처리하였다. 가식 3일후선발된묘목들을화분에옮겨심고온실 ( 연평균상대습도 61~77%, 온도 14.6~29 ) 에서관찰하였다. 계속적인동일위치의근원경측정을위해지상에서 1cm 위치에흰페인트로표시하여그곳을연속측정하였다. 식재 2주후에초기생장을측정하였고굴취전에수고와근원경을다시측정하였다. 2009년 10월초에뿌리가손상되지않도록화분에서묘목을굴취한후흐르는물에토양을씻어준후 65 항온기에서건조시켰다. 1주일후지상부와지하부로나누어건중량을측정하고뿌리와잎의양분분석을실시하였다. 실험데이터는 SAS 9.2를이용하여분산분석을실행하여유의성을살펴보았다. 결과및고찰 (Results and Discussion) 2009 년도에시비스트레스를처리해주지않은화분토양에서편백의수고생장은전년도포지에서의시비경험에따른차이가없었지만, 시비스트레스처리화분토양에서는전년도시비경험이있는편백의수고생장이더좋게나타났고, 근원경에서는전년도시비경험에따른유의한차이를발견할수없었다 년도처리에따른수고생장율은무처리에서 13%, 과시비스트레스처리에서 29% 로, 약 2 배이상의생장율차이를보였다

302 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Figure 1. Height after inducing fertilizer stress. Figure 2. Root collar diameter after inducing Vertical bars are standard errors of mean(n=6). fertilizer stress. Vertical bars are standardd errors of mean(n=6). 지상부와뿌리의건중량은 2009년도에과시비스트레스환경을만들어준토양에서전년도에시비경험이있던편백의지상부로물질축적량이무처리보다많았다. 또한뿌리건중량은 2009년도에과시비스트레스환경을만들어준토양에서뿌리발달을저하시켜무시비환경보다바이오매스축적량은적게나타나지만, 과시비스트레스처리에서전년도에시비경험이있던편백의뿌리건중량이시비경험이없던것보다더많음을볼수있다. Quoreshi 와 Timmer(2000) 에따르면지수적시비 (exponential fertilization) 는식물의양분요구가낮은수준인기간동안체내에염분축적가능성을줄여줌으로써시비양이식물의요구에부합하기때문에, 과시비로인한뿌리손상피해를최소화시키는것을도울수있다는이론으로설명될수있다. Figure 3. Aboveground dry weight after inducing fertilizer stress. Vertical bars are standard errors of mean(n=6). Figure 4. Root dry weight after inducing fertilizer stress. Vertical bars are standard errors of mean(n=6)

303 June 7-8, 2011 잎과뿌리에서질소와인의농도는전년도시비경험에의한유의한차이가보이지않았고, 시비스트레스처리에서질소와인의함량이대체적으로높았는데, 질소의함량은유의한차이를보이지않았다. 하지만 2008 년도에시비를경험하지않고 2009 년도에과시비스트레스처리를해준화분에서자란편백뿌리에서인의체내농도가유의하게높았는데, 이것은발근에관여하는인이양분흡수를위한뿌리발달을위해뿌리에분포하는비율이높은것이거나흡수에비해낮은생장율로인하여단위무게당농도는증가한것일수도있다. Table 1. Nutrient concentrations(g kg -1 ) of leaf and root after inducing fertilizer stress. Standard errors are in parenthesis(n=6). Elements fertilizer-induced stress None Fertilization previous-fertilization treatment Control Treatment Control Treatment Leaf N 11.2(2.1) 10.4(0.9) 12.3(1.6) 12.2(1.2) P 1.7(0.1) 1.8(0.1) 1.6(0.2) 1.9(0.2) Root N 8.7(1.3) 8.4(1.0) 13.3(3.2) 10.9(1.6) P 1.9(0.3) 2.0(0.2) 3.2(0.4) 2.3(0.5) 이처럼묘포장에서시비를경험한편백묘목들이산지에서과시비스트레스를받을경우그것에대한내성을가짐으로써시비를경험하지않은묘목들보다수고성장이더좋고, 뿌리손상이적으며바이오매스축적량비율이높은것을알수있다. 즉, 묘포장에서시비를경험한묘목들이시비를경험하지않은것들보다시비경험에의한내성으로인하여과시비스트레스에더잘대응하여산지식재성공률을높이는방법이고이는산불지초기복원에필요한건전묘목을생산할수있는유용한방법임을이번연구를통해알수있었다. 인용문헌 (References) 박병배, 변재경, 김우성, 성주한 묘포에서질소, 인, 칼륨비료주기가물푸레나무, 들메나무, 잣나무, 전나무의생장및양분에미치는영향. 한국임학회지 99(1): 85~

304 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 변재경, 김용석, 이명종, 손요환, 김춘식, 정진현, 이천용, 정용호 시비수준에따른소나무, 낙엽송, 자작나무, 상수리나무묘목의생장변화. 한국임학회지 96(6): 693~698. 이영진, 이미향, 이경학, 손영모, 서정호, 박인협, 손요환 편백조림지에서영급이바이오매스확장계수와줄기밀도에미치는영향. 한국임학회지 95(1): 50~54. Balaya, J., Dianda, M., Wilson, J., Ouedraogo, S. and Sanon, K Predicting field performance of five irrigated tree species using seedling quality assessment in Burkina Faso, West Africa. New Forests 38: Birge, Zonda K. D., Jacobs, Douglass F., Salifu, and Francis K Exponential nutrient loading as a means to optimize bareroot nursery fertility of oak species. USDA Forest Service Proceedings RMRS-P-43. Deans, J.D., Mason, W.L., Cannell, M.G.R., Sharpe, A.L. and Sheppard, L.J Growing regimes for bare-root stock of Sitka spruce, Douglas fir and Scots pine. 1. Morphology at the end of the nursery phase. Forestry Supplement 62: Imo, M., Timmer, V.R Nitrogen uptake of mesquite seedlings at conventional and exponential fertilization schedules. Soil Science Society of America Journal 56(3): Park, B.B., Cho, M.S., Lee S.W., Yanai R.D., Lee D.K Minimizing nutrient leaching and improving nutrient use efficiency of Liriodendron tulipifera and Larix leptolepis in a container nursery system. New Forests(in press). Salifu, K.F., Timmer, V.R Optimizing nitrogen loading of Picea mariana seedlings during nursery culture. Canadian Journal of Forest Research 33: Timmer, V.R Exponential nutrient loading: a new fertilization technique to improve seedling performance on competitive sites. New Forests 13:

305 June 7-8, 2011 Abstract Dynamics of Advanced Seedlings of Pinus koraiensis in three Stand Types 세가지임분유형에서잣나무전생치수의동태 Hee Moon Yang*, Joo Hwan Sung, Suk Kwan Kim Korea forest Research Institute, Seoul , Korea This study was conducted to investigate the dynamics of advanced seedlings on different stand types to understand potential of natural regeneration and to consider applicable possibility of intensively managing technique to promote natural regeneration of Pinus koraiensis in Korea. The study site was classified into three stand types; thinned, about 30 years old Korean white pine plantation(stand type A), thinned, about 50 years old plantation (stand type B), and non-thinned about 50 years old plantation(stand type C). The stand types showed different stand and environment characteristics by stem density, species composition and stand development. In summer 2005, the beginning of this study, investigated seedlings of Korean white pine were 2,125, 18,650 and 55,725 individuals / ha in stand type A, B and C, respectively. The ratios of generation and death of advanced seedlings were 0.81 and 0.38, 0.62 and 0.49, and 0.33 and 0.84 in stand type A, B and C, respectively. Density and annual growth of survived seedlings during investigated period were 1,200 and 0.06, 10,600 and 0.20, and 10,625 individuals / ha and 0.02mm / yr in stand type A, B and C, respectively. 요약 한국잣나무림에서의잣나무천연갱신가능성을타진하고자간벌되었던 30 년생잣나무조림지 ( 임분유형 A), 50 년생잣나무조림지 ( 임분유형 B), 그리고전혀간벌되지않았던 50 년생잣나무조림지 ( 임분유형 C) 등 3 개의임분유형에서갱신치수밀도와동태를파악하였다. 임분들은간벌처리의유무와임령에따라임목밀도, 수종구성, 그리고식생발달에있어각각다른구조와환경특성으로보였다. 이연구가시작된 2005 년여름에조사된갱신치수는임

306 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 분유형 A, B, C 에서각각 ha 당 2,125, 18,650, 55,725 개체였다 ( 간벌되지않은지역의밀도가매우높았음 ). 갱신율과고사율은임분유형 A 에서 0.81 과 0.38, 임분유형 B 에서 0.62 와 0.49, 그리고임분유형 C 에서 0.33 과 0.84 였다. 전체조사기간동안생존한갱신치수의밀도와 1 년간생장량은임분유형 A, B, C 별로각각 1,200 개체 /ha 와 0.06mm, 10,600 개체 /ha 와 0.20mm, 그리고 10,625 개체 /ha 와 0.02mm 로조사되었다. Keyword : 잣나무 (Pinus koraiensis), 갱신, 생물다양성연구 서론 (Introduction) 잣나무의자연적인분포지는일본의일부지역부터한국, 중국북동부, 러시아시베리아지역이지만, 한국에서는일제시대와치산녹화시기에제주도를제외한남한지역에주요조림수종으로식재되어왔다. 남한지역에잣나무조림지의면적은 340 천 ha 로전체조림지면적의약 19% 를차지하고있다. 잣나무조림지에서의천연갱신은매우어렵다는것이일반적인이해이지만, 주변활엽수림에서의자연적인치수발생은모수림과의거리와비례하여풍부한것으로파악되고있으며, 잣나무조림지에서도간벌등밀도변화이후에풍부한치수발생이관찰되고있다. 따라서잣나무조림지의임분구조에따라잣나무천연갱신가능성이다를것으로판단되며, 이를구명하여잣나무천연갱신을촉진할수있는집약적인산림관리기술의적용가능성을고찰하는것이이연구의목적이다. 재료및방법 (Methodology) 연구지는여러간벌처리에의해인공림의생물다양성및생태계변화과정을파악하고, 생태적관리기법을개발하고자선정된춘천인공림과생물다양성연구지에서수행되었으며, 간벌처리가실연되기전인 2005 년여름, 2005 년가을, 2006 년봄, 2006 년여름까지의조사자료를기반으로하였다. 대상지의잣나무조림지는간벌되었던 30 년생잣나무조림지 ( 임분유형 A), 50 년생잣나무조림지 ( 임분유형 B), 그리고전혀간벌되지않았던 50 년생잣나무조림지 ( 임분유형 C) 등 3 개임분유형으로구분되며, 각임분유형은다른구조적, 환경적특성을보였다 (Table 1). 각각의임분유형별로 5m x 5m 크기의조

307 June 7-8, 2011 사구를각각 16 개씩을설치하였으며, 치수의밀도와발생및고사등동태를모니터링하였다. Table 1. Site conditions of stand types in the study area. Topography & Environmental features Stand Canopy Floor Mean Mean Soil Relative Altitude Slope Mean Types Aspects Closure closure Temp. Humid. moist. Illum. (m) ( ) LAI (%) (%) ( ) (%) (%) (%) A W 438~ B W 403~ C NW, W 329~ 결과및고찰 (Results and Discussion) 2005 년여름, 연구초기에잣나무치수밀도는 A 임분유형이 2,125 개체 /ha, B 임분유형이 18,650 개체 /ha, C 임분유형이 55,725 개체 /ha 로 50 년생임분에비해 30 년생임분의밀도가극히낮았다 (Figure 1). 1 년간의밀도변화는간벌이수행되었던 A 와 B 임분유형에서는각각 44% 와 7% 가증가하였으나, 간벌이수행되지않았던 C 임분유형은 51% 감소하였다. 갱신치수의 1 년간발생율은 A 임분유형 81%, B 임분유형 38%, C 임분유형 62% 인데반해고사율은각각 49%, 33%, 84% 였으며, 1 년간생존한잣나무치수는 1,200, 10,600, 10,625 개체 /ha 였다. 잣나무조림지의임분유형별로갱신치수의밀도, 발생, 고사등은임령과간벌유무등에의해기인된임분구조와환경요인에영향을받는것으로판단되며 (Johnstone, 1981), 특히, 간벌처리등임분밀도조절이갱신치수생육에영향을미쳐 1 년이내의초기고사를방지하는데효과적일것으로판단된다

308 International Forest Fire Symposium on Commemorating the International Year of Forests Individ./ha Sm 2005yr Wt 2005yr Sp 2006yr Sm 2006yr Stand Type A Stand Type B Stand Type C Figure 1. Change on abundance of Pinus koraiensis seedlings in three stand types 인용문헌 (References) Johnstone, W.D Pre-commercial thinning speeds growth and development of lodgepole pine: 25 year results. Inf. Rep. NORX-237. Canadian Forest Services Northern Forestry Research Centre, Edmonton, Alta

309 June 7-8, 2011 Synersistic effects of a landslide risk in forest fire damaged area of Namwon city using aerial photographs 항공사진을이용한전북남원시산불피해지의산사태위험도상승효과분석 Choongshik Woo*, Ho-joong Youn, Chang-woo Lee, Ki-hyung Park and Kyongha Kim KFRI, 57 hoegi-ro, Dongdaemun-gu, Seoul, , Korea Abstract Woocs77@gmail.com This study was conducted to know how much increase a landslide risk caused by forest fire. The object area is forest fire damaged area in Namwon city, Jeonbuk province. A landslide risk before forest fire damage was based on the landslide hazard map of Korea Forest Service. A landslide risk after forest fire damage was analyzed by extracting landslides occurred in 2010 from aerial photographs. Synergistic effect a landslide risk in forest fire damage area was appeared to about 32% by the accuracy of the landslide hazard map showed to researches. 요약 전북남원의산불피해지에서발생한산사태지역을대상으로산불피해이전의산사태위험도와산불피해이후의산사태위험도의차이를분석하였다. 산불피해이전의산사태위험도는산림청에서제공하는산사태위험지도에서추출하였으며, 산불피해이후의산사태위험도는 2010 년발생한산지토사재해구역을항공사진에서추출하여구하였다. 기존연구에서나타난산사태위험지도의정확도를기준으로산불피해지의산사태위험도상승효과는약 32% 증가하는것으로나타났다

310 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 서론 (Introduction) 일반적으로여름철태풍및집중호우시발생하는산사태는산불피해지에서일반산지에서보다발생확률이높은것으로알려졌다. 이것은산불로인해산지에존재하는산림이훼손되고토양의점착성이떨어져토양의유실이급속하게진행되기때문이다 ( 박경등, 2009). 산불로인하여토사침식과동시에산사태발생이얼마나증가할것인지에대한정량화가필요하다. 현재까지산사태위험도를판단할수있는대표적인기준으로산사태위험지판정표가있다 ( 최경, 1998). 산사태위험지판정표에제시된인자중임상이산불피해지를반영할수있는부분이지만대부분임상과영급으로분류되어있어적용이쉽지않다. 따라서본연구에서는 2005년대형산불이발생한전북남원시산동면의산불피해지를대상으로 2010년발생한산지토사재해 74 공간자료와산림청의산사태위험지도를이용하여산불피해시나타나는산사태위험도의상승효과를분석하였다. 방법 (Methodology) 전북남원시산동면일대를 2010 년 9 월항공사진을촬영하여정사보정을거쳐단일모자이크영상을제작하고, 육안판독에의해산지토사재해발생구역을추출하였다 (Figure 1). Figure 1. Aerial photographs after landslide occurrence in September 2010 and landslide hazard map of Korea Forest Service (White line of central image means the landslide zones) 74 산지토사재해는산사태와토석류등을총칭함

311 June 7-8, 2011 추출한산사태공간자료를이용하여산불피해지와일반산지에서발생한산사태면적을각각산출하였으며산사태위험지도와중첩하여 1, 2 등급지의비율을산출하였다. 결과및고찰 (Results and Discussion) 산림청통계자료를이용하여이지역에발생한산지토사재해를분석한결과남원시일대에서 2010 년 8 월 13~18 일까지국지성집중호우로인해약 40ha 의산지토사재해가발생하였는데, 이중산불피해지에서약 18ha 가발생하였다. 즉, 전체산지토사재해피해중약 45% 가산불피해지역에서발생한것으로나타났다. 이것을일반산지과산불피해지로구분하여산지토사재해발생면적을각면적으로나누어단순발생비율을산출해보면일반산지는 0.05%, 산불피해지는 10% 로일반산림대비산불피해지의산지토사재해발생비율이 200 배높아진다 (Figure 2). 이중 2005 년대형산불이발생했던산동면목동리일대에서는산지토사재해가남원시전체의 17%, 전체산불피해지의약 39% 가발생하였다. 여기서항공사진을이용하여산지토사재해의실제발생면적을분석한결과산림청집계면적보다약두배많은 12.5ha 로나타났다. Figure 2. The area of soil sediment disaster and occurrence rate through comparing with general mountainous area and forest fire damaged area Figure 3 의 (a) 와같이산림청산사태위험지도와남원지역의산사태공간자료를이용하여산사태발생위치에서의산사태위험지도의등급분포를알아보았으며기존에같은방법으로연구된전북무주 진안 장수지역과강원평

312 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 창지역의결과와비교하였다 (Woo et al., 2008; Youn et al., 2009). (a) (b) Figure 3. (a) Overlay rate of hazard map and actual landslides of three areas occurred in the 2000s (b) Synergistic effect of a landslide risk in Namwon using average accuracy of hazard map in two areas(except Namwon) 전북무주 진안 장수지역과강원평창지역은산불피해지가아닌일반산지에서발생한산사태로남원의산불피해지에서발생한산사태와는특성이다르다. 또한산사태위험지도는산불로인해변화된상황을반영할수없어남원산불피해지와직접비교하기는어렵다. 그러나산불피해지보다위험도가낮은일반산지에서나타난산사태위험지도의정확도 75 를최소위험도로가정하여실제발생한산사태공간자료와산사태위험지도를등급분포를분석하였다. Figure 3 의 (a) 에서남원산불피해지역은산사태위험 1 등급이없고대부분 2, 3 등급이분포하고있어산사태위험지도의정확도는 55% 로나타났다. 그러나 (b) 에서와같이다른두지역과같은분포를갖을경우, 즉산불피해로산사태위험도가증가하여현재의 2, 3 등급이 1, 2 등급으로상승하여평균정확도 87% 에가까워진다면산사태위험도상승효과는약 32% 가증가하는것으로판단할수있다. 산불피해에의한산사태위험상승효과를더욱정확히알기위해서는실제산불지역에서발생한산사태를조사하여산불피해지에대한산사태위험도판정방법이연구되어야한다. 그리고이를바탕으로제작된위험지도를이용하여기존의산사태위험지도와의비교를통해분석되어야한다. 본연구는이러한연구를시작하기위한기초연구이며향후앞서언급한것같이산불피해지를위한맞춤형산사태위험지도를개발하도록노력할것이다. 75 산사태발생위치에나타난산사태위험지도 1, 2 등급비율

313 June 7-8, 2011 인용문헌 (References) Choi, K., 1998, Technical manual for conserving soil & water, Korea Forest Service, p416. Park, K., Lim, J. and Lee, Y., 2009, The Surface Geology Changes in Large Forest Fire Affected Areas, Korea Forest Research Institute. Youn, H., Lee, C., Woo, C., Jeong, Y. and Lee, C., 2009, Development of debris flow mechanism and predict method of sediment area, Korea Forest Research Institute. Woo, C., Youn, H., Lee, C. and Jeong, Y., 2008, Analysis of influence factors of forest soil sediment disaster using aerial photographs, J. Korean Env. Res. & Reveg. Tech. 11:

314 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Estimation on Forest Fire Potential Hazard by Fuel Load Change of Forest Strata in Forest Stand 산림내층위별연료량변화와임내산불잠재위험성평가 Myoungsoo Won 1*, Youngjin Lee 2, Kyosang Koo 1, Myungbo Lee 1 1 Div. of Frest Disaster Management, Korea Forest Research Institute, dongdaemun-gu, South Korea 2 Div. of Forest Resource, Kongju National University, Kongju, South Korea Abstract mswon@forest.go.kr This study performed a field survey of pine forest and oak kind (Quercus acutissima and Quercus variabilis) broadleaf forest in southern Chung-Nam Province (including Daejeon), Gongju, Yesan and Cheongyang. Upper trees in southern Chung-Nam province indicated that ground fuel loads of pine trees stand including bark, stem xylem, live branches and leaves ranges from 7.11 ton/ha to ton/ha, which accounts for 58.5% to 90.9% in each part of fuel loads that contribute to upper class fuel loads out of total amount. Oak trees also demonstrated similar fuel loads and proportions, which are 10.7~ ton/ha and 69.0~95.1% on upper class, and 0.82~4.19 ton/ha and 0.5~1.9% on middle class respectively. Three parameters (fuel moisture content (FMC), height to the live crown base (CBH) and surface fire intensity (SFI)) were calculated for potential risk estimation on stands on the condition of that fire spreads successfully from surface to crown. These parameters were introduced for assessment of potential fire risk by estimating critical surface fire intensity (CSI), which determines crown fire transition. 요약 산림내산불잠재위험성평가를위해충남지역을대상으로소나무림과참나무림의층위별연소물량의분포를추정하기위하여각임분의층위별연소물량을 10m 10m 방형구내의상층 ( 수고 8m 이상 ), 중층 (8m 이하 ), 관목층, 지표층 ( 초본, 낙엽, 낙지 ) 을대상으로총 36 개소를조사하였다. 조사결과를이

315 June 7-8, 2011 용하여단위면적당 (ha) 바이오매스량 ( 연소물량 ) 을추정하였다. 혼효림의층위별연소물량은소나무림과참나무림에서얻어진결과를 1/2 씩합산하여추정하였다. 분석결과충남지역소나무임분의지상부연소물량중수피를포함한줄기목질부, 생지부, 생엽부를모두합친상층의연료량은 7.11~ ton/ha 로서각각의총량중 58.5~90.9% 를차지하였고영급이증가할수록상층연소물량이증가하는경향을나타내었다. 참나무임분의상층연료량은 10.7~ ton/ha 로서각각의총량중 69.0~95.1% 를차지하였다. 본연구결과에서얻어진층위별연소물량의공간분포를이용하여충남지역의임내의잠재위험성평가를위해지표화에서수관화로전이되는조건으로연료습도 (FMC), 지하고 (CBH), 지표화강도 (SFI) 등 3 가지파라미터를산출하여, 수관화전이를결정하는지표화강도 (CSI) 를추정하여연료량변화에따른임내산불잠재위험성을평가하였다 Introduction Fire behavior and fire severity are determined by characteristics and vertical and horizontal continuity of diverse fuel level, and fire danger on a particular stand affects on fire behaviour based on potential amount of fuel (Russel et al., 2004). Fuelbed strata has various representation of combustion environment, fire spread rate and fire impact. In addition, crown layer, ladder fuels and shrub layer impact on crown fire, and herb layer, surface fuel and woody fuels have an effect on surface fire (Snadberg and others 2001). There is a limitation of fire potential danger assessment on the basis of a single method for preparation and correspondence of yearly repeated fires. Therefore, this study performed an estimation of burning quantity on each layer of pine and oak forest in Chung-Nam, South Korea by using 4th digital forest type map and field survey data. Then, performance of fuel load map production about upper, middle, shrub and surface layer in the forest was carried out to estimate the fire risk. Materials and Mehtods This study performed a field survey of pine forest and oak kind (Quercus acutissima and Quercus variabilis) broadleaf forest in southern Chung-Nam Province (including Daejeon), Gongju, Yesan and Cheongyang. Fossil-wood survey was introduced in pine forest and broadleaf forest (Quercus mongolica and Quercus variabilis), which covers 3 plots in each age class (Ⅰ- Ⅵ) (36 plots in total) with 10m 10m plot size as the two forest types show similar characteristic

316 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 of stand in each age class. Diameter at the breast height (root-collar caliper for under age class 2), age of trees, tree height and number of trees in a plot are measured on ground section (upper, middle tree) in each plot. Middle trees were cut down at the closest ground height to measure tree height and fresh weight (weight of fuel). In order to estimate surface fuel load, fresh weight measuring was conducted. It includes shrub, herb, fallen leaves and branches on existing pine and broadleaf tree stand. Size of square plots were designed as 2m 2m for shrub layer and 0.5m 0.5m for herb, fallen leaves and branches at average area 3 repeated times. After every specimen was cut and measured, fresh weight was measured in the field. Among selected specimens, one third were separated and packed to deliver to laboratory. They were dried in a dry oven until specimen temperature reached 85. Then, dry weight calculation was conducted at the laboratory with the utilization of two variables (tree height and diameter at the breast height) to estimate upper tree fuel load. The calculation employed The Wood Resource Assessment Program (Korea Forest Research Institute, 2004), which was developed by Korea Forest Research Institute. On the basis of dry weight (fuel load) calculation in each layer on the quadrat, amount of biomass estimation per unit area was performed. Finally, this field survey was introduced to the ArcGIS 9.3 program for building up a spatial data set. This procedure took advantage of the fuel load according to structure of the stand in each layer, which is based on classification of coniferous forests, broadleaf forests and mixed forests in each age class from the 4th digital forest type map. In addition, the data set evaluates fire potential risk based on Van Wagner's (1977) Crown Fire Theory. Results and Discussion 1. Fuel load of stand in each layer as age class changes Upper trees in southern Chung-Nam province indicated that ground fuel loads of pine trees stand including bark, stem xylem, live branches and leaves ranges from 7.11 ton/ha to ton/ha, which accounts for 58.5% to 90.9% in each part of fuel loads that contribute to upper class fuel loads out of total amount. It demonstrates that the fuel loads increased as age class increased. Also, middle class trees represented an increase of fuel loads as age class increased. The fuel loads were between 0.55 ton/ha and 3.53 ton/ha, which accounts for 0.53~3.53% of the total amount. On the other hand, a shrub layer fuel loads change as age class increased mostly indicated opposite to the other classes. Shrub layer fuel loads ranged from 0.06 to 1.91 ton/ha, which comprises between 0.1 and 15.7% from total ground fuel loads. Surface layer, which includes a herb, leaves and

317 June 7-8, 2011 fallen leaves and branches, showed fuel loads 0.04~1.12 ton/ha, 1.11~17.92 ton/ha and 0.91~6.99 ton/ha respectively. Each of them accounts for 0.01~9.2%, 6.1~15.8 and 1.3~7.5%. The result of shrub layer represents that fuel loads of herb decreased as the age class increased, while fallen leaves and branches increased fuel loads at the same condition. Oak trees also demonstrated similar fuel loads and proportions, which are 10.7~ ton/ha and 69.0~95.1% on upper class, and 0.82~4.19 ton/ha and 0.5~1.9% on middle class respectively. The same trend as the shrub layer in the pine trees occurred. Fuel loads ranged from 0.14 to 0.35 ton/ha, which accounts for 0.1~1.6%, with fluctuation of fuel loads. Relative proportions of herb (0.05~1.15 ton/ha), fallen leaves (2.95~6.64 ton/ha) and branches (0.32~6.81 ton/ha) fuel loads out of total ground fuel loads, were 0.02~7.7, 2.2~19.6% and 20.~3.1% respectively. The oak trees showed a tendency towards decrease of herb fuel loads and increase of fallen leaves and branches fuel loads as the age class increased. 2. Estimation of potential fire risk on stands from fuel loads change In general, intense surface fires play a role in drying upper class fuel (Alexander 1998) and crown fuels (referred to as canopy fuels, or aerial fuels) imply hanging fuels (vine, moss, live leaf, branch etc) on trees. These fuels are mostly fine or live fuels, which are less than 0.25 inch. Crown fuel is a biomass necessary for crown fire that can spread on to shrubs and trees in lower classes or to other crowns. Shrub/small tree stratum provides continuity of fire from surface fuel to crown fuel, which eventually impacts on crown fire as ladder fuels and as

318 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 it enhances intensity of fire. Three parameters (fuel moisture content (FMC), height to the live crown base (CBH) and surface fire intensity (SFI)) were calculated for potential risk estimation on stands on the condition of that fire spreads successfully from surface to crown. These parameters were introduced for assessment of potential fire risk by estimating critical surface fire intensity (CSI), which determines crown fire transition. A support theory for the assessment is that crown fire transition can be predictable when SFI is higher than CSI or remains to only surface fire when SFI is equal or lower than CSI. A formula for the theory is described below: CSI = CBH 1.5 ( FMC) 1.5 Where, CSI = critical surface fire intensity required for crowning (KW/m), CBH = height to the live crown base (m), and FMC = foliar moisture content (%). A spatial composition of CBH and surface layer (fallen leaves and branches, herbaceous layer) fuel loads in southern Chung-Nam province is described in Figure 2. As can be seen in Figure 3, the figure describes changes of CSI in response to FMC and CBH changes. It reveals that a higher CSI figure is required for the successful transition from surface fire to crown fire when CBH increases. Moreover, crown transition CSI estimation result is described Figure 4, which is about a spatial distribution when we assume that crown fuel moisture content is 50%, which represents that it is suitable for spring dry period moisture content. Based on Figure 4, it is suggested that potential fire risk should be reduced before the crown transition occurs by performing activities such as thinning, pruning and elimination of lower class fuel when CSI figure is low, which possibly results in faster crown transition process. In addition, verification process will be performed with potential fire risk assessment result by using amount of heat release data in each part of wood, which is obtained for experimental purposes, and the spatial distribution of fuel loads in each class, which is acquired by this study in the future

319 June 7-8, 2011 Fig. 2. (left) CBH and (right) distribution of surface fuel loads in southern Chung-Nam province Fig. 3. CSI figure as CBH of pine tree stand changes in southern Chung-Nam province

320 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Fig. 4. CSI distribution in southern Chung-Nam province when crown fuel moisture content is 50 References 손영모, 이경학, 정세경, 김성호, 이우균 임목자원분석론 - 측정 생장모델 평가. 국립산림과학원연구신서제 4 호. pp36. Alexander M. E Crown fire thresholds in exotic pine plantations of Australasia. PhD thesis, Australian National University, Canberra. 228pp. K.G. Hirsch Canadian Forest Fire Behavior Prediction (FBP) System: user's guide. Special Report 7. Canadian Forest Service Northwest Region Northern Forestry Centre. pp Russell T. Graham, Sarah McCaffrey and Theresa B. Jain Science Basis for Structure to Modify Wildfire Behavior and Severity. USDA Forest Service RMRS. General Technical Report RMRS-GTR-120. USDA Forest Service. A strategic Assesment of Forest Biomass and Fuel Reduction Treatments in Western States USFS, Research and Development, In partenership with the Western Forestry Leadership Coalition. Hal E. Anderson Aids to Determining Fuel Models For Estimating Fire Behavior. USDA Forest Service Intermountain Forest and Range Experiment Satation Ogden, UT General Technical Report INT

321 June 7-8, 2011 Jamdes K. Brown and Thomas E. See Downed Dead Woody Fuel and Biomass in the Northern Rocky Mountains. USDA Forest Service Intermountain Forest and Range Experiment Station. General Technical Report INT-177. Russell T. G., Sarah M., and Theresa B. J Science Basis for Changing Forest Structure to Modify Wildfire Behavior and Severity. USDA Frest Service Rocky Mountain Research Station. General Technical Report RMRS-GTR-120. Sandberg. D.V, R.D. Ottmar and G.H. Cushon Characterizing fuels in the 2l st Century. lnternational Journal of Wildland Fire 10:1-7. Van Wagner, C.E Conditions for the start and spread of a crown fire. Canadian Journal of Forest Research. 7:

322 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Analysis of spatial distribution for forest fire damaged areas in East-coast of Gangwon province 강원도동해안산불피해지역의공간분포분석 Byung-oh YOU *, Dong-hwan PARK, Sung-ho KIM, Jun-seob KIM Division of forest resources information, Dept. of forest & climate change, 57 hoegi-ro, Dongdaemun-gu, Seoul, , Korea Abstract Forest fires have been threatened to natural resources, endangered species, properties and even to human lives. Efficient management of forest fires requires a complete understanding of the environmental and human related activities, as well as complicate spatial relationships among them. And understanding the impact of fire requires detailed knowledge about where fires occur, how they alter landscape patterns and precise calculations of fire regime parameter. In this work, a reconstruction of the perimeters of fires that occurred during was made by means of aerial photographs, forest cover type map aided by field inspection in study area located in East-coast of Gangwon province. Therefore, this study is to provide and classify the terrain, forest cover type, topographical factors, and to analyze spatial distribution for forest fire areas by applying the GIS/RS technology. 요약 산불의효율적인관리를위해서는산불발생에영향을미치는인자와이들사이에서발생하는복잡한공간적관계에대한상세한정보가요구된다. 본연구는 년동안에발생한강원도동해안산불피해지의공간적특성을정량화하고산림재해에대한기초자료를제공하기위하여수행되었다. 대부분의산불피해지는고도 100m 이하, 경사도는 에서가장큰면적분포를보이는것으로나타났으며, 소나무림 Ⅱ 영급에서가장큰피해면적을보였다. 동해안산불피해지는임분구조, 지형과기후조건그리고특히가파른지형에상당히민감한특성을보이는것으로나타났다

323 June 7-8, 2011 Introduction Massive forest fires have caused considerable environmental damage and adversely affected the social, economic and cultural aspects of forest sustainability Forest fire may profoundly alter the structure of the landscape (Dansereau and Bergeron, 1993). Landscape patterns can affect ecological processes (Turner, 1989), among others, the spread of disturbances (Turner et al., 1989), and, in particular, of fire. Consequently, much of the attention has focused in understanding the interactions between landscape patterns and fire since, ultimately, fire occurrence may be strongly linked to such patterns (Davis and Burrows, 1994). Mapping past fires is only possible through reconstruction based on aerial photography or other remote sensing procedure. It has been shown that, depending on the source of ignition, fires may occur in different places, and affect different types of vegetation (Vazquez and Moreno, 1998a). A prefferd occurrence of fire on certain locations within a given landscape means that average figures of fire regime parameters would be misleading, since some areas will burn more frequently than others. Knowing this is important for management, both in terms of risk evaluation and in terms of the consequences for the ecosystem. Therefore, understanding the ultimate role of fire requires that we know what patterns it creates, particularly with regard to the spatial distribution of burned areas, as these may affect future fire propagation, and also of where fires actually occur. Mapping fires, and determining the relationships between topographic features or other characteristics of the terrain and burned areas is important to evaluate the impacts of fire. The objective of this work is to map the fires that occurred in an area of East-coast in Gangwon province and determine where fires actually occurred in relation to topographic features (elevation, slope, aspect) or other characteristics of the terrain. Material and Methodology We used the FGIS information of all fires that occurred during the period in forest fire damaged areas, Sokcho-si(2004), Gangneung-si(2000, 2004), Samcheok-si(2000), Yangyang-gun(2005), Donghae-si(2000), Goseonggun(1996, 2000) of Gangwon province. In order to analyze the areas affected by fires as a function of their topographic attributes, we constructed a digital elevation model (DEM) of this burned areas. We digitized the elevation lines at meter intervals from a 1:25,000 topographic map from this burned areas. The resulting vector files were rasterized to an image. From this image, and by means of interpolations, we obtained the DEM from which, once filtered, we

324 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 derived the corresponding layers of elevation, slope, aspect. Results and Discussion Topographic features of forest fire damaged areas Burned areas were not distributed proportionately to the available areas of forest territory according to their topographic features. Most of damaged area was much bigger at lower elevations (below 100m), at steep slope-angles (from 20 to 30 ), and at various aspects. Since 1996, total eight times forest fires recently have been occurred in East-coast of Gangwon province. Forest fire has been occurred with frequency in The biggest forest fire was that for 17,097ha in Samcheok-sin Samcheok-si at 2000, it is the largest damagedd area that approximately 5,882ha at As a result of analysis of spatial distribution for forest fire occurred on elevation 50~100m and 8,630ha occurred on slope-angles 20~ 30. Figure 1. Distribution of forest fire damaged areas in Samcheok-si and Donghae-si Characteristics of forest type in forest fire damaged areas Pine forests damaged around 9,500ha weree the major damaged forest type. Age class Ⅱ(from 11 to 20 years) has been mostly damaged among the forest type D. Based on the information by the 4 th forest type map DB, 90% of the burned area was coverd by trees, i.e., it was a wooded-forest of pine types described above, although very young pine stands are also frequently considered as wooded- forest. Figure 2. Damaged area by age class in Samcheok-si and Donghae-si

325 June 7-8, 2011 Trends of forest fire in East-coast area Forest fire damaged areas in East-coast of Gangwon province is highly susceptible to forest fires because of stand structure and topographic and climatic conditions. The forest is currently characterized by thick growth because of inadequate thinning and remaining debris, and its thick layer of fallen leaves catches fire and burns easily. Forest fires also spread rapidly in these mountainous areas (fires spread eight times faster on steep slopes than to level ground). References Dansereau, P.R., Bergeron, Y., Fire histoy in the southern boreal forest of northwestern Quebec. Can. J. For. Res. 23, Davis, F.W., Burrows, D.A., Spatial simulation of fire regime in Mediterranean-climate landscapes. In: Moreno, J.M., Oechel, W.C. (Eds.), The role of fire in Mediterranean-type Ecosystems. Ecological Studies, Vol Springer, New York, pp Turner, M.G., Landscape ecology: the effect of pattern on process. Ann. Rev. Ecol. Syst. 20, Turner, M.G., Gardner, R.H., Dale, V.H., O Neill, R.V., Predicting the spread of disturbance across heterogeneous landscapes. Oikos 55, Vázquez, A., Moreno, J.M., 1998a. Patterns of lightning-, and people-caused fires in peninsular Spain. Int. J. Wildland Fire 8(2),

326 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 The Spatial Change of Forest Fire Occurrence during the Spring Season between 1990s and 2000s 1990 년대와 2000 년대봄철산불발생의공간적변화분석 Suk-Hee Yoon 1, Myoung Soo Won 1*, Kyo Sang Koo 1, Kyong Ha Kim 1, Woo Kyun Lee 2 1 Korea Forest Research Institute, 57 hoegi-ro, Dongdaemun-gu, Seoul, , Korea 2 Department of Environmental Science and Ecological Engineering, Korea University, Seoul, , Korea mswon@forest.go.kr Abstract For efficient forest fire management in Korea, we need to minimize the damage of forest fire and manage them systematically. Therefore, This study aims to analyze the spatial change of forest fire occurrences during the spring season(february to April) between 1990s and 2000s. To analyze the frequencies of forest fire occurrence, we used a comprehensive database from the forest fire inventory from 1991 to 2009 and analyzed the spatial patterns of fire occurrence at the interval of ten days between 1990s and 2000s. As a result of analysis, early April showed the highest frequency of forest fire occurrence both 1990s and 2000s. Compared to the 1990s and 2000s, the regional change of forest fire showed the most frequent fire events around Chungcheong province. Especially extra 27 fires increased in Daejeon city, and the second most frequent fire had more than 10 fires in Jeolla province and Incheon. However, the number of fire frequency decreased by 12 fires at the end of April in Hongcheon county(the province of Gangwon). This is the largest drop over the study period. 요약 본연구에서는집중적으로발생하는봄철 (2 4월) 산불을효율적이고체계적으로관리하기위하여 19년간 ( 년) 발생한봄철산불을 1990년대와 2000년대로나누어순기별산불발생빈도공간패턴을분석하고연대별에따른순기별산불발생빈도변화패턴을분석하였다. 분석결과 1990년대와

327 June 7-8, 년대에가장높은산불발생빈도를나타낸순기가 4 월상순으로같았지만이에대한지역별산불발생빈도가한반도서쪽지역인충청지역을중심으로대전에서 27 건으로가장많이증가하였고, 호남및인천지역에서 10 건이상증가한것으로분석되었다. 그러나강원홍천지역에서는 4 월하순에최대 12 건이감소한것으로나타났다. Introduction 우리나라에서는 1991년부터 2009년까지연평균 448건의산불이발생하였고, 이중에서 73% 에해당하는산불이봄철건조기인 2월과 4월사이에집중적으로발생하였다 ( 산림청, 2009). 또한최근 5년간 (2001~2005년) 발생한산불 2,715건중봄철산불은 1,824건으로전체의 67% 를차지하고있다 ( 원명수등, 2006). 따라서본연구에서는집중적으로발생하는봄철 (2 4월) 산불을효율적이고체계적으로관리하기위하여 19년간 ( 년) 발생한봄철산불을 1990년대와 2000년대로나누어순기별산불발생빈도의공간적인발생패턴을분석하고 1990년대와 2000년대의산불발생빈도변화를분석하고자하였다. Methodology 먼저국립산림과학원에서개발된산불연구정보시스템을활용하여 19 년간 (1991~2009 년 ) 발생한산불발생위치를연속지적도의 PNU 코드 (19 자리수 ) 와매칭하여산불발생위치포인트테이터를지적의중심점으로하여공간자료화하였다. 구축된산불발생위치자료는 1990 년대와 2000 년대로나누어봄철순기별따른 165 개시 군별산불발생빈도데이터를행정구역별폴리건과포인트정보를결합한후 ArcGIS Ver.9.2 프로그램을이용하여 1km 1km 공간해상도로 IDW(Inverse Distance Weighting, 보간법 ) 분석을실시하였다 년대와 2000 년대로나누어순기에따른각각의산불발생빈도분석결과는다시 1990 년대와 2000 년대의지역별 순기별각래스터별로시군의공간적빈도변화차이를분석하였다. Results and Discussion 1990 년대대비 2000 년대의봄철을순기별산불발생빈도변화를분석한결과, 3 월중순 >4 월상순 >3 월하순순으로 2000 년대의산불발생빈도가증가한것으로분석되었다 년대에는 1990 년대에비하여 3 월중순부터 4 월상순까지의산불발생지역의공간적분포가넓게나타났다 년대에비하여 2000 년

328 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 대의산불발생빈도변화가가장많이증가한지역은 3 월하순과 4 월중순에대전에서 27 건으로나타났다. 또한 5 개광역시는산불발생빈도가전부증가되는경향을나타냈다. 반면에 4 월에집중적으로발생했던강원지역중강원홍천에서 4 월하순에 12 건의산불이감소하여가장큰감소폭을보였다 년대는 1990 년대에비하여 2 월초순부터영남과강원영동지역에서증가하는패턴을보이면서 3 월중순부터는전국적으로발생빈도변화량이커졌다. 3 월하순으로접어들면서강원지역을제외하고전지역에서증가되었고 4 월초순에는충청 ( 대전포함 ) 과호남 ( 광주포함 ) 지역을중심으로 10 건이상증가하는산불발생빈도변화를나타냈다. 따라서본연구분석결과, 1990 년대와 2000 년대에가장높은산불발생빈도를나타낸것이 4 월상순으로같았지만이에대한지역별산불발생빈도가한반도서쪽지역인충청지역을중심으로호남지역과인천이높게나타나는경향을보였다. Figure 1. Change of forest fire frequency of ten-days intervals from February to April between 1990s and 2000s. References Korea Forest Service Korea Forest Service, Statistical yearbook of forest fire. Won, M. S., Koo, K. S. and Lee, M. B An analysis of forest fire occurrence hazards by changing temperature and humidity of ten-day intervals for 30 years in spring. Korean Journal of Agricultural and Forest Meteorology 8(4):

329 June 7-8, 2011 Chemical Composition of Precipitation and Wet Deposition in Forest Area, Gangwon. 강원산림지역에서의강수화학성특성 Heejung Youn*, SeungWoo Lee Korea Forest Research Institute, 57 hoegi-ro, Dongdaemun-gu, Seoul, , Korea Abstract To look into the degree of the long distance movement of the air pollution within Mt. Kyebang, Hongcheon one of the clean regions in Korea, major ions, components of collected precipitation, were analyzed by using the ionchromatography after measuring ph and electrical conductivity. The quality assurance of chemical composition data was checked by considering the ion balance and electrical conductivity. Also, the ph of precipitation showed in the ph5.0~5.5 range, less than the ph5.6 in which the standards of the acid precipitation. On the whole, the more precipitation got, the less ion concentration became: It could showed what it influenced on the fluctuation of the precipitation ph according to fluctuation and neutralization of the according materials. In the meantime, as composition ratios of wet deposition, SO 4 2- and nss-so 4 2- of anions, NH 4 + and nss-ca 2+ of cations accounted for high percentage. Especially, as the concentration of SO 4 2-, the artificial pollutant, and Ca 2+, the soil substance, showed high, Mt. Kyebang, Hongcheon having barely fixed air-pollutants sources was most likely to be influenced by the pollutants moved from long distances. 요약 대기오염은인구밀집지역및산업물질에의한인위적오염원과산불및해염입자등에의한자연적오염원에의하여영향을받는다. 특히배출오염원이적은산림내에서는자연적오염원과인위적오염원에의한장거리이동에의하여대기오염에영향을받을수있다. 본연구에서는강원도내비오염지역인홍천계방산지역내의대기오염물질의유입정도를알아보고자이지역내의강수를채취하여산도 (ph) 및이온분석을실시하였다. 분석자료에대한이온수지와전기전도도 (EC) 검토에서신뢰

330 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 성이확인되었고홍천계방산지역내강수산도 (ph) 는산성우의기준이되는 ph5.6 이하의범위인 ph5.0~5.5 범위에집중적으로나타났다. 대체적으로강수량이많을수록이온농도가감소하는경향을나타내었고산성원인물질의증감과중화물질에따라강수산도 (ph) 의증감에영향을미치는것을알수있었다. 한편, 강수중이온성분의조성비율을살펴보면음이온은 SO 4 2- 와 nss-so 4 2-, 양이온은 NH 4 + 와 nss-ca 2+ 가많은비율을차지하였다. 특히대기오염물질배출원이거의없는홍천계방산지역에서 SO 4 2- 와 Ca 2+ 의농도가높았던결과를통해이곳의강수화학성이장거리이동성물질에의해영향을받고있음을추정할수있다. Introduction Evaluating acidity of precipitation is essential to understand the influence of air pollution on forest ecology. As local and overseas researches on ioncomposition by precipitation reported, it is known that the long-distance movement of air polluting elements and yellow dust do affect Korea. For this, chemical researches of precipitation composition are under study based on urban area and coastal areas including the Westsea, yet studies on long-distance movement for mountainous regions with rather lower pollution level have rarely performed. Thus, this research has targeted the less polluted region of Mt. Kyebangin Hongcheon, Gangwon-do to measure the influx of pollutants caused by the long distance movement with precipitation ph analysis as well as ioncomposition and the balance. Methodology The collected samples were cold-stored under 4, and then filtered through 0.45μmmembrane filter to measure EC and ph per each. The concentration of anion(no - 3,Cl -, SO 2-4 and cation(nh + 4,Ca 2+,Mg 2+,K +,Na + ) were analyzed utilizing ion-chromatography. The ph average is calculated converting the ph value into hydrogen ion concentration, precipitation being weighted average, and reconverting into ph; then the reliability verification of the precipitation analysis data was carried to minimize the margin of error. There liability verification for ion- exchange resin and EC (electrity conductivity) balance was performed by checking of the monthly average value of total samples

331 June 7-8, 2011 Results and Discussion The Reliability Verification on Precipitation Analysis Data The total linear slope between total anion concentration and total cation concentration was , close to1, and R 2 marked (P<0.001) to prove there liability of analysis results, while the balance between cation and anion showed relatively fine. The total of cation marked relatively higher than one of anion, which appears to be influenced by other ions with out anion analyzed element, including HCO - 3 and organic acid such as formic acid and acetic acid. The calculated value of electrical conductivity showed higher than the measured value of electrical conductivity; yet judging from the total slope marked and R 2 was (P<0.001), the significance level is high and the balance between measured electrical conductivity and calculated electrical conductivity shows comparably fine. The ph Distribution of Precipitation The ph appearance-frequency of measured precipitation shows that over 80% of precipitation on the region of Mt.Kyebang, Hongcheon located between the range of ph 5.0~5.5. The Relation between ph and Precipitation Precipitation is measured between the range of 2.0~115.6mm with average 51.88±54.82mm. It shows ph widely spread at precipitation below 50mm. This appears the high influence of fine particle in the air and polluted gas materials. The Ion-Composition by Precipitation The ion-composition ratio of monthly precipitation in 2008 shows anion in the order of SO 4 2- >nss- SO 4 2- >NO 3 >Cl -, while cation in the order of NH 4 + > nss- Ca 2+ >Ca 2+ > Na + >Mg 2+, K +, The major base element NH 4 + ion is followed by Ca 2+ high as 25%. Among the precipitation elements, Ca 2+ is caused from ocean, but also from artificial pollutant such as pavedroad, as well as natural pollutant of soil evaporated from surface to the air by wind or yellow dust caused by the long- distance movement from desert area. Considering that there is a few artificial pollutant in Hongcheon area with comparatively less traffic and less pollutant rate nearby, the major influence is presumed from yellow dust moved long distance. Also with the fact that SO 4 2- takes 33% of anion, and nss SO

332 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 takes 31%. We expected that the major cause comes from fire smoke, artificial pollution by SO 4 2-, yet Hongcheon of the clean area is affected by long distance movement of air pollution substances occurred out side of the region. References Ahmed, AFM., Singh, R.P. and Elmubarak, A A chemistry of atmospheric precipitation at the Western Arabian Gulf Coast. Atmospheric Environment 24(12): Christian, E.J Air chemistry and Radio activity. Academic Press. pp Fujita, S., Ichikawa, Y., Kawaratani, R. and Tonooka, Y Preliminary inventory of sulfur dioxide emissions in East Asia. Atmospheric Environment 25: Johnston, J., William,Jr., Shriner,D.S., Klarer,C.I. and Lodge,D.M Effect of rain ph on senescence, growth, and yield of bush bean. Environmental and Experimental Botany 22(3): Lee, B.K., Hong. S.H. and Lee, D.S Chemical composition of precipitation and wet deposition of major ions on the Korean peninsula. Atmospheric Environment 34:

333 June 7-8, 2011 Trends of Timber Products in East Coast Forest Fire Area after 2000 in Korea 2000 년이후동해안산불지역의목재생산추이 Abstract Seongyoun Lee* Korea Forest Research Institute, Seoul, , Korea The east coast forest was damaged by forest fire in April 2000 in Gangwondo and the area burnt was about 23,794ha. This was about 1.7% of forest area in Gangwon-do. The damaged timber was 1,539 thousand m3, was 97% of domestic timber production (1,592 thousand m3 in 2000). After the forest fire on the east coast in 2000, timber production in Gangwon-do decreased by After that, timber production is rising gradually. After 2006, with the rise in domestic timber production, timber has been produced more than 300 thousand per year in Gangwon-do. By regional groups, timber production in Gangneung and Samcheok was double in 2000 compared with other years then the production has increased steadily. In 2001, timber production in Donghae was 10 thousand m³ which was five times more than in 2000, however, after that became similar to production in the average year. It was probably caused by supplying wood which can be used as timber since damaged tree had been removed. 요약 2000 년 4 월에강원도동해안지역의산불로약 23,794ha 의산림이피해를받았다. 이는강원도산림면적의약 1.7% 에해당되며, 당시피해축적은 1,539 천m3으로, 2000 년국내재생산량인 1,592 천m3의 86% 에해당된다 년동해안산불이후강원도의목재생산은 2003 년까지감소를보이다가이후점차증가를보이고있는것으로나타났다 년이후우리나라의국내재생산량이증가되면서강원도에서는매년 300 천m3이상을생산하는것으로나타났다. 지역별로보면, 강릉과삼척지

334 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 역의경우 2000 년에는예년의 2 배에달하는생산을보였으며이이후에는꾸준히증가하는추세를보이고있다. 동해시에서는 2001 년에 2000 년의 5 배에달하는약 10 천m3을생산하기도하였으나이후예년수준으로나타났다. 이러한현상은동해안산불로피해목제거를하면서용재로서사용이가능한목재를공급하여나타난측면도있는것으로추정된다. 서론 (Introduction) 우리나라는정부수립이후 1996 년에최대규모인강원도고성산불 (3,762 ha ) 등 5,368 ha에달하는산불피해발생하였으며, 2000 년 4 월에는고성, 동해, 강릉, 삼척울진산불이동해안지역을따라거의동시다발적으로발생하여건국이래최대의 23,794 ha초대형산불피해가나타났다. 우리나라의목재생산량은 2009 년현재 3,176 천m3로, 2000 년도동해안산불로피해를받은산림피해재적량인 1,539 천m3의약 2 배에해당된다. 강원도에서는동해안산불이발생한 2000 년을중심으로 2003 년까지감소를보이다가이후증가를나타내고있다. 이연구는 2000 년동해안산불피해가발생한지역을중심으로행정구역별로목재생산과관련된통계자료를이용하여 1995 년부터 2009 년까지목재생산의변화를분석하였다. 재료및방법 (Methodology) 동해안산불지역의목재생산의추이는임업통계연보와임산물생산통계및해당시군통계에서자료를확보할수있는생산량및생산액의자료를이용하여분석하였다. 시계열자료는 1995 년부터 2009 년까지의 data 를이용하였으며, 이중임상별자료를확보할수있는경우에는이자료를이용하여생산액과생산액의추이도함께분석하였다. 한편, 산림청의통계자료와지방정부의통계자료와일치하지않는경우에는산림청의통계자료를이용하였으며, 일부통계 data 의경우에는해당지방정부의통계를이용하여분석하였다. 결과및고찰 (Results and Discussion) 2000 년동해안지역의산불피해면적은총 23,794 ha로영급별로살펴보면,

335 June 7-8, 2011 Ⅰ영급이 3,348ha, Ⅱ영급이 3,280ha, Ⅲ영급이 7782ha, Ⅳ영급이 8446ha, Ⅴ영급이 848ha이다. 시군별로보면, 고성군이 2,696ha, 강릉시가 1,447ha, 동해시가 2,224ha, 삼척시가 17,097ha, 울진군이 310ha로나타났다. 고성군은대부분 Ⅰ 영급이피해를입었으며, 강릉시는 Ⅳ영급, 동해시도 Ⅳ영급에서피해를많이받은것으로나타났으며, 삼척시는 Ⅲ~ Ⅳ영급에서피해가많은것으로나타났다. 울진군은 Ⅱ~ Ⅲ영급에서피해면적이많은것으로나타났다. 표 1. 시군별영급별산불피해면적 단위 : ha 구분 계 Ⅰ영급 Ⅱ영급 Ⅲ영급 Ⅳ영급 Ⅴ영급 계 23,794 3,438 3,280 7,782 8, 고성 2,696 1, 강릉 1, 동해 2, , 삼척 17,097 1,161 2,942 5,781 6, 울진 한편 2000 년동해안지역의산불피해축적은총 1,539 천m3로영급별로살펴보면, Ⅱ 영급이 123 천m3, Ⅲ 영급이 461 천m3, Ⅳ 영급이 853, Ⅴ 영급이 101 천m3이다. 시군별로보면, 고성군이 52 천m3, 강릉시가 134 천m3, 동해시가 163 천m3, 삼척시가 1,171 천m3, 울진군이 18 천m3로나타났다. 산불피해축적중용재로서의가치가있다고판단되는 Ⅲ 영급이상이약 92% 에해당된다 표 2. 시군별영급별산불피해축적 단위 : m3 구분 계 Ⅰ영급 Ⅱ영급 Ⅲ영급 Ⅳ영급 Ⅴ영급 계 1,539, , , , ,217 고성 52, ,949 3,936 0 강릉 134, ,113 45,856 76,237 3,272 동해 163, ,940 27, ,302 2,300 삼척 1,170, , , ,147 95,645 울진 18, ,760 10,565 2, 년동해안산불이발생한강원도의목재생산량의추이를보면그림 1 과같이산불이발생한 2000 년부터 3 년간감소하는경향을보이다가 2003 년이후목재생산은증가하는모습을나타내고있다

336 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 천m 산불피해를입은지역의목재생산은그림 2 에서보는바와같이삼척시의경우산불발생이후 4 년간은정체현상을보였으나이후약간증가하는추이를보이고있다. 강릉시의경우산불발생이후 2003 년까지는절반수준으로감소를보이다가이후약간증가하는추이를보이고있으며, 고성군의경우산불발생에도불구하고 2006 년까지는증가를보이다가이후감소하는경향을나타내고있다. 동해시의경우산불발생이후증가하였다가 2003 년이후부터는정체현상을보이고있다. 천m3 100 고성강릉동해삼척

337 June 7-8, 2011 인용문헌 (References) 동해안산불피해지공동조사단 산불피해지의건전한자연생태계복원및항구적인산림복구계획수립을위한동해안산불지역정밀조사보고서 Ⅱ p. 산림청 임업통계연보. 산림청 임산물생산통계. 산림청 동해안산불백서 Ⅰ, 401p. Rin Won Joo, Byung Heon Jung, Seong Youn Lee, Jae Soo Bae, Kyongha Kim, Sang-Yoel Han, Jong-Hwan Lim, Kyeong-hak Lee, Rheehwa Yoo Trends and Projections for the Forest Sector in the Republic of Korea. Korea Forest Research Institute. 118p

338 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Monitoring of Vegetation Recovery in the East Coastal Forest Bunt Area of Korea using Multi-temporal Landsat Images Abstract 다시기 Landsat 영상을이용한동해안산불지역의식생회복모니터링 Seungho Lee*, Hyunkook Cho, Eunsook Kim, Cheolmin Kim Korea Forest Research Institute, Seoul, , Korea Using multi-temporal Landsat TM data, NDVI and MSI changes over times were analyzed according to forest type and damage class in order to examine the vegetation recovery after forest fire. NDVI values were rapidly decreased and MSI values were increased after containment of the fire and gradually increased over times. NDVI and MSI curves of coniferous forest - especially red pine forest - yielded the lowest and highest values respectively. In 2007, NDVI and MSI recovery rate was within about 98% and 93% respectively. It was shown that recovery of MSI was small in severely damaged area. Topographical characteristics of short-term recovery area was different from that of long-term recovery area. 요약 산불이후의식생회복을조사하기하여다시기 Landsat TM 위성영상을이용하여 NDVI 와 MSI 의변화를분석하였다. 산불발생에이후, NDVI 값은빠르게감소하고 MSI 는증가하였으며그후점차값이회복되었다 년현재, 산불미피해지와비교했을때 NDVI 를통해추정한산불피해지의식생활력도는전체적으로약 98% 회복이되었으며, MSI 를통해추정한수분스트레스는약 93% 회복되었다. 임상별로는회복력의차이가뚜렷하게나타나지않았으나,

339 June 7-8, 2011 수분스트레스는산불피해강도가클수록회복률이작은것으로나타났다. 단기간에빠른식생회복을보인산불피해지는장기간에큰식생회복을보인지역과비교했을때, 서로다른지형특성을보였다. Introduction The objective of this study was to detect vegetation recovery over times after forest fire occurred in Samcheok city in April 2000, using multi-temporal Landsat image data obtained in 1989, 2000, 2001 and Also, topographic factors that affect vegetation recovery are analyzed according to recovery time. In order to examine the vegetation changes over times, NDVI and MSI values were derived from time series image data sets. Methodology Multi-temporal Thematic Mapper(TM) image data obtained from Landsat-5 were used for this study. The forest fire over the study area was broke out in April. Four analysis images created in 1998, 2000, 2001 and 2007 were analyzed in this study. All images were geometrically referenced to Korean TM(Eastern) projection. Bands 1 through 5 and 7 were converted to at-satellite planetary reflectance. Figure 1. Satellite image data used for change detection NDVI(Normalized Difference Vegetation Index) and MSI(Moisture Stress Index) changes over times were analyzed according to forest type and damage class in order to examine the vegetation recovery after forest fire. NDVI is a typical index to assess the vegetation vitalirty(eq. 2). MSI is used to assess

340 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 water content within vegetation canopy(eq. 2). Also this index has been shown to be very effective for mapping forest damage(vogelmann et al., 2009). Relative NDVI and relative MSI were used to analyze relative recovery rate of damaged area compared to undamaged area. 100 Eq.1 Eq.2 Eq.3 Results and Discussion - Interpretation of Forest Damage by Image Differencing The differences of vegetation change were distinctively appeared in both band3 and band4 after containment of the fire. NDVI value was rapidly decreased and MSI value was increased after fire. This changes were gradually recovered since that time (Figure 2). Figure 2. Image differencing between pre-fire and post-fire image

341 June 7-8, Short-term and Long-term Recovery Short-term and Long-term recovery area of severely damaged area were selected and topographical characteristics of these areas were examined. Short- to term recovery was operated in low elevation area, low slope area compared total severely damaged area. And long-term recovery was operated in eastern and southern slope. Main impact factor of short-term recovery was elevation and slope related to connectivity and impact factor of long-term recovery was aspect related to solar radiation. - Vegetation Recovery of Burnt Forest NDVI and MSI curves of coniferous forest - especially red pine forest - yielded the lowest and highest values respectively. In fact, most of the red pine forest in study area were prone to forest fire and severely damaged(figure 3). In 2007, NDVI and MSI recovery rate according to forest type was within 97~99% and 92~95% respectively. NDVI and MSI recovery rate according to damage class was within 98~999 % and 90~97% respectively. It was shown that recovery of MSI was small in severely damaged area. Figure 3. Relative NDVI and relative MSI change according to forest types Figure 4. Relative NDVI and relative MSI change according to damage class

342 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Figure 5. Topographical characteristics of short/long-term recovery area References Lee, S.H., C.M. Kim and H.K. Cho Vegetation change detection of forest burn area using Multi-temporal Landsat TM and ETM+ Data. Korean Journal of Remote Sensing 20(1): Vogelmann, J.E., B. Tolk and Z. Zhu Monitoring forest changes in the southwestern United States using multitemporal Landsat Data. Remote Sensing of Environment 113:

343 June 7-8, 2011 Mapping and Evaluation of Forest Carbon Stock derived from the National Forest Inventory Data and Landsat Imagery 국가산림자원조사와위성영상자료를이용한산림탄소축적량주제도제작및평가 Jong-Su Yim 1*, Il-Bin Jung 1, Yong-Jin Kim 2, Hyun-Kook Cho 1, Man-Yong Shin 2 1 Division of Forest Resources Information, Korea Forest Research Institute, Seoul , Korea 2 Department of Forest, Environment, System, Kookmin University, Seoul , Korea. jsyim@kookmin.ac.kr Abstract As carbon sink, forests play a critical role. This study was implemented to produce a forest carbon stock map at the county level and to asses the produced map. Field plot data from the national forest inventory were used as training (surveyed in ) and as an independent dataset (surveyed in 2009) data. In order to produce the thematic map, the k-nearest neighbor (knn) technique was applied by combining field data and Landsat TM. Forest carbon stock maps were produced with k=5 and their accuracy were evaluated with the independent dataset. It was resulted that the relative root mean square error and mean deviation for overall data-set were estimated to be 38.5% and -1.8 Cton/ha, respectively. 요약 기후변화와관련하여탄소흡수원으로써산림의역할은점차증가되고있다. 본연구는산림탄소축적량의공간적분포를나타낼수있는산림탄소축적량을제작하고검증하기위하여수행하였다. 국가산림자원조사에서수집된야외표본점자료를훈련 ( ) 및검증자료 (2009) 롤이용하였으며, Landsat 영상과통합하여 k-nearest neighbor 기법에의해시군단위의산림탄소축적량주제도를생산하였다 (k=5) 년에수집된 78 개표본점자료를이

344 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 용하여주제도를검증한결과전체상대추정오차율은 38.5%, 평균편의는 -1.8 Cton/ha 로나타났다. Introduction In the context of climate change, the role of forests as carbon sink is expanded. A national forest inventory has been implemented to assess forest resources in Korea. To efficiently manage forest carbon stock, it is a necessary of its thematic map. Forest carbon map can be produced by combining field plot data and ancillary information such as satellite data or aerial photo which covers including non-observed areas across an area of interest. The objective of this study was to produce a forest carbon stock map and to evaluate a produced map. Methodology Material In this study, field data from the 5 th National Forest Inventory (NFI) were used for mapping and verification of forest carbon stock. The NFI-field plots consisting of 4 sub-plots were systematically distributed over the entire country with a grid size of 4km. Table 1 shows the information of field plots and satellite image used in this study. Field plots surveyed in years were employed to predict forest carbon stock at unmeasured plots as training dataset and those in surveyed in 2009 year were used to evaluate a produced forest carbon stock map. Forest carbon stock per plot was computed with BCF/BEF and CCF by forest type. Landsat TM-5 that has been popularly used forest resources assessment over a large area was acquired at May and June The image was geometrically corrected with a digital map. Table 6. Number of field plots and satellite imagery for mapping forest carbon stock for each study site Classification Landsat TM Study site Yangpyong Wonju Seongju Acquisition date 25. May June 2009 Num. of training plots 2,804 2,804 2,784 Num. of plots for evaluating

345 June 7-8, 2011 k-nearest Neighbor(k-NN) technique The k-nn technique is mostly used to produce thematic maps by combining field plot data and satellite data. Table 3 shows the equations and estimators used in the k-nn method. Table 7. Equations and estimator used in the k-nn technique. Euclidean distance Distance-weighting Carbon stock estimator : spectral values at target plot i, : spectral value at reference plot i, m : number of bands, and : distance between target and reference plots on spectral feature Mapping and Evaluation In order to assess the k-nn estimates for different number of k, a cross validation was used. RMSE, relative RMSE, and mean deviation were computed as following equations in Table 3. An optimal value of k for each county was selected. For evaluating k-nn estimates, independent field data (n=78) surveyed in 2009 were used. Table 8. Assessment Statistics used in this study. RMSE RMSEr Mean deviation RMSE RMSEr MD where and are observed value and estimated value at the plot i, is a mean of estimates, and n is number of field plots used

346 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Results and Discussion Forest carbon stock map From a leave-one-out assessment, an optimal value of k (k=5) was selected and carbon stock maps were produced. With digital forest maps, non-forest areas were eliminated (Figure 1). (a) Yangpyeong county (b) Wonju city (c) Seongju county Figure 1. Forest carbon stock map for each study site (k=5) Evaluation of forest carbon stock map A number of field plots used for evaluating per county were a few from 18 to 31. In the case of Seongju, the relative RMSE and MD were the largest (41.8% and 10.4 Cton/ha, respectively) as shown in table 4. As a result, those of overall dataset were 38.4% and -1.8 Cton/ha. Table 9. Accuracy assessment with field data surveyed in 2009 (n=78) Classification No. of field plots RMSE (Cton/ha) Relative RMSE(%) MD (Cton/ha) Yangpyeong Wonju Seongju overall

347 June 7-8, 2011 Comparision of Small Aera Estimation Techniqes for Providing Forest Growing Stock Volume from NFI5 Field Data 국가산림자원조사자료를활용한소면적산림통계산출기법비교 Il-Bin Jung 1*, Jong-Su Yim 1, Sung-Ho Kim 1, Jong-Chan Kim 1, Man-yong Shin 2,Won-Sung Han 3 1 Div. of Forest Resources Information, Korea Forest Research Institute 2 Dept. of Forest Environment System, Kookmin Univ. 3 National Forest Cooperative Federation Korea Forest Research Institute, 57 hoegiro, Dongdaemun-gu, Seoul, , Korea jungilbin.leohunter@gmail.com Abstract The 5 th National Forest Inventory (NFI5) was implemented in 2006~2010 for providing forest resources statistics. Forest information at county level is impoitant to forest policy making. Since a number of field plots in a county are insufficient, small area estimation (SAE) has been mostly applied. In this study, three small area estimation methods; direct estimation (DE), synthetic estimation using field data within neighboring a county (SNE) and within a boundry based on distance from a county of interest (GE), were compared to evaluate forest statistic at county level. As it turnd out GE could be provide more reliable information. 요약 우리나라에서는보다과학적인전국단위산림기본통계생산과임목자원및산림환경생태자원정보의수집을목적으로 2006 년부터 2010 년까지제 5 차국가산림자원조사를수행한바있다. 본연구에서는원활한정책수립과통계적수요가많은시군구의산림통계산출을위하여보조정보의획득이어려운우리나라의실정에맞추어소면적통계추정방법인합성추정법의적용가능성을모색하고자한다. 이에경기도지역, 31 개시군구의표본점자료를바탕으로직

348 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 접추정법과합성추정법 ( 이웃시군기반및거리버퍼기반 ) 을비교함으로서보다효율적인합성추정의적용방안을모색하였으며, ha 당임목축적과상대추정오차를비교한결과거리버퍼기반합성추정법이다른방법보다정확한추정이가능한것으로나타났다. Introduction In Korea, the 5 th National Forest Inventory (NFI5) was implemented in 2006~2010 for evaluating forest resources statistics. Although the NFI5 was designed to provide forest resources statistics at the national level, forest statistics at county level that is important information for NFI users and decision-making should be derived from NFI data. SAE has been populary applied to provide information for sub-population that did not consider sampling design. Synethic estimation is a method of SAE and used ancillary data from neighboring subpopulation that has homogenous properties to a sup-population of interest. Methodology In this study, growing stock volumes (V/ha) per plot level, were calculated by NFI5 permenat sampled plot (NFI5 PSP) dataset in Geonggi-do that was collected during 2006~2010. NFI5 PSP design is consisting 4 sub-plots as shown in fig 1. Figure 1. NFI5 PSP design Three estimation methods were applied; direct estimation (DE), synthetic estimation using field data in neighboring counties (SNE) and geo-spatial estimation (GE). Calibration approaches wewe used for SNE and GE, respectively, as shown in fig

349 June 7-8, 2011 Figure 2. Calibration methods using GIS in case of Yongin (left: SNE, right: GE) Number of NFI5 PSP (NP) by estimation methods is Table 1 that is main dataset to estimate V/ha and relative standard error (RE). Reasonable distance of GE was selected to closed RE 10% for each county. Table 1. NP at county level for each estimation method in Geonggi-do County DE SNE NP GE County GE DE SNE 1km 2km 1km 2km 23km 24km Gapyung 177 1, Anyang Goyang Yangju Kwacheon Yangpyung 138 1, Kwangmyung Yeoju Kwangju Yeoncheon Guri Osan Gunpo Yongin Kimpo Uiwang Namyangju Uijungbu Dongducheon Icheon Bucheon Paju Sungnam Pyungtaek Suwon Pocheon Siheung Hanam Ansan Hwasung Ansung NP Results and Discussion RE variation by estimation method and GE with increasing distance is fig 3. As it results, RE is dcreasing during buffer distance increasing but, it is a slight difference

350 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Figure 3. RE variation by methods and GE with increasing distance distance Some county that was not inventoried or designed need more distance for estimate reasonable RE for example Osan, Kwangmyung. But Gapyung and Yangpyung has reasonable statistics using DE. Table 2. Comparison of estimation results for each small area estimation method County DE SNE GE NP V/ha RE NP V/ha RE RD NP V/ha RE Gapyung % % % Goyang % Kwacheon % Kwangmyung % Kwangju % % Guri % Gunpo % Kimpo % % Namyangju % % Dongducheon % Bucheon % Sungnam % % Suwon % Siheung % Ansan % Ansung % % Anyang % Yangju % % Yangpyung % % % Yeoju % % Yeoncheon % % Osan % Yongin % % Uiwang % Uijungbu % % Icheon % % Paju % Pyungtaek % % Pocheon % % Hanam % % Hwasung %

351 June 7-8, 2011 The GE is a more useful method for small area estimation. However, to provide more reasonable forest statistic, it should be needed to integrate various forest inventory data such as urban forest and forest management inventory data

352 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Analysis on Land-use and Land-cover Change using Medoum Resolution Satellite Image 중해상도위성영상을이용한토지이용및토지피복변화분석 Jung-Soo Lee, Dong-Hwan Park, Phua, Mui-how, Byung-Oh You, Hyun-Kook Cho* Korea Forest Research Institute, 57 hoegi-ro, Dongdaemun-gu, Seoul, , Korea. Abstract LULCC (Land Use and Land Cover Change) is one of the most important driving forces of disturbances on the Earth s surface. This study discusses the detection of land cover changes in Kangwon Province, Korea, using multitemporal Landsat TM and ancillary data. We compared the performance of Normalized Burn Ratio (NBR) and Normalized Difference Water Index (NDWI) from multisensor data in the image differencing method for detecting the disturbances between 1999 and We performed geometric and radiometric corrections for the multitemporal Landsat TM data satellite data. The ancillary data, built into GIS database, consists of spatial forest management data, administrative boundary, forest type and road networks. Furthermore, the ARD and burned area masks were acquired from interpretation of aerial-photos and ground-truthing. The accuracy of the detect rate in land cover change area was better in the order of NBR and NDWI. As the best detection method, the NBR differencing was successful in capturing the disturbances between with an accuracy of 89%. The spatial distribution of disturbances was also analyzed. 요약 LULCC(Land Use and Land Cover Change) 는지구표면교란의중요한영향력중하나이다. 본연구는 Landsat TM 영상과보조자료를이용하여강원도지역의토지피복변화지분석을실시하였다. 정규탄화지수 (NBR) 와정규수분지수 (NDWI) 를이용하여 1999 년과 2000 년사이의변화지역을검출하여비교하였으

353 June 7-8, 2011 며, Landsat TM 영상의기하보정, 방사보정을실시하였다. 보조자료는 GIS database 로구축된산림경영자료, 행정경계, 임상도, 임도망도를사용하였으며또한, 현장조사와항공사진으로구축된 ARD 와산불자료를이용하였다. NBR 과 NDWI 지수를사용하여변환된영상의토지피복변화지분석결과 NBR 의일치율이약 89% 로가장높게나타났다. Introduction This study examined the possibility of locating the areas where land-use and land-cover are changed using the vegetation indices (NBR, NDWI) drawn from the Landstat TM images of two different timings. Methodology ArcGIS 9.3 and ENVI were used for the data analysis of Landstat TM satellite image data. For the statistical analysis of areas with changed land-cover, Excel 2007 and SPSS were used. A preparation work and standardization work on the Landstat TM satellite image data of two different timings were done. A coordinate correction was done so that the object-site administrative boundary and forest fire damage area boundary would be overlapped. Areas without change in trees, forest and buildings were identified from the two images. Total 15 locations without change were arbitrarily selected (pixel value 5 for each) and image correction was done on those (Figure 2). Regarding the extraction indices selection for the areas with land-cover change, the indices used in the previous research literature on the landcover change area extraction using satellite image were used first. Then the indices used in the land-use change were chosen (Table 1). The images of two timings were reconciled and the values of changed locations were extracted as land-cover change areas. Then the extracted areas in the land-cover change areas were compared with GIS information and analyzed using average and standard deviation. As last, a Kappa analysis was done to assess the accuracy. Results and Discussion A consistency analysis was done on the satellite image and land-cover changed areas using the NBR and NDWI indices converted from Landstat TM satellite image data of two different timings. Images were converted to NBR and NDWI and grouped for year 2000 and

354 International Forest Fire Symposium on Commemorating the International Year of Forests The changed areas were extracted. Areas that have values lower than the value that is subtracted of standard deviation from average at the interval of 0.5 times were chosen as the changed areas. The consistency is the percentage of extracted pixels on the forest fire damaged area. In the section between μ-1*δ and μ-1.5*δ, NBR was 89%, the highest, and NDWI was 76%. As the section was enlarged, the extracted change areas decreased and the consistency also had decreasing trend. Among the indices, the consistency decrease of NDWI was bigger than that of NBR. The Khat analysis on the Landsat image was done. In the section of μ- 2.5*δ or lower, the consistency was the highest with NBR 65% and NDWI 47%. A filtering was done at the location where the standard deviation with highest consistency of NBR was 2.5 times. Khat indices all decreased. On the other hand, a filtering at the location where the standard deviation is 1 gave the result that both NBR and NDWI increased close to 50%. Therefore, it is believed that NBR index is effective in the extraction of areas that will change from current forest area to non-forest area in the future. Table1. Detection index Items Band used Formula Remark NBR Band 4, 7 NDWI Band 4, 5 Band4 Band7 Band4 Band7 Band4 Band5 Band4 Band5 Grading of forest fire damage, preparation of forest fire damage map Change in plants growth, change in water stress, locating the wet areas Figure 1. Study method

355 June 7-8, 2011 Forest Cover Change Detection for Mt. Bogdkhan and Tuul River Basin in Mongolia 몽골복단산과툴강지역의산림면적변화 Jung Hwa Chun* Korea Forest Research Institute, Seoul, , Korea Abstract According to the Mongolian Ministry of Nature and Environment, the total forest area in Mongolia has decreased by 1.2 million ha during the last twenty years (about 60,000 ha per year). UNEP report on land cover assessment for Mongolia states the major cause of forest loss and degradation in Mongolia is forest fire. Even though low precipitation in Mongolia creates high fire hazard, about 90 percent of forest fires are caused by human activities. The results of satellite image classification and field surveys showed forest loss especially on the forest edge with easy human access for both of two study areas (Mt. Bogd khan protection forest and Tuul river basin) due to forest fire followed by illegal cutting. Forest area in Tuul river has decreased by 19.1 % between 1990 and On the contrary, forest area in Mt. Bogd Khan rather increased by 7.5 % due to naturally regenerated young forest. This shows the important role of appropriate protection area as well as the rehabilitation of devastated land for the conservation of forest ecosystem in a country where restoration in difficult due to harsh climatic conditions. 요약 몽골환경부통계에따르면몽골의전체산림면적은지난 20 여년간 120 만 ha ( 연간 6 만 ha) 가감소한것으로나타났다. 몽골의토지피복평가및모니터링에관한 UNEP 의보고서 (UNEP, 1998) 에서는몽골지역임상의변화를야기하는주된원인을산불로보고있다. 몽골의극히적은강수량이산불의위험도를증가시키는요인이기는하지만산불원인의 90% 는지역주민의경제활동과관련되어있다. 위성영상과현지조사를병행하여보그드한산림보호지역과툴강유역의산림변화를분석한결과두지역모두사람이접근하기쉬운

356 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 임연부에서산림의훼손이나타났다. 툴강유역의산림의 1990 년에서 2001 년사이에약 19.1% 가감소한것으로나타났다. 반면보그드한지역의산림은천연갱신된유령림으로인해 7.5% 가증가하였다. 이러한결과는극심한기후조건으로인해복원이어려운국가의경우황폐지조림못지않게적절한보호지역의역할이중요함을단적으로보여준다. 서론 (Introduction) 몽골환경부통계에따르면몽골의전체산림면적은지난 20 여년간 120 만 ha ( 연간 6 만 ha) 가감소한것으로나타났다. 몽골의토지피복평가및모니터링에관한 UNEP 의보고서 (UNEP, 1998) 에서는몽골지역임상의변화를야기하는주된원인을산불로보고있다. 몽골의극히적은강수량이산불의위험도를증가시키는요인이기는하지만산불원인의 90% 는지역주민의경제활동과관련되어있다. 그러나, 산불외에도지역에따라서는벌채및과도한방목등의원인으로장기간에걸쳐상당한면적의반건조지역의관목림을포함한산림지역이초지및황폐지로전환되었으며, 현시점에도이러한변화는계속되고있다. 또한몽골에서의산림파괴는극히적은강수량, 긴겨울등의극심한기후조건과맞물려사막화를가속화하고있으며지역주민의생활에위협요인이되고있다. 본연구에서는비록과거영상이기는하나몽골의산림보호지역과비보호지역의산림피복변화를살펴보고자하였다.. 재료및방법 (Methodology) 2003년발생한 Landsat 7의 scan line corrector 고장이후로활용도가낮아지기는하였으나대면적을대상으로상대적으로높은해상도를지닌 Landsat 영상은경년변화탐지에매우적합하다. 본연구에선자원탐사위성인 Landsat 5(TM) 의 1990년 9월 10일에촬영된 , (Path-Row) 지역, Landsat 7(ETM+) 의 2001년 8월 31일에촬영된동일지역의서로다른두시기의영상을이용하였으며중점연구대상지의지도및임상도등의참조자료를이용하였다. 중점연구대상지의식생변화탐지를위하여본연구에서는식생의변화탐지에보다유용한것으로알려진 Tasseled Cap 변환기법을이용하였으며정규식생지수 (NDVI) 의비교를통하여산림의변화양상을파악하고자하였다. 또현지조사를통해 ground truth를확보함으로써분석의정확도를높이고자하였다

357 June 7-8, 2011 결과및고찰 (Results and Discussion) 년각종참조자료와의비교를통해다사무감독분류및감독 분류를실시하고산림지역만을추출하여 1990 년의산림지역과 2001 년의산림 지역의차이를분석한결과는표 1. 에나타나있다. 표 년연구대상지역의산림면적변화단위 : (ha) 지역 연도 증감 Mt. Bogd Khan 19,706 21, ,419(7.5%) Tuul river basin 15,494 12,530-2,964(19.1%) 분석결과 Mt. Bogd Khan 지역은 1990 년에비해오히려약 7.5% 정도산림면적이증가한것으로나타났는데이러한결과는 Mt. Bogd Khan 지역이절대보전지역으로서타지역에비해잘보전되었음을보여주고있다. 그러나산림지역의임연부에서는비록소규모이기는하나산불과벌채등의훼손이일어난것으로나타났다. 현장조사결과를토대로하여분석한결과훼손지의규모는그리크지않은것으로나타났으며훼손의주원인은산불이발생한후의벌채또는임연부에서의소규모벌채인것으로판단되었다. 실제적으로산림면적이증가한것으로분석된원인은과거에임내에존재하던유령림의수고생장및직경의진계생장, 수관의밀도증가및상층을우점하는중 장령림의수관울폐도증가에서기인한것으로판단된다. Tuul 강상류지역의경우상당부분의산림이초지및나지로변화하여지속적으로훼손되어온것으로판단되었다. 특히임연부의산림훼손이심각한것으로나타났으며현재생육하고있는임분들역시활력도가낮아진것으로나타나산림지역의보호및조림을통한복원이시급한것으로판단되었다. 본연구결과는몽골과같이극심한기후조건으로인해산림의복원이어려운국가에서는적절한보전지역의설정및관리가조림등을통한복원못지않게중요한일임을단적으로시사한다. 인용문헌 (References) UNEP LAND COVER ASSESSMENT AND MONITORING, Mongolia. UNEP/ Environment Assessment Programme for Asia and the Pacific

358 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Idea of Rifling Application of Water Mist Nozzle for Forest Fire Extinguishing 산불진화용미분무수노즐의강선적용에관한아이디어 Choi Seok Hwan*, Kim Yong Sik, Ahn Byoung Jun, Lee Oh Kyu, Lee Soo Min, Choi Don ha Korea Forest Research Institute, Seoul, , Korea Abstract This Study is about application idea of rifling groove onto the inner surface of water mist nozzle for extinguishing forest fire. This apparatus has been already adopted to the linear reactor of Supercritical Water Saccharification Pilot Plant, KFRI. These experiments have suggested somewhat meaningful advantages as to spreadability, driving distance and so on of firefighting water. 요약 본연구는국립산림과학원바이오에너지연구과가연구하고있는초임계수산화분해장치의반응기에채용되었던내부강선 ( 腔線 ; Rifling) 에관련된연구를응용한것으로서미분무수소화장비의핵심부품인분무노즐의설계에아이디어를제공하고자하는것이다. 실험결과적어도미분무수노즐내부의설계형태를변화시키는것만으로도방화수의퍼짐성과도달거리에있어서어느정도유의한효과를거둘수있음을보이고있다

359 June 7-8, 2011 서론 (Introduction) 미분무수 ( 微噴霧水 ; Water Mist fire Extinguishing) 소화설비는미세한수적 ( 水滴 ; 물방울 ) 으로이루어진분무를소화 ( 消火 ) 에활용하는것으로서과거에는국제해사기구의노력에힘입어상업용선박등에주로사용되던것이최근에는전산실ㆍ엔진룸ㆍ주방과같은육상시설에서도많이도입되고있다. 한편미분무수소화설비혹은장비가산불진화용으로사용되는예는많이알려지지않고있는데이는미분무수진화가제한된공간에서의진압방법으로발전된데다가여타화재와는다른 산불의고유한특성과진화방법 에따른것이라고생각할수있다. 여기에서는미분무수설비의소화성능변수에대한깊은고찰보다는이에응용할수있는기계적인아이디어를살펴보고자한다. 연구방법 (Methodology) 미분무수소화설비라함은고압으로가압된물이헤드 (Head) 를통하여분사되어소화하는소화설비를말하는데분무소화설비는미세한수적 ( 水滴 ) 이화염및주변기상의온도에의해급격하게증발하고팽창하면서열을흡수하고화원 ( 火源 ) 주변을적셔냉각을하는방식이다. 미분무수방식은최소설계압력에서헤드로부터방출되는방화수 ( 防火水 ) 입자중 99% 의크기가 400micrometer 이하로분무되는소화방식을말하고미분무수노즐 (Water Mist Nozzle) 은미분무수를만들수있도록설계된하나이상의오리피스 (Orifice) 를갖고있는특수목적기구를이르는데특히미분무소화설비는수적을급속히체적팽창시켜화원주변의산소를밀어내고연료 - 산소혼합기의농도를떨어뜨려열복사피드백을차단하면서소화를진행하는냉각 - 질식효과를가지고있다고알려져있다. 여기에서이러한소화원리에영향을미치는가장중요한변수로열전달표면적의증가와이에따른열전달및증발현상에영향을주는수적의크기분포와분무밀도, 분무운동량등을들수있다. 이러한미분무소화설비는미분무노즐, 압력용기, 섹션밸브, 펌프유니트등으로구성되는데특히핵심부품인미분무노즐은 IMO Res. MSC. 265(84) 에따라노즐성능시험및소화성능시험을통해형식승인을얻어야하고특히 Back Draft 또는 Flashover 등의현상을막을수있도록채용에있어서적절한노즐형태와가압방식의선택이필요하다

360 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 한편국립산림과학원바이오에너지연구과는초임계상태에도달한고온고압의물 (Waer) 을이용하여목분슬러리 (Slurry) 를분해하는반응기를보유하고있는데이러한반응기는선형, 반구형, 원형튜브형등다양한형태가있는바, 특히선형의반응기에있어서는고온고압의물리상태에서입자 (Granule) 형태를지닌유체가반응기내부를난류확산 (Turbulent Diffusion) 등의거부거동이없이매우신속하게이동하여야하므로그내부형태가중요하다. 이러한거동을해결하기위하여국립산림과학원이출원하여등록한대한민국특허제 에서는선형의반응기내부에특수한강선 ( 腔線 ; Rifling) 을넣어서목분슬러리를비롯한매우작은수적들과내부유체를효과적으로배출하는방법을제시하고있는데, 여기에서초임계상태에도달한물이강선반응기를통과하여고압상태에서해압되어대기중으로배출되는물리거동은산불진화용호스노즐이분무상태의소화제를토출하는것과비슷한양상을보이는데특히전술한소화원리에영향을미치는중요변수등이유사하게거론될수있다. 따라서이연구에서는이들방식의차이점과개별특성을모두고려하지는않고, 다만대기중에미분무수가노출되기직전까지의도관 (Pipe) 의강선 (Rifling) 과그회전에따른분사압력, 방출거리, 분무량의영향을살펴보았다. 아래 Fig.1은일반미분무노즐과시험용으로제작한강선형미분무노즐로서, 특히 Fig.2에는내부에강선을채용한미분무노즐의내부구조모식을표현하였다. 또한 Fig.3은각노즐의실제단면을보이고있다. Fig.1. 강선형미분무 노즐 ( 상 ) 과일반미분무노즐 ( 하 )

361 June 7-8, 2011 Fig.2. 강선형미분무노즐의내부형태모식 Fig.3. 일반형미분무노즐내부단면 ( 좌 ) 와강선형미분무노즐내부단면 ( 우 ) 강선은노즐의내부에새겨진나선형의홈 (Groove) 으로서고압으로토출되는방화수가화원 ( 火源 ) 을겨냥하여대기중으로통과하도록유선 ( 流線 ) 을만드는역할을하는데, 이때강선에서튀어나온부분인강선등 (Land) 과들어간부분인강선홈 (Groove) 의형태는직선형, 회전형등몇가지의변형이있을수있고강선등과강선홈의절삭높이의차이는가공전의노즐내부의직경에따라변이가있을수있다. 강선의형태는 4 조우선 (4/Right, 4R), 4 조좌선 (4/Left, 4L), 6 조우선 (6/Right, 6R), 6 조좌선 (6/Left, 6L), 8 조우선 (8/Right, 8R), 8 조좌선 (8/Left, 8L) 등다양하게시도될수있는바, 강선의회전정도 (The Rate of Twist) 를이르는강선회전량 (Rifle Twist) 은 Emanuel G. Greenhill 의 강선회전량에대한경험식 ( 일명 Greenhill Formula) 을적용하였다. 반응기의내부의연마, 곧리밍 (Reamming) 은토출되는미분무수의마찰압력에의하여도강선등과강선홈이손상되지않는보통수준인일반총강 ( 銃腔 ) 의연마수준으로하였다. 분사압력은순수한물을방화수로사용하여

362 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 토출압 300bar 이상의환경에서압력을점진적으로높여가며비거리 ( 飛距離 ; Driving Distance) 를측정하는방식으로실험하였다. Fig.4는토출압을조절하기위해사용한고압펌프 FELUWA hose diaphragm piston pumpp (Type: ZGL 30/130 - K 36 SM 235/190 HD) 를보인것이고 Fig.5는강선및리밍작업을보인것이다. Fig.4. 방화수토출펌프 Fig.5. 강선및리밍작업 결과및 고찰 (Results and Discussion) 산불현장의환경과산불진화의방법은그양상이매우복잡다단하지만만약근거리의지표화나지중화의경우진압을하기위한방화수의살포방식에서는방화수의토출압력과퍼짐성 (Spreadability) 이중요하다고할것이다. 이런측면을고려할때기본적으로일반적인미분무노즐과 4조우선 (4/Right, 4R) 강선을채용한미분무노즐의방화수토출압과비거리, 그리고발사형태 (Parabolic Shape) 을살펴볼수있을것이다. 아래 Fig.6은각노즐을사용했을경우의발사형태를나타낸것이고 Fig.7은각노즐의토출압과미분무수적의최대도달비거리를나타낸것이다

363 June 7-8, 2011 Fig.6. 강선형노즐사용 ( 좌 ) 와일반형노즐사용 ( 우 )

364 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Fig.7. 토출압변화에따른각미분무수노즐수적의최대도달거리 이는실제산불진화현장에서사용하는기계화산불진화시스템에적용한결과는아니지만적어도미분무소화설비에있어서미분무노즐내부의설계형태를변화시키는것만으로도어느정도유의한효과를거둘수있을것이라는단서를제공하고있다. 추후심도있는연구가진행된다면각산불의형태에따른효과적이고적절한, 보다발전된형태의미분무노즐의설계가가능할것으로기대한다. 참고문헌 (References) Myung Bae Kim: A State of the Art for Water Mist Fire Protection Systems and Introduction of NFPA 750. Journal of Korean Institute of Fire Sci. & Eng. 2000, Vol.1, No2.: 2-5 Tae Hwan Kim, Sang Yeong Myeong: Development for the Water Mist Package Having a Low Pressure Nozzles on Wooden Cultural Properties and Temples. Journal of Korean Institute of Fire Sci. & Eng. 2010, Vol.24, No4.: Ha-Sung Kong: A Study of the Suppression System based on the Fire Protection System the Korea Cultural Property due to the Forest Fire - About Water Mist System. Journal of Korean Institute of Fire Sci. & Eng. 2007, Vol.21, No4.: K. Y. Bae, H. T. Chung, H. B. Kim, I. S. Jung, C. Kim : Thermal Numerical

365 June 7-8, 2011 Simulation on Fire Suppression Characteristics through Mobile Mist Spray Nozzles. Journal of Korean Society for Power System Engineering. 2009, Vol.13, No5.: Kang-Youl Bae, Hee-Taeg Chung, Hyoung-Bum Kim: Numerical Analysis on Interaction between Fire Flame and Water Mist according to the Variation of Nozzle Performance. Journal of Mechanical Science and Technology. 2007, Vol.21: Jihyun Kwark, Young-Han Kim: Fire Suppression evaluation of the Water mist nozzle complying with IMO Res. MSC 265(84). Journal of The Korean Society of Marine Engineering. 2010:

366 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Effects of Forest Fire on the Water Quality in Forest Catchments Abstract 산림유역에서산불이계류수질에미치는영향 Hyung Tae Choi *, Choong Hwa Lee Korea Forest Research Institute, Seoul, , Korea This study was conducted to investigate the effects of forest fire on the water quality. The burned catchment had been damaged by forest fire broken out in April, An unburned catchment was selected to compare with the burned catchment. Water quality monitoring had conducted from 2000 to 2005, and monitoring parameters were BOD, COD, TN, TP, Cl -, NO 3 -, Na +, Mg 2+ and SS. The periodic mean concentrations of each water quality parameters measured in the burned catchment were higher than those in the unburned catchment during the whole experimental period. Especially, there were big differences between the periodic mean concentrations of TN, TP, NO 3 - and SS under the burned and unburned catchment. Also, there were wide differences between mean concentrations of each parameter under the burned and unburned catchment in However, those differences were smaller in a lapse of time. In particular, there weren't distinctive differences between mean concentrations of each parameter under the burned catchment and those under the unburned catchment.. 요약 본연구는 2000 년 4 월산불이발생한강원도삼척시근덕면일대의산불유역을대상으로산불에따른물환경의변화를밝힘으로써산불발생지역의물환경보전대책수립을위한자료를제공하고자수행하였다. 연구를수행하기위하여산불발생유역과산불비발생유역의각유역출구에서 2000 년부터 2005 년까지계류수질변화를모니터링하였다. 모든수질인자들의평균농도는산불비발생유역에비해산불발생유역이높아산불로인해수질이악화된것으로판단되며, 이중전질소 (TN), 총인 (TP),

367 June 7-8, 2011 NO 3 -, 부유물질량 (SS) 의농도가현저하게증가하였다. 또한, 계류수질의경시적변화를분석한결과, 생물화학적산소요구량 (BOD) 과화학적산소요구량 (COD) 은산불발생후 4 개월이경과한 8 월부터두지역간에차이가거의없는것으로나타나산불에의해교란되었던계류수질이산불발생후 4 개월이경과한시점에서뚜렷하게안정화되는것으로분석되었다. 서론 (Introduction) 2000 년에동해안에발생된산불로강원도고성군, 강릉시, 동해시, 삼척시및경상북도울진군에걸쳐 23,448ha 의산림이피해를입었다. 이는서울여의도면적의 78 배에이르는막대한면적으로우리나라총산림면적의 0.36% 에이른다. 이들 5 개지역가운데삼척지역이 16,751ha 로가장피해가컸다 ( 임업연구원, 2000; 이경재, 2000; 정연숙등, 2000a; 2000b). 본연구는강원도삼척시에위치한산불발생유역과산불비발생유역의계류수질변화를조사, 분석함으로써산불에따른물환경변화특성을구명하고, 산불발생유역의물환경보전대책수립을위한자료를확보하고자수행하였다. 재료및방법 (Methodology) 산불에의한물환경변화를구명하기위하여강원도삼척시근덕면궁촌리에소재한산불발생유역과개산리에소재한산불비발생유역을조사유역으로선정하였다 (Figure 1). 산불발생유역은유역면적이 61ha로 2000년 4월 7일부터 15일까지 9일간삼척시근덕면궁촌리일대에발생한산불로인하여 13,033ha의산림피해를받은산불지역의일부이고, 산불비발생유역은삼척시노곡면개산리의일부유역으로유역면적은 392ha이다 (Figure 2). 두지역의모암은화강편마암이었으며, 산불비발생유역의임상은침엽수 Ⅲ~ Ⅳ영급이었다. 경사는산불비발생유역이 5~30 이었으며, 산불발생유역은 10~25 이었다

368 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Figure 1. The location of experimental catchments. Figure2. Topographic maps of experimental catchments. 산불에따른계류수질변화를구명하기위하여산불발생유역및산불비발생유역의각유역출구에서 2000 년부터 2005 년까지계류수질모니터링을실시하였다. 생물화학적산소요구량 (BOD), 화학적산소요구량 (COD), 전질소 (TN), 총인 (TP), 양이온 (Na +, Mg 2+ ) 및음이온 (Cl -, NO 3 - ), 부유물질량 (SS) 등의분석을위해시료를 Ice Box 에보관한후실험실에가져와 0.45μm 의필터에각각 2 회통과시킨후양이온중 Na + 는 Frame Photometer 로, Mg 2+ 는 EDTA 적정법으로분석하였다. Cl -, NO 3 - 은 Ion Chromatography 로분석하였다. 부유물질은강우시와강우직후에채취한시료를실험실에가져와 200 ml의시료에대해지름 47mm 의여과지 (GF/C) 로여과한다음오븐에넣어 105 에서 2 시간건조시킨후그질량차이를측정하여농도를계산하였다. 결과및고찰 (Results and Discussion) 2000 년 6 월부터 9 월까지 4 개월간산불발생유역과산불비발생유역 ( 대조유역 I) 에서 36 차례에걸쳐측정한생물화학적산소요구량 (BOD), 화학적산소요구량 (COD), 전질소 (TN), 총인 (TP), 양이온 (Na +, Mg 2+ ) 및음이온 (Cl -, NO 3 - ), 부유물질 (SS) 을분석한결과는 Table 1 과같다

369 June 7-8, 2011 Table 1. Comparisons between water quality of the burned and unburned catchment in Item Burned catchment Average (Min.~Max) ( mg /l) Unburned catchment Average (Min.~Max) ( mg /l) BOD 0.85 (0.00~3.04) 0.69 (0.00~2.62) COD 6.43 (2.60~18.62) 3.07 (0.60~6.91) TN 3.10 (0.60~11.28) 0.91 (0.05~3.82) TP 0.08 (0.01~11.28) 0.02 (0.00~0.06) Cl (2.34~8.97) 5.09 (1.98~8.19) NO (0.13~1.94) 0.36 (0.05~0.63) Na (5.29~11.84) 5.83 (4.07~8.20) Mg (0.70~8.78) 1.90 (0.89~5.09) SS (1.90~288.50) 1.59 (0.00~8.93) Table 1 과같이산불이발생한지역에서계류수의 BOD, COD, TN, TP, 양이온및음이온, 부유물질농도는강우와함께임지로부터계류로유입되는유출물에의한영향을많이받는것으로나타났다. 측정기간동안생물화학적산소요구량 (BOD) 은산불발생유역에서평균 0.85(0~3.04) mg /l, 산불비발생유역에서평균 0.69(0~2.62) mg /l 로산불발생유역이산불비발생유역에비해다소높은농도를보였으나, 두유역모두하천수질환경기준상수원수 1 급수의범위내에포함되어오염되지않은청정한산림내계류수를나타내었다. 그러나측정기간동안의화학적산소요구량 (COD) 은산불발생유역에서평균 6.43(2.60~18.62) mg /l, 산불비발생유역에서평균 3.07(0.60~6.91) mg /l 로산불발생유역이산불비발생유역보다평균약 2.1 배높아수질이악화된것으로분석되었는데, 이는산불로인하여재등이강우와함께계류로유입되어기인한결과로판단된다 ). 또한, 총질소는산불발생유역에서평균 3.10(0.60~11.28) mg /l 로산불비발생유역에서의평균 0.91(0.05~3.82) mg /l 의약 3.4 배에달하였으며, 총인은산불발생유역에서평균 0.08(0.01~0.57) mg /l 로, 산불비발생유역평균 0.02(0~0.06) mg /l 의 4 배로써산지사면에서유출되어계류로유입되는총질소, 총인의양이증가하여계류수질이악화된것으로나타났다. 아울러계류수질오염의지표라할수있는음이온가운데특히개벌등에의해식생이제거되었을때계류로유출되는양이급격히증가하는환경악화의중요한지표인 NO 3 - 의경우측정기간동안산불발생유역에서는평균 1.17(0.13~1.94) mg /l 로산불비발생유역의 0.37 (0.05~0.63) mg /l 에비해 3.2 배높아산불로인해계류수질이악화되는것으로분석되었다

370 International Forest Fire Symposium on Commemorating the International Year of Forests 년 6 월과 10 월, 2004 년 7 월과 8 월, 2005 년 6 월에각각 1 회실시한산불발생유역과산불비발생유역 ( 대조유역 II) 의계류수질조사결과는 Table 2 와같다. 조사유역모두 먹는물수질기준 및 하천수질환경기준 Ⅰ 급수 범위이내를나타냈다 (ph, DO, Cl -, SO 4 2- ). 그러나 Cl -, Na + 이온농도는산불발생유역이산불비발생유역보다다소높게나타났는데, 이는각연도별유사한경향으로산불발생유역이해안에인접해있어해풍과함께해수속의염분이유입되었기때문인것으로판단된다. 또한, 전기전도도 (EC) 역시산불발생유역이산불비발생유역보다높게나타났다. Table 2. Comparisons between water quality of the burned and unburned catchment during 2003 to EC DO Cl NO 3 SO 4 Na + NH + 4 K + Mg 2+ Ca 2+ Year ph (μs/ cm ) ( mg /l) ( mg /l) Burned Catchment Unburned Catchment Drinking water standard River water quality standard ( Ⅰ ) Streamwater quality standard ~ ~ ~ <250 <10 < < < < <5 <10 <10 <5 <0.5 <2 <2 <10 인용문헌 (References) 이경재 산불로인해파괴된동해안지역생태계복원. 자연보존 110: 임업연구원 동해안산불산림피해조사결과및산림복구방향. 산림청임업연구원. 17 쪽. 정연숙, 노찬호, 오현경, 이규송. 2000a. 산불전식생구조가산불후식생복원에미치는영향. 한국생태학회초록집. 한국생태학회. 58 쪽. 정연숙, 노찬호, 오현경, 이규송. 2000b. 동해안산불피해생태계의효과적인자연복원기법. 자연보존 110:

371 June 7-8, 2011 The Effect of Microbial Compost on Soil and Vegetation in Forest-fire Areas 미생물퇴비가산불이후토양과식생에미치는효과 Jeong Soo Park 1*, Im-kyun Lee 2, Eun Ju Lee 1 1 School of Biological Science, Seoul National University, Seoul , Korea 2 Department of Forest Conservation, Korea Forest Research Institute, Seoul, , Korea Abstract Jspark097@gmail.com The forest located in Yongdong area has been experiencing fire repeatedly during spring dry season in Korea peninsula. A forest fire incinerates plant materials, humus, and detritus, which often cause loss of surface soil materials during rainy season. These events may be detrimental to establish a stable community structure. We studied the effect of microbial compost on vegetation reintroduction and soil properties in compost treated and non-treated plot (12m X 12m) which was set up three different sites. When we compared the soil properties, organic matter, water content, and EC were higher in compost treated plot than non-treated plot. And we found that vegetation cover and number of species were also increased in the treated plot. These results suggest that vegetations in the treated plot supplied soil organic matter whereby soil loss was decreased rather than non-treated plot. The results of soil organism activity using fluorescein diacetate (FDA) hydrolysis show that treated plot has higher microbial activity than non-treated plot, and the results of microbial biomass C also revealed that treated plot has higher biomass than non-treated plot except for Gangneung site. It was found from the results that microbial compost may be positive effect on soil stabilization and vegetation reintroduction. 요약 영동지역산림은봄철건조기마다산불을자주겪고있다. 산불은식물체와다양한토양유기물을태우고이로인해우기에토양유실이일어나게된다 (Morris SE, 1987). 이러한과정은군집구조안정화에부정적인영향을미치게된다. 산불이일어났던세개지역에미생물퇴비처리

372 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 방형구와비처리방형구 (12 m X 12 m) 를설치하여미생물퇴비에의한식생도입과토양환경에미치는영향을연구했다. 토양특성을비교한결과유기물함량과수분함량, EC 수치가비처리구에비해증가한것을발견할수있었다. 식생의경우도처리구가식생피도와종수에서비처리구에비해높게나타났다. 이는처리구에서식생피복도가높아지면서토양유실이감소되어토양특성의차이가발생한것으로보인다. 토양미생물활성을측정한결과도처리구가활성이높았으면미생물생물량의경우는강릉지역을제외하고처리구가비처리구에비해높았다. 위의결과를통해미생물퇴비가토양의안정화와식생의도입에긍정적인영향을미쳤음을예상할수있었다. 서론 (Introduction) 동해안산림은봄철건조기마다산불을자주겪고있다. 또한마사토를기반으로한표토층은산불후우기가되면토양표면을덮어주는식물층이빈약하기때문에토양유출현상이심하다. 현재까지는주로조림이나자연천이에의존해서자연적인회복이되기를기다렸으나조금더적극적인복원방법으로유용한미생물을산불직후투입해줌으로써미생물의균사에의한표토층토양유실을최소화하고미생물에의한유기물흡착및분해를통한토양개량으로식물종자의발아및유식물의생장을촉진함으로써산불피해지의조기회복을기대할수있다 (Hart SC, 2005). 재료및방법 (Methodology) 적용한미생물제제는갯벌에서식하는미생물인광합성세균 (Rhodopseudomonas palustris), 유산균 (Lactobacillus plantarum), 효모 (Saccharomyces cerevisiae) 로이루어졌다. 고상제제는위의미생물로발효퇴비로만든것이다. 사용한미생물은그자체가많이증식하는것이아니라토양에들어가면유용영양염류활용도를높이고자체토양내유용미생물의활동을촉진할목적으로선택하였다. - 강릉대조 (KC), 강릉처리 (KT) : 강원도강릉시사천면청솔공원 2005 년 5 월 12 일남쪽사면 12m X 12m (144m2) 지역에액상미생물제제를 40L, 고상미생물제제를 40kg 을일차적용하였다 년 8 월 20 일현장확인후복원이느린곳에고상미생물제제를 20kg 추가처리하였다. 처리효과를확인하기위해처리구에서왼쪽으로 10m 떨어진곳에동일한면적을대조구로선택하였다

373 June 7-8, 삼척대조 (SC), 삼척처리 (ST) : 강원도삼척시길곡리 2005 년 5 월 12 일남쪽사면 15m X 15m (225m2) 지역에액상미생물제제를 40L, 고상미생물제제를 40kg 을일차적용하였다 년 8 월 20 일현장확인후복원이느린곳에고상미생물제제를 20kg 추가처리하였다. 처리효과를확인하기위해처리구에서왼쪽으로 10m 떨어진곳에동일한면적을대조구로선택하였다. - 울진대조 (UC), 울진처리 (UT), 비산불 (NF) : 경북울진군현종산 2007 년 5 월 11 일현종산남서쪽사면에 12m X 12m (144m2) 지역에액상미생물제제및고상미생물제제를처리구 1 (UT1) 에는 40kg, 40L 를각각처리하고, 처리구 2 (UT2) 에는 20kg, 20L 각각처리했다. 8 월 20 일현장확인후복원이느린곳에고상미생물제제를 20 kg 추가처리하였다. 처리효과를확인하기위해처리구에서왼쪽으로 10m 떨어진곳에동일한면적을대조구로선택하였다. 또한비산불지역대조를위하여같은현종산내산불피해를입지않은소나무림을정해서현장외대조구로활용하였다. 소나무숲은직경 30 센티미터내외의성숙한숲을선택하였다. 결과및고찰 (Results and Discussion) 강릉, 삼척지역의경우처리구와대조구를비교했을때대부분의토양특성이처리구가대조구에비해양호하게나타났다. 특히삼척의처리구는대조구에비해토양수분함량, 유기물함량, EC, 양이온값들이높게나타났다. 울진지역은비산불지역이산불지역에비해토양특성이양호하게나타났으며대조구와처리구간의차이도 EC 와양이온의경우 2 배가량차이를보였다 (Table 1). 이는처리구에서식생의피복도가월등히높아토양유실이감소되면서토양특성의차이가발생한것으로보인다. Table 1. Soil properties in control(c) and treated (T) plot Sampling Soil EC AP Mg 2+ K + Ca 2+ WC (%) OM (%) Bulk ph site sample (us/cm) (mg/g) (mg/g) (mg/g) (mg/g) 강릉 C T 삼척 C T C 울진 T NF WC: water content, OM: organic matter, AP: available phosphate

374 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Fluorescein diacetate hydrolysiss assay를이용해미생물효소활성의측정한 2009년결과에서는강릉지역은대조구와처리구간의차이가미미하고삼척지역은처리구 (ST) 가대조구에비해미생물활성이두배가량높았다. 2010년의결과도삼척처리구가대조구에비해높은측정값을보였다. 울진지역의경우대체로 2009년보다처리구가대조구에비해높은미생물효소활성을보였다 (Fig. 1). 이를통해유효미생물의처리와식생의피도증가로미생물효소활성이높아졌음을확인할수있었다. Fumigation Extraction 방법을이용해토양미생물의생물량 (microbial biomass C) 을측정결과에서는건조토양 1g 당가장높은 microbial biomass C 를가지고있는토양은비산불지역 (NF) 으로 2009년과동일했고다음으로울진처리구 (UC) 었다. 가장낮은생물량은보인지역은강릉처리구 (KT) 로 NF2에비해 4배가량낮게측정되었다 (Fig. 2). 토양미생물효소활성은토양유기물함량, 수분함량과높은상관관계를보였으며, 토양미생물생물량도유기물함량과높은상관관계를보였다. 이상의결과를통해유효미생물처리지역의식생생육이증가하고피도값이높아짐으로해서토양환경이개선되어미생물의효소활성과생물량이증가된것으로추측할수있었다. Fig. 1. Total microbial activity at each site

375 June 7-8, 2011 Fig. 2. Microbial Biomass C at each site 인용문헌 (References) Morris SE, Moses TA ( 1987) Forest fire and the natural soil erosion regime in the Colorado Front Range. Annal Assoc Am Geogr 77: Choromanska U, DeLuca TH (2002) Microbial activity and nitrogen mineralization in forest mineral soils following heating: evaluation of postfire effects. Soil Biol Biochem 34: Hart SC, DeLuca TH, Newman GS, MacKenziee MD, Boyle SI (2005) Post-fire vegetative dynamics as drivers of microbial community structure and function in forest soils. For Ecol Manage 220:

376 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Allocating Initial Attack Resources in the Republic of Korea with a scenario based optimization model 최적화모델을이용한진화자원배치 Yohan Lee 1, Heidi J. Albers 1, Byungdoo Lee 2*, Myung Bo Lee 2 *, and Kyung Ha Kim 2 1 Oregon State University, Corvallis, OR 97333, U.S. 2 Korea Forest Research Institute, Seoul, , Korea Byungdoo@forest.go.kr Abstract This study explores the optimal spatial allocation of initial attack resources in the Republic of Korea. We use spatially explicit GIS-based information on the ecology, fire behavior, and economic characterizations important in Korea. The data include historical fire events in the Republic of Korea from 1991 to 2009, suppression costs, and spatial information of forest fire extent. Interviews with forest managers inform the range of objective functions and policy goals we address in the decision model. Based on this information, we create a modified optimization program to explore the Korean initial attack resource allocation decisions with a range of policy goals. The disparity of optimal fire policy exists between regions depending on their own characteristics (i.e., fire behavior, terrain, budget constraints, and policy goals). We conduct sensitivity analysis by varying parameters systematically. The information about the relative importance of components of the setting helps to identify rules of thumb about initial attack resource allocations in particular ecological or policy settings. Introduction Over recent decades, large wildfires have posed significant budgetary challenges to fire management agencies charged with protecting human life, property, and natural resources from fire hazards. Since 1990, the areas burned by large wildfires have increased dramatically. For example, in the Republic of Korea, the recent fires such as the Donghae fire (2000) and the Samchuk fire (2000) in Gang-won province burned huge areas and caused substantial

377 June 7-8, 2011 suppression costs and property losses. Further, the synchrony of large wildfires across broad geographic regions often contributes to budget shortfalls when suppression costs exceed the congressional funds appropriated for suppression. Fire managers attempt to deploy suppression resources to stations and dispatch such resources to successfully contain fires before unacceptable costs and damages occur. Once a fire ignite, a strong and fast initial attack76 (IA) action is most effective at containing a fire quickly and preventing it from escaping and causing substantial losses. Thus, deploying IA resources to satisfy the expected demands for fire suppression is critical to achieving the fire manager s goal. However, IA resources are costly, so fire managers must allocate them efficiently, especially in the era of declining budgets. In our research, the main questions are as follows: 1) How do fire managers allocate IA resources to bases efficiently within a limited budget in the Republic of Korea? 2) How do major parameters in the model (e.g., budget, dispatch rule) affect the allocation decision of IA resources? 3) What is the optimal spatial allocation of helicopters (i.e., helitack), a major resource, for IA in the Republic of Korea? How does the optimal spatial allocation of IA resources change under alternative management objectives? The main purpose of this dissertation is to inform fire managers on how to effectively deal with changes in the suppression budget and how those changes will affect their measures of IA performance. Method 1. Study Area Our study area in the Republic of Korea comprises the whole landscape of the country. The forest area of the Republic of Korea is around 6.4 million ha (i.e., approximately 64 % of the total land area). Forest land is highly susceptible to forest fires because the area is currently characterized by thick growth owing to insufficient fuel treatment and its thick layer of fallen leaves catches fire and burns easily. Forest fires can spread rapidly in these mountainous areas (fires 76 Initial attack is the action taken by fire-fighting resources that are first to arrive at a fire location

378 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 spread much faster on steep slopes than they do on level ground). Spring is the most dangerous season (i.e., March, April, and May) because it is dry with high winds; 68% of forest fires occur in spring. As discussed, the causes of forest fires are mainly anthropogenic (KFS, 2007). Conifers cover 2.7 million ha, broad leaves 1.7 million ha, and mixed forest 1.9 million ha (KFS, 2005). Local administrative units (i.e., Si and Gun) are responsible for private forests and mountain villages in their areas, and these units regard mountain villages as an important policy arena. 2. Model To address research questions for the Korean case study, we construct a scenario optimization model with two stages that, first, decide the amount of resources assigned to each station at the beginning of a fire season, and then, decide the amount of resources dispatched from each station to each fire during a fire day. Our optimization model employs the standard response framework, which is defined as the required amount of resources that can reach the fire within a pre-defined response time. Because the fire suppression system can limit the number of escaped fires, which is likely to be large, by detecting fires soon after they ignite and quickly dispatching an IA force, the set-covering style model is commonly adopted in IA planning (Haight and Fried 2007; MacLellan and Martell 1996). Scenario optimization is widely used to model uncertainty in the parameters of location models by specifying a set of scenarios that represent the possible realizations of unknown parameters to determine a robust solution that can perform well across all scenarios (Haight and Fried 2007). In our optimization model, data include the locations of fire stations and possible locations of fires along with times required for travel between stations and fires. Ignition uncertainty is characterized by a set of fire scenarios, each listing the location and intensity of fires that could occur in a single day. To construct the Korean fire scenarios, we use the developed stochastic Korean fire simulation model (Lee et al. 2011). This simulation model includes fire occurrence modules for generating fire scenarios and contains random variables for whether and how many fires occur on a given day, along with the location and the ignition time for each fire. We use the Korean fire simulation model to generate fire scenarios across the Republic of Korea forest areas during a fire season. Each scenario represents a day in which a combination of fire count, location, ignition time and behavior (e.g., rate of spread) occur. Using these fire scenarios, we construct a scenario-based optimization program to explore Korean IA resource allocation decisions with a range of policy goals

379 June 7-8, 2011 The parameters of the fire scenarios in the optimization model are derived from information in the fire days obtained from the stochastic simulation. Each fire scenario represents a single fire day and includes a list of fire locations where fires occur along with the number of helicopters required in the standard response to each fire. Through the cluster analysis that uses GIS to detect spatial patterns of forest fire, Lee and Lee (2009) classify administrative district into 5 clusters by fire susceptibility. We assume that the number of helicopters required at a fire location is an increasing function of fire cluster level. The number of helicopters required in the standard responses, by cluster, is decided by the estimated mean number of helicopters deployed, using historical data for 19 years (MacLellan and Martell 1996). The objective of our optimization model is to minimize fire-related suppression costs and losses subject to the requirement that expected fires receive a standard response. Using this optimization model, we investigate the spatially optimal allocation of IA resources such as helicopters among 9 stations administered by the KFS across a heterogeneous landscape in the Republic of Korea with two targeting goals: population centers and important ecological locations (i.e., endangered species habitats). For the analysis, we will construct a spatially explicit optimization model that accounts for all potential fire areas by minimum administrative unit (i.e., 225 Si, Gu, and Gun in about a 10 mile 10 mile area) in the landscape using GIS. To our knowledge, no study has yet focused on the problems of the deployment and dispatch of IA resources in the Republic of Korea. Korean Initial Helitack Model Our initial helitack model is a mixed-integer programming model that specifies how many helicopters to deploy at each helicopter base in the Republic of Korea and how those helicopters satisfy daily helicopter demands across multiple fire locations efficiently. Model Assumptions Our model is based on several simplifying assumptions. (1) The fireload index in each fire day is independent of its value on the previous day. Although the current fire and its intensity can affect future fire events in practice, we assume that the fireload index in each fire scenario is unrelated to another scenario

380 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 (2) Representative Fire Locations (RFL) in Korean case are defined as the centers of 228 distinct administrative places, which is a minimum fire planning unit, across the Republic of Korea excluding its islands. The size of each administrative region is about 10 miles 10 miles. Thus, a helicopter can cover the whole area of an administrative region within a few minutes. (3) The standard response is defined as the required amount of resources that can reach the fire within 30 minutes for helicopters. In fact, helicopters require a stand-by time (e.g., 5 15 minutes). I didn t account for this stand-by time in our model. (4) Each helicopter can respond to only one fire simultaneously on any given day. However, this assumption is later released because air resources can be used on multiple fires per day if those resources are well distributed over the course of the day. To find the optimal spatial allocation of IA resources, the objective function and constraints can be represented mathematically as follows. Index i and I denote the index and set of kinds of fire resources; j and J the index and set of fire stations; k and K the index and set of potential fire locations; s and S the index and set of fire scenarios. Decision Variables y j = binary variable; 1 if station j is open, 0 otherwise; x ij = integer variable for amount of resource type i deployed at station j; d ijks = integer variable for amount of resource type i at station j that is dispatched to fire location k during fire day s; and z ks = binary variable; 1 if fire location k receives a standard response during fire day s; 0 otherwise. Parameters = a weight (0 1); = fixed cost of opening station j; c i = annual cost of operating resource type i; = weight of an uncovered fire at representative fire location k;

381 June 7-8, 2011 = probability that fire scenario (fire day) s occurs; r isk = amount of resource type i required at location k during fire day s; t ijk = response time of resource type i from station j to location k; B = annual budget limit for IA; Cap ij = capacity of resource type i at station j; T ik = maximum response time of resource type i to location k; and N ik = set of stations for resource i from which resources can reach location k within the maximum response time; i.e., N = j t < T. { } Objective Function The objective function to be minimized is the total annual cost of operating resources and the expected number of fires that do not receive the standard response within the threshold time range (Equation 1). Minimize: 1 1 (1) In equation 1, the first two terms indicate the fixed cost of opening stations and the total annual cost of helicopters deployed, and the last term represents the total expected number of fires that do not receive the standard response with a weight. The variables y and x are first-stage variables, as they have to be decided upon before the outcome of the stochastic variable s is observed. The variable z, which is a binary variable, is a second-stage variables, as they can be calculated after the outcome of s is known. D represents the protection priority to get an IA response first by weighting a high cost if a fire occurs at fire location k. p denotes the probability of a fire scenario. Instead of estimating the probability of the occurrence of each fire day, I assume that each scenario is equally likely (Haight and Fried 2007; MacLellan and Martell 1996). Constraints In inequality 2, the first constraint requires that the total annual cost of operating suppression resources for the IA is constrained to less than or equal to the budget limit. ik ijk i i I j J c x i ij B (2)

382 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 In inequality 3, the constraint represents the capacity of each station. x Cap ij ij y j for all i I and j J (3) In inequality 4, the constraint requires that the amount of resources dispatched from each station during each fire day is less than or equal to the amount of resources deployed at the station. dijks k K x ij for all i I, j J and s S (4) In the inequality 5, the last constraint represents the condition for whether a fire receives a standard response during stage two. Whether a fire receives a standard response (zks = 1) implies that the number of each type of resource that is within the standard response time and dispatched to the fire, d j N ijks ik, is greater than or equal to the amount of resources required, riks. However, if riks = 0, there is no fire at location k during fire day s and zks = 1, thereby necessitating no resource commitment. z r ks isk j N ik d ijks for all i I, k K and s S (5) 3. Application Our application in the Republic of Korea focuses on the deployment of primary helicopters among 8 stations, assuming that other resources such as hand crews and fire engines are retained in their current locations. The application involves 100 fire scenario days of potential fire days, each with fires occurring at different locations. For the application, we use spatially explicit GIS-based data on the ecology, fire behavior, and economic/cost characterization important in the Republic of Korea from Korea Forestry Research Institute. The information on potential fire locations is used to calculate traveling time from fire-fighting resource bases to RFLs in the Republic of Korea. Using the Korea fire simulation model of Lee et al. (2011), we generate 100 fire scenarios. We also use the current data on helitack from Forest Aviation Headquarters of Korea Forest Service (KFS) (website: ). Annual costs of initial attack resources from KFS are utilized as unit costs in the application

383 June 7-8, 2011 Results and Discussion The curve showing the tradeoff between number of helicopters deployed and expected number of fires per 10+ fire days (days on which 10 or more fires occurred) not receiving the standard response has a convex shape where noncoverage declines at a decreasing rate as number of helicopters deployed rises (Figure 1). The points on the curve represent non-dominated solutions and their relative performance with respect to the two objectives. For each non-dominated solution, improvement in one objective cannot be achieved without simultaneously causing degradation in the value of the other objective. As a result, the points represent a frontier below which there were no better solutions. Figure 1. Tradeoff between number of helicopters deployed and expected number of fires that do not receive a standard response The optimal deployment of helicopters depends on the objective function weight. If minimizing the number of helicopters deployed is most important (i.e., α=1), the choice is solution A in which the expected number of fires not receiving the standard response is equal to the average daily fire frequency of (Table 1). As more weight is given to minimizing the number of uncovered fires, more helicopters are deployed (solution D), the expected number of fires left uncovered

384 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 is 3.15 (25% of the average number of fires per 10+ fire day). Increasing the number of helicopters from 13 to 27 (solution I) reduces the number of uncovered fires to 0.08 (1% of the daily average). The slope of the tradeoff curve, which represents the gain in daily number of fires covered per unit increase in number of helicopters deployed, is relatively steep between solutions A and G (0.40 fires/helicopter). Between solutions G and I, the slope is relatively flat (0.06 fires/helicopter). Optimal solutions concentrate helicopters in stations that are close to RFLs with the highest fire loads or populated area (i.e., Gang-won, Kyung-buk, Kyung-nam, and Kyung-gi). Table 1. Number of helicopters deployed per station and number of fires that do not receive a standard response from spatial optimization a Case Num. of Helicopters Num. of fires not covered Helicopter Bases J1 J2 J3 J4 J5 J6 J7 J8 A B C D E F G H I a The dispatch level is defined as 5 categories: 1: 1 helicopter; 2: 1 helicopter; 3: 2 helicopters; 4: 2 helicopters; 5: 3 helicopters. Using historical data, we conduct a sensitivity analysis on dispatch rules. The new dispatch rule is built based on the previous dispatch experiences. When we apply the new dispatch rule in our optimization model, the results show that more helicopters are required to achieve the goal to provide a standard response to a fire location quickly. For example, to reduce the number of uncovered fires to 0.08 fires (1% of the daily average), 45 helicopters should be deployed to 8 stations. Compared to the previous case (Table 1), 12 helicopters (26%) are increased to get the same results in terms of the number of uncovered fires. With this setup, we find that up to 54 helicopters along with the other initial attack resources contain 99% of the fires. Incrementally reducing the number of

385 June 7-8, 2011 primary helicopters from 54 to 14 reduces the containment rate to 25%. The number of contained fires is sensitive to where the helicopters are deployed, which shows that the spatial allocation of initial attack resources is critical to contain a fire quickly. Table 2. Number of helicopters deployed per station and number of fires that do not receive a standard response from spatial optimization with a different dispatch rule Case Num. of Helicopters Num. of fires not covered Helicopter Bases J1 J2 J3 J4 J5 J6 J7 J8 A B C D E F a Based on the historical data, the dispatch level is newly defined as 5 categories: 1: 0 helicopter; 2: 1 helicopter; 3: 2 helicopters; 4: 3 helicopters; 5: 4 helicopters. The results of this study suggest how the consideration of spatial heterogeneity with different policy preferences affects the optimal allocation of the IA resources during a fire season. Further, I expect to discern which factors (fire behavior, terrain, budget constraints, policy goals, etc.) drive differences in optimal policy between the two regions by varying parameters systematically in the sensitivity analysis. This study create a foundation for future work by establishing the application of the stochastic simulation optimization framework to other settings, creating a platform to explore other policy goals, and building the capacity for sophisticated forest land risk management in Korea. References Haight, R. G. & Fried, J. S. (2007). Deploying Wildland Fire Suppression Resources with a Scenario-Based Standard Response Model. INFOR, 45(1): Lee, B. and M. B. Lee. (2009). Spatial Patterns of Forest Fires between 1991 and Journal of Korean Institute of Fire Science and Engineering,

386 International Forest Fire Symposium on Commemorating the International Year of Forests (1): Lee, B., Lee, Y. H., Lee, M.B., Albers, H.J. (2011). Stochastic Simulation Model of Fire Occurrence in the Republic of Korea. Journal of Korean Forestry Society (in Press) Lee, Y., Lee, K., and Kim, S. (2006). Severity of Forest Fire according to Forest Stand Structure. KFRI Journal of Forest Science, 69, MacLellan, J. I. & Martell, D. L. (1996). Basing airtankers for forest fire control in Ontario. Operations Research, 44,

387 June 7-8, 2011 Eco-toxicity of firefighting foams assessed in aquatic and semi-aquatic compact biotopes Kohei Otsuka, Reina Inokuchi, Yuki Ishizaki, Kazuya Uezu, Bart Dewancker and Tomonori Kawano* Graduate School of Environmental Engineering, The University of Kitakyushu, 1-1 Hibikino, Wakamatsu-ku, Kitakyushu, Japan Abstract Based on the observation in compact biotopes mimicking the fresh water environments and wetland, both the acute and long-term eco-toxic impacts of two fire-fighting foams were assessed. Spraying of synthetic detergent-based foam formula was shown to be more toxic compared to soap-based formula and mock water treatment. Introduction A variety of chemicals including foaming agents are used to aid in the protection of forest resources from the wildland fires (both woodland and grassland fires) (USDA Forest Service, 1994). The fire-fighting foams are formulations composed principally of surfactants, and act by increasing water efficiency. These chemicals are rapidly gaining acceptance as effective and efficient tools in several countries such as United State, Australia (Rawet et al., 1996; Adams and Simmons, 1999). Also, in Japan, Kitakyushu City Fire and Disaster Management Department (FDMD) and Tokyo Fire Department, followed by other local fire-related authorities nation-wide, have been using some firefighting foams in the urban fire controls since 1999 (Kitakyushu City Fire and Disaster Management Department, 2005). In case of wildland fire managements, much greater amounts of fire-fighting chemicals might be emitted to the environments compared to ordinary urban fire controls. Therefore possible ecotoxicities of these chemicals should be tested before designing the chemical fire fighting strategies in the fields. However, until recent date, the potential ecological impacts of such chemicals have not been thoroughly assessed (USDA Forest Service, 1994). Therefore, certain research efforts for obtaining the reliable data of ecological risk

388 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 assessments are highly required for safely managing the use of such chemicals at the ecologically acceptable range. In Japan, possible application of such chemical foams in wildland fire managements is now being considered. Prior to actual use in the real fire-fighting field exercise in Japan, there is an emerging need for testing the impacts of these chemicals on the living organisms composing the typical landscapes or ecosystems in Japan. For above purpose, in recent years, our group in Kitakyushu have been proposing and conducting a series of bioassays for assessing and comparing the toxicities of commercially available and newly developed fire-fighting foams and their components (chiefly synthetic and fatty acid-based detergents). The biological materials used in our recent bioassays mostly focused on the acute toxicity and biodegradability of fire-fighting chemicals include a tiny fresh water model fish, Oryzias latipes (Lin et al., 2006; Kawano et al., 2007a; Mizuki et al., 2007); aquatic protozoa such as Paramecium bursaria (Kadono et al., 2006a; Goto et al., 2007, 2008; Mizuki et al., 2007) and Paramecium caudatum (Kadono et al., 2006b), germinating seeds of rice plants (Oryza sativa. L.), a key component in the semi-aquatic landscape in Japan (Kawano et al., 2006, 2007b), and microbial system in activated sludge (Mizuki et al., 2010). In the present report, we wish to describe our latest attempt for assessing the long-term impacts of fire-fighting foams in aquatic (fresh water environment) and semi-aquatic (wetland) eco-systems, by employing the biotope-based observation. Experimental Compact biotopes mimicking the aquatic and semi-aquatic eco-systems were set using plastic containers filled with water and/or soil mixture in May, Then, model aquatic organisms such as a tiny fish (Oryzias latipes), an aquatic microbe, green paramecia (Paramecium bursaria), aquatic and semiaquatic plants such as Common Water Hyacinth (Eichhornia crassipes) and rice plants (Oryzias latipes) were installed. Concomitantly, lavae of other organisms such as small sinistral, air-breathing freshwater snail (Physella acuta) and insects chiefly dragon flies were also installed. Water or two fire-fighting foams (1. Phoschek, a commercially available internationally known formula containing synthetic surfactant cocktails as active ingredients, and 2. Miracle-foam, a recently released Japanese formula chiefly consisted of fatty acid salts or soaps) were sprayed above the compact biotopes in September, The latter foam formula was developed by our recent research

389 June 7-8, 2011 (Mizuki et al., 2007, 2010). Concentrations of fire-fighting agent in pre-foaming mixtures were 0.5 % (w/v) and 1 % (w/v), respectively. Onto 1 m 2 of model biotope, 3 L of water or fire-fighting mixture was sprayed (thus foam spread) after compressing the air in the fire-fighting cylinders. Acute toxicity of fire fighting agents especially plants, fish and microbe was assessed several times within initial 2 weeks. Long-impacts of chemical treatment was assessed in April 2011, based on the changes the populations of small snails and lavae of dragon flies. Results and discussion The details of our results are shown on our poster. Briefly, among three different treatments, namely water, the synthetic foam (Phoschek) and the soapbased foam (Miracle-foam), acute toxicity in Paramecium bursaria, Orysias latipes, and Eichhornia crassipes was highest in the synthetic foam-treated biotopes. While blooms of populations in small snails (Physella acuta) and larvae of dragon flies were observed in water-treated and the soap-based formula-treated biotopes 7 months after the sprays, no single larva of insect or snail could be found in the synthetic foam-treated biotopes. Our demonstration encourages more studies on the impacts of fire-fighting activities or strategies against the eco-system when applied to wildfire exercises. We are now developing a novel fire-fighting formula with less impacts on the ecosystem with modified formulae of soap-based fire-fighting foam. Further tests (to be conducted partly this year) are required for materializing such ideal firefighting formulae. References Adams, R. and Simmons, D. (1999) Ecological effects of fire fighting foams and retardants, Ref 1001, In: Conf. Proc. Austr. Bushfire Conf (Eds. Lunt, I., Green, D. G. and Lord, B.), Published by the Schl. of Environ. Info. Sci. and Johnstone Centr., Charles Sturt Univ., Albury, Australia. Goto, K., Kadono, T. and Kawano T. (2008) Use of natural mineral waters as the sources of diversified natural waters worldwide for testing the eco-toxicity of detergents using green paramecia. ITE-IBA Letters 1 (2): Goto, K., Lin, C., Kadono, T., Hirono, M., Uezu, K. and Kawano, T. (2007) Ecotoxicity of a soap component (sodium oleate) and a synthetic detergent

390 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 cocktail using green paramecia assayed in natural water samples from East Asia. Journal of Environmental Engineering and Management. 17 (6): Kadono, T., Uezu, K., Kosaka, T. and Kawano, T. (2006a) Altered toxicities of fatty acid salts in green paramecia cultured in different waters. Zeitschrift fur Naturforschung 61c (7,8): Kadono, T., Uezu, K., and Kawano, T. (2006b) Confirming the altered toxicities of fatty acid salts in Paramecium caudatum cultured in different waters. ITE Letters 7 (6): Kawano, T., Kadono, T., Matsuoka, N., Tamura, T. and Uezu, K. (2007) Development of soap-based fire-fighting agents less toxic to germinating rice (Oryza sativa L.) seeds. ITE Letters 8 (5): Kawano, T., Kadono, T., Matsuoka, N., Tamura, T. and Uezu, K. (2006) Possible ecological risk assessment of commercial fire-fighting foams using germinating rice (Oryza sativa L.) seeds. ITE Letters 7 (4): Kawano, T., Lin, C., Kadono, T., and Uezu, K. (2007a) Ecological risk assessment of fire-fighting chemicals using medaka fish (Oryzias latipes) in different water conditions. ITE Letters 8 (3): Kitakyushu City Fire and Disaster Management Department (2005) Research and development of ecologically acceptable fire suppression formula for urban fires (article in Japanese), Gekkan Shobo (Monthly Fire Fighting by Tokyo Ho-Rei), No. 313 (2005-9), 1-8. Lin, C., Kadono, T., Yoshizuka, K., Uezu, K. and Kawano, T. (2006) Assessing the eco-toxicity of novel soap-based fire-fighting foam using medaka fish (Oryzias latipes, Red-orange variety) adopted to river and sea water conditions. ITE Letters 7 (5): Mizuki, H., Toyomura, M., Uezu, K., Yasui, H., Kawano, T., Akiba, I., Kawahara, T., Hatae, S., Sakamoto, N., Akiyama, M., Mizota, C., Umeki, H. and Yamaga, K. (2010) Microbial degradation of a soap-based fire-fighting agent in activated sludge. Journal of Environmental Engineering and Management 20(2): Mizuki, H., Uezu, K., Kawano, T., Kadono, T., Kobayashi, M., Hatae, S., Oba, Y., Iwamoto, S., Mitsumune, S., Owari, M., Nagatomo, Y., Umeki, H. and Yamaga, K. (2007) Novel environmental friendly soap-based fire-fighting agent. Journal of Environmental Engineering and Management 17 (6): Rawet, D., Smith, R. and Kravainis, G. (1996) A comparison of water additives for mopping-up after forest fires, Intern. J. Wildland Fire, 6, USDA Forest Service (1994) Wildland Fire suppression, In: Ecological Risk Assesment, USDA Forest Service, USA

391 June 7-8, 2011 A Combustion Analysis of Surface Fuel Burning Experiment According to Density Variation Abstract 밀도변화에따른지표연료의연소특성분석 Eungsik Kim, Janghwan Kim*, Donghyun Kim, Byungdoo Lee Dept. of Safety & Health Engineering, Hoseo University. Baebang Asan Chungnam Korea. eskim@hoseo.edu Dept. of Safety & Health Engineering, Hoseo University. Baebang Asan Chungnam Korea. tio5283@hanmail.net Korea Forest Research Institute, Seoul, , Korea k3d3h1@forest.go.kr Korea Forest Research Institute, Seoul, , Korea byungdoo@forest.go.kr The type and intensity of Forest fire is depending on total fuel amounts contained in forest. Almost 70% of Korea is mountainous area and generally surface of these areas have many mountains and valleys. In Korea forest fire begins with surface fire near agricultural area by human error. Combustion characteristics of surface fuel according to its mass densities are the object of this paper. The fallen leaves of Quercus variabilis and Pinus densiflora, each of which is the major type of coniferous and deciduous tree respectively are used as surface fuel. The data such as combustion temperature, mass loss rate, flame height, duration of combustion and velocity of hot gas are obtained and it is to be used for the spread model and for the basic parameters for risk assessment of fire. In natural state the depth of surface fuel is generally about 5~10 cm deep and some places such as low wooded valley and crevice in the rock can have even thicker depth. In natural condition the mass of surface fuel per unit volume of cubic meters is between approximately 10~32 kg/m 3, so the densities of fuel sample in the basket are set to from 9.6 to 31.8 kg/m 3 and the diameters of the basket are changed to 20, 30, 40 and 50 cm. Almost over 150 fuel samples were burnt for data collection. Pilot ignition is carried on the top of the fuel. To find the temperature distribution from the bottom of basket up to 1.5 m, 32 thermocouples are installed in grid style and DAQ collects 32 temperature data from each

392 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 thermocouple per second. As a result in case of Pinus densiflora mass loss rate, duration of flame, flame height and combustion time become larger as the mass density and diameter of basket increase, on the other hand Quercus variabilis shows saturation characteristics in mass loss rate and flame height. Velocity of hot gas is proportional to flame height. 요약 산불의형태및강도는임내의연료물량, 수종등에따라변화한다. 우리나라와같이경사가급하고, 구릉지가많은지형에서의산불은대부분이지표화로시작하여수관화로전이되는특징을가지고있다. 따라서초기연소물질인굴참나무와소나무낙엽을이용하여밀도에따른지표연료의연소특성분석을하였다. 본연구에서는산림내지표연료층과유사한조건으로굴참나무와소나무낙엽을지름이각각 20, 30, 40, 50cm 높이 10cm인원통형바스켓에직경과지표연료의단위면적당밀도를달리하여 5회이상의반복실험을통하여총 160여회이상실험하였다. 실험장치의전체시스템은온도측정을위하여 Ø1.6mm K-type 열전대는총 32개를설치하였다. 바스켓중심바닥면, 중앙, 시료표면에열전대 3개, 바스켓중앙시료표면으로부터수직방향 25cm간격으로열전대 6개, 바스켓양쪽외곽에서수직방향 25cm간격으로 14개, 바스켓양쪽수평방향 25cm간격에서 8개, 주위온도측정용 1개를격자모양으로설치측정하였다. 질량감속도는 load cell을이용 1g단위로 1초간격으로측정하였으며, 기체유속은바스켓직상부 1.5m에 Kanomax사고온풍속계를고정설치하여기체유속 (0~25m/sec), 기체온도 (0~500 ) 를동시에측정하였다. 화염높이는캠코더로녹화하여 1초단위로재생화염의높이를 5cm 간격으로판독, 소염시점까지측정하였다. 결론적으로침엽수종낙엽의경우밀도와지름의증가함에따라질량감소속도, 화염지속시간, 화염의높이그리고연소시간은증가한반면, 활엽수종낙엽의경우질량감소속도와화염높이는증가하다가감소하였으며화염지속시간과연소시간은증가하였다. 또한, 기체유속및온도는화염높이가커질수록증가하는경향을나타내었다. 재료및방법 (Methodology) 실험은산림내지표연료층의두께및밀도와비슷한조건에서하기위해소나무와굴참나무낙엽을지름이각각 20, 30, 40, 50cm 높이 10cm 인바스켓에넣고, 온도분포, 질량감소속도, 화염의높이, 연소시간, 화염지속시간, 열방출속도, 기체유속및온도등의연소특성을측정하였다. Table 1 은바스켓의

393 June 7-8, 2011 직경과지표연료의단위면적당무게를달리하여만든시료의구성표로동일바스켓크기를기준으로하여중량변화따른밀도비는 1 : 1.7 : 2.3 : 3.3 이며, 동일밀도의경우바스켓크기에따른연료중량비는 1 : 2.25 : 4 : 6.25 이다. 각시료의경우 5 회이상의반복실험을통하여총 160 여회이상측정하였으며, 1 초단위로측정데이터를수집한후분석하였다. Table 1. Density and weight of sample filled in each basket Density(kg/m 3 ) Diameter of basket 20cm 30cm 40cm 50cm g 67.5g 120g 187.4g g 112.5g 200g 312g g 157.5g 280g 436.3g g 225g 400g 624g 결과및고찰 (Results and Discussion) Figure 1 는소나무낙엽과굴참나무낙엽에대한밀도변화에따른 MLR 의실험결과로, 소나무낙엽의밀도변화에따라 MLR 가최대약 3.5~6 배정도증가하였으며, 굴참나무낙엽은 MLR 가최대약 4.8~6.5 배정도증가하다가 15.9kg/ m3에서감소하는경향을보이고있다. 또한동일한밀도조건에서소나무낙엽이굴참나무낙엽에비해빠른질량감소속도를나타내었으며, 이는두시료간의평균열량값차이와낙엽의형태에따라공기유입량의차이에서기인된것으로판단된다 Mass Loss Rate(g/s) cm-P.D. 30cm-P.D. 40cm-P.D. 50cm-P.D Mass Loss Rate(g/s) cm-Q.V. 30cm-Q.V. 40cm-Q.V. 50cm-Q.V Density(kg/ m3 ) Density(kg/ m3 ) (a) Pinus densiflora (b) Quercus variabilis Figure 1. Mass loss rate versus density change of sample

394 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Figure 2(a) 는소나무낙엽의밀도가증가함에따라화염지속시간이서서히증가함에비해, Figure 2(b) 는화염지속시간의폭차이가크며, 특히지름이 50cm 바스켓의화염지속시간이최대약 2 배 ( 밀도 22.2kg/ m3에서 31.8kg/ m3 ) 증가함을확인하였다. 특히굴참나무낙엽의밀도가증가함에따라화염의높이는감소한반면연소지속시간은증가하였다. 그이유는밀도가증가함에따라공기유입량이감소로연소속도및화염높이가감소하는것으로사료된다 time(sec) 20cm-P.D. 30cm-P.D. 40cm-P.D. 50cm-P.D time(sec) 20cm-Q.V. 30cm-Q.V. 40cm-Q.V. 50cm-Q.V Density(kg/ m3 ) 인용문헌 (References) Density(kg/ m3 ) (a) Pinus densiflora (b) Quercus variabilis Figure 2. Flame continuation time versus density change of sample. Albini, F.A. (1980), " Thermochemical Properties of Flame Gases from Fine Wildland Fuels ", USDA Forest Service Research Paper INT-243. Dupuym, J.L., Marechal, J., and Morvan, D. (2003), Fires from a cylindrical forest fuel burner : combustion dynamics and flame properties, combustion and flame, pp Kim, D. H. (2004), " Analysis of the effect factors Cause by a scale of Forest Fire Damaged Area", Journal of Korean Forest Society, Vol.1, No. 0. Kim, E.S. (1997), " A Study on Forest Fire Spreading Prediction Program ", Korea Forest Service Research Paper. Kim, J.H., Kim, E.S., Lee, M.B., Kim, D.H., and Park, H.J. (2007), "A Combustion Analysis of Surface Fuel Burning Experiment According to Density Variation" Korean Society of Safety Conference. Kim, J.H., Kim, E.S., Lee, M.B., Kim, D.H., and Park, H.J. (2008), "A Study of Calorie Analysis Methods about Surface Fire Fuel ", Journal of Korean Institute of Fire Science & Engineering, Vol. 22, No. 3, pp

395 June 7-8, 2011 Changes of Fuel Consumption in forest by Thinning and Pruning 간벌및가지치기방법에따른임내연료량변화 Kyosang Koo 1*, Sungcheol Jung 1, Youngho Kang 1, Kyongha Kim 1, Junho Park 2 1 Korea Forest Research Institute, 57 hoegi-ro, Dongdaemun-gu, Seoul, , Korea 2 Gyeongsangnam do Forest Environment Research Institute, 386 Sumokwon-ro, Jinju, , Korea Abstract kyosang@forest.go.kr This study is for minimizing the damage about the large fire, and for investigating the effects of fire brake forests according to the method of thinning and pruning among silvicultural systems which are able to prepare or confront the forest fire damage. Fuel amounts of Pinus rigida stands treating thinning were as about half time as them of the standard plot which was not done. In case of the fuel consumption each treating plots after pruning, the standard plot was the lowest because the fuel consumption of the lower layer were increased by the thick growth of Sasa borealis according to treating pruning. Therefore, to reduce the fuel consumption, it has been thinking that periodical weeding on the lower layer should be needed after doing pruning. As a result of analyzing the fuel consumption of litter layer affecting a lot to breaking forest fires, the higher the intensity of thinning was, the lower the fuel consumption would be after thinning in Pinus rigida stands. However, the fuel consumption of litter layer after pruning in Pinus koraiensis stands tended to increase because of the thick growth of the lower layer. 요약 본연구는대형산불에대한피해를극소화시키고, 산불재해에대비 대응할수있는임업적시업방법중간벌및가지치기방법에따른내화수림대조성효과를구명하고자한다. 리기다소나무림의간벌처리에따른연료량은처리를하지않은대조구보다약 1/2 정도낮게나타났다. 가지치기처리후처

396 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 리구별연료량은대조구가가장적게나타났는데, 이는가지치기처리에따른조릿대밀생으로하층연료량이증가하였기때문이다. 따라서, 연료량을감소시키기위해서는가지치기처리후하층의풀베기작업이주기적으로실시해야할것으로사료된다. 산불발생에큰영향을미치는낙엽층의연료량분석결과, 리기다소나무림의간벌처리후연료량은간벌강도가높을수록연료량은낮아지는경향이나타났다. 그러나, 잣나무림의가지치기처리후낙엽층연료량은가지치기후하층의밀생으로연료량이증가하는경향이나타났다. Introduction 최근우리나라에서도이상고온, 건조등기후변화에따른건조일수, 산림내지피물등연소물질증가등으로산불의대형화및전국동시다발성산불발생이증가하는추세이다. 대형산불에대한피해를극소화시키고, 산불재해에대비 대응할수있는임업적시업방법개발이필요하다. 이러한임업적시업방법중내화수림대조성은산불발생으로인한피해가큰소나무단순림을대상으로산림내임도나작업로를이용하는방법, 주변임상을고려하여침엽수와활엽수 ( 참나무류 ) 가혼효되도록유도하는방법, 임분을자연지형을이용하여 10 20m 폭으로내화수림대를조성하는방법등이있다. 본연구는이러한임업적시업방법중임내간벌및가지치기방법에따른내화수림대조성효과를구명하여대형산불예방과피해저감을위한기초자료로활용하고자한다. Methodology 연구대상지는경남거창군위천면상천리 (N , E ) 에위치하고있으며, 리기다소나무조림지는 1979년 3,000본 /ha을조림하여 1991 년도에 1,500/ha로간벌을실시한후잡관목이울폐하였으며해발고 668m, 경사도 5 20 였다. 잣나무조림지는 1990년도에조림하였고수령은 28년이며, 해발고 410m, 경사도 5 10 였다. 내화수림대조성을위한임업적시업방법은리기다소나무 (Pinus rigida, P.R.) 림은 100본 /ha, 200본 /ha, 300본 /ha으로간벌처리하였으며, 잣나무 (Pinus koraiensis, P.K.) 림은수고대비 50% 와 70% 가지치기를실시하였다. 임내처리별효과를분석하기위하여각각의처리후임내연료량을분석하였다

397 June 7-8, 2011 Results and Discussion 리기다소나무림의간벌강도후처리구별전체연료량은입목밀도가높은대조구에서 52.2ton/ha 로높게나타났으며, T 100 에서낮게나타났다. 하층연료량은대조구 (31.4ton/ha)>T 300(20.6ton/ha)>T 200(18.7ton/ha)>T 100(16.6ton/ha) 순으로나타났다. 간벌강도처리에따른연료량은처리를하지않은대조구보다약 1/2 정도낮았다. 가지치기처리구별전체연료량은 P 50(92.2ton/ha)>P 70(87.5ton/ha)> 대조구 (58.7ton/ha) 순으로대조구가가장적게나타났는데, 이는가지치기처리에따른조릿대밀생으로인한하층의연료량이증가하였다 (Wang et al. 1995). 따라서, 연료량을감소시키기위해서는가지치기처리후주기적으로하층의풀베기작업을실시해야할것으로사료된다. 산불발생에큰영향을미치는낙엽층의연료량분석결과, 리기다소나무림의간벌강도후연료량은대조구 (31.1ton/ha)>T 300(20.6ton/ha)>T 200(18.7ton/ha)>T 100(16.2ton/ha) 순으로, 강도가높을수록연료량은낮아지는경향이나타났다. 잣나무림의가지치기후연료량은 P 50(19.7ton/ha)>P 70(9.8ton/ha)> 대조구 (8.2ton/ha) 순으로, 가지치기후하층의밀생으로연료량이증가하는경향이나타났다. Figure 1. Fuel consumption of layers at pruning plots Figure 2. Fuel consumption of layers at pruning plots Figure 3. Fuel consumption of litter layer at thinning plots Figure 4. Fuel consumption of litter layer at pruning plots

398 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 References Wang, J.R., Simard, S.W. and Kimmins, J.P Physiological responses of paper birch to thinning in British Columbia. Forest Ecology and Management 73;

399 June 7-8, 2011 Scattering of tsunami wave from a forest 산림에의한지진해일파의산란 Se-Myong Chang 1*, Zhenqing Zhang 1, Tae Wan Park 1 Hojoong Youn 2, Byungdoo Lee 2 1 Kunsan National University, Jeonbuk, , Korea smchang@kunsan.ac.kr 2 Korea Forest Research Institute, Seoul, , Korea rhkim@forest.go.kr Abstract A tsunami originated in a deep sea due to the tectonic activity such as earthquake is the transport phenomena of great scale energy, which makes much damage to sea-bounded area since its wave height becomes higher approaching the sea shore. Therefore, the tsunami wave is a kind of shallow water wave, and the characteristics similar with the shock wave in gas dynamics should be noted so remarkably that there exists a sharp discontinuity before and after the wave front. Actually there is a mathematical similarity between the equations of shallow water waves and gas dynamics. In this research, the existing code concerning the computation of unsteady shock waves is improved to simulate the propagation of tsunami waves. With this code, reflection, diffraction, and scattering of an incident shock wave are studied in the view point of physics, and the basic parameter analysis of stand density and the distance between trees is made for the application in the future. 요약 지진해일은심해에서지진등의지각활동을통하여발생하는거대에너지의전달현상이며, 해안에가까워질수록파고가높아져연안지방에많은피해를준다. 따라서지진해일파는일종의천수파이며, 파동선단을전후하여파고의날카로운불연속이존재한다는점에서기체역학에서연구대상인충격파와유사한특성이있다. 실제로천수파에관련된방정식에는기체역학방정식과서로수학적상사성이존재한다. 본연구에서는기존비정상충격파계산코드를개량하여지진해일파의전파를계산할수있도록하였다. 이러한

400 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 코드를이용하여입사하는충격파의반사, 회절, 산란등의물리현상을탐구하고, 임분밀도및수간거리에대한기초파라미터해석을실시하였다. 서론 (Introduction) 지진해일파 ( 일명쓰나미 ) 는지진과같은심해지각활동으로발생하며, 해안선에가까운천해지역에전파되었을경우, 갑자기높은파고의파도가몰아쳐많은피해를준다. 이러한쓰나미의피해는최근일본도호쿠지방의지진을통하여우리사회에많은경각심을주었다. 본연구에서는이러한지진해일파의방재를위한해안방재림의설치타당성에대해논하고자한다. 직관적으로살펴볼때, 쓰나미는파동의일종이며, 이는거대에너지의전파이므로, 일반파동과마찬가지로장애물 ( 산림 ) 을만났을때에너지의일부를잃어버릴것이라예측할수있다. 따라서본연구에서는적절한수학적모델링을통하여지진해일파의해안천해지역전파와산림에의한반사, 회절, 산란등을다루고자한다. 방법 (Methodology) 본연구에서는전산유체역학 (CFD: computational fluid dynamics) 기법을사용하여다음과같은가정에서문제를해결하고자한다. 1) 지진해일에의한나무의변형혹은파괴는무시한다. 2) 수관 (crown) 을제외한줄기만을기둥으로모델링한다. 3) 지진해일은불연속한한개의비선형파동으로가정한다 (Riemann 문제 ). 4) 지진해일파는국부적으로수심이거의일정한영역을전파한다고가정한다. 5) 유동과산림의마찰에의한감쇠는무시한다 ( 즉, 파동에너지는보존된다.). 이러한가정아래, 다음과같은방정식을주어진초기조건에대해풀이하고자한다. uh 0 (1)

401 June 7-8, 2011 h : 파고 u : 유동속도 g : 중력가속도 (= 9.8 m/s 2 ) t : 시간좌표 x : 공간좌표 uh u h gh (2) 식 (1), (2) 는기체역학에서발생하는충격파 (shock wave) 와수학적유사성 (analogy) 이있다고알려져있다. 즉, 파고 (h) 를기체밀도 (ρ) 로바꾸고, 식 (2) 의우변을압력 (p) 의구배항으로처리한다면기체역학에서의연속방정식과운동량방정식에해당한다. 식 (1), (2) 의시스템은무차원화하면다음과같은단일파라미터의영향을받는다. Fr (3) 식 (3) 을 Froud 수라고하는데, 기체역학에서 Mach 수와같은역할을한다. 특성속도를비교해보면, 파고 3~4 미터정도의지진해일의특성속도는음속 (speed of sound) 의약 1/100 단위에해당됨을알수있다. 따라서본연구에서는 Fr=2.45 인경우만으로관심영역을집중하기로한다 ( 파고비는 3.0:1). 기체역학의 Rankin-Hugonite 조건과같이, 지진해일파에서도적절한좌표변환과식 (1), (2) 와같은보존법칙 (conservative laws) 을이용하면, 다음과같은관계식을얻을수있다 ( 아래첨자 1 은파동앞, 2 는파동뒤의영역을의미함 ): 1 8Fr 1 1 (4) u 1 Fr gh (5) 따라서식 (4), (5) 를이용하면, 주어진파동파라미터 Fr 에대하여초기조건을정할수있다. 결과및고찰 (Results and Discussion) 본문제는계산영역이매우크기때문에일단가로 300 미터, 세로 2.2~7 미터의계산영역을설정하였다. 산림이균일하다고가정하면, 이러한계

402 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 산영역이세로방향으로무한히반복된다고볼수있다. 따라서계산영역의수간대칭선에대칭영역조건을부여하였다 (Figure 1). 2.2 미터는남해물건리의임분밀도 (220 본 /ha) 로부터도출한일반적수간거리이며, 7 미터는이보다 10 배임분밀도를증가 (2000 본 /ha) 시켰을경우의수간거리이다. 흉고직경은각각 15 와 30 센티미터를비교하였다 (Table 1). Table 1. Computational parameters Simulation Distance b/ trees (m) Diameter of Trees (m) Length of Forrest (m) Reduction in Flow velocity Reduction in Wave height I % 21.5 % II % 22.1 % III % 21.5 % IV % 21.5 % Fig. 1. Computational domain Fig. 2. Computational result (Simulation I) 지진해일파는계산영역좌측 40 미터에서출발하여 50 미터지점부터시작되는숲에도달한다. 숲의영역길이는 100 미터를표준으로, 150 미터까지를계산했다. 본연구에서의계산은비정상유동에대한것이므로, 각시간대별로 Figure 2 와같은장면을얻을수있다 ( 파동출발 38 초뒤의모습을확대한것임, 격자개수는약 21 만개 ). 나무가장애물로작용하여후류 (wave, 그림의파란부분 ) 를형성하고있음을확연히알수있다. Table 1. 의 I 번시뮬레이션의경우, 숲내부의수간대칭선에서후류의파고감쇄는약 19 센티미터 (6.3%), 최종파동선단에서유속에대해서는 3.7%, 파고에대해서는 21.5% 의감쇠효과가있음을보여준다. Table 1 의데이터를보면다음과같은잠정적인결론을얻을수있다

403 June 7-8, ) 파고에대한감쇠효과는어떤경우에대해서도거의비슷하다. 이는시뮬레이션에서파동의총에너지가보존된다고가정하였으므로, 파동산란에의해후류에서복원되는지진해일파의강도는거의일정하기때문으로보인다. 2) 유동속도에대한감쇠효과는임분밀도가커질수록, 즉수간거리가작아질수록크고, 나무의흉고직경이커지면감쇠효과는커진다. 그러나숲의영역이넓어진다고해서감쇠효과가커지는것은아니었다. IV 번시뮬레이션의경우숲을투과한파동의속도가증가해있는것을발견할수있었는데, 이는빽빽한산림이오히려노즐 (nozzle) 처럼작용하여유동속도를증가시키는효과 ( 벤추리효과 ) 를가져왔다. 따라서흉고직경이큰나무를이용하여임분밀도가큰숲을조성한다면, 비교적작은면적의숲으로도지진해일에효과적으로대응할수있는해안방재림을조성할수있음을알수있었다. 인용문헌 (References) Moscow University Physics Bulletin(ISSN ), Normalized Shallow Water Equations (T.G. Elizarova and M.V. Afanasieva). Vol. 65, No. 1, pp , John Willey and Sons(ISBN ), Linear and Nonlinear Waves(G.B. Whitham)

404 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Development of Algorithm for Analyzing Priority Area of Forest Fire Surveillance Using Viewshed 가시권분석을이용한산불감시우선지역분석체계개발 Byungdoo Lee, Gyesun Ryu*, Sunyoung Kim, Kyoungha Kim Korea Forest Research Institute, Seoul, , Korea Abstract The algorithm for priority area of forest fire surveillance to enhance the effectiveness of fire detection was developed in this study. Viewshed analysis and fire occurrence probability maps were overlaid to decide the priority of forest fire surveillance. Viewshed was calculated with factors such as the heights of fire detection system, the heights of smoke column and the terrain roughness. Fire occurrence probability map was built using a logistic regression model. This result can be used in allocation of fire prevention resources and selection of suitable point for new fire detection system. 요약 산불감시활동에의한탐지확률을높이고, 감시자원의효율적인이용을위해서는산불감시우선지역에대한분석이요구된다. 따라서산불감시우선지역을추출하기위해가시권분석과산불발생확률분석을실시하였으며, 중첩을통해가중치를부여하였다. 가시권분석은탐지확률과관련된감시자원의높이, 산불연기높이, 지형의굴곡도 (roughness) 에따른유효가시거리인자를다르게하여실시하였다. 산불발생확률은로지스틱회귀분석모형과연료, 기상, 지형인자및토지피복, 접근성인자 DB 를이용하여분석하였다. 개발된산불감시우선지역분석체계는산불감시자원의효율성제고를위한기초자료로활용될수있을것으로예상된다

405 June 7-8, 2011 서론 (Introduction) 우리나라의산불은매년건수와피해면적이증가하고있으며이로인한산림생태계피해와진화비용등의사회적비용도증가하는추세에있다. 우리나라는대부분의산불이입산자실화나논밭두렁소각등인위적인요인에의해발생하고있기때문에산불예방활동이우선적으로강조되고있다. 산불의조기발견을위해각국에서는인력과기계화를이용한다양한탐지시스템을갖추고있다. 우리나라는무인감시카메라와감시탑이산불감시활동의큰비중을차지하고있는데, 산림청의 2006 년자료에따르면전국적으로총 762 대의무인감시카메라가설치되어운영중이다. 하지만산림청의 2006 년무인감시카메라감시실적자료를보면논밭두렁소각등의실적을제외하고직접적으로산불을탐지한경우는평균 4% 로매우낮게나타나산불감시시설의효율성에대한체계적인분석이필요한실정이다. 따라서본연구는우리나라의상황에맞는산불감시시스템평가체계를개발하기위한기초사례연구로서감시시설별가시권분석과산불발생확률지도와의중첩분석을통하여대상지역에대한산불감시우선지도를생성하였다. 재료및방법 (Methodology) 경상북도에서 3 번째로산불발생확률이높은봉화군를연구대상으로하였다. 가시권분석은 ArcGIS Viewshed analysis 를이용하여, 가시지점의높이 (OFFSETA) 는 11.6m (= 감시탑높이 10m+ 사람의평균적시선높이 1.6m) 로, 대상물의높이 (OFFSETB), 즉산불연기높이를 10m 로, 유효가시거리 (RADIUS2) 를 25km 로설정하였다. 산불발생확률지도는선행연구 ( 안상현외, 2005) 를통하여생성하였으며, 관련인자는임상, 고도, 경사, 산림과의거리, 묘지와의거리를이용하였다. 최종산불감시우선지도를도출하기위해가시권지도와산불발생확률지도각각을최소값이 0, 최대값은 1 이되도록퍼지변환을하였다. 산불감시우선지도는다음식을이용해산불발생확률이높지만가시성이낮은지역에가장높은가중치를주고, 반면산불발생활률의낮지만가시성이높은지역에가중치를 0 으로부여하였다. [Surveillance Priority] = 0.25 * [Fire Occurrence Probability] * (1 - [Visibility])

406 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 결과및고찰 (Results and Discussion) 봉화군의가시권분석결과, 1 개이상의감시시설로부터감시되는면적은전체의 21% 로나타났으며, 퍼지변환된가시성의평균값은 0.07 로매우낮게나타났다. 봉화군의산불발생확률은평균 0.57 이며, 산지가많이분포하는북부지역보다는인구가많이분포하는중남부지역의산불발생확률이높게나타났다 (Figure 1). 최종적으로도출된산불감시우선도는 Figure 2 와같으며, 평균 0.8 이상의우선도를나타내고있다. 봉화군의가시성이전체적으로매우낮기때문에우선도가높게나타나고있으며, 또한산불발생확률의패턴과비슷하게나타나고있다. 추후가중치의변화에대한민감도분석이요구된다. 본연구에서개발된산불감시우선지역분석체계는산불감시자원의효율성제고를위한기초자료로활용될수있을것으로예상된다. Figure 1. Visibility and Fire Occurrence Probability Map Figure 2. Surveillance Map

407 June 7-8, 2011 인용문헌 (References) 안상현, 원명수, 김동현, 강영호, 이명보, 이시영 의성군지역산불발생및대형화위험지역구분, 한국지리정보학회지. 제 8 권제 2 호. Catry, F.X., Almeida, R. M. and Rego, F.C Produçcãao de Cartografia de Visibilidades para Portugal Continental. A Importâancia da sua Utilizaçcãao na Vigilâancia Contra Incêendios Florestais. Silva Lusitana 12(2): Catry F.X., Damasceno P., Silva J.S., Galante M., Moreira F Spatial distribution patterns of wildfire ignitions in Portugal. Proceedings of the 4th International Wildland Fire Conference. CD Rom (10 pp.). Seville, Spain. Catry F.X., Rego F.C., Santos T, Almeida J, Relvas P Forest fires prevention in Portugal - Using GIS to help improving early fire detection effectiveness. Proceedings of the 4th International Wildland Fire Conference. CD Rom (12 pp.). Seville, Spain. Rego F.C., Catry F.X Modelling the effects of distance on the probability of fire detection from lookouts. International Journal of WildLand Fire. 15:

408 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Evaluating the feasibility of wood damaged by forest fire on solid biomass fuel 산불피해목의고형바이오연료로이용가능성평가 Soo-Min Lee*, Dong-Hyun Kim, Don-Ha Choi, Seok-Hwan Choi, Byoung-Jun Ahn, Yong-Sik Kim, Oh-Kyu Lee Korea Forest Research Institute, Seoul, , Korea Abstract This study was performed to evaluate the feasibility of wood damaged by forest fire as solid biofuels. The degree of woods damaged by radiant heating instrument at 900 o C were classified as seriously damaged, moderately damaged and slightly damaged. As results, average values for calorific value and ash content were 4,810 kcal/kg and 0.21%, respectively. Thus, there is no difference in calorific value and ash content among the wood samples. In addition, C, H, and N contents were 50.8% (w/w), 6.6% (w/w) and 0.05% (w/w), respectively, from elemental analysis of wood samples. Therefore, these results might imply that wood damaged by forest fire has great potential feasibility of raw material for wood pellet. 요약 본연구에서는산불피해로부터발생되는피해목에대하여바이오매스연료로의이용가능성을평가하기위해국내주요수종인소나무를대상으로실험하였으며, 산불피해와유사한처리를가한후연료로서의특성을평가하였다. 직경 28cm의소나무수간부위에대하여피해정도심 중 경으로구분하여복사열처리를하였다. 발열량, 함수율, 회분, 원소분석등의항목을분석함으로써산불피해목의바이오매스연료자원으로서의특성및자원이용가능성을평가하였다. 소나무의전건기준고위발열량은 4,810kcal/kg으로매우우수한것으로분석되었으며, 회분은 0.21% 로매우낮았으며, 탄소함량은 50.8%, 수소는 6.6% 로분석되었다. 질소는 0.05% 로분석되어매우우수한에

409 June 7-8, 2011 너지특성을지니고있음을확인하였다. 1. 서론 (Introduction) 지구온난화및기후변화에따라전세계적으로산림재해가증가하고있다. 특히 2000 년과 2002 년국내동해안에서발생한대형산불등국내에서도산불피해지로부터발생되는산불피해목의목재로서의이용가능성에대한접근이시도되고있다. 국립산림과학원에서 1997 년부터 2006 년에걸쳐산불피해목에대한재질변화연구를통해과학적인자료를구축한바있다. 본연구에서는산불피해목의바이오에너지이용가능성을평가하기위해산불피해정도에따른시료를제조하고, 발열량, 회분및주요원소 (C, H, N 등 ) 에대한에너지관련특성을측정하고결과를비교하였다. 2. 재료및방법 (Methodology) 2.1 공시재료직경 28cm, 길이 1.2m의소나무 (Pinus densiflora) 를이용하여 30cm 길이로 3등분하여공시재료를제조하였다. 시료의산불피해를묘사하기위해피해도에따라심, 중, 경으로구분하였다. 심은수피를포함하여변재 1cm 내부까지연소, 중은수피전체연소, 경은수피두께의 1/2연소를가정하여시료를제조하였다. Table 1. Sample classification. Sample Diameter Thickness of bark Damage Damage range No. (cm) (cm) 1 Seriously Bark and 1cm-thick of sapwood Moderately Bark Slightly Half thickness of bark 산불피해에따른시료제조산불피해는 Table 1에서제시한방법분류에따라국립산림과학원에서보유하고있는착화성시험기를이용하여처리하였다. 간단하게설명하면, 900 의복사열을시료표면에처리하여피해도에따라연소정도를가감하여수행하였다

410 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Figure 2. Prepared sample. 2.3 결과분석준비된시료는국립산림과학원고시제2009-2호 ( 목재펠릿품질규격 ) 에서제시하고있는시험방법에따라함수율, 회분, 발열량을분석하였으며, 국립산림과학원보유원소분석기 (EA1112A, Thermo, USA) 를이용하여원소분석을수행하였다. 3. 결과및고찰 (Results and Discussion) 3.1 연소정도에따른시료의표면변화연소정도에따라표면에수피에생성된정도를구분하는것은쉽지않은것으로판단되었다 (Figure 3. Lower line). 이는산불에의한수목에피해가발생해도목부의대부분에는영향을미치지않기때문으로판단된다

411 June 7-8, 2011 Figure 3. Samples from heat radiation treatment. (from left; seriously, moderately, slightly. Upper line: cross, Lower line: longitudinal) 3.2 함수율, 발열량, 회분분석결과산불피해처리를통해제조된시료의중간부위에서 3cm 두께의함수율은감소하였으며, 처리정도에따라함수율에차이가발생하는것을확인할수있다. 실험조건에서산불피해처리를위한시료는길이가 30cm로비교적짧기때문에처리시간이지속됨에따라함수율제거효과가있었던것으로판단된다. 반면, 발열량이나회분에서는시료의처리조건에따른차이를확인할수없었다. 발열량은매우우수하여목재펠릿으로서의이용가능성이높은것으로사료된다. Table 2. The comparison of moisture content, calorific value and ash content. Moisture Calorific value Sample No. (%) (kcal/kg) Ash (%) , , , 원소분석결과 Table 3. 에서제시된원소분석결과에서는에너지이용을위한주요원소 함량에차이가없는것으로사료된다

412 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Table 3. The results of C, H, N elemetal analysis Sample No. C(%) H(%) N(%) 참고문헌 (References) 산불피해지생태계변화조사, 국립산림과학원연구보고

413 June 7-8, 2011 Chemical properties of a Soap-based Class A Foam for Wildfire Takayoshi Kawahara 1 *, Shuichi Hatae 1, Isamu Akiba 2, Kazuya Uezu 2 1 R&D Department, Shabondama Soap Co., Ltd , Minamifutajima, Wakamatsu-ku, Kitakyushu City, Fukuoka Japan 2 The University of Kitakyushu 1-1 Hibikino, Wakamatsu-ku, Kitakyushu, Japan kawahara@shabon.com Abstract Our group has developed an environmentally friendly Class A foam for wildfire. We used eco-friendly, naturally derived soap and a highly biodegradable chelating agent to create samples that were evaluated for foaming, foam stability, and low-temperature stability. These chemical properties are dependent on the chelating agent and the component of fatty acid potassium (potassium oleate, potassium laurate), from which the soap is composed. We found that the foaming and foaming stability were equal or greater than the generally utilized EDTA, even when a highly biodegradable chelating agent was used. We also discovered an optimum fatty acid composition that demonstrates significant foaming and foaming stability, as well as a fatty acid composition that displays significant lowtemperature stability. Therefore, we demonstrated that selecting the optimum chelating agent and fatty acid composition makes it possible to develop a firefighting foam for wildfires that has low environmental impact. Introduction In 2007, our group (Kitakyushu City Fire and Disaster Management Bureau, University of Kitakyushu, etc.) successfully developed and began marketing a Class A foam for structural fire (Mizuki et al., 2007; Mizuki et al., 2010). That foam is characterized by 2 factors, a low toxicity to aquatic life and high biodegradability. The key to this special firefighting foam, which has a low impact on the environment, is soap. This is because soap instantly reacts to minerals (Ca 2+, Mg 2+, etc.) that exist in the natural environment to become harmless metallic soap. It loses its interfacial activity and immediately loses toxicity

414 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Another reason for soap s low environmental impact is the fact that it is broken down in a matter of days, and thus has a very high degree of biodegradability. Our next step was to begin developing a Class A foam for wildfire based on the Class A foam for structural firefighting we developed using soap. Agents used to extinguish wildfires significantly affect the environment since a large volume of extinguishing agent is dispersed over the natural environment. For that reason, it is desirable for extinguishing agents used in forest fires to have a low toxicity to aquatic creatures and a high degree of biodegradability. Consequently, soap is believed to be most suitable as the principle component for firefighting foam. Although soap has low toxicity and is highly biodegradable, the metallic soap produced through soap s specific reaction inhibits the production and retention of foam. As a result, the foaming and foam stability, which are important to firefighting foam, are reduced. Thus, preventing the production of metallic soap is key to stopping the reduced performance of firefighting foam. The most effective way to prevent the production of metallic soap is the use of a chelating agent. The commonly used chelating agent, EDTA comes to mind, but it has poor biodegradability and there are concerns over its environmental impact. Therefore, we searched for a chelating agent that is highly biodegradable and has a low environmental impact. In addition, soap and highly biodegradable chelating agents have an extremely strong mutual interaction and gel even at low concentration. It was presumed that low-temperature stability would deteriorate because of this specific gel-forming reaction. In light of this, our group researched the impact of the foaming properties and foam stability of firefighting foam that uses environmentally friendly, highly biodegradable chelating agents. Foaming properties and foam stability were assessed using expansion ratio and rate of reduction, which are testing standards for Class A firefighting foam in Japan. The pour point was used to assess lowtemperature stability when a highly biodegradable chelating agent is utilized. Since we also believed that the foaming properties, foam stability, and lowtemperature stability are impacted by the fatty acid potassium component that comprises the soap, we worked to develop a Class A foam for wildfire by optimizing the fatty acid composition. Experimental Materials A sample was created that mixed 40% soap components (potassium laurate, potassium oleate, diluent), 40% chelating agent, and 20% diluent. We used 5 types

415 June 7-8, 2011 of chelating agent: Ethylendiaminetetraacetic acid (EDTA), L-glutamate diacetate (GLDA), L-asparticaciddiacetic acid (ASDA), Methylglycineacetic acid (MGDA), and [S,S]-ethylediaminedisuccinic acid (EDDS). Foaming test The foaming were assessed using expansion ratio conforming to Japan s testing standards for Class A firefighting foam. 40mL of the firefighting foam was diluted in 4L of tap water set to 20 and poured into a fire extinguisher. Nitrogen gas was injected at 0.85MPa. After 10min., the foam mixture was placed in a foam-collection container. Its height was measured and the expansion ratio was found using the following equation: Expansion ratio = (foam height[m]) x (cross section of foam-collection container [m 2 ]) / (aqueous fire-extinguishing solution volume[m 3 ]) Foam stability test Foam stability was assessed using rate of reduction conforming to Japan s testing standards for Class A firefighting foam. The foam stability test measured the time it took 25% (1L) of the foam produced to return to an aqueous solution. Low-temperature stability test A 45ml sample was placed in a test tube to measure the pour point, and cooled using an ethanol bath maintained at -40 to -50. The test tube was removed each time the temperature dropped 2.5 and carefully tilted to confirm whether the sample surface was fluid. If fluidity was not seen even when the test tube was held horizontally for 5sec., the pour point was established at that temperature Results and Discussion ASDA, MGDA, GLDA, and EDDS are chelating agents believed to have a favorable degree of biodegradability. Both the expansion ratio and rate of reduction were shown to be high for ASDA, MGDA, EDTA, and GLDA, in that order. Therefore, even chelating agents with good biodegradability perform equally or better than the commonly used EDTA

416 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Next, we optimized the composition of potassium oleate and potassium laurate. For EDTA, various fatty acid compositions maintained a high expansion ratio and rate of reduction, but ASDA, MGDA, and GLDA had both a high expansion ratio and rate of reduction when potassium oleate was around 60%. Even among highly biodegradable chelating agents, both the expansion ratio and rate of reduction were high for ASDA and MGDA, and an optimum ratio between potassium laurate and potassium oleate was demonstrated. MGDA and GLDA, chelating agents with typically good biodegradability, were used for the pour point. The ratio of fatty acids was altered to research the optimum composition of potassium oleate and potassium laurate. For either chelating agent, when the ratio of potassium oleate was increased, the pour point was lowered. The pour point was at its lowest when the ratio of potassium oleate was 60%. Moreover, the pour point rose as the ratio of potassium oleate was increased, leading to the discovery that a 60:40 ratio between potassium oleate and potassium laurate had the highest fluidity. Based on the above, by selecting the chelating agent with the best biodegradability and the fatty acid composition, we were able to develop a Class A foam for wildfire that has high foaming and foam stability, plus a high degree of low-temperature stability. References Mizuki, H., Uezu, K., Kawano, T., Kadono, T., Kobayashi, M., Hatae, S., Oba, Y., Iwamoto, S., Mitsumune, S., Owari, M., Nagatomo, Y., Umeki, H. and Yamaga, K. (2007) Novel environmental friendly soap-based fire-fighting agent. Journal of Environmental Engineering and Management 17 (6): Mizuki, H., Toyomura, M., Uezu, K., Yasui, H., Kawano, T., Akiba, I., Kawahara, T., Hatae, S., Sakamoto, N., Akiyama, M., Mizota, C., Umeki, H. and Yama ga, K. (2010) Microbial Biological Degradation of Soap-Based Fire- Fighting Agent in Aquatic Condition. Journal of Environmental Engineering and Management 20 (2):

417 June 7-8, 2011 Effectiveness of Rehabilitation Treatments for Forest Restoration in Post-wildfire Area: A Case on a Hillside slope in Eastern Coastal Forest Area, Republic of Korea 산불피해지산림복원을위한사면복구처리효과분석 - 동해안산불피해사면을대상으로 - Chang-woo Lee 1*, Jung Il Seo *, Ho-joong Youn 1, Kyongha Kim 1 1 Department of Restoration Forest, Korea Forest Research Institute, Seoul, , Korea. 2 Department of Geosciences, Oregon State University, 104 Wilkinson Hall, Corvallis, Oregon 97331, USA Abstract leecw@forest.go.kr In the spring of 2000, a large-scale wildfire burned the biggest mountain range in the eastern coastal area in Republic of Korea. Although various rehabilitation treatments were often applied to minimize surface erosion and restore vegetation cover after the wildfire, few studies have quantified their effectiveness. The aim of our study was to evaluate effectiveness of rehabilitation treatments, and discuss considerations on their applications in post-wildfire areas. We based vegetation coverage, rainfall intensity (regulating surface erosion) and sediment yield data from 7 rehabilitation treatments (direct seeding, hydroseeding, tree planting, terrace-sodding, stone masonry, vegetation sack and burned-log barrier) compared to untreated field data. We found that our post-wildfire area with the lack of a sprouting source (caused by high wildfire severity, eroded sandy soil and steep slope) needed to allow approximately four years or more for physical stability throughout natural recovery of native vegetation. In general, wildfires destroy plant community and create a subsurface water-repellent layer, and thus rainfall interception by vegetation and litter layer is reduced and soil infiltration is restricted, thereby increasing overland flow and sediment yield. Although vary, the restoration time period, however, was reduced by application of the rehabilitation treatments. Our findings suggest considerations of the vegetation conditions, slope and location of treatment sites and the potential

418 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 structure problem of treatments, and this should be important for forest restoration in post-wildfire areas. 요약 한국동해안지역은 2000 년봄대규모산불로인해많은산림이소실되었다. 이러한산불피해지역토양침식의최소화와복구를위한다양한사면처리가이루어지고있으나, 그효과에대해명확한구명이이루어지지않고있다. 본연구에서는산불피해지에다양한사면처리공법을적용하여그효과분석및고찰을실시하였다. 총 7 가지의사면처리공법 ( 파종, 종자뿜어붙이기, 일반조림, 선떼붙이기, 산돌쌓기, 녹색마대공, 불탄나무편책공 ) 으로처리한사면과대조구에대한토사유출량, 강우량, 강우강도, 식생피도등을분석하였다. 그결과높은피해강도, 토양침식, 급경사등악조건의사면은자연복원력을회복하는데약 4 년혹은그이상의시간이필요한것으로나타났다. 일반적으로산불은지표식물을파괴하고강우침투를억제함으로써지표수를증가시키고토양침식을증가시킨다. 하지만사면처리공법의적용기간동안토양침식은명확히줄어드는것으로나타났다. 또한산불피해지복구를위해서는지피식생의상태, 사면경사, 사면처리의물리적구조등이충분히고려되어야할것으로판단되었다. Introduction Worldwide, numerous studies have been conducted on rehabilitation treatments, such as grass seeding (Pinaya et al. 2000; Robichaud et al. 2006), log barriers (Raftoyannis and Spanos 2005; Wagenbrenner et al. 2006), wood chips (Buchanan et al. 2002; Kim et al. 2008), mulching (Dodson and Peterson 2010) and fertilizers (DeBano 1989). Based on the results in these studies, there is no doubt that these rehabilitation treatments in post-wildfire areas can be important in minimizing the risk of flooding and sedimentation. However in Republic of Korea, most of previous studies (e.g., Chun et al. 2003; Lee et al. 2004) have concentrated on initial (or short-term) changes of physical, chemical and ecological characteristics caused by wildfires compared to unburned areas, and few studies have quantified effectiveness of various rehabilitation treatments based on field investigations of vegetation recovery and sediment production in post-wildfire areas. The main aim of this study was to evaluate the effectiveness of rehabilitation treatments for surface erosion control and vegetation recovery after a severe wildfire in the eastern coast area, Republic of Korea. The specific

419 June 7-8, 2011 objectives were to: (i) evaluate effectiveness of various rehabilitation treatments on vegetation recovery and surface stability, and (ii) discuss considerations on their applications for forest restoration in post-wildfire areas. In this study, we applied 7 rehabilitation treatments, which are currently used in burned forest areas in Republic of Korea. Methodology This study was conducted on the hillside slope in Mount Geombong (681.6 m), which is located in the Nogock, Wondeock, Samcheock, Gangwondo, Republic of Korea (Figure 1). The hillside slope is underlain primarily by not only sedimentary rocks but also metamorphic rocks of Cretaceous age, and its surface is mainly covered by eroded sandy soil (Seo et al. 2010). Of all plots settled, the 7 plots, except for 1 untreated control plot, were established by rehabilitation treatments, which are commonly performed to restore post-wildfire area in Republic of Korea. These are (i) direct seeding, (ii) hydroseeding, (iii) tree planting, (iv) terrace-sodding, (v) stone masonry, (vi) vegetation sack and (vii) burned-log barrier. These rehabilitation treatments were conducted after cutting of burned trees within plot boundaries, while the burned trees in the untreated control plot were remained (Figure 2a). In this study, we investigated daily rainfall data collected at the Imwon Meteorological Observatory closet to study site, which is released on website ( of Samchoek Emergency Management Agency. From this dataset, we used all rainfall data during investigation period (i.e., from June to October for 5 years ( )) to confirm effects of rainfall intensity on variation in surface erosion in post-wildfire area. To estimate sediment yield caused by surface erosion in post-wildfire areas, we collected sediments that were trapped by catchment barrels at the bottom of 8 plots. Here to avoid overestimation of sediments which can be caused by constructions of rehabilitation treatments, we excluded the data collected at the first time in 2002 from the dataset. Finally, the sediment biomass data of all plots were expressed as a sediment yield per unit plot area (SY, g/m 2 ). To reveal the explanatory parameters that were most influential to surface erosion in this study, we used a generalized linear model (GLM) with a Gaussian error distribution and identity link function. The responsible variable was SYs in each elapsed-year of each rehabilitation treatment plot. The explanatory variable chosen to explain the SYs was the rainfall parameter expressed by the cumulated daily rainfall greater than or equal to x mm (DRc amm). Here, the a value was varied from 0 mm (annual rainfall) to 150 mm at 10-mm intervals (i.e., 0, 10, 20,, 130, 140 and 150 mm). Model selection was performed by the best-subset procedure based on

420 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Akaike s Information Criterion (AIC), and the regression model(s) with the AIC value ranging from the smallest value (x) to x+2 was considered the best-fit model for the measured variation in the data (Burnham and Anderson 2002). Two-factor repeated measures analysis of variance (RM ANOVA) was used to evaluate significant differences in SYs according to the rehabilitation treatments and the elapsed-years, with multiple sampling times in each plot as the repeated factor. Here based on the result in previous analysis, all SY data were expressed as a SY per unit rainfall (unit SY, g/m 2 /mm) after division by the DRc amm selected in each plot. When the effects of the rehabilitation treatments and the elapsed-years on unit SY were significantly different on the two-factor RM ANOVA, Tukey- HSD multiple comparison was performed. Here P < 0.05 was considered to indicate statistical significance for all tests. The normality of the distributions was tested using the Kolmogorov Smirnov test. All statistical analyses were performed using the statistical language R version ( Results and Discussion Model selection using AIC in the GLM revealed that rainfall intensities of the best-fit models explaining variations in SYs in all plots increased generally along the spectrum of elapsed-year but their increasing trends were slightly different according to rehabilitation treatments (Table 1). In the untreated control plot, DRc 10MM was the best predictor explaining SY in the 1st investigatedyear, and models of DRc 0MM and DRc 20MM were similarly influential; DRc 20MM was the best predictors explaining SY as well as DRc 10MM in the 2nd investigated-year; DRc 20MM was the best predictor explaining SY in the 3rd investigated-year; DRc 40MM was the best predictor explaining SY, followed by DRc 30MM in the 4th investigated-year; and DRc 40MM was the best predictor explaining SY in the 5th investigated-year and model of DRc 50MM was equally well supported. The best-fit models selected in the direct seeding plot and the tree planting plot had very similar trends with the untreated control plot, except for in the 3rd investigated-year that models of DRc 30MM and/or DRc 40MM were selected. In the plots of the terrace-sodding and the stone masonry, models that have relatively higher rainfall intensities compared to the untreated control plot were selected in the 1st, 2nd and 3rd investigated-year but similar models with those in the untreated control plot were selected in the 4th and 5th investigated-year. However, the best fit models explaining SYs in the plots of the hydroseeding, the vegetation sack and the burned-log barrier generally showed higher rainfall intensities than those in the untreated control plot all over the investigated-year. Especially, in the 3rd investigated-year, the models of DRc 50MM and/or DRc 60MM with high

421 June 7-8, 2011 rainfall intensities were already selected as influential predictors explaining SY in these plots. The result of two-factor RM ANOVA showed that annual mean values in unit SYs in all plots decreased generally along the spectrum of elapsedyear; the unit SYs are relatively larger in order of the 1st, 2nd, 3rd, 4th and 5th investigated-years, although this trend has partially non-significant differences (Table 2). The unit SYs, however, were quite differed as a function of rehabilitation treatment. Especially in the 1st investigated-year, whereas the unit SYs in the untreated control plot and the direct seeding plot were greater than or similar with 1.0 g/m 2 /mm, the units SYs in the plots of the hydroseeding, the tree planting, the vegetation sack and the burned-log barrier were less than 0.5 g/m 2 /mm. These differences according to the rehabilitation treatments were reduced with the passing of year, and the unit SYs in all plots, except for the untreated control plot (approximately 0.11 g/m 2 /mm), were less than 0.1 g/m 2 /mm in the 3rd investigated-year. The unit SYs in all plots were further reduced and finally reached below 0.03 g/m 2 /mm in the 5th investigated-year, although not significantly different from the 3rd investigated-year to the 5th investigated-year. Table 1. The models selected in GLM for rainfall intensity controlling surface erosion according to rehabilitation treatment and elapsed-year. Structure of the model selected in GLM 1st year 2nd year 3rd year 4th year 5th year Model ΔAIC Model ΔAIC Model ΔAIC Model ΔAIC Model ΔAIC ntreated control ~ DRc 10MM ~ DRc 0MM ~ DRc 20MM ~ DRc 20MM ~ DRc 10MM 1.1 ~ DRc 20MM ~ DRc 40MM ~ DRc 30MM 1.3 ~ DRc 40MM ~ DRc 50MM irect seeding ~ DRc 10MM ~ DRc 20MM ~ DRc 0MM ~ DRc 10MM ~ DRc 20MM 0.4 ~ DRc 30MM ~ DRc 40MM ~ DRc 40MM ~ DRc 50MM ydroseeding ~ DRc 30MM ~ DRc 30MM ~ DRc 50MM ~ DRc 40MM 1.2 ~ DRc 50MM ~ DRc 60MM ~ DRc 70MM ~ DRc 60MM ~ DRc 70MM ree planting ~ DRc 20MM ~ DRc 10MM 0.3 ~ DRc 20MM ~ DRc 40MM ~ DRc 30MM 0.5 ~ DRc 40MM ~ DRc 40MM ~ DRc 50MM errace-sodding ~ DRc 20MM ~ DRc 30MM 1.6 ~ DRc 30MM ~ DRc 30MM ~ DRc 40MM ~ DRc 40MM ~ DRc 50MM tone masonry ~ DRc 20MM ~ DRc 30MM ~ DRc 30MM ~ DRc 40MM 0.9 ~ DRc 40MM ~ DRc 50MM 1.2 ~ DRc 40MM ~ DRc 50MM ~ DRc 30MM 0.9 egetation sack ~ DRc 20MM ~ DRc 30MM 0.6 ~ DRc 40MM ~ DRc 50MM ~ DRc 60MM 1.4 ~ DRc 50MM ~ DRc 60MM ~ DRc 70MM ~ DRc 60MM ~ DRc 70MM urned-log barrier ~ DRc 30MM ~ DRc 40MM ~ DRc 50MM 1.8 ~ DRc 60MM ~ DRc 60MM ~ DRc 70MM ~ DRc 60MM ~ DRc 70MM Note: The responsible variable (y) is the sediment yield (SY, g/m2) and the explanatory variables are the rainfall parameters expressed by the cumulated daily rainfall greater than or equal to x mm (DRc amm, mm). The regression model(s) with the AIC value ranging from the smallest value (x) to x+2 was considered the best fit model (Burnham and Anderson 2002)

422 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Table 2. Differences in sediment yield per unit rainfall according to rehabilitation treatment and elapsed-year. Sediment yield per unit rainfall ( 10 1 ) (Unit SY, g/m 2 /mm) 1st year 2nd year 3rd year 4th year 5th year Untreated control ± 4.67 a 5.36 ± 1.04 abcd 1.06 ± 0.15 cd 1.12 ± 0.39 cd 0.27 ± 0.17 cd Direct seeding 9.71 ± 2.87 ab 3.06 ± 0.79 bcd 0.80 ± 0.15 cd 0.83 ± 0.28 cd 0.25 ± 0.15 cd Hydroseeding 3.95 ± 1.23 bcd 0.69 ± 0.15 cd 0.42 ± 0.11 cd 0.11 ± 0.04 cd 0.09 ± 0.05 d Tree planting 4.90 ± 1.62 bcd 2.34 ± 0.77 bcd 0.50 ± 0.12 cd 0.92 ± 0.44 cd 0.10 ± 0.05 d Terrace-sodding 6.03 ± 1.78 abcd 5.73 ± 1.24 abcd 0.56 ± 0.12 cd 0.87 ± 0.23 cd 0.29 ± 0.15 cd Stone masonry 7.66 ± 3.26 abc 3.34 ± 0.85 bcd 0.81 ± 0.18 cd 0.62 ± 0.17 cd 0.25 ± 0.10 cd Vegetation sack 2.45 ± 0.67 cd 1.26 ± 0.37 cd 0.44 ± 0.13 cd 0.61 ± 0.21 cd 0.12 ± 0.06 d Burned-log barrier 0.78 ± 0.21 cd 0.77 ± 0.24 cd 0.30 ± 0.14 cd 0.17 ± 0.08 cd 0.05 ± 0.02 d An application of rehabilitation treatments should lead to increasing vegetation recovery and subsequent surface stability for forest restoration in postwildfire areas. This is particularly important to understand an effect of wildfires in regulating not only the existence and maintenance of stream-dwelling organisms but also the transfer and residence time of sediment for aquatic habitat formation. Moreover, our findings also provide important information to establish watershed management strategies when considering disaster prevention in densely populated areas. However, there are many knowledge gaps in understanding effectiveness of post-wildfire rehabilitation treatments at the watershed scale. For example, this study does not cover a range of various forests (species, seral stage, growth rate, artificial versus natural, harvested versus non-harvested) and/or geological conditions in the source watersheds. Hence, effectiveness of various rehabilitation treatments discussed in this study may not be available to various watersheds having different natural conditions. Furthermore, the chronological sequence from recruitment to fluvial export of sediment in post-wildfire areas was not fully examined. The sediment recruited into small streams in post-wildfire areas are fluvially exported to downstream rivers, repeating the cycle of transport and storage, and thus a time lag between initial recruitment (in small streams adjacent to post-wildfire areas) and fluvial export (in downstream rivers) should vary according to an application of rehabilitation treatments. The role of the rehabilitation treatments on the chronological sequence of sediment produced in post-wildfire areas is absolutely necessary to establish sediment budgets considering various geomorphic disturbances at the watershed scale. References Chun KW, Seo JI, Yeom KJ, Cha DS, Kim KN, Inoue S, Ezaki T, 2003, Variations of suspended solid and sediment yield in forest fire area: effects of coverage and afforestation. J Rainwater Catchment Syst 9:

423 June 7-8, 2011 DeBano LF, 1989, Effects of fire on chaparral soils in Arizona and California and postfire management implications. In: Berg NH (ed.) Fire and watershed management (Gen Tech Rep PSW-109), US Department of Agriculture Forest Service Pacific Southwest Forest and Range Experimental Station, Sacramento, CA: pp Dodson EK, Peterson DW, 2010, Mulching effects on vegetation recovery following high severity wildfire in north-central Washington State, USA. For Ecol Manag 260: DOI /j.foreco Lee CW, Lee CY, Kim JH, Youn HJ, Choi K, 2004, Characteristics of soil erosion in forest fire area at Kosung, Kangwondo. J Korean For Soc 93: (in Korean with English abstract) Pinaya I, Sato B, Arias M, Díaz-Fierros F, 2000, Revegetation of burnt areas: relative effectiveness of native and commercial seed mixtures. Land Degrad Dev 11: DOI /(SICI) X(200001/02)11:1<93::AID- as pos LDR380>3.0.CO;2-U Raftoyannis Y, Spanos I, 2005, Evaluation of log and branch barriers t-fire rehabilitation treatments in a Mediterranean pine forest in Greec e. Int J Wildlandd Fire 14: DOI /WF04031 Robichaud PR, Lillybridge TR, Wagenbrennerr JW, 2006, Effects of postfire seeding and fertilizing on hillslope erosion in north-central Seo JI, Chun KW, Kim SW, Kim MS, 2010, Rainfall pattern regulating surface Washington, USA. Catena 67: DOI /j.catena erosion and its effect on variation in sediment yield in post-wildfire area. J Korean For Soc 99: (in Korean with English abstract) Wagenbrenner JW, MacDonald LH, Rough D, 2006, Effectiveness of three p ost-fire rehabilitation treatments in the Colorado Front Range. Hydrol Process 20: DOI /hyp

424 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Introduction and growth of Pinus densiflora at the naturally regenerated forest stands after forest fire 산불피해지내천연갱신림에서소나무의이입과생장 Abstract Joo-Hoon Lim*, Chorong Ahn, Dong-Hun Ji, Yong-Ho Jeong Department of Forest Conservation, Korea Forest Research Institute, Seoul, , Republic of Korea Sixteen years after the onset of natural regeneration on the burned area at Goseong, Quercus spp. are dominant at the mountain slope and valley while the Pinus densiflora showed an tendency to inhabit the mountain ridge and rocky land. The purpose of this research is to investigate the influence of a forest fire by examining the number of Pinus densiflora introduced and the growth of it according to the various intensity and frequency of fire. Thus, this research surveyed the number of the introduced pines and annual growth. The research result revealed that most Pinus densiflora were established between 1995 and 1999 on the area which was burned once in On the other hand, most Pinus densiflora were introcuced between 2005 and 2009 on the area which was bured twice in 1996 and There has been more differentiation in tree height at onetime damaged area than two-time damaged area. The annual growth of Pinus densiflora increased sharply between 3 to 8 years after establishment, and stabilized from 9 to 11 years after establishment. 요약 강원도고성산불피해지를자연복원시켜 16 년이경과한현재사면부와계곡부는참나무류가우점하는반면능선부및암석지에서는소나무가서식하는양상을보였다. 본연구에서는산불의강도와횟수가다른지역의소나무이입과생장을조사하여산불의영향을알아보고자하였다. 각조사지에서소나무의이입된개체수, 수령에따른수고, 연년수고생장량을조사하였다. 그결과 1 회피해지에서는 년에, 2 회피해지에서는 년에가장많은개체가유입되었으며 1 회피해지는 2 회피해지에비해수고의분화가많

425 June 7-8, 2011 이일어났다. 소나무의연년수고생장량은발생 3-8 년에는급격히증가하며 8-11 년에는유지되는경향을보였다. 향후숲의발달에관한지속적인연구가필요할것으로생각된다. 서론 (Introduction) 강원도고성은 1996 년과 2000 년산불피해를받으면서사면부와계곡부에서우점수종인소나무를맹아생장력이강한참나무류가대체하는양상을보이고있다. 반면, 능선부와암석지처럼토양이척박하여활엽수의생장이어려운곳에서는소나무실생묘가도입되고있다 ( 임과안 2011). 산불은식생에물리적피해를입힐뿐아니라주변환경을변화시켜식생의개체군이입과생장에영향을미칠것으로사료된다. 본연구는산불피해지내천연갱신림에서산불이소나무이입과생장에미치는영향을알아보기위해수행되었다. 재료및방법 (Methodology) 조사지역은강원도고성군죽왕면산림청시험림내자연복원지의능선부 5 곳이다. 각조사지에능선을따라 50m( 조사구 5 는 40m) 를설정하고이를중심으로양쪽으로 10m 씩을조사구로설정하였다. 조사구 1, 2, 3, 4(2 회피해지 ) 는 1996 년과 2000 년산불로전소된지역이며조사구 5(1 회피해지 ) 는 1996 년지표화피해를입은지역이다. 조사구내의소나무위치, 수령, 마디길이, 결실여부등을조사하였다. 또한이를바탕으로각조사지에서 1995 년부터 2009 년까지 15 년간이입된소나무개체수를 5 년단위로분석하였으며소나무수령에따른수고를능선별로비교하였다. 마지막으로다섯곳의조사구에서발견된모든소나무개체의연차별연년수고생장량을조사하였다. Fig. 1. 조사구의위성영상사진

426 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 결과및고찰 (Results and Discussion) 각조사구의 1996 년부터 2009 년사이에이입된소나무개체수를 5 년단위로분석한결과조사구 5 는 1996 년산불을전후로개체이입이가장많으며이후감소하였다. 반면조사구 1, 3, 4 는 2005 년에서 2009 년사이에이입된개체가가장많았다. 이와달리조사구 2 는 년사이에개체이입이없는데이것은조사구내의참나무와의경쟁으로인한것으로사료된다. 수령별수고가가장큰곳은조사구 4 였으며조사구 2 는수고생장이저하된양상을보였다. 조사구 4 의능선부주변은참나무류밀도가높지않은반면조사구 2 의능선부주변은높은참나무류밀도를보여경쟁으로인한수고생장의차이인것으로보인다. 조사구 5 는다른조사구와는달리수령이많은개체의비율이높고수령에따른수고의분화가많이일어난것으로나타났다. 소나무의연차별연년수고생장량을조사한결과발생 1-3 년까지는정체되어있으나 3 년부터 8 년까지는연간수고생장량이 8.6cm 에서 22.2cm 로매년가파르게증가하였다. 이후 8 년부터 11 년까지는거의변화가없었으나 12 년부터는대폭증감하는양상을보인다. 12 년이후의연년수고생장량은조사구 5 의특징을반영하는것으로사료된다. 수령이 년사이인개체는총 62 개체로그중조사구 1-4 에해당되는것은 13% 에해당하는 8 개체뿐이기때문이다. 국내에서산불로인해숲이전소된지역인강원도는소나무림이전소된이후참나무림으로갱신되고있으며소나무는참나무가서식하기에어려운척박한능선부및암석지를중심으로분포하고있다. 향후숲의발달은참나무와소나무의경쟁에의해결정될것으로사료되나이와관련된연구는거의진행되지않고있는실정이다. 척박한토양환경과건조한기후의강원도 ( 임 2000) 에서참나무가수세를유지할지, 소나무가다시우점식생이될지지속적인연구가필요할것으로판단된다

427 June 7-8, 2011 Fig. 2. 각조사구의 5 년간유입개체수 Fig. 3. 각조사구의소나무수령에따른수고 Fig. 4. 소나무의연차별연년수고생장량 인용문헌 (References) 산불후자연복원지소나무침입에미치는영향인자분석 한국환경복원기술학회춘계학술대회집 동해안산불과기상 임주훈. 한국농림기상학회지. 2(2):

428 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Policy of Forest Fire Prevention and Analysis of Forest Fire during Chosun Dynasty Abstract 조선시대산불방지정책및산불현황 Donghyun Kim*, Youngho Kang Korea Forest Research Institute, Seoul, , Korea This study surveyed historical records on outbreak and features of forest fires during Chosun Dynasty s 518 years in being and analyzed the Annals of the Chosun dynasty; The Diaries of the Royal Secretariat, archives from Records of the Border Defense Council, The Compilation of Ministry Proposals, Posthumous Records of King Cholchong etc. Forest fires were most prevalent and extreme during King Hyunjong(14 cases) and King Sunjo(13 cases) due to strong wind, and the biggest-ever forest fire broke out in the fourth year of King Soonjo(1804) in the east coast of Korea in Kangwon province. The fire had resulted in 61 fatalities and 2,600 destroyed houses. Forest fire in the east coast of Korea, Kangwon province, in the 13 th year of King Hyeonjong(1672) is recorded to have caused the highest number of deaths, 65people. The most frequent cause of forest fires during Chosun Dynasty was unidentified(42 cases), followed by accidental fire(10 cases), arson or lightning(3 cases), fire during hunting(2 cases), play with fire by children, destruction of patty fields and dry fields by fire and house fire (1 case respectively). By region, 56% of forest fires erupted in the east coast(39 cases) and this was followed by the west sea(9 cases), Seoul and central region(8 cases) and the southern part of Korea(7 cases). By season, spring was found to be most vulnerable to forest fire as it accounted for 73% of the total amounting to 46 cases. Behind were summer(11 cases), winter(6 cases) and autumn(0 case). Specifically, most forest fire broke out on April and May, which is the same as today. Archives and literature indicate that the person who involved in forest fire by accidental as well as arson had to be punished by banishment, expulsion from government office and public hanging. Also, officials in charge of the region that suffered forest fire were subject to reprimand. In conclusion, risk and gravity of forest fires were evident during the Chosun Dynasty as specified in historical archives and share many similarities with today s forest fires in terms of the duration and regional patterns

429 June 7-8, 2011 요약 본연구는역사기록서분석을통한조선왕조시대 518년동안의산불발생및특성, 그리고정책에관한기록을고찰하였다. 본연구에활용된연사기록서는조선오앙조실록, 승정원일기, 국역비변사등록, 각부청의서존안, 철종행장등으로조선시대왕조별로산불이기록된문헌을조사, 분석하였다, 역사기록문헌에서기록된산불내용은발생개요, 산불종류, 산불피해등에관한사항과함께송전의산불예방조치사항등이포함되어있다. 조선왕조역사기록서를분석한결과, 산불발생건수가많았던왕조는현종 (14건) 과순조 (13건) 로강풍으로인한산불피해가극심한것으로기록되었고최대산분피해는순조 4년 (1804) 에발생한강원도동해안산불로사망자 61명, 민가 2,600호가소실되었다. 또한최대인명피해가발생한산불은현종13년 (1672) 강원도동해안산불로 65명이사망하였다. 조선시대산불발생원인은원인미상 (42건) > 실화 (10건) > 방화, 낙뢰 (3건) > 수렵입화 (2건) > 어린이불장난, 논밭두렁소각, 가옥화재 ( 각 1건 ) 로나타났으며지역별산불발생은동해안지역이 39건 (56%) 으로가장많이발생한것으로나타났다. 계절별산불발생은봄철기간산불이 46건 (73%) 으로가장많은것으로나타났으며연중산불이가장많이발생한기간은 4 5월로현재의산불위험시기와유사하다. 산불관련자에대한처벌의경우, 방화자및실화자에대해서는유배, 관직박탈, 효시등의처벌기록과관리지역책임자에대한문책이기록되어있다. 결론적으로조선왕조시대의경우에도산불위험성과심각성에대해여러문헌에서기록되어있는것을알수있고현재우리나라산불발생기간및지역패턴과유사한것으로나타났다. 서론 (Introduction) 조선시대에도역대왕조별로산불발생상황이어떠하였는지를구명해보기위하여조선왕조실록, 승정원일기등참고문헌을통하여산불발생과관련된사항을추출한후분석하여당시의산불발생특성은어떠했으며, 산불종별, 계절별특성및산불발생위험지역구분, 강풍과산불발생과의관계등을분석하여조선시대의산불특성을구명해내고, 당시대에발생된산불이우리에게주는교훈을도출해보고자한다. 또한산불과관련된행정및처벌에관한사항을정리하였고본연구를통해조선왕조시대산불정책을토대로현우리나라산불정책방향을되짚고향후, 발전시켜나갈산불정책을마련하는데도움이되고자한다. 본연구에서사용된연도와날짜는현재의양력으로변경하여기술하였으며원문에대한번안문은음력으로정리하였다. 미기록된산

430 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 불발생일자는관련문서분석으로통해표기하였다. 재료및방법 (Methodology) 조선왕조시대의산불정책을분석, 정리하기위해서는공식적인역사기록서를토대로이루어져야한다. 본연구에서분석된역사기록서는조선왕조실록, 승정원일기, 대전회통, 고려사절요, 경국대전, 대전속록, 조선후기산림정책사, 국역비변사등록등에대해산불정책과관련된내용을분석, 정리하였다. 본연구에활용된역사기록서분석은지난 40 년간한국고전번역원에서구축한 DB 자료를토대로이루어졌다. 일부역사기록서및고전에대해서는번역작업이진행중이며번역되지않는기록연대에대해서는조선후기산림정책사 ( 배재수외 3, 2002), 조선사외사 ( 차성찬, 1947), 대전속록 ( 조선총독부중추원, 1935) 드의자료를토대로보완하였다. 본조사분석에서나온내용을바탕으로산불예방분야와구휼등민심안정분야로나누어정리하였다. 분석한조선시대역사기록서에서아직발견하지못한산불기록이남아있어향후이에대해지속적으로문헌검색및분석이진행될필요가있음을밝혀둔다. 결과및고찰 (Results and Discussion) 왕조별산불현황산불발생원인을살펴보면, 총 63건의기록에서산불발생원인이명확히기록되어있지않은건수가 42건으로가장많고이후실화에의한산불발생이 10건으로기록되어있다. 방화에의한산불과여름철낙뢰에의한산불이각각 3건으로기록되었고군대에서사냥을위해불을놓은사례가 2건기록되어있고기타어린이불장난, 논밭두렁소각, 가옥화재에의한산불이각각 1건씩기록되어있다. 조선시대산불발생기록에서산불이발생한일자를기록한내용은전체 63건의산불발생기록중에서 19건만이정확한산불발생일자가기록되어있다. 표 1은실제산불발생일자와함께산불발생이기록되지않은내용에대해기록된날짜를함께정리하여나타낸것이다. 따라서분석자료는실제산불발생일자와차이가있음을밝혀둔다. 표 1에서정리한계절별산불발생은봄철 (46건, 73%) > 여름철 (11건, 17.5) > 겨울철 (6건, 9.5%) > 가을철 (0건, 0%) 로봄철기간산불발생이가장많은것으로나타났다. 현재산림청에서운영하고있는산불조심기간인봄철산불조심기간 (2월 1일 5월 15일 ) 과가을철산불조심기간 (11월 1일 12월 15일 ) 에발생한산불발생은 51건으로전체 63건

431 June 7-8, 2011 의약 81% 를차지하고있다. 여기서여름철산불은낙뢰로인한산불발생의특이성으로인해기록이남겨져있는것으로판단된다. 또한 12 월과 1 월겨울철에도산불이발생한것으로기록되었다. Figure 1. Cause of Forest Fire during Chosun Dynasty Figure 2. Numbers of Forest Fires by Each Month 조선시대산불정책조선왕조시대산불방지정책은크게예방분야와민생안정에관한정책으로구분할수있었고산불예방분야에서는산불발생가능지역과산불보호대상지역으로구분하여예방정책을시행하였다. 산불발생가능지역으로는화전경작산림지역, 강무장등사냥터, 봉수대주변산림지역, 병해충구제를위한입화가능지역이며왕실묘주변, 사고 ( 史庫 ), 소나무육림을위한산림인금산또는봉산, 도성숲은산불보호대상지역에포함되어산불방지정책을시행하였다. 민생안정에관한정책에서는산불피해지역백성을위로하고산불피해지복구를위해어사를파견하여구휼토록정책을펼쳤다. 특히, 산불예방정책시행을위반하여산불을발생시킨죄에대해서는엄하게책임을물어귀양에서부터사형에이르는처벌을내렸다. 결론적으로조선왕조시대에도산불의심각성으로인해조정및지방관서등에서여러방지정책들이시행되었다는것을알수있다. 인용문헌 (References) 조선왕조실록, 국역비변사등록,

432 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 승정원일기, 철종행장, 윤선용, " 강원도강릉군오대산에발생한화재에사고를지킨승도 ( 僧徒 ) 와민인 ( 民人 ) 등에게상금을예산외지출해달라는청의서제 112호 ". 각부청의서존안제11책 68장 a면 (1899). 塩田正洪. 조선산림사료제1집, p. 705(1934). 국사편찬위원회, 조선왕조실록소개, sillok.history.go.kr/. 산림청, 2010년임업통계년보, 제40호, p. 494(2010). 계명대학교한국학연구원, 조선전기도성사산 ( 四山 ) 의관리에관한연구, 한국학논집제40집, pp 쪽 (2010). 차성찬 (1974). 조선사외사-봉화. 명성사 pp 배재수외3인 (2002). 조선후기산림정책사, 금조절목이문. pp 조선총독부중추원 (1935). 대전속록. 288p. 김종서, 신서원 (2004). 고려사절요 ( 상 ). 838p

433 June 7-8, 2011 The analysis for cognition of forest fire danager of structure in Wildland-Urban interface 산림인접지내건물의산불위험에대한인식분석 Si-Young Lee 1, Houng Sek Park 2*, Byung-Doo Lee 3, Kyo-Sang Koo 3 1 The Professional Graduate School of Disaster Prevention, Kangwon National University, Joongang-ro1 Samcheok-si, Gangwon-do, , Republic of Korea 2 Dongkuk University, Pil-Dong 3Ga Chung-ku, Seoul, Republic of Korea parkhs08@naver.com 3 Korea Forest Research Institute, Seoul, , Korea Abstract The facilities in wildland-urban interface was a main reason of forest fire ignition and a main subject of protection. It is very important to decide the method of administration and the order of priority for protection of life and property. In this study, we researched a survey about classification and priority of facilities in wildland-urban interface. In results, the recognition rate of high possibility facilities for forest fire occurrence was high, but, the recognition rate of facilities which need more intensive protection management was low. The formulation of facilities classification in wildland urban interface and promotion of this system were needed. The officer recognized rice paddy and levee, graveyard and orchard as high possibility facilities for forest fire occurrence. And they recognized cultural assets and housing as facilities which need more intensive protection management. 요약 산림인접시설은산불의발화원이자주요보호대상으로써, 이에대한관리와소방우선순위의결정은인명과재산의보호를위해매우중요하다고할수있다. 본연구에서는이에대한설문조사를실시하여인접시설의대상물과우선순위에대한조사를실시하였다. 설문조사결과산불발생취약대상에대한인지도는높으나, 산불보호대상에대한인지도는낮으므로, 이에대한분류체계수립과교육이필요하며, 산불관련공무원들은논밭두렁, 묘지, 과수

434 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 원을취약대상으로인지하며, 문화재와천연보호림을보호대상으로인식하고, 주요피해대상은문화재와주거시설이라고제시하였다 서론 (Introduction) 최근산불은선진국에서부터개발도상국까지최근인구증가와인구의도시집중에따른도시의확장으로인해, 기존도시지역지역에인접한외곽산림, 녹지를도시로편입하여, 개발하여왔다. 이러한녹지대는거주민의건강, 경관을위해쾌적한환경을제공하게되지만, 도시에서발생한화재가산불로전이되거나, 산불이도시의건물로쉽게옮기게하는매개체가되고있다. 최근 2009 년발생한호주산불은이지역에서의산불로인해극심한인명피해를발생시키기도하였다. 우리나라도과거동해안산불, 양양산불등대형산불사례에서민가및건물이소실되는피해를입은바있으며, 특히, 원자력발전소와공공시설이위협받기도하여, 이를위해특별한소방대책을수립하기도하였다. 따라서, 본연구에서는산불발생시인명과재산을효과적으로보호하고, 효율적인진화및방재대책을수립하기위한대책중하나로인접지에서의산불발생시우선적으로진화및보호해야할대상을파악하고, 산불이발생할수있는취약대상을지정함으로써, 산불감시의우선순위와산불보호를위한우선순위를지정하고자하였다. 이러한산불위험지도의제작의사전단계로써, 산림인접위험시설물관리기법의개발에앞서, 실제관련종사자의의식조사를통해산림인접위험시설물에대한의식수준을조사하였다. 재료및방법 (Methodology) 2.1 조사지역및방법 설문대상지역은산림청과거산불통계를활용하여경상북도를선정하였다. 산림청발간산불통계에의하면, 경상북도지방은지난 20 년간가장많은산불이발생하였다. 으며, 해당지역의경제사회적여건에따라, 울진군 ( 중소도시형 ), 봉화군 ( 영농형 ), 경주시 ( 도시형 ) 의세시, 군을선정하여 2010 년 4 월부터직접방문과면담을통해설문조사및전문가면담을실시하였다. 조사시전통적으로산불을담당해온지방자치단체의산림부서외에일반적인화재를진화하는소방공무원을동일한설문지로조사함으로써, 화재와관련된의식을비교할수있도록조사하였다

435 June 7-8, 2011 설문지구성은공무원의근무년수, 담당년수등의일반적인특성을파악할수있는문항과산림인접지역의취약대상과산불보호대상에의식에대한내용으로하였으며, 설문의측정방법은변수의특성을구명하기위해명목척도방법과 5 점리커트척도방법을이용한간격척도방법을이용하였다. 소방공무원들에대한설문조사는산림담당업무를담당하는공무원과의의식차이의분석을위해동일한문항으로구성하였다. 조사대상지역내군청, 면사무소, 국유림관리소등을직접방문하여조사의요지를설명한후설문지를배부하여, 현장에서직접개인적으로설문지에기입하는자기기입방식 (Self- administered) 을시행하였다. 대상설문지는산림과직원 40 명, 국유림관리소직원 25 명, 소방공무원 80 명으로총 145 부를분석에이용하였다. 2.2 분석방법 산림담당공무원과소방공무원의근무년수와같은일반적인사항과산림내및인접지역의산불발생취약대상및산불보호대상관련질문대한답변들을통계패키지 Statistical Packagefor social Science(S.P.S.S) 을사용하여빈도분석을실시하였다. 결과및고찰 (Results and Discussion) 3.1 산림인접취약대상및산불보호대상시설에대한인지여부 산림담당공무원과소방공무원들의해당관리지역의산림내혹은산림인접지역에서의산불발생취약대상과산불보호대상지역에대한인지여부를조사한결과, 담당공무원의해당관리지역내의산불발생취약대상에대한인지도는높았으나 ( 평균인지도 3.74), 산림인접산불보호대상에대한인지도는낮은것으로조사되었다 ( 평균인지도 2.72). 이것은담당지역내의산불예방을위해취약대상에대해서는자체적인분류체계와계획을가지고있는데반해, 산불보호대상에대해서는상대적으로계획이미비하였기때문인것으로사료된다. 따라서관리지역내의산불보호대상에대한분류체계수립및교육을통해, 산불발생시보호대상중요도에따른체계적이고효율적인지역단위산불진화계획의수립이필요할것으로판단되었다

436 International Forest Fire Symposium on Commemorating the International Year of Forests 산림인접산불발생취약대상물과보호대상물의인식에대한설문조사결과 산림담당공무원들과소방공무원들이산불발생이가장잘일어날것이라생각한 2 가지시설물과산불로부터보호되어야할가치가있다고생각하는 2 가지시설물을선정하도록설문한조사결과는다음표 1, 2 와같다. 또한산불이발생하면가장많은피해가예상되는시설 2 가지를선정하도록한설문조사결과는다음표 3 과같다. 설문조사결과담당공무원들은산불발생취약대상으로논ㆍ밭두렁을지목하였으며, 그다음으로산림관련공무원들은묘지를지목하였으며, 소방공무원들은과수원을산불발생취약대상으로인식하고있었다. 이는최근논ㆍ밭두렁및과수원에서발생하는영농산불이다발함에따라이러한대상을주요산불발생취약대상으로인식하는것으로사료된다. 또한, 산불보호대상에대한인식조사결과공무원들은사찰등문화재가가장우선적인보호대상으로인식하고있었으며, 그다음으로천연보호림을보호대상으로인식하고있었다. 이는최근양양산불과남대문화재와같이문화재의화재로인한소실시사회적영향이크기때문에우선적인보호대상으로인식하는것으로보인다. 또한, 산불발생시피해가예상되는시설에대한인식조사결과산림관련공무원들은문화재시설이가장피해가예상된다고지적한것에비해, 소방공무원은주거시설이가장큰피해를입을것으로예상하였다. 따라서산림인접보호시설에대한중요도설정시문화재와주거시설에대해진화계획시우선권을부여해야할것으로판단된다. Table 1. The recognition for damaged facilities 대상 지방자치단체공무원국유림관리소소방서소방공무원 N( 명 ) 백분율 (%) N( 명 ) 백분율 (%) N( 명 ) 백분율 (%) 묘지 논ㆍ밭두렁 과수원 무속행위지 송전탑

437 June 7-8, 2011 Table 2. The recognition for protected facilities 지방자치단체소방서소방국유림관리소대상공무원공무원 N( 명 ) 백분율 (%) N( 명 ) 백분율 (%) N( 명 ) 백분율 (%) 사찰등문화재 화기물보관소 ( 주유소, 가스충전소, LPG 저장시설등 ) 천연보호림 자연휴양림 중계탑 Table 3. The damage expectation of facilities 대상 지방자치단체공무원국유림관리소소방서소방공무원 N( 명 ) 백분율 (%) N( 명 ) 백분율 (%) N( 명 ) 백분율 (%) 농림시설 주거시설 산업시설 군사시설 문화재시설 결론 (Conclusion) 조사결과지자체의산림담당공무원이전체근무년수나담당업무종사년수가국유림관리소와소방관보다다긴것으로나타났으나, 실제진화경험은소방공무원이높은것으로조사되었다. 또한, 담당구역내의산림내혹은산림인접한산불발생취약대상과산불시우선적으로보호해야할대상에대한인지도조사결과, 세기관모두 3.5 이상을기록하여, 어느정도인지하고있는것으로조사되었으며, 국유림관리소, 지자체산림과, 소방서순으로인지도가조사되었다. 하지만, 산불발생취약대상에대한인지도에비해산불보호대상에대한인지도는낮으므로이에대한표준화된분류체계수립과교육이필요할것으로사료된다. 또한, 산림내혹은산림인접한시설물중산불발생위험시설이라고생각되는시설물에대한설문조사결과세기관모두논 밭두렁이가장위험한취약대상물이라고인식하고있으며, 그다

438 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 음으로묘지를지적한반면, 소방서는과수원을취약한시설로인식하고있는것으로조사되었다. 또한, 산불발생시우선적으로보호해야할대상으로사찰및문화재로답변하였으며, 그다음으로천연보호림을우선적으로보호해야할대상이라고답변하였다. 이는양양산불과여타산불로인한문화재의파괴에대한사회적영향이라고사료된다. 산불발생시피해가예상되는산림내혹은산림인접지의시설에대한설문조사결과산림관련기관 ( 지자체산림과, 국유림관리소 ) 은문화재시설이주요피해를입을것이라고답변한반면, 소방공무원은주거시설이많은피해를입을것이라고답변하였다. 이러한이유는산림담당공무원은실제대형산불시의경험으로인한결과로사료되며, 소방공무원은일반화재로인한출동경험과대상지역가옥등의주거시설이실제산림과인접해있기때문인것으로사료된다. 따라서, 이와같은설문조사결과를바탕으로산림내혹은인접지역의취약시설과보호시설에대한개념정립의기초자료를제공할수있을것으로예상되며, 지역단위산불방재계획수립시진화및보호우선수위설정에중요한자료로활용될거이라사료된다. 향후보다많은지역의조사를통해세부적인우선순위설정과제반지식을구축하는것이필요하다하겠다. 인용문헌 (References) 강영옥. (2004). 서울시방재지도작성연구 ( 화재위험지도를중심으로 ). 서울시정개발원. 212pp. 김정욱, 김지영, 김용일, 유기윤. (2007). Where 2.0 서비스를이용한재해지도작성에관한연구. 한국공간정보시스템학회춘계학술대회논문집 : 소방방재청 재해지도작성기준등에관한지침. 소방방재청고시제 호 ( ) 산림청 (2010) 년산불통계연보. 산림청 (2006). 산불관리통합규정

439 June 7-8, 2011 The Fertilization Effect on the Growth and Foliage Nutrients of Containerized Quercus acutissima at the Forest Fire Damaged Area 동해안산불피해지에서시비처리에의한굴참나무용기묘생장과엽내양분변화 Jae Kyung Byun 1*, Byung Bae Park 2, Dong Hun Ji 1 1 Division of Forest Restoration, Korea Forest Research Institute, 57 hoegi-ro, Dongdaemun-gu, Seoul, , Korea. Bjk2754@forest.go.kr 2 Division of Forest Ecology, Korea Forest Research Institute, 57 hoegi-ro, Dongdaemun-gu, Seoul, , Korea. bbpark@forest.go.kr Abstract The purpose of this study was to investigate the fertilization effects on the growth of containerized Quercus acutissima at the forest fire damaged land. Seedlings aged 1-0 were planted in April 2004 at the forest fired area close to eastsea of Korea. We tested two fertilization treatments: one was different NPK combinations such as control, 3:4:1, 6:4:1, 3:8:1 and 6:8:1 and the other was the level of applications of slow released fertilizer such as no application, 30g, 60g, and 90g/plant. The order of height at NPK combinations was 6:4:1 > 3:8:1 > 3:4:1> 6:8:1> control. As fertilization increased, the height significantly increased. Based on the results, we can recommend 90g NPK = 6:4:1 combination as the best fertilization method in this forest fire damaged area. There were no significant differences in leaf nutrient concentrations and chlorophyll contents. This means fertilization effect occurred in the experiments because trees grew with similar tissue nutrient concentrations. 요약 이연구의목적은동해안산불지역에서비료배합비율및시비량이내화수종인굴참나무용기묘의초기활착및생장에미치는영향을구명하는것이다. 1 년생 (1-0) 용기묘를 2004 년 4 월에동해안산불지에식재하고, 비료배합비율시험을위해무시비구, 3:4:1(N:P:K), 6:4:1, 3:8:1 및 6:8:1 처리를수행하였

440 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 으며, 시비량시험은 1 주당 UF 완효성복합비료 30g 시비구를표준구로 60g 구 (2 배량구 ), 90g 구 (3 배량구 ) 를처리하였다. 굴참나무수고생장량은 6:4:1 처리구 > 3:8:1 처리구 > 3:4:1 처리구 > 6:8:1 처리구 > 무시비구순이었다. 굴참나무의수고생장은시비량이증가할수록뚜렷이높아져각처리간유의적생장차이가있었다. 따라서산불피해지에굴참나무용기묘의초기생장증대를위해서는 N:P:K=6:4:1 비율로완효성 3 배량 (90g) 시비를적용하는것이바람직하다. 시비조합및시비량에따른엽내양분농도및엽록소함량은처리간에뚜렷한차이는없었다. 이것은특정양분의결핍혹은과다현상이없는것으로, 양분농도는일정수준을유지하면서시비량에따라생장이증가했다는것은시비에의한효과로볼수있다. Introduction 전국적으로크고작은산불로인하여매년많은면적의산림이소실되고있고, 또한산림이울창해지면서대형산불로이어지고있다 년 4 월에강원도동해, 삼척지역에서발생한동해안산불은 23,794ha 란엄청난면적의피해가있었다. 산불의확산은여러가지요인이있지만동해안지역과같은소나무단순림은대형산불로이어지는원인이되고있다. 따라서참나무류와같은활엽수종으로내화수림대를조성하여산불확산방지가필요하다. 산불은지상부 biomass 와산림내유기물층을제거하여직 간접적으로산림토양의물리화학적성질을변화시킨다. 또한산불피해지는토양침식, 토사유출및산사태발생의우려가상존해있고양분의용탈이심해건전한산림으로의조성이어려워조기복구방법과산림의생산성을지속적으로유지하기위한산림토양관리가이루어져야한다. 본연구는비료배합비율및시비량이내화수종인굴참나무용기묘의초기활착및생장에미치는영향을구명을위하여수행하였다. Methodology 2000 년도 4 월에발생한동해안산불피해지인강원도삼척시원덕읍임원리 163 임반다소반일대의 LTER site 의내화수림대조성지에서실시하였다 년 4 월에시험지를구획하고굴참나무용기묘 (1-0) 를식재하였으며, 시험지에대한토양단면조사를실시한후토양시료를채취하여물리화학성을분석하였다. 적정비료배합비율을구명하기위한시비수준별시험은선행 pot 시험에서처리효과가낮은 2:2:1( 질소 : 인산 : 칼륨 ) 시비구와 3:4:2 시비구를제외하

441 June 7-8, 2011 고무시비구, 3:4:1, 6:4:1, 3:8:1 및 6:8:1 시비구등 5 개처리를하였다. 질소비료는요소, 인산비료는용과린, 칼륨비료는염화가리를이용하여배합하였다. 시비량시험은 1 주당 UF 완효성복합비료 30g 시비구를표준구로 60g 구 (2 배량구 ), 90g 구 (3 배량구 ) 를두었다. 시비방법은등고선방향으로식재목에서약 10cm 떨어진곳에반원형시비를하였다. 시비처리전에수고와근원경에대한기초생장조사를실시하였으며, 2006 년까지매년 10 월에수고와근원경에대한연간생장을조사하였다. 또한처리별로잎을채취하여양분함량과엽록소함량을분석하였다. 처리별수고및근원경생장량에대한유의성검정은분산분석 (Anova) 을이용하였다. Results and Discussion 1. 수고및근원경생장 (1) 시비수준별시험수고는질소시비량이많은 6:4:1 처리구가가장높아 87.1cm 생장하여무시비구 36.2cm 에비하여 2.41 배생장하였다. 6:4:1 처리구는다른 4 개모든처리구와유의적생장차이가있었다 (Figure 1). 질소와인산시비량이 2 배많은 6:8:1 처리구는생장이높지않았는데, 6:4:1 처리구및 3:8:1 처리구보다유의적으로낮았다. 이들처리구는잎이약간황변하는것으로보아과다시비에의한피해가있는것으로생각된다. 굴참나무수고생장량은 6:4:1 처리구 > 3:8:1 처리구 > 3:4:1 처리구 > 6:8:1 처리구 > 무시비구순이었다. 근원경생장은수고생장과마찬가지로 6:4:1 처리구에서가장높았으며다른 4 개처리구모두와유의적생장차이가있었다. 또한 6:8:1 처리구는수고생장과마찬가지로 6:4:1 처리구와 3:4:1 처리구보다유의하게낮았다. 처리별근원경생장은 6:4:1 처리구 > 3:4:1 처리구 > 3:8:1 처리구 > 6:8:1 처리구 > 무시비구순이었다. 무시비구생장률을 100 으로볼때각각 164%, 152%, 131%, 124% 의생장률을보여생장량이가장높은 6:4:1 처리구는무시비구에비하여 1.64 배더생장하였다. (2) 시비량시험굴참나무의수고생장은시비량이증가할수록뚜렷이높아져각처리간유의적생장차이가있었다. 수고는완효성복비표준구, 2 배량구, 3 배량구에서각각 26.6, 71.8, 92.7cm 이었으며, 생장량이가장많은완효성복비 3 배량구는표준구에비하여무려 66.1cm 더높았다. 처리별수고생장률은표준구를 100 으로볼때 2 배량구및 3 배량구는각각 270%, 348% 가생장하여 3 배량구는표

442 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 준구에비하여 3.5배더생장하였다. 굴참나무의근원경역시시비량증가에따라생장량이뚜렷이높아지는경향을보여각처리간에유의적생장차이가있었다. 처리구별 ( 표준구, 2배량구, 3배량구 ) 생장량은각각 5.6, 11.5, 17.2mm이었으며, 생장량이가장많은완효성복비 3배량구는표준구에비하여 11.6mm 더자랐다. 굴참나무의근원경생장률은완효성복비표준구를 100으로볼때 2 배량구와 3배량구에서각각 205%, 307% 생장하였다. 이와같이삼척지역산불피해지에서굴참나무의시비효과는매우크게나타났다. 따라서산불피해지에굴참나무용기묘의초기생장증대를위해서는 N:P:K=6:4:1 비율로완효성 3배량 (90g) 시비가적용되어야할것이다. 2. 엽내양분및엽록소함량 (1) 시비수준별시험굴참나무묘목의엽내질소, 인, 칼륨농도는무시비구보다시비처리구에서높았다. 질소와인은 3:8:1 처리구에서가장높았으며칼륨과마그네슘농도는 6:8:1 처리구에서가장높았다. 엽록소함량은엽내양분농도와같이 3:8:1 처리구에서가장높았으나, 엽내양분농도및엽록소함량모두처리간에유의적인차이는없었다. (2) 시비량시험굴참나무묘목의엽내질소농도는완효성복비표준구, 2 배량구및 3 배량구에서각각 1.77%, 1.96% 및 2.13% 로시비량이많아질수록농도가높아완효성복비 3 배량구가가장높았다. 인농도는완효성복비표준구, 2 배량구및 3 배량구에서각각 0.31%, 0.35% 및 0.34%, 칼륨농도는 1.14%, 1.19% 및 1.25% 로시비량이많을수록점차높아지는경향이있었다. 또한칼슘농도는 1.11%, 1.14% 및 1.15% 이었고마그네슘농도는 0.18%, 0.28% 및 0.23% 로마그네슘 3 배량구를제외하고는시비량이많을수록농도가높아지는경향이있었다. 엽록소함량은완효성복비 2 배량구에서 3.89mg/g 로가장높았으며표준구는 3.26mg/g, 3 배량구는 2.31mg/g 으로나타났다. 시비량별엽내양분농도및엽록소함량은처리간에유의적인차이는없었다. 시비량이많아져도처리간에엽내양분의농도가비슷한것으로보아특정양분의결핍혹은과다현상은없는것으로보이며, 양분농도는일정수준을유지하면서시비량에따라생장이증가했다는것은시비에의한효과라고생각된다

443 June 7-8, 2011 Study on Regional Spatial Autocorrelation of Forest Fire Occurrence in Korea 우리나라산불발생의지역별공간자기상관성에관한연구 Moon-Il Kim 1, Woo-Kyun Lee 1*, Hanbin Kwak 1, Raesun Jung 1, Taejin Park 1, Sangchul Lee 1 1 Department of Environmental Science and Ecolgical Engineering, Korea Univeristy, Seoul , Korea *Corresponding author(leewk@korea.ac.kr) Abstract Forest fire in Korea has been controlled by local government, so that it is required to understand the characteristics of regional forest fire occurrences for the effective management. In this study, to analyze the patterns of regional forest fire occurrences, we divided South Korea into nine zones based on administrative boundaries and performed spatial statistical analysis using the location data of forest fire occurrences for The spatial distributions of forest fire were analyzed by the variogram, and the risk of forest fire was predicted by kriging analysis. As a result, forest fires in metropolitan areas showed strong spatial correlations, while it was hard to find spatial correlations of forest fires in local areas without big city as Gangwon-do, Chungcheongbuk-do and Jeju island. 요약 우리나라산불관리의주체는관할시도행정자치단체이므로, 산불을효율적으로관리하기위해서는지역별산불발생특성에대한이해가필요하다. 본연구에서는지역별산불발생패턴을분석하기위해서, 행정구역도를기반으로전국을 9 개로분할하고, 각권역에서 1991 년부터 2008 년까지발생한산불의위치 data 를사용해서공간통계적분석을실시하였다. 산불발생자료의공간분포는베리오그램을통해분석하였고, 도출된공간자기상관성을가지고크리깅기법을통해산불발생위험을예측하였다. 분석결과, 대도시를포함하는권역에서는산불이강한공간상관성을가지고있었지만, 강원도, 제주도, 충청북도등의대도시를포함하지않는지역에서는산불의공간상관성이낮은것으로나타났다

444 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Introduction Forest fire is one of the typical forest disasters that damages forests extensively in a short time(choi et al., 2006). Forest fire makes sustainability management impossible by causing variety of biological, economical and social damages such as loss of forest products, decrease in biodiversity and decrease in the number of tourists. Furthermore, according to a recent research, the danger of forest fire is going bigger due the decline in humidity and the number of days with precipitation that are preceded by the increase of average temperature which is caused by global warming(mccoy and Burn, 2005; Sung et al., 2010). Therefore, it is necessary to study more about ways to prevent forest fire. In this study, the entire of South Korea was classified into 9 areas using the administrative map and the pattern of forest fire occurrence was analyzed by each area. Through the variogram which is one of the Geostatistics, the spatial autocorrelation of forest fire was studied. Since this method could expect the variance of point by exploring the spatial distribution of variables and analyze influence of factors along with the distances, spatial pattern and relationship could be explained. Materials and Mehtods 3. Study area and Materials Forest fire occurrence history was collected by Korea Forest Service(KFA). This daily fire point contains information of occurred time and address. The address of fire occurrence was geo-coded to let data have spatial coordinate by using a land registry map from Korea Cadastral Survey Corporation. The point data was converted to density data with 5km quadrats for quantitativeness. The reason we fixed the size of quadrat to 5km is to minimize an error on converting point to quadrat count because the average of polygon area of land registry map which was used for geo-coding was within 25 square kilometer. 4. Spatial Autocorrelation Forest fire in order to analyze the characteristics of the spatial, spatial autocorrelation analysis was conducted for each region. Spatial autocorrelation is the correlation among values of a single variable strictly attributable to their relatively close locational positions on a two-dimensional(2-d) surface, introducing a deviation from the independent observations assumption of classical statistics(griffith, 2009). In this study, spatial autocorrelation were analyzed using semi-variogram. The semi-variogram is customarily used in the variogram and therefore, semi-variogram is used to replace variogram in this study. The variogram is a quantitative descriptive statistic that can be graphically represented in a manner which characterizes the spatial continuity(i.e. roughness) of a data set. The mathematical definition of the variogram is described below:

445 June 7-8, 2011 r h 1 2n z x z x h x = vector of spatial coordinates(with components x, y for our 2D example), z(x) = variable under consideration as a function of spatial location, h = lag vector representing separation between two spatial locations, and z(x+h) = lagged version of variable under consideration. The variogram us the plot of variance at each separation distance. When the variance no longer increases, the constant variance is called the sill, and the distance at which the variance approaches the variance is called range of regionalized variable. This range is a neighborhood within which all locations are related to one another(davis, 1986). The spatial structures can be quantified by this range. In theory the variogram value at the origin should be zero. If it is significantly different from zero for lags very close to zero, then this variogram value is referred to as the nugget. The nugget represents variability at distance smaller than the typical sample spacing, including measurement error. Variogram model is divided into various models by depending on the presence sill and nugget. In this study, constructed spatial data was analyzed by a spherical model. As a result of the analysis of spatial correlation through empirical variogram, the tendency of spatial correlation was reached to spherical model. In addition, compared with gaussian and exponential model, the error of empirical variogram was the smallest among them. Results and Discussion 1. Regional spatial autocorrelation According to the experimental results, the presence of metropolitan areas (Seoul, Busan, Daejeon, Daegu, Gwangju), the spatial correlation was evident (Table 1). In the case of Busan Ulsan Geongsangnam-do, one city is adjacent to another. Therefore the spatial autocorrelation was slightly wider than other regions. On the other hand, Gangwon, Jeju Island, Chungcheongbuk-do area aren t shown a clear spatial autocorrelation. When a lot of fires occur, Sill was relatively higher values. Especially in the case of Seoul Incheon Gyeonggi area was got high nugget value. The significance of these results, variations of between near local areas are so high. It implied that the variation of the fire occurrence was large at a close distance

446 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Table 1. Spatial autocorrelation by region. (a) Seoul, Incheon, Gyeonggi-do, (b) Chungcheongbuk-do, (c) Gangwon-do, (d) Daejeon, Chungcheongnam-do, (e) Daegu, Gyeongsangbuk-do, (f) Busna, Ulsan, Gyeongsangnam-do, (g) Jeollabuk-do, (h) Gwangju, Jeollanam-do, (i) Jeju Region Nugget Range(meter) Partial Sill (a) , (b) , (c) , (d) , (e) , (f) , (g) , (h) , (i) , Prediction of Regional Forest Fire Risk Finally, vulnerability map was same as the figure 4. Comparing with the actual forest fire data in figure 1, the results was accurately expected. This research had a advantage on forest fire management because maps, as a polygon data, could expect the risk on the unrisen areas in comparison to the raw data

447 June 7-8, 2011 Fig 1. Regional Forest Fire Risk Map. (a) Seoul, Incheon, Gyeonggi-do, (b) Chungcheongbuk-do, (c) Gangwon-do, (d) Daejeon, Chungcheongnam-do, (e) Daegu, Gyeongsangbuk-do, (f) Busna, Ulsan, Gyeongsangnam-do, (g) Jeollabuk-do, (h) Gwangju, Jeollanam-do, (i) Jeju Conclusion This research, each administrative district has shown difference of spatial correlation. Especially, the administrative district including major cities has strong spatial

448 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 group. However, it is hard to identify spatial correlation in Jeju and Kangwon etc. which distribute regular population. Therefore, forest fire prevention strategy should be planned by district with considering spatial correlation. This study predicted risk of forest fire through only contemplating the forest fire without regard to variables that effect to forest fire. However, this study has performed to predict forest fire, regardless of variables of forest fire and land cover. Therefore, further study should be supplemented by considering various variables and land cover. References Choi, S.P., Kim, D.H., Lee, S.K The Abstraction of Forest Fire Damage Area using Factor Analysis from the Satellite Image Data. Korea Spatial Information System Society. 14(1): Davis, J Statistics and Data Analysis in Geology(2nd ed.). John Wiley & Sons, 637pp. Griffith, D.A Spatial Autocorrelation. Elsevier Inc. McCoy, V.M., and Burn, C.R., 2005, Pontential Alteration by Climate Change of the Forest-Fire Regime in the Boreal Forest of Central Yukon Territory. 58( 3): Sung, M.K., Lim, G.H., Choi, E.H., Lee, Y.Y., Won, M.S., Koo, K.S Climate Change over Korea and Its Relation to the Forest Fire Occurrence. Atmosphere 20(1):

449 June 7-8, 2011 A Research on the Fire Extinguishment Resource Utilization of the Fire Department Specialized in Forest Fire 산불전문진화대의진화자원활용실태에대한설문조사연구 Jung-Hun Kim¹*, Jaey-oun Shin¹, Byungdoo Lee², Kyongha Kim² ¹Hoseo University, Asan, , Korea ² Korea Forest Research Institute, Seoul , Korea Abstract kjh0789@hoseo.edu On a national scale Forest Fire Extinguishment Ground Teams that are involved a forest fire consist Forest Fire Specialty Extinguishment Teams (city, county, borough) and Forest Fire Extinguishment Teams (national forest). We choose tree area to survey the experience of forest fire extinguishment, average time of arriving at the scene, operating and using equipment situationally (beginning, middle, end), main force equipment when go to put out a forest fire, where the fresh water area for water supply. When forest fire ll be reported workers assign to arrival the scene 40 min at the latest. Main force equipment when go to put out a forest fire is fire-rake followed by the extinguish car, packpump followed by the power pump. Main force equipment when go to put out a forest fire is fire rake followed by the extinguish car, pack-pump followed by the power pump. Operating and using equipment are constructing a firebreak(67%) and fire-rake(69%) at the beginning, putting out a forest fire(88%) and power pump(45%) at the middle, cleaning the remainder of fire(60%) and packpump(27%) at the end. Reading the ranking of equipment by regional groups, Gyeongju hold helicopters. 요약 전국적으로산불현장에투입되는산불전문진화대 ( 지상진화대 ) 은크게 2 팀으로산불전문예방진화대 ( 시, 군, 구 ), 산불진화대 ( 국유림 ) 팀으로구성되어있

450 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 으며, 3 개소지역을선정하여산불진화경험, 현장도착평균소요시간, 상황 ( 초기, 중기, 후기 ) 에따른작업및주사용장비, 산불진화시주력장비, 용수공급시담수지등에대한설문을실시하였다. 산불발생시늦어도 40 분이내 (96.1%) 현장도착할수있도록배치되어있었으며, 주력장비에대한설문조사는불갈퀴 > 진화차량, 등짐펌프 > 동력펌프순으로, 상황에따른작업및주사용장비로는초기에는방화선구축 (67%, 작업 ) 과불갈퀴 (69%), 중기에는산불진화 (88%, 작업 ) 와동력펌프 (45%), 후기에는잔불정리 (60%) 와등짐펌프 (27%) 가가장높은순위로나타났다. 지역별장비순위를보면헬기를보유한경주의경우산불진화시헬기 (67.8%) 에의존하는경향을보였다. 서론 (Introduction) 우리나라는전국토의 65% 가산으로둘러싸여있어주 5 일근무제가정착되면서산을찾는사람들의발걸음이국내어느산에서도끊이지않고있는실정이며, 2007 년부터 2009 년 12 월 31 일까지의산림청통계에따르면전국적으로 1,377 건의산불이발생하였으며피해면적이 1,837.88(ha), 피해액 4,705,809 ( 천원 ) 으로나타나산불관리의필요성이심각하다는것을보여주고있다. 이는연평균피해면적 3,635(ha) 으로우리나라산림면적의국토의 0.057% 의산림이훼손되고있었다. 특히경북지역의산불발생건수가 280 건으로가장높게나타났으며피해면적또한, (ha) 로가장크게나타났다. 산불은성능좋은진화자원과수준높은진화인력의다양성을요구하는역동적인재난사고이다. 진화자원은지상및공중진화자원으로구분되며, 지상진화자원은지상진화대원, 수작업도구, 차량등으로, 공중진화자원은항공기 ( 고정익, 헬기 ) 로구분할수있고, 자원을분류하는것은사고담당자가진화목적에알맞은적합한자원을운용하도록명령을내릴수있도록하기위해서다. 그러나최근우리나라산불현장에서보면민, 관군등의인력과장비가많아동원되어도동원된진화자원의효율적인배분과지휘가미흡하다는지적이많았다. 효율적인산불진화를위해서는여러종류진화자원의가용성과유용성을대해산불현장지상진화행동매뉴얼과각종진화자원들의최적배치와문제점에대한많은연구들이시행되고있다. 따라서본연구에서는진화자원중지상진화자원의실태조사및현황을알아보기위하여 3 개소지역을선정하여설문 & 현장기존기록조사를하여산불진화에경향및문제점을알아보고자한다

451 June 7-8, 2011 재료및방법 (Methodology) 전체산불발생의 21.96% 를차지하고있는경북지역 ( 경주, 봉화, 울진 ) 의산불예방전문진화대및산불진화대 ( 국유림 ) 를직접방문하여산불전문예방진화대의산불진화경험, 신고접수후현장도착평균시간, 상황 ( 초기, 중기, 후기 ) 에따른작업및주사용장비, 산불진화시주력장비, 용수공급시담수지등에대한설문조사를실시하였다. 설문대상 (Table 1) 은경주, 봉화, 울진 3 개소지역의산림공무원및산불예방전문진화대, 산불진화대 ( 국유림 ), 감시원, 의용소방대의인원수를표로나타낸것이다. 이중산불에관련하여전문적이고지속적인활동을하고있는산불예방전문진화대및산불진화대 ( 국유림 ) 에소속되어있는 175 명에대하여산불진화경험에대하여설문조사를하였다. Table 1. Forest fire brigade in three location Gyeongju Bonghwa Uljin total 산림공무원 산불예방전문진화대 산불진화대 ( 국유림 ) 감시원 의용소방대 ,920 결과및고찰 (Results and Discussion) 설문조사대상자의경우산불진화의경험이 1~3 회의경우 28%, 10 회이상의대원이 58% 로산불진화에전문성을가진대상자들로설문조사를습득하였으며, 산불발생시, 0~30 분이내 (69.5%), 0~40 분이내 (96.1%) 로늦어도 40 분이내에산불현장에도착할수있도록배치되어있었다. 현사용중인주력장비대한설문조사의결과불갈퀴 (27.1%), 등짐펌프 (26.7%), 진화차량 (25.1%), 동력펌프 (21.1%) 로비슷한성향을볼수있었으며, 용수공급에대한담수지의경우하천및냇가 (49%), 호수 (31%), 강 (14%) 순으로나타내었다. 또한산불진화작업을초, 중, 후기작업을단계적으로설문조사한결과 (Table 2.) 로초기작업은방어선구축, 중기작업은산불진화, 후기작업은잔불정리로나타내었으며, 현재사용중인장비중살수장비인동력펌프의경우주력장비에는포함이되나실사용되는경우가적은것으로나타났으며산불진화

452 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 시진화차량과헬기에의존하는경향을보이고있다. 헬기를보유하고있는경주시의경우헬기진화 (67.8%) 에상당부분을차지하였다. 따라서살수가가능한장비중산불진화에큰비중을차지하는장비들을보유시그장비에의존하여다른장비들의활용성이떨어질것으로예상되었다. Table 2. Operating and using equipment situationally(beginning, middle, end) The beginning The middle The end Works 방화선구축 (67%) 산불진화 (26%) 지대물제거 (5%) equipments 불갈퀴 (69%) 진화차량 (14%) 동력펌프 (11%) Works 산불진화 (88%) 방화선구축 (7%) 가지제거 (4%) equipments 동력펌프 (45%) 진화차량 (24%) 헬기 (12%) Works 잔불정리 (60%) 산불진화 (38%) 방화선구축 (2%) equipments 등짐펌프 (27%) 불갈퀴 (27%) 진화차량 (25%) 인용문헌 (References) Korea Forest Service. 2007~2009. Statistical Yearbook. Korea Forest Service. Kyu-Chool Choi, Soon-Man Youn, A Study on Improving System Plan for the Raising Efficiency of Forest Fire Extinguishing, KIFSE, Fall Conference of the Korean Society of Safety, pp (2008). Si-Young Lee, Construction plan of ground fire fighting conduct manual on forest fire occurrence site KIFSE Vol.21 No.3. pp.8-14(2007)

453 June 7-8, 2011 Differences in bird communities by postfire restoration 산불후복원방법에따른조류군집의차이 Eun-Jae Lee 1*, Seung Hun Son 2, Joo-Hoon Lim 3, Woo Shin Lee 1, Shin Jae Rhim 2 1 Department of Forest Sciences, Seoul National University, Seoul, , Korea 2 School of Bioresource and Bioscience, Chung-Ang University, Ansung, , Korea 3 Department of Forest Conservation, Korea Forest Research Institute, Seoul, , Republic of Korea lejae2@snu.ac.kr Abstract We examined differences in bird communities in relation to characteristics of habitat structure in a pine forest, Samcheok, South Korea. An unburned stand, a stand burned 7 years earlier and then naturally restored, and a stand where Japanese red pine Pinus densiflora seedlings were planted after the fire were used for the survey. Habitat structure was dramatically changed by postfire silvicultural practices. Number of stand trees, shrubs, seedlings, snags, and vegetation coverage were significantly different among study stands. We made 1,421 detections of 46 bird species during 23 separate line transect surveys per stand between February 2007 and December The mean number of observed bird species and individuals, bird species diversity index (H0), and Simpson s diversity index (Ds) were highest in the unburned stand and lowest in the pine seedling stand. There were more species and individuals of forest-dwelling birds in the unburned stand than both burned stands. 요약 강원도삼척산불피해지역에서복원방법에따른서식지구조의변화및이에따른조류군집의차이를파악하였다. 서식지구조의경우특히교목, 관목, 치수, 고사목의수및엽층별피도량에있어서산불및산불후복원방법에따라크게차이를보였다. 산불미피해지와산불피해후조림지, 자연복원지에서 2007 년 2 월부터 2008 년 12 월까지총 23 회에걸쳐서선조사법을통해졸

454 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 류조사를실시하였다. 그결과총 46 종이 1,421 회에걸쳐서관찰되었으며, 관찰된평균조류종수, 개체수, 종다양도지수및 Simpson 다양도지수모두산불미피해지에서높게나타났고, 조림지에서가장낮게나타났다. 특히산림성조류의경우에는산불피해지에비해산불미피해지에서높은관찰종수및개체수를보였다. Introduction Forest fire is one of the major disturbance agents in forest ecosystems and plays a major role in determining landscape patterns by creating large mosaics of various habitats. Fire affects vegetation, structure, and the wildlife community. Postfire practices, such as removal of woody debris, downed trees, and snags, inadvertently alter a wide variety of habitat structure and wildlife communities, along with those targeted by management activities (Hutto and Gallo 2006). Those forest management activities may result in significant habitat loss on reduction in richness and diversity of forest bird species, and shifts in community composition and relative abundances (Robinson and Robinson 1999). However, changes in wildlife communities in pine forest after a forest fire have been poorly documented and discussed, but concerns over these changes are increasing in Korea, partly as a result of the frequency of large-scale forest fires in South Korea over the last decade (Lee et al. 2008). Also, there have been very few studies of changes in bird communities in relation to changing vegetation structure by forest fire and postfire silvicultural practices in pine forest on the Korean peninsula. We hypothesized that bird species abundance and diversity and bird community composition would be affected by forest fire and postfire silvicultural practices. In this study, we compared bird communities in different postfire silvicultural management stands and the unburned (control) stand in relation to differences in vegetation structure within a pine forest. Here we discuss the implications for managing burned pine forests in South Korea in relation to the bird community

455 June 7-8, 2011 Methodology Our study area was located in the coniferous forest of Samchuck, Gangwondo Province, South Korea. The most dominant tree species is Japanese red pine Pinus densiflora. Mongolian oak Quercus mongolica and cork oak Q. variabilis also occur in the study area (Lee et al. 2008). We selected three types of stands within that pine forest: an unburned stand, a stand burned 7 years earlier and then naturally restored, and a stand in which Japanese red pine P. densiflora seedlings were planted. To assess the effect of forest fire and silvicultural practices on bird abundance and community composition, we determined species abundance and diversity using line transect surveys (Bibby et al. 2000). The surveys were conducted 23 times in each stand, 1 3 times every other month, between February 2007 and December 2008, between 0500 and 0900 h. Parallel transect lines spaced at 50-m intervals were established in each stand. All birds heard or seen were recorded, along with the estimated distance, and only birds estimated to be within the 50-m belt were used in the analysis. All observed birds landing or hovering but not those flying across the study area were recorded. Stand characteristics were measured at 100 randomly selected points within a 5-m diameter circle on the survey route in each stand. For each tree, shrub, and woody seedling within a circle, we recorded species and number. The volume of coarse woody debris (CWD) was recorded. We classified vertical layer into understory (0 1 m), midstory(1 2 m), sub-overstory (2 8 m), and overstory (8 m). Coverage was classified into the following four categories based on the percentage of cover in each vertical layer, following Lee et al. (2008): 0 (coverage percentage = 0%), 1 (1 33%), 2 (34 66%), and 3 (67 100%). Results and Discussion There were significant differences in forest structure among stands. Overstory vegetation coverage was highest in the unburned stand and lowest in the planted Japanese red pine seedling stand after the fire. Sub-overstory and midstory vegetation coverage were highest in the naturally restored stand and lowest in the pine seedling stand. However, understory vegetation coverage was highest in the pine seedling stand than the others. There were more snags, higher volume of downed CWD, and number of shrub stems in the naturally restored stand. Also, number of tree stems was highest in the unburned stand. Number of woody seedlings were highest in the pine seedling stand and lowest in the unburned stand. In the unburned stand, the overstory vegetation was more

456 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 developed than in both burned stands. Understory vegetation of the pine seedling stand was more developed than in the other stands. We made 1,421 detections of 46 bird species during 23 line transect surveys in each stand between February 2007and December There were 717 individuals of 33 species in the unburned stand, 466 individuals of 32 species in the naturally restored stand, and 235 individuals of 29 species in the pine seedling stand. Mean number of species, individuals, bird species diversity index, and Simpson s diversity index were highest in the unburned stand and lowest in the pine seedling stand. Forest structure was dramatically changed by fire and postfire silvicultural practices (Lee et al. 2008). Dense understory cover is related to lower tree density in pine forests (Moore and Deiter 1992), presumably due to increased light, water, and nutrient availability under sparser tree canopies. Although burning may reduce understory cover initially, over time, increases in that understory cover would be expected. The amount of woody debris is also influenced by the type of treatment, whether it is removed, piled, or scattered. However, postfire silvicultural practice is expected to result in immediate reductions in woody debris (Converse et al. 2006). Vegetation structure and composition are key factors determining habitat selection in birds, and successional changes in habitats through time result in corresponding changes in bird communities. Forestry activities may result in significant habitat loss on the breeding grounds of forest dependent birds, reductions in bird richness and diversity, and shifts in bird community composition and relative abundance(andrew et al. 2004). References Andrew PJ, Erica N, Dennis RV Effects of selection cutting on bird communities in contiguous eastern hardwood forests. J Wildl Manage 68: Bibby CJ, Burgess ND, Hill DA, Mustoe SH Bird census techniques, 2nd edn edn. Academic Press, London. Converse SJ, Block WM, White GC (2006) Small mammal population and habitat responses to forest thinning and prescribed fire. For Ecol Manag 228: Hutto RL, Gallo SM The effects of postfire salvage logging on cavitynesting birds. Condor 108: Lee EJ, Lee WS, Rhim SJ Characteristics of small rodent populations in post-fire silvicultural management stands within pine forest. For Ecol

457 June 7-8, 2011 Manag 255: Moore MM, Deiter DA Stand density index as a predictor of forage production in northern Arizona pine forests. J Range Manage 45: Robinson WD, Robinson SK (1999) Effects of selective logging on forest bird populations in a fragmented landscape. Conserv Biol 13:

458 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Development of U-ICT Forest Disaster Management System Based on LBS(Location Based Services) Abstract LBS 기반 U-ICT 산림재해관리시스템개발 Donghyun Kim*, Changwoo Lee Korea Forest Research Institute, Seoul, , Korea LBS (Location Based Service) based on GPS is utilized a variable parts of a traffic, a control of distribution, an insurance, a security and so on. This system is also very useful for disaster management to response a crisis situation like as forest fire and landslide. In order to guide the safe extinguishment duty of fire fighter and the safe of residents, and acquire its exact location information in case of a large scale of forest fire and rapidly landslide, it is very important to monitor the real time coordination data. Forest disaster as forest fire and landslide need to secure the safe for fire fighters and presidents nearby disaster area. In this study is a related for developing the Smartphone applications of forest disaster that consist functions of real-time monitoring, information transmission. 요약 GPS 기반의위치관제시스템은현재교통, 물류관제, 보험, 보안등다양한분야에서활용되고있다. 재난 재해관리를방재분야에서도다양한형태의위기상황에대응하기위해재난 재해종류및규모별로다양한정보자원을이용해야하고인적 물적자원을활용하여빠르게대응해야한다. 산불및산사태는진화대원의안전한작업유도와주변민간인의보호를위해재해현장정보를신속히파악하기위해실시간으로모니터링, 정보전송등의기능이필수적이다. 본연구에서는스마트폰및 GPS-CDMA 단말기를이용한산불, 산사태등의산림재해를사례로한 자연재해관제시스템 개발에대해제시하였다

459 June 7-8, 2011 서론 (Introduction) 재해 재난관리관제시스템은재해 재난에대하여예방, 대비, 대응등단계별활동을지원하기위한정보시스템을말한다. 여러형태의재해 재난에대하여다양한정보자원을이용해야하고여러임무를맡고있는인적자원과물적자원을활용하여빠르게대응해야한다. 기존의방재시스템은여러시스템으로개발되어개별적으로운용되어통합상황분석과즉각적인위험경보및대응에다소어려움이따른다. 본연구에서는재해 재난중산불현장상황관리시스템과산사태위험관리시스템개발에대한연구내용을발표하고자한다. 먼저우리나라산불은최근 10 년 ( ) 간매년 478 건이발생하여 1,161ha 의산림피해를입고있다. 여기서산불발생은대부분입산자실화 (43%), 논밭두렁소각 (17%), 담뱃불 (10%), 쓰레기소각 (9%), 성묘객실화 (6%), 기타방화의심등에의한산불 (15%) 등인위적원인에의해발생되고있어산불감시활동에있어 IT 를이용한적극적이고신속한대응체계가필요하다. 산림청에서는산불감시원배치와산불감시카메라운영을통해감시활동을벌이고있다. 전국에설치된산불무인감시카메라는약 500 여대가설치되어운영하고있으나음영지역이많고산림및산림인접지에서의소각행위에대한방지활동이어려워산불감시원의예방감시활동역할이더욱중요하다. 하지만일선시군에서는산불감시원의산불예방, 감시활동사항을모니터링하기가어려워적극적인산불감시를위한활동관리가어려운실정이다. 따라서산불감시원의활동을모니터링하여산불취약지및음영지역이발생하지않도록하기위한효율적인관리가필요하다. 한편산지토사재해의경우매년약 700ha 의산사태가발생하고있으나재해발생장소에대한파악이어려워기초자료의축적이어려운실정이다. 재해위치에대한환경입지인자등의정확한파악은향후재해예측및재해재발을위한사방구조물설치에중요하다. 하지만재해발생지로의접근이어렵고다양한정보취득을위한장비를가지고산지에서다양한조사를실시하기란쉬운일이아니다. 따라서 GPS, 수치지도등다양한지형, 임상정보를현지에서손쉽게취득하고 DB 를실시간으로전송 저장할수있는시스템구축이시급하다. 이에본연구에서는스마트폰및 GPS-CDMA 단말기를이용하여산불과산사태에 산림재해관제시스템 개발및위험관리시스템구축방향을제시한다

460 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 재료및방법 (Methodology) 스마트폰을이용한재해재난관제시스템 재해 재난관제시스템은방재업무를효율적으로수행하기위해구축하는것으로재난정보의수집, 예방활동그리고즉각적인대응을위한실시간정보공유, 명령전달을위한예방 - 대비 - 대응단계별활동지원최신정보관리시스템을말한다. 예방단계활동은 GIS, GPS 등을이용한예방활동방향과정책을도출하고, 인적자원, 물적자원등의효율적자원관리를통한실시간경보체계구축을통해대비시스템구축을지원한다. 대응단계활동은 ICT 기술을활용하여효율적인상황파악과대응시필요한의사결정을위해신속하고정확한정보전달체계를구축한다. 재해 재난관제시스템핵심기능의원활한수행을위해서첨단정보통신기술기반의전략적아키텍처가필요하다. 그림 1 은재해 재난관제시스템블록도를나타내며위치지시장치 (GPS-CDMA 단말기 ) 에서재난위치정보를 SMS 메시지로이동통신사서버로전송후전송된데이터를관제시스템에서이동통신사서버에접속하여재난정보를수신한후항공영상과 GIS 수치지도에정보를제공하는것을기반으로한다. 본연구에서는 LBS 기반으로재난재해위치정보및현장상황정보전송및공유효울화를위해재해위치관제시스템과연동하는스마트폰애플리케이션을개발하였다. (a) (b) (c) Figure 1. Block Diagram for Disaster Management System (a), Flowchart for Sending a Message of Disaster Information (b), Flowchart to support the Disaster Information (c)

461 June 7-8, 2011 결과및고찰 (Results and Discussion) 산불 산불현장진화대원위치관리및산불상황관리를위해개발한스마트폰용산불현장관리시스템애플리케이션을개발하였다. 애플리케이션의주요기능은산불발생위치전송, 사진촬영전송, 동영상촬영전송, 상황보고, 간이확산예측, 산불조사야장작성, 산불등급판정등으로구성되어있다. 이러한정보는 Web_ 항공영상기반산불위치관제시스템과연동하여산불을실시간관리, 운영할수있다. Figure 2 는스마트폰산불현장운영애플리케이션의메인화면및사진촬영을위한기능구현화면이다. 산사태발생또는토석류피해지에대해필요한정보 ( 발생위치, 임상상태, 피해상황, 계류개황, 기타지형및지질정보등 ) 를수치지도를통해현장에서손쉽게파악하고기록된자료가디지털자료로입력될수있는시스템을스마트폰에탑재할수있는시스템을개발하였다. 주요기능으로는 GPS 를연계하여정확한피해지위치파악, 수치지도를탑재함으로써다양한지형및지질정보를화면터치만으로입력하고주변환경에대한정보를파악할수있다. 또한이러한정보를실시간으로메인서버로전송할수있는시스템을구축함으로써재해실무담당자들이현장에서조사한내용을공유하고연구및행정에활용할수있다. Figure 2 는스마트폰산불현장운영애플리케이션의메인화면및사진촬영을위한기능구현화면과산지토사재해조사애플리케이션의메인화면이다. (a) (b) (c) Figure 2. The Main Scene of Forest Fire Management Tool in Smart Phone Application (a), The Function of Taking Picture included GPS Information (b), The Main Scene of Soil Sediment Disaster Survey Tool in Smart Phone Application (c)

462 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 인용문헌 (References) 산림청, 2010 산불통계연보. 손기천, 최영열, 박충식 (2008). 정보융합기반지능형재난방재시스템프레임워크 한국 IT 서비스학회 2008 추계학술대회. 김동현 (2010), 산불예방및초기대응을위한효율적산불감시운영관리시스템 국립산림과학원산림과학정보 216 호

463 June 7-8, 2011 Liming effects on soil condition of a forest fire burned stand in southern Korea 산불갱신임분의토양상태에미치는석회처리효과 Lee Seungwoo*, Ji Donghun, Lee Choonghwa, Jeong Jinhyun Division of Forest Restoration, Korea Forest Research Institute, Seoul, Korea soilloverlee@forest.go.kr Abstract After forest fire, a Machilus thunbergii stand around an industrial complex in the southern Korea was limed with coarse granular dolomite of 0.0, 2.0, 2.5, 3.0, 3.5, and 4.0 ton/ha. Soil chemical properties and respiration in rooting zone were monitored during 4 years. Soil ph, N, total P increased, whereas exchangeable Al 3+ content decreased with the increase of liming amount. C/N ratio of soil did not show significant difference among plots with different treatment amount of dolomite. Dolomitic liming increased decomposition rate of organic matter and the result increased organic carbon and total nitrogen. The accererated decomposition would be originated from the combined effect of the increased soil ph and higher microbial activity due to that. 요약 남부지역공단주변에서산불후의후박나무갱신림을대상으로석회고토의양을달리하여처리하고토양화학성과토양호흡량의변화를측정하였다. 석회고토처리량이증가함에따라토양산도, 전질소, 유효인산함량이같이증가한반면치환성알루미늄은반대의경향을보였다. 토양내탄질률은석회고토처리량에따라차이를보이지않았다. 석회고토처리가유기물의분해를촉진시켜토양중의유기물함량과전질소함량을같이증가시켜나타난결과로추정된다. 이러한유기물분해환경의개선은토양산도교정과미생물활성증가효과의복합적인작용에의한것으로판단된다

464 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Introduction Reforestation or rehabilitation after forest fire, especially under strong acidic load, needs a countermeasure such as broad-leaved trees planting and/or liming to alleviate the negative effects of soil degradation. This study was carried to examine the effects of dolomitic liming on the restoration of a Machilus thunbergii stand near an industrial complex through investigating soil properties and tree growth as recovery indicators. Methodology In 1998, fired forest of 33ha at Mt. Youngchui in Yeosu-city, located in southern Korea, was restored by Machilus thunbergii together with dolomitic liming and others were remained in grassland mainly with Miscanthus sinensis. For liming treatment (900m 2 /plot), coarse granule dolomite was used as a soil ameliorator in 6 doses (0, 2.0, 2.5, 3.0, 3.5, 4.0 dolomite ton/ha). We monitored the change of soil chemical properties and soil respiration in Results and Discussion Soil Chemical Properties & Soil Resperation Liming increased soil ph with increasing dolomite dose amount. For soil target ph (to 5.5), 3.23 CaMg(CO 3 ) 2 ton per ha was required, which was doubled than calculated amount. Liming enhanced the pools of available C and N without change of C/N ratio in mineral soil. It means that the dolomitic liming for the restored stand contributed effectively to nutrition recycling in soil. Also phosphorus availability in soil increased, but not potassium, on the other side decreased labile Al 3+, a plant-toxic material. Soil respiration rate increased with liming dose, which indicates root growth and increased soil microbial activity after the liming

465 June 7-8, 2011 Figure 1. The change of soil ph, total nitrogen and C/N ratio in top soils after liming application Figure 2. The change of Soil exch-al and soil resperation rate after liming application Restoration of forest having very acidic soil and continuing acid loading needs to combat soil acidification risk before and/or after replantation. In this study, dolomitic liming increased soil ph together with decomposition rate of organic matter resulting in increased organic carbon and total nitrogen. The rapid decomposition would be originated from the combined effect of the increased soil ph and higher microbial activity. Liming with dolomite, an alkaline products, showed several positive effects on restored Machilus thunbergii stand after forest fire. Firstly a rise in soil ph and reduction in concentration of exchangeable Al would provide soil ecological benefits. Consequently because this study site is closed to heavy acid deposit sources, the lime application would mitigate the possibilities of negative effects of long-term atmospheric acid deposition

466 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 References Clair, T., Hindar, A., Liming for the mitigation of acid rain effects in fresh waters: a revies of recent results. Environmental Reviews 13: Hindar, a., Wright, R.F., Nilsen, P., Laessen, T., Hogberget, R., Effects on stream water chemistry and forest vitality after whole-catchment application of dolomite to a forest ecosystem in southern Norway. Forest Ecology and Management 180: Huettl, R.F., Zoettl, H.W., Liming as a mitigatin tool in Germany s declining forests-reviewing results from former and recent trials. Forest Ecology and Management 61: Lundstrom, U.S., Bain, D.D., Talylor, A.F.S., van Hees, P.A.W., Effects of acidification and its mitigatin with lime and wood ach on forest soil processes: a review. Water Air Soil Pollutin Focus 3: SFA, Action plan to counteract soil acidification and to promote sustainable use of forestland. In, Meddelande Swedish Forest Aegncy, 40pp. Westing, O., Zetterberg, T., Recovery of acidified streams in forests treated by total catchment liming Water, Air, & Soil Pollution: Focus 7:

467 June 7-8, 2011 The flora of tree pathogens in forest-fired regions in Gangwon province in Korea 강원도산불피해지역의수목병원균상조사 Sanghyun Lee 1*, Seungkyu Lee 1, Sangtae Seo 1, Jongkyu Lee 2 1 Division of Insect Pests & Diseases, Korea Forest Research Institute, Seoul, , Korea 2 Tree Pathology and Mycology Lab, Kangwon National University, , Chuncheon Korea shlee4@forest.go.kr Abstract The flora of forest-fired stands in Gangwon province, Korea were surveyed. Powdery midew was annually occurred on black locust(robinia pseudo-acasia) and oak(quercus spp.) trees, and rust, Tubakia leaf spot, powdery mildew, and Endothia canker were often occurred on oak trees damaged by fire. Especially in 1 st LTER site, pine gall rust was observed on the alternate host, oak tree as well as pines. Pine gall rust would be a potential threat in pine forests with oak trees as a understory vegetation. In artificial plantations of Pinus thunbergii in Gosung, 32 kinds of disease including needle cast were reported on 20 tree species. Pine needle cast was severely occurred until 2003, but the trees were gradually recovered and showed good growth of trees since In Japanese white birch plantations, disease rate by brown leaf spot was annually reported about 30-50%. Severely damaged leaves were early defoliated, and tree vitality would be very low. Thus, it is strongly recommended that the disease should be managed especially in plantations. Maple tar spot, which is a biological indicator of air pollution, was also reported annually in LTER site located in Wonduk, Samcheok. The fruting bodies of Rhizina undulata, the pathogen of Rhizina root rot, were observed in pine stands after forest fire for 2 years, but these were not observed any more since This may due to the loss of vitality of the pathogen by dry conditions in spring season and dense distribution of understory vegetations including brake(pteridium aquilinum var. latiusculum), Artemisia princeps var. orientalis, Black locust. In LTER site in Samcheok, anthracnose of Korean weigela(weigela subessilis), leaf spot of Kudzu vine(pueraria lobata), and witches -broom of Rhus chinensis were severe, but others mimic

468 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 요약 강원도산불피해지의수목병원균상을조사한결과, 아까시나무와참나무류에매년흰가루병이심하게발생되고있었으며, 산불피해로수세가약해진참나무류에녹병, 튜바키아점무늬병, 흰가루병, 황색줄기마름병등다양한병해가발생되고있었다. 제 1 영구조사구에서는특히소나무혹병과중간기주인참나무류에잎녹병이발생하고있어서소나무혹병발생지에하층식생으로참나무류가분포하고있는임지에서는소나무혹병의피해확산이우려되었다. 고성조림지의수목병원균상을조사한결과, 해송조림지에서는해송잎떨림병등 20 개수종에 32 병종이관찰되었으며, 해송잎떨림병은 2003 년까지심하게발생되었으나 2004 년부터는점차회복되어병해를관찰할수없었으며, 생장도초기보다는회복된것으로관찰되었다. 또한자작나무조림지에서는갈색점무늬병의피해가매년 30~50% 정도조사되었으며, 심하게발병된자작나무의경우잎이조기낙엽되고수세가쇠약하게되므로조림지에서는이병이발생되지않도록주의해야할것이다. 삼척원덕읍에위치한 LTER site 에서는공기오염의생물학적지표로이용되는단풍나무타르점무늬병이매년관찰되었으며, 산불후소나무림에서많이발생되는리지나뿌리썩음병의자실체인파상땅해파리버섯이 2000 년 4 월에발생한산불후 8 월부터관찰되기시작하여 2 년동안계속관찰할수있었으나, 2002 년도부터봄철건조와고사리, 쑥, 아까시나무등하층식생이너무밀식되어병원균의활력이떨어져자실체가발생되지않는것으로판단되었다. 삼척 LTER site 에서는병꽃나무탄저병, 칡겹둥근무늬병, 붉나무빗자루병등의피해가심하였고, 다른병해의피해는경미한수준이었다. 서론 (Introduction) 산불로인한피해는재산상의문제뿐만아니라살아있는동식물과인간에게큰고통과피해를주며, 후손들에게물려주어야할아름다운숲과환경가치를잃게하여국가적으로도큰손해를끼친다 년고성산불이후 2000 년강원도고성군에서부터경상북도울진군에이르는동해안지역에발생된대형산불은산불이얼마나무섭고두려운존재인지를깨우쳐준큰교훈이되었으며, 산과더불어살아가는농민들의삶과지역경제까지위축시켜버렸다. 산불이발생한산림에서는일부생물에게유익한환경제공과병해충억제, 관목과낙엽층제거등의효과도있지만생태계의교란과수목에 2 차피해가발생되어병해충에대한내성이급격히떨어진다. 특히산불이발생된

469 June 7-8, 2011 지역에서는소나무, 해송, 낙엽송등침엽수에리지나뿌리썩음병피해가가장우려되며, 참나무류에는부후가발생되어재질손실을가져온다. 산불이발생된산림에서수목병해에대한연구는리지나뿌리썩음병에대해서만보고되었을뿐다른병해는보고된기록이거의없는실정이므로, 산불발생후나타나는수목병원균의경시적변화에대한연구가향후다른연구에기초적인자료로제공되었으면하는바램이다. 재료및방법 (Methodology) 산불피해지인고성지역의제 1, 2 영구조사구, 송지호주변의잣나무, 해송, 자작나무조림지와 2000 년도산불피해지인삼척 LTER site 에서수목병원균상을조사하였다. 조사방법은조사구내를순회하면서이상증상목의병징조사와함께병해발생정도를기록하고시료를채집하였다. 채집한시료는표징과병원체의특징등을현미경으로관찰한후참고문헌과비교하여병원균을동정하였다. 결과및고찰 (Results and Discussion) 고성제 1, 2 영구조사구의수목병원균상을조사한결과, 아까시나무와참나무류에흰가루병과병꽃나무탄저병이매년발생되는것으로조사되었으며, 특히참나무류에는녹병, 점무늬병, 튜바키아점무늬병, 흰가루병등다양한병해가발생되고있었다. 참나무류의튜바키아점무늬병은 2001 년도에발생되기시작하여 2003 년부터매년피해가심해지는것으로조사되었으며, 참나무류에는흰가루병과튜바키아점무늬병이가장피해를크게주는병으로조사되었다. 제 1 영구조사구에서는소나무혹병이많이발생되었으며, 혹병이발생되면바람이나폭설등에의해쉽게가지가부러지며, 이혹은매면비대해져서 30 cm이상의혹으로도자라기도한다. 혹병은봄과가을에소나무의줄기와가지에녹병포자기와녹포자기를만들며, 여름포자퇴와겨울포자퇴는참나무잎에형성한다. 참나무에는피해가그리심각하지않지만어린소나무가감염되면몇년내로고사된다. 제 1, 2 영구조사구에서병해피해로는녹병이 17 개수종에발생되었으며, 이중붉나무와조록싸리의녹병피해가심한것으로조사되었다. 해송, 자작나무, 잣나무조림지의수목병원균상을조사한결과, 조림지에서아까시나무흰가루병, 참나무류흰가루병과튜바키아점무늬병피해가전체적으로심하게발생되어있었으며, 해송조림지에서는 20 개수종 32 병종이관찰되

470 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 었다. 해송조림지내해송잎떨림병은 2000 년부터피해가심하였으나, 2004 년부터는점차회복되어조림지에서병해를관찰할수없었으며, 해송생장도초기보다는생육이양호한것으로관찰되었다. 따라서산불발생후해송을조림할경우 3~4 년정도관리를잘해주면생장이좋을것으로판단되었다. 한편자작나무조림지에서는갈색점무늬병피해가매년발생되고있었으며, 갈색점무늬병은잎을조기에떨어뜨리고수세를약화시키므로경관조림시에는병이발생되지않도록각별히주의해야할것으로판단되었다. 자작나무갈색점무늬병피해율은매년 30~50% 로조사되었으며, 심하게발병된자작나무의경우 6 월말에도대부분의잎이낙엽지고건전목에비해수세가쇠약한것으로조사되었다 년삼척산불지역인원덕읍 LTER site 를조사한결과, 2001 년에는단풍나무타르점무늬병등 24 개수종 28 병종이관찰되었으며, 산불발생후소나무림에서가장많이눈에띄는리지나뿌리썩음병자실체인파상땅해파리버섯이많이발생되어있었다. 리지나뿌리썩음병의병원균인 Rhizina undulata 는소나무의뿌리를침해하여나무를고사시키는병으로알려져있으며, 리지나뿌리썩음병의피해는군상으로나타나고, 매년조금씩불규칙한원형으로확산된다. 병원균포자는고온에서만발아하는특성을가지고있어, 산불발생지, 모닥불자리, 해안사구림에서많이발생한다. 또한소나무뿌리가리지나뿌리썩음병에감염되면처음에는땅가의잔뿌리가흑갈색으로썩고, 점차굵은뿌리로확대되어뿌리전체가흑갈색으로변한다. 병든부분에는방사상의흑갈색균사속이나타나며, 분비된수지와토양입자가섞여딱딱한덩어리가형성된다. 소나무의뿌리가감염되면수세가약해지고잎이누렇게변하면서소나무가말라죽는다. 산불발생후삼척 LTER site 에서는 6 월중순경까지는리지나뿌리썩음병의자실체를확인할수없었으나, 8 월경부터발생되기시작하여 2 년동안계속관찰할수있었다. 하지만 2002 년도부터는자실체가발생되지않았는데, 이는봄철건조와고사리, 쑥, 아까시나무등하층식생이너무밀식되어리지나뿌리썩음병원균의활력이떨어진것으로판단되었다. 한편산불발생후 LTER site 에서매년관찰된단풍나무타르점무늬병은공해에약한특징이있어공기오염의생물학적지표로이용되고있으며, 도시의조경수나정원에서는거의발생되지않는병해로알려져있다. 삼척 LTER site 에서는병꽃나무탄저병, 칡겹둥근무늬병피해가심하였으며, 물푸레나무, 아까시나무, 참나무류등많은수종에흰가루병이발생되고있었다. 흰가루병은수목에치명적인병해는아니지만, 거의모든식물을침해하고심하게발생되면광합성부족, 양분의탈취등으로생육이위축되고수량감소를초래한다. 삼척 LTER site 의수목병원균상은산불발생후어느정도안정화되어가는추세였으며, 참나무류와붉나무에서병해가많이발생되는것으로보아산불발생후피해를가장많이받는수종인것으로판단되었다. 일부참나무류에는맹아가고

471 June 7-8, 2011 사하고지제부가변색되어부후가진행중인참나무가많이관찰되었으므로지속적인조사의필요성이있는것으로판단되었다. 인용문헌 (References) Bakshi, B. K., and S. singh Heart rot in trees. Academic Press: Callan, Brenda E Rhizina rot of Conifers, Forestry Canada, Forest Insect and Disease Survey. Forest Pest Leaflet No 564p Horikoshi, T., T. Tateishi and F. Takahashi Changes of fungus flora after fires in Pinus densiflora forest. Trans. Mycol. Soc. Japan 27:

472 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Changes of photosynthetic characteristics of Japanese larch(larix kaempferi) seedlings raised under different CO 2 concentration 서로다른 CO 2 농도에따른낙엽송의광합성특성변화 Donghun Ji*, Jaekyung Byun, Jinhyun Jeong Korea Forest Research Institute, Seoul, , Korea goon76@forest.go.kr Abstract Larch (Larix kaempferi) is an early successional conifer, and one of the most popular species for reforestation in northern Japan from s. Recently, we can expect natural regeneration after the disturbances of timber harvest or typhoon damages because mother trees are reaching to the age of reproduction. However, success of regeneration of seedlings depends on the size of forest gap because of light demanding traits in larch species. Therefore, we should evaluate the light use traits of photosynthesis in larch seedlings treated with shading in combination of nitrogen supply. We simulated environmental condition in a forest gap for light use characteristics of regenerated larch seedlings. When long shoot needles started to elongate, three-year-old seedlings were treated with three levels of light conditions (i.e. 0%, 50% and 80% cut) with and without nitrogen supply (2 N kg/hr yr). After 80 days treatments, we determined light and CO 2 depending photosynthetic rate was measured with an open gas flow system (LI- 6400) and composition of chlorophyll a and b, ratio of chlorophyll to nitrogen for acclimation ability to shading condition. Moreover, we examined the relationship between needle morphology or anatomy (A mes /A) and net photosynthetic rate. Relative growth rate of the seedlings was also determined in volume of stem. Based on these parameters, we will discuss the plausible understanding of forest environment for natural regeneration of larch seedlings. 서론 (Introduction) 자연교란은산림의천연갱신을촉진하는요인중하나로서최근에는벌

473 June 7-8, 2011 채등의인위적교란과태풍등의자연교란후천연갱신에의한산림재생이기대되고있다. 유라시아북동부에는낙엽송속수종이광범위하게분포하고있으며우리나라에서는 년대부터광범위하게식재된낙엽송이재생산연령에도달하였으며장령림에는벌목이나수확으로인하여임내의상층목이소개되어있는곳이있어천연갱신이기대되는지역이많다. 본연구에서는양수로알려져있는낙엽송의치수가임상의피음이나소개된곳에서자연발생하여어떤생장을하는지를생리적, 형태적반응을조사하여천연갱신이가능한환경을수목의생리적변화통하여해명하는것을목적으로하였다. 또한, 최근질소침착이증가하는경향을주목하고광부족과질소침착의복합적인영향을구명하는데주목했다. 재료및방법 (Methodology) 재료는 3 년생낙엽송묘목을대상으로피음격자를이용하여차광조건 0%, 50%, 80% 의 3 조건, 질소처리는처리와무처리의 2 단계조건으로실시하였으며질소량은 3kgN ha -1 yr -1 을처리하였다. 생리적특성을조사하기위해 LED light source 와 CO 2 injector system 이부착된휴대용광합성측정기 (Li-6400, Li Cor) 를사용하여광합성변화를측정하였고, 측정후 sample 을채취하여 DMSO 법으로엽록소함량을측정하였다. 결과및고찰 (Results and Discussion) 낙엽송묘목의근원경생장에는피음처리의효과가나타나지않았다. 이는생장에필수요소인광을흡수하기위해서수고생장을우선시하는경향이있다. 피음의정도가높아지면광합성작용이억제되어광합성산물이감소하기때문에결과적으로직경생장은감소하는경향을나타낸것으로사료된다

474 International Forest Fire Symposium on Commemorating the International Year of Forests 2011 Figure 1. Amount of height and root collar diameter growth of Larix kaempferi seedlings Figure 2. Effect of shade treatment on the photosynthetic rate of Larix kaempferi seedlings 낙엽송묘목의광 -광합성곡선 (Light curve) 을측정한결과광포화의광합성속도 (Psat) 는차광율 0%>50%>80% 의순으로높았으며질소첨가의영향은무차광 (0%) 에서만나타났다. 광을보충하여운반하는엽록소량과조성은식물의광합성능력과밀접한관계가있다. 특히집광부위와결합한엽록소 b량은피음의정도가높아질수록증가하여집광능력이증가하는것을의미한다. 또한, 광이감소하면할수록 Chl/N 의비율이증가하지만양수인낙엽송도침엽은광이부족한상황이되면질소를엽록소에분배하는것을알수있다. 그러나, 50% 이상의피음조건에서는광포화에서의광합성속도는질소가존재하더라도특히낮은값을나타낸다. 즉낙엽송치수는차량율이높은광조건에서도생장이가능할것으로생각된다. 그러나광합성관련단백질합성에필수요소인질소가토양중에충분히포함되어있다하더라도피음조건에서는환원에너지가부족하기때문에충분히사용이어렵다. 따라서양수인낙엽송치수는차광율이 50% 이상인환경에서는갱신이어려울것으로사료된다