<B8F1C2F72E687770>

Size: px

Start display at page:

Download "<B8F1C2F72E687770>"

현주 김
6 years ago
Views:

1 Journal of the Korean Society of Agricultural Engineers Vol. 57, No. 4, pp. 1~9, July 2015 DOI : ISSN eissn 식생가뭄반응지수 (VegDRI) 를활용한위성영상기반가뭄평가 Satellitebased Hybrid Drought Assessment using Vegetation Drought Response Index in South Korea (VegDRISKorea) 남원호 * Tsegaye Tadesse ** Brian D. Wardlow *** 장민원 ****, 홍석영 ***** Nam, WonHo Tsegaye Tadesse Brian D. Wardlow Jang, MinWon Hong, SukYoung Abstract The development of drought index that provides detailedspatialresolution drought information is essential for improving drought planning and preparedness. The objective of this study was to develop the concept of using satellitebased hybrid drought index called the Vegetation Drought Response Index in South Korea (VegDRISKorea) that could improve spatial resolution for monitoring local and regional drought. The VegDRISKorea was developed using the Classification And Regression Trees (CART) algorithm based on remote sensing data such as Normalized Difference Vegetation Index (NDVI) from MODIS satellite images, climate drought indices such as Self Calibrating Palmer Drought Severity Index (SCPDSI) and Standardized Precipitation Index (SPI), and the biophysical data such as land cover, eco region, and soil available water capacity. A case study has been done for the 2012 drought to evaluate the VegDRISKorea model for South Korea. The VegDRISKorea represented the drought areas from the end of May and to the severe drought at the end of June. Results show that the integration of satellite imageries and various associated data allows us to get improved both spatially and temporally drought information using a data mining technique and get better understanding of drought condition. In addition, VegDRISKorea is expected to contribute to monitor the current drought condition for evaluating local and regional drought risk assessment and assisting droughtrelated decision making. Keywords: classification and regression trees algorithm (CART); data mining technique; drought assessment; normalized difference vegetation index (NDVI); vegetation drought response index (VegDRI) Ⅰ. 서론 기후변화로인한전세계적인온난화현상으로인하여극심한가뭄의발생빈도가높아지고사회적, 경제적, 환경적피해가대형화되어가는추세이다 (Wilhite et al., 2000). 국립해양기상청 (National Oceanic and Atmospheric Administration, * National Drought Mitigation Center, University of Nebraska Lincoln, Lincoln, NE, USA ** National Drought Mitigation Center, School of Natural Resources, University of NebraskaLincoln, Lincoln, NE, USA *** Center for Advanced Land Management Information Technologies, School of Natural Resources, University of NebraskaLincoln, Lincoln, NE, USA **** Department of Agricultural Engineering, Institute of Agriculture & Life Science, Gyeongsang National University, Jinju, Republic of Korea ***** National Academy of Agricultural Science, Rural Development Administration Corresponding author Tel.: Fax: mwjang@gnu.kr Received: January 13, 2015 Revised: April 27, 2015 Accepted: April 29, 2015 NOAA) 이발표하는재해유형별경제손실액보고서 (Smith and Katz, 2013) 에의하면가뭄에의한경제적손실은홍수에비해 2 배이상이며, 1980년이후가뭄의발생빈도는전체재해빈도중 14 % 에해당하지만그피해액은전체재해피해액 25 % 를차지하고있다 (Nam et al., 2014). 이처럼가뭄은자연재난중홍수, 산사태등기타재난과비교하여영향면적이넓고상대적으로장기간에걸쳐영향을받기때문에, 전세계적으로다양한가뭄지수가개발되어가뭄의심도 (severity) 를정량화하고, 가뭄의전조를감지하며가뭄이진행되는영향범위를파악하는연구가진행되었다 (Hayes et al., 2004; Wilhite et al., 2007; Nam et al., 2012a). 가뭄의평가를위해일반적으로사용하는가뭄지수는기상학적가뭄지수인표준강수지수 (Standardized Precipitation Index, SPI) 와강수량및유효토양수분량에근거하여가뭄을판단하는파머가뭄심도지수 (Palmer Drought Severity Index, PDSI) 및강수량과증발산의변동성을고려한표준강수증발산지수 (Standardized Precipitation Evapotranspiration Index, SPEI) 를사용하고있다 (Hayes et al., 2011; Hunt et al., 2014; Svoboda et al., 2015). 이지수들은적용대상지역의지형및기상특성, 자료수집의제한성등을고려하여다양한유관기 한국농공학회논문집제 57 권제 4 호,

2 식생가뭄반응지수 (VegDRI) 를활용한위성영상기반가뭄평가 간에서가뭄상황을모니터링하기위한공간지도형태로가뭄지수를제공하고있다 (Nam et al., 2012b). 하지만제공되고있는가뭄지수의공간분포는지점자료를기반으로내삽기법 (interpolation) 을통해재산정된지도로공간해상도측면에서조악한해상도를갖고있는한계점이있다 (Tadesse et al., 2005). 이와같은한계점을보완하기위하여주기적이고동일한정확도로지상자료의획득이가능하다는측면에서인공위성을활용한가뭄분석연구의필요성이대두되었다. 재해관리분야에서원격탐사기술은재해발생을인지하고발생지역의재해진행과피해정도를신속하게제공할수있다는점에서효용성이높다 (Hayes et al., 2011). 특히가뭄은국지적으로영향을미치는다른재해들과는다르게광범위한지역에발생하고식생발달과밀접한관계를갖기때문에위성영상으로부터식생의발달정도를평가하고식생지수를계산함으로써가뭄의크기를계량화하는연구가수행되었다 (Tadesse et al., 2015). 인공위성을활용한가뭄분석의기본개념은가뭄으로인한식생의활성도저하및지표면온도의상승으로이어지는현상을위성영상으로부터포착하여가뭄현상을파악하는것이다 (Brown et al., 2008). 위성영상을이용할경우접근이용이하지못한지역의조사가수월하며, 장기적인변화관측이나환경감시등에유용하고시공간적으로변화가심한요소의관측과광역적접근이가능하다 (Tadesse et al., 2005; Shin et al., 2006; Swain et al., 2011). 국외의경우국립해양기상청, 미국항공우주국 (National Aeronautics and Space Administration, NASA), 국립가뭄경감센터 (National Drought Mitigation Center, NDMC, 등국공립기관에서원격탐사기술 (remote sensing) 과위성영상을활용하여식생의변화를정기적으로예측하고, 가뭄예보및모니터링시스템을구축하여웹을통하여주기적보고하고이를정책적으로활용하고있다 (Tadesse et al., 2010; Otkin et al., 2014). 국내의경우정규식생지수 (Normalized Difference Vegetation Index, NDVI) 및식생상태지수 (Vegetation Condition Index, VCI) 를활용하여기후학적물수지에근거하는광역적인가뭄분석을수행하였다 (Shin and Kim, 2003; Shin and Eoh, 2004). Jang et al. (2007) 은북한전역의 NDVI를분석함으로써 1998년부터 2001년까지북한의가뭄평가를수행하였으며, Park and Kim (2009) 및 Ahn et al. (2014) 은기상및기상학적가뭄지수와 NDVI와의상관성을제시함으로써가뭄평가를위한 NDVI의활용성을제시한바있다. Sur et al. (2014) 은다중분광센서인 MODIS (MODerate resolution Imaging Spectroradiometer) 를활용하여잠재증발산량과실제증발산량의비를이용한가뭄지수인 Evaporative Stress Index (ESI) 를국내에적용하였다. 또한 NDVI를활용하여논피복지도제작및작물의재배면적및생육, 생산량등을모니터링하는연구가진행되고있다 (Kwon et al., 2005; Jeong et al., 2011; Hong et al., 2012; Na et al., 2012). 하지만 NDVI의경우동일한식생상태일지라도영상별 / 계절별로다르게평가될수있고, 동일한영상에서도지역에따라동일한식생상태가다르게평가될수있기때문에, NDVI를직접적으로가뭄평가의객관적인기준으로활용하기에는한계가있다 (Jang et al., 2007). 본연구에서는위성영상을이용한식생정보및기후정보, 토지피복, 고도, 이용가능수분량등의생물물리학적정보를활용한식생가뭄반응지수 (Vegetation Drought Response Index in South Korea, VegDRISKorea) 를제시하고, 국내의적용성검증을위하여 2012년에발생한가뭄의시공간적가뭄상황을분석하였다. Ⅱ. 재료및방법 1. 식생가뭄반응지수 (VegDRISKorea) 의개발 Vegetation Drought Response Index (VegDRI) 는 NDMC 와미국지질조사국 (United States Geological Survey, USGS) 에서개발한식생가뭄모니터링지표로써위성영상기반의단일자료를활용한가뭄분석의한계점을극복하기위하여기상학적가뭄지표및토지피복, 생태지역등의생물물리학적정보를활용한가뭄지표이다 (Tadesse et al., 2005; Brown et al., 2008). 현재 VegDRI 가뭄지표는 NDMC ( unl.edu) 및 USGS ( 의가뭄모니터링시스템을통해 2 주간격으로미국전역의가뭄상황을제공하고있다. VegDRI는적용대상지역의활용가능한기후인자및가뭄지표와획득가능한생물물리학적정보등을고려하여적용지역에따라차별화된가뭄지표를개발할수있다 (Tadesse et al., 2014). 본연구에서는남한지역에적합한기후요소및가뭄지표, 생물물리학적정보를활용하여기존의 VegDRI 가뭄지표를변형하여남한지역에적용가능한 VegDRISKorea 가뭄지표를제시하였다. Fig. 1은 VegDRI SKorea의방법론을도시한것으로위성영상정보, 기후정보, 생물물리학적정보의세가지입력자료를바탕으로데이터마이닝 (data mining) 기법을활용하여규칙기반회귀모형 (rulebased regression tree model) 을적용한후격자기반의 VegDRI 값을산정한다. 2 Journal of the Korean Society of Agricultural Engineers, 57(4),

3 남원호 Tsegaye Tadesse Brian D. Wardlow 장민원 홍석영 Fig. 1 VegDRISKorea methodology including database, model development (regressiontree rules generation), and map generation (adjusted Tadesse et al., 2010) 2. 입력자료가. MODIS 위성영상자료 MODIS는 NASA의지구관측시스템 (Earth Observation System, EOS) 프로젝트일환으로 Terra (EOS AM1) 위성및 Aqua (EOS PM1) 위성의주센서이다 (Park et al., 2013). 본연구에서는관측주기가짧고관측폭이넓어육지와해양의광역현상변화관측에적합한 Terra 위성및 Aqua 위성의 MODIS 자료를사용하였으며, 육지 (land) 연구를위하여만들어진 MOD13 및 MYD13 (vegetation indices) NDVI 자료를이용하였다 ( 합성주기가 16 일단위로작성된 250 m 공간해상도를갖는 2001년부터 2013 년까지의 NDVI 자료를사용하였으며, 적설의영향으로인하여지표면의정확한관측이불가능한겨울철의자료는분석에서제외하였다. 분석기간은 4월 (4월 7일4월 22일 ) 부터 10 월 (10월 16일10월 31일 ) 까지 16 일단위의 13 개의기간을설정하였다. NDVI는식생의유무나지표의녹색도 (greenness) 를정량화하기위하여사용되고있는식생지수의하나로서, 광역의 식생특성을파악하기위해서유용한측정방법으로널리사용되고있다 (Kim and Kim, 2010). 식생활력도를나타내는지표인 NDVI는식 (1) 과같이가시광선파장영역과근적외선파장영역을조합하고정규화하여계산되며, 1.0에서 1.0까지의무차원값으로나타낸다 (Shin et al., 2010). (1) 은근적외선파장으로 MODIS 밴드 2, 는가시관선의적색파장으로 MODIS 밴드 1이다. VegDRISKorea 모델은위성영상인자로서각시기별과거의누적식생상태 (accumulated NDVI) 의정규화값을의미하는 SSG (Standardized Seasonal Greenness) 를사용하며, 식 (2) 와식 (3) 과같이산정된다 (Brown et al., 2008; Tadesse et al., 2010). (2) 한국농공학회논문집제 57 권제 4 호,

4 식생가뭄반응지수 (VegDRI) 를활용한위성영상기반가뭄평가 Table 1 Input data for the VegDRISKorea model Data type Data set name Acronym Source Satellite Standardized seasonal greenness SSG MODIS Terra NDVI Climate Standardized precipitation index SPI Meteorological Stations from Korea Selfcalibrating palmer drought severity index SCPDSI Meteorological Administration (KMA) Digital elevation model DEM USGSEROS Biophysical Land cover LC USGS Land Cover Institute (LCI) Soil available water capacity AWC Rural Development Administration (RDA) Ecological regions ECO The Nature Conservancy (3) SG (Seasonal Greenness) 는분선기간내각연도별 NDVI 의누적값으로산정되며, 와 는각각초기 / 후기생육시기 (growing season), 는생육시기를의미한다. 와 는 2001년부터 2013년까지산정한각생육시기별 SG의평균 (average) 과표준편차 (standard deviation) 를나타낸다. 나. 가뭄지표및생물물리학적자료 VegDRISKorea 모델의기후인자로서전세계많은지역에서가뭄을모니터링하기위한도구로활용되고있는 SPI와 SCPDSI (Self Calibrating Palmer Drought Severity Index) 를사용하였다. 1965년 Palmer에의해개발된 PDSI는기상학적인자인강수량과기온을이용하여가뭄을정량적으로평가할수있는가뭄지수로서, 강우량및기온, 일조시간등의자료를사용하여잠재증발산량을추정한후, 대상지역의실제강우량과기후학적으로필요한강우량의차를계산하여가뭄지수를산정한다 (Svoboda et al., 2002). PDSI는온도변수를포함하여증발산량의영향을고려하지만토양의수분보유함량 (available water capacity) 등여러변수값을필요로하고, 경험적인상수가지역특성을반영하지못하기때문에 (Nam et al., 2015), Wells et al. (2004) 은 PDSI의경험적인상수가해당지역의특성을바탕으로자가보정될수있도록수정한 SCPDSI를제안하였다. SPI는가뭄은상대적으로물의수요에비해물의부족을유발하는강수량의감소에의해시작된다는것에착안하여개발된기상학적가뭄지수로써, 강수량만을입력자료로사용하고정규화과정을통해적용지역에대한다양한시간척도의표준화된값을제공한다. 본연구에서는 2001년부터 2013년까지도서지역을제외한전국의 54 개기상관측소에대한 2 주단위의 SCPDSI와 36 주단위의 SPI를산정하였다. VegDRISKorea 모델의생물물리학적인자는수치표고 모델 (Digital Elevation Model, DEM), 토지피복지도, 유효토양수분량, 생태지도 (Ecological regions) 를사용하였다 (Table 1). 생태지도는환경조건의변동성을고려하여유사한생태계및환경자원과지리학적정보로구분된지도로서남한지역의경우상록수림 (evergreen forest), 혼합림 (mixed forest), 활엽수림 (deciduous forest) 으로구분된다 (Olson and Dinerstein, 2002). DEM은 90 m 해상도의 SRTM (Shuttle Radar Topography Mission) 자료, 토지피복지도는 km 해상도의 USGS 자료를사용하였으며 (Broxton et al., 2014), 모든생물물리학적인자는 250 m 해상도로변환 (resample) 하였다. 미국 VegDRI의경우위성영상자료로서평균계절별녹색도비율 (percent annual seasonal greenness) 을적용하고있으며, 이밖에도관개지역 (Irrigated locations) 자료및남방진동지수 (Southern Oscillation Index, SOI) 등을활용하고있다. 3. 데이터마이닝기법을이용한분류규칙개발데이터마이닝기법은물리적인변수들의패턴과상관성을분석하기위한기법으로사용되고있으며, 의사결정나무 (decision tree), 비선형회귀분석 (nonlinear regression analysis) 과군집분석 (cluster analysis) 등이있으며, 기존의자료들의분석을통하여목표변수의예측이가능하다는장점을갖고있다 (Kim and Park, 2010). Tadesse et. al. (2004) 은다양한변수들이복잡하게연계되어있는가뭄이식생상태에미치는영향을데이터마이닝의기계적인알고리즘으로가뭄의특성을분석하였다. VegDRISKorea 모델에서는남한지역에적용가능한가뭄지수및지형적인특징이고려된변수들과식생을표현하는변수들과의상관분석을이용하여시공간적인가뭄지수를산정하였다. VegDRISKorea 모델의규칙기반회귀분석을위하여위성영상자료 (SSG), 가뭄지표 (SPI, SCPDSI), 생물물리학적자료 (DEM, LC, AWC, ECO) 의 7 가지입력자료를기상관측소지점별로추출하였다. 자료의형태상위성영상 4 Journal of the Korean Society of Agricultural Engineers, 57(4),

5 남원호 Tsegaye Tadesse Brian D. Wardlow 장민원 홍석영 Table 2 VegDRISKorea classification scheme and class value ranges (Brown et al., 2008) VegDRI Class Value range Extremely wet Greater than 4.00 Severely wet 3.00 to 3.99 Moderately wet 2.00 to 2.99 Slightly wet 1.00 to 1.99 Near normal 0.99 to 0.99 Mild dry 1.99 to 1.00 Moderately dry 2.99 to 2.00 Severely dry 3.99 to 3.00 Extremely dry Less than 4.00 자료및가뭄지표와같은데이터의연속성을갖는변수와토지이용, 생태지역과같은범주형변수로구분할수있다. 각지점별로추출된데이터를이용하여의사결정나무기법의하나인 CART 알고리즘 (Classification And Regression Tree algorithm) 을적용한후 (Rulequest, 2013), 수집데이터간의규칙기반회귀식 (rulebased regression tree equation) 을생성하였다. 규칙기반의회귀분석을위하여종속변수로서가뭄심도에대한다양한기준을갖고있는 SCPDSI를사용하였기때문에 (Brown et al., 2008), VegDRISKorea 가뭄지표의분류기준은 SCPDSI와동일하다 (Table 2). 식 (4) 는 8월 29일부터 9월 13일까지생육시기의분류규칙을예로도시한것이다. 각시기별, 규칙별로도출된회귀식을각격자에작용하여연도별 / 시기별격자기반 (rasterbased) 의 VegDRI 가뭄지도를생성하였다. if Ⅲ. 적용및고찰 1. 식생가뭄반응지수를활용한가뭄분석 2012년봄에는전국적으로평년대비강수량이부족한상황에서중부지역을중심으로극심한가뭄이발생하였다. 전국적으로 5월평균강수량은 36.2 mm로예년평균의 36.4 % 로관측되었고, 경기서부와충남서해안지역은강수량예년비가 20 % 에불과했다. 6월상순과중순의강수량은 10.6 (4) mm, 19.4 mm로평년대비각각 33 %, 41 % 에불과했고, 하순에는 44.7 mm로평년대비 66 % 를기록했으며, 하순후반에는장마전선의영향으로전국에비가내려가뭄이해갈되었다 (Ahn et al., 2014). 본연구에서는경기, 충남, 전북지역을중심으로발생한 2012년봄기간의극심한가뭄상황을 VegDRI의공간분포를통해분석하였다. Fig. 2는 2012년 4월부터, 10월까지의 VegDRISKorea 공간지도로써가뭄상황의공간분포변화양상을나타낸것이다. 4월중순부터경기북부, 강원지역부터약한가뭄 (mild dry) 이발생하였으며, 5월부터 7월까지내륙을포함한경기서부와충남서해안지역에극심한가뭄 (severely dry) 이발생하였다. 8월에는국지적인강수및태풍의영향으로인해 7 월과비교하여가뭄이해소된것을확인할수있다. Table 3과 Table 4는시기별 / 도별 VegDRISKorea의약한가뭄과극심한가뭄의발생지역의비율을정리한것이다. 경기도지역의경우 5월초순과 6월초순에약 69 % 지역에서약한가뭄상태가나타났으며, 충남지역의경우 5월초순부터 7월하순까지약 79 % 지역에서가뭄상황이발생하였다. 강원지역은 2012년전기간에걸쳐약 35 % 지역에서가뭄상황이지속되었으며, 9월중순에약 61 % 지역에서약한가뭄상황이발생하였다. 강원지역의경우, 9월 14일부터 29일까지속초, 대관령, 춘천, 인제, 홍천기상관측소의평균강수량이 101 mm로평년과유사하였지만, 생태지역이주로혼합림으로구분되고유효수분량이작은지역이분포되어있어가뭄상황에민감하게반응한것으로판단된다. 극심한가뭄의경우경기, 강원, 충남지역에 5월초순에서 6월하순까지발생하였으며, 특히충남지역의경우약 28 % 지역에서극심한가뭄상태가발생하였다. 2012년봄가뭄사상기간에대한분석결과인공위성자료기반의 VegDRISKorea 가뭄지표가남한지역의가뭄사상분석에적합한것으로판단된다. 2. 기상학적가뭄지수와의비교 / 분석본연구에서는기상학적가뭄지수인 SPI, SCPDSI와 VegDRISKorea의가뭄분석의공간변동성을비교하였다. Fig. 3와 Fig. 4는 Fig. 2의 VegDRI 가뭄지표와동일한기간에대한 36 주단위의 SPI와 2 주단위의 SCPDSI의가뭄지수공간지도이다. 현재국내에서적용되고있는 SPI의공간분포와비교한결과 VegDRI의공간분포와유사한경향이나타났다. SCPDSI의경우가뭄의발생범위에대한차이가발생하였지만, 전체적인가뭄의공간분포는유사한결과가도출되었다. SPI 가뭄지표의경우강우량만을입력자료로이용하기때문에산정방법이단순하여여러기관에서널리적용이되고 한국농공학회논문집제 57 권제 4 호,

6 식생가뭄반응지수 (VegDRI)를 활용한 위성영상 기반 가뭄 평가 Fig. 2 Time series drought map of VegDRISKorea in 2012: (a) Apr. 7Apr. 22, (b) Apr. 23 May 8, (c) May 9 May 24, (d) May 25 June 9, (e) June 10 June 25, (f) June 26 July 11, (g) July 12July 27, (h) July 28Aug. 12, (i) Aug. 13Aug. 28, (j) Aug. 29Sep. 13, (k) Sep. 14Sep. 29, (l) Sep. 30Oct. 15, (m) Oct. 16Oct. 31 Table 3 Drought ratios below mild dry for VegDRISKorea according to eight provinceslevel Administrative district (a)1) (b) (c) (d) (e) (f) (g) (h) (i) (j) (k) (l) (m) Gyeonggido Gangwondo Chungcheongbukdo Chungcheongnamdo Jeollabukdo Jeollanamdo Gyeongsangbukdo Gyeongsangnamdo units: percentage (%) 1) (a) Apr. 7Apr. 22, (b) Apr. 23 May 8, (c) May 9 May 24, (d) May 25 June 9, (e) June 10 June 25, (f) June 26 July 11, (g) July 12July 27, (h) July 28Aug. 12, (i) Aug. 13Aug. 28, (j) Aug. 29Sep. 13, (k) Sep. 14Sep. 29, (l) Sep. 30Oct. 15, (m) Oct. 16Oct. 31 Table 4 Drought ratios below severely dry for VegDRISKorea according to eight provinceslevel (a)1) (b) (c) (d) (e) (f) (g) (h) (i) (j) (k) (l) (m) Gyeonggido Gangwondo Chungcheongbukdo Chungcheongnamdo Jeollabukdo ,3 Jeollanamdo Gyeongsangbukdo Gyeongsangnamdo Administrative district units: percentage (%) 1) (a) Apr. 7Apr. 22, (b) Apr. 23 May 8, (c) May 9 May 24, (d) May 25 June 9, (e) June 10 June 25, (f) June 26 July 11, (g) July 12July 27, (h) July 28Aug. 12, (i) Aug. 13Aug. 28, (j) Aug. 29Sep. 13, (k) Sep. 14Sep. 29, (l) Sep. 30Oct. 15, (m) Oct. 16Oct Journal of the Korean Society of Agricultural Engineers, 57(4),

7 남원호 Tsegaye Tadesse Brian D. Wardlow 장민원 홍석영 Fig. 3 Time series drought map of 36weeks SPI in 2012 from (a) Apr. 7Apr. 22 to (m) Oct. 16Oct. 31 Fig. 4 Time series drought map of 2weeks SCPDSI in 2012 from Apr. 7Apr. 22 to (m) Oct. 16Oct. 31 있으며, 현재까지 통용되고 있는 가뭄지수들 중 적용성이 가 장 높다 (Hayes et al., 2004; Wilhite et al., 2007). 하지만 SPI, SCPDSI는 지점관측 자료를 기반으로 각 지점의 결과 값을 이용하여 내삽기법을 적용하여 산정한 공간지도이기 때문에 공간적인 정확성이 낮으므로 시/군/구 단위의 행정단위의 가 뭄 분석을 할 수 없는 한계점이 발생한다. VegDRISKorea 가뭄지표의 경우 250 m의 공간 해상도 를 갖고 있으며, 소단위 행정단위의 가뭄상황에 대한 분포를 확인 할 수 있기 때문에, 지역적 규모에서 기상자료 및 기후학 적 가뭄지표가 제공할 수 없는 가뭄 상황에 대한 정보를 제공 할 수 있다. 또한 지역별 시기별 식생 수준 및 가뭄 상황을 객 관화할 수 있는 기준으로 활용가능하며, 가뭄 확산의 시공간 적 특성을 파악하기 위한 가뭄지도 작성 및 향후 가뭄에 대한 방재 대책에 활용할 수 있을 것으로 판단된다. Ⅳ. 결 론 본 연구에서는 MODIS NDVI 위성영상을 이용한 식생 정 보 및 SPI, SCPDSI의 기존 가뭄지수, 토지피복, 고도, 이용 가능수분량 등의 생물물리학적 정보를 활용하여 CART 알고 리즘으로부터 수집된 데이터간의 분류 규칙을 생성하고, 격 자기반의 식생가뭄반응지수 (VegDRISKorea)를 제시하였 한국농공학회논문집 제57권 제4호,

8 식생가뭄반응지수 (VegDRI) 를활용한위성영상기반가뭄평가 다. 또한국내의적용성검증을위하여 2012년에발생한가뭄의시공간적가뭄상황을분석하고, 기존의기상학적가뭄지수와비교분석하였다. VegDRISKorea을이용한가뭄분석을통해경기, 충남, 전북지역을중심으로발생한 2012년봄기간의극심한가뭄상황을반영한결과가도출되었다. 본연구의결과는가뭄의실시간감시를위하여원격탐사의활용및다양한정보를통합하여복합적으로반영하는격자기반의고해상도가뭄정보를제공함으로써행정구역별가뭄방재재해대책을세울수있을것으로사료된다. 또한향후위성영상자료의장기적인자료를축척하여활용한다면가뭄분석의신뢰성있는결과를도출할것으로판단된다. 사사이논문은농촌진흥청연구사업 ( 세부과제번호 : PJ ) 및 2013년도정부 ( 교육부 ) 의재원으로한국연구재단의지원을받아수행된기초연구사업임 (2013R1A6A3A ). REFERENCES 1. Ahn, S.R., J.W. Lee, and S.J. Kim, Analysis of 2012 spring drought using meteorological and hydrological indices and satellitebased vegetation indices. Journal of Korea National Committee on Irrigation and Drainage 21(1): 7888 (in Korean). 2. Brown, J.F., B.D. Wardlow, T. Tadesse, M.J. Hayes, and B.C. Reed, The vegetation drought response idex (VegDRI): a new integrated approach for monitoring drought stress in vegetation. GIScience & Remote Sensing 45(1): Broxton, P.D., X. Zeng, D. SullaMenashe, and P.A. Troch, A global land cover climatology using MODIS data. Journal of Applied Meteorology and Climatology 53: Hayes, M.J., O.V. Wilhelmi, and C.L. Knutson, Reducing drought risk: bridging theory and practice. Natural Hazards Review 5(2): Hayes, M., M. Svoboda, N. Wall, and M. Widhalm, The Lincoln declaration on drought indices: universal meteorological drought index recommended. Bulletin of the American Meteorological Society 92: Hong, S.Y., J. Hur, J.B. Ahn, J.M. Lee, B.K. Min, C.K. Lee, Y.H. Lee, K.D. Lee, S.H. Kim, G.Y. Kim, and K.M. Shim, Estimating rice yield using MODIS NDVI and meteorological data in Korea. Korean Journal of Remote Sensing 28(5): (in Korean). 7. Hunt, E.D., M. Svoboda, B. Wardlow, K. Hubbard, M. Hayes, and T. Arkebauer, Monitoring the effects of rapid onset of drought on nonirrigated maize with agronomic data and climatebased drought indices. Agricultural and Forest Meteorology 191: Jang, M.W., S.H. Yoo, and J.Y. Choi, Analysis of spring drought using NOAA/AVHRR NDVI for North Korea. Journal of the Korean Society of Agricultural Engineers 49(6): 2133 (in Korean). 9. Jeong, S.T., K.C. Jang, S.Y. Hong, and S.K. Kang, Detection of irrigation timing and the mapping of paddy cover in Korea using MODIS images data. Korean Journal of Agricultural and Forest Meteorology 13(2): 6978 (in Korean). 10. Kim, G.S., and J.P. Kim, Analysis of spatialtemporal variability of NOAA/AVHRR NDVI in Korea. Journal of the Korean Society of Civil Engineers 30(3B): (in Korean). 11. Kim, G.S., and H.G. Park, Estimation of drought index using CART algorithm and satellite data. Journal of the Korean Association of Geographic Information Studies 13(1): (in Korean). 12. Kwon, H.J., S.C. Shin, and S.J. Kim, Climatic water balance analysis using NOAA/AVHRR satellite images. Journal of the Korean Society of Agricultural Engineers 47(1): 39 (in Korean). 13. Na, S.I., J.H. Park, and J.K. Park, Development of Korean paddy rice yield prediction model (KRPM) using meteorological element and MODIS NDVI. Journal of the Korean Society of Agricultural Engineers 54(3): (in Korean). 14. Nam, W.H., J.Y. Choi, S.H. Yoo, and B.A. Engel, 2012a. A realtime online drought broadcast system for monitoring soil moisture index. KSCE Journal of Civil Engineering 16(3): Nam, W.H., J.Y. Choi, S.H. Yoo, and M.W. Jang, 2012b. A decision support system for agricultural drought management using risk assessment. Paddy Water Environment 10(3): Nam, W.H., M.J. Hayes, D.A. Wilhite, T. Tadesse, M.D. Svoboda, and C.L. Knutson, Drought management and policy based on risk assessment in the context of climate change. Magazine of the Korean Society of Agricultural Engineers 56(2): 215 (in Korean). 17. Nam, W.H., M.J. Hayes, D.A. Wilhite, and M.D. Svoboda, Projection of temporal trends on drought characteristics using the standardized precipitation evapotranspiration index (SPEI) in South Korea. Journal of the Korean Society of Agricultural Engineers 57(1): 3745 (in Korean). 8 Journal of the Korean Society of Agricultural Engineers, 57(4),

9 남원호 Tsegaye Tadesse Brian D. Wardlow 장민원 홍석영 18. Olson, D.M., and E. Dinerstein, The Global 200: priority ecoregions for global conservation. Annals of the Missouri Botanical Garden 89(2): Otkin, J., M. Shafer, M. Svoboda, B. Wardlow, M. Anderson, C. Hain, and J. Basara, Facilitating the use of drought early warning information through interactions with agricultural stakeholders. Bulletin of the American Meteorological Society, in press, doi: /BAMSD Park, J.K., B.S. Kim, S.Y. Oh, and J.H. Park, Applicability of vegetation indices from Terra MODIS and COMS GOCI imageries. Journal of the Korean Society of Agricultural Engineers 55(6): 4755 (in Korean). 21. Park, J.S., and K.T. Kim, Evaluation of MODIS NDVI for drought monitoring: focused on comparison of drought index. The Journal of GIS Association of Korea 17(1): (in Korean). 22. Rulequest, An overview of Cubist. RuleQuest Research Pty Ltd, St Ives, NSW, Australia. available at rulequest.com/cubistwin.html. 23. Shin, H.J., R. Ha, M.J. Park, and S.J. Kim, Estimation of spatial evapotranspiration using the relationship between MODIS NDVI and morton ET for Chungjudam watershed. Journal of the Korean Society of Agricultural Engineers 52(1): 1924 (in Korean). 24. Shin, S.C., and C.J. Kim, Application of normalized difference vegetation index for drought detection in Korea. Journal of the Korean Water Resources Association 36(5): (in Korean). 25. Shin, S.C., and M.S. Eoh, Analysis of drought detection and propagation using satellite data. Journal of The Korean Society of Hazard Mitigation 4(2): 6169 (in Korean). 26. Shin, S.C., J. Soo, K.T. Kim, J.H. Kim, and J.S. Park, Drought detection and estimation of water deficit using NDVI. Journal of the Korean Association of Geographic Information Studies 9(2): (in Korean). 27. Smith, A.D., and Katz, R.W., US billiondollar weather and climate disasters: data sources, trends, accuracy and biases. Natural Hazards 67: Sur, C.Y., K.J. Kim, W.J. Choi, J.H. Shim, and M.H. Choi, Drought assessments using satellitebased drought index in Korea: southern region case in Journal of The Korean Society of Hazard Mitigation 14(3): (in Korean). 29. Svoboda, M., D. LeComte, M. Hayes, R. Heim, K. Gleason, J. Angel, B. Rippey, R. Tinker, M. Palecki, D. Stooksbury, D. Miskus, and S. Stephens, The drought monitor. Bulletin of the American Meteorological Society 83(8): Svoboda, M.D., B.A. Fuchs, C.C. Poulsen, and J.R. Nothwehr, The drought risk atlas: enhancing decision support for drought risk management in the United States. Journal of Hydrology, in press, doi: /j.jhydrol Swain, S., B.D. Wardlow, S. Narumalani, T. Tadesse, and K. Callahan, Assessment of vegetation response to drought in Nebraska using TerraMODIS land surface temperature and normalized difference vegetation index. GIScience & Remote Sensing 48(3): Tadesse, T., D.A. Wilhite, S.K. Harms, M.J. Hayes, and S. Goddard, Drought monitoring using data mining techniques: a case study for Nebraska, USA. Natural Hazards 33: Tadesse, T., J.F. Brown, and M.J. Hayes, A new approach for predicting droughtrelated vegetation stress: integrating satellite, climate, and biophysical data over the U.S. central plains. ISPRS Journal of Photogrammetry & Remote Sensing 59: Tadesse, T., B.D. Wardlow, M.J. Hayes, and M.D. Svoboda, The vegetation outlook (VegOut): a new method for predicting vegetation seasonal greenness. GIScience & Remote Sensing 47(1): Tadesse, T., G.B. Demisse, B. Zaitchik, and T. Dinku, Satellitebased hybrid drought monitoring tool for prediction of vegetation condition in eastern Africa: a case study for Ethiopia. Water Resources Research 50: Tadesse, T., B.D. Wardlow, J. Brown, M. Hayes, M. Svoboda, B. Fuchs, and D. Gutzmer, Assessing the vegetation condition impacts of the 2011 drought across the U.S. southern great plains using the vegetation drought response index (VegDRI). Journal of Applied Meteorology and Climatology 54: Wells, N., S. Goddard, and M.J. Hayes, A selfcalibrating palmer drought severity index. Journal of Climate 17: Wilhite, D.A., Hayes, M.J., Knutson, C., and Smith, K.H., Planning for drought: moving from crisis to risk management. Journal of the American Water Resources Association 36(4): Wilhite, D.A., M.D. Svoboda, and M.J. Hayes, Understanding the complex impacts of drought: a key to enhancing drought mitigation and preparedness. Water Resources Management 21: 한국농공학회논문집제 57 권제 4 호,

인문사회과학기술융합학회

인문사회과학기술융합학회 Vol.5, No.5, October (2015), pp.471-479 http://dx.doi.org/10.14257/ajmahs.2015.10.50 스마트온실을 위한 가상 외부기상측정시스템 개발 한새론 1), 이재수 2), 홍영기 3), 김국환 4), 김성기 5), 김상철 6) Development of Virtual Ambient Weather Measurement