< C0E5BFC1C0E72E666D>

Transcription

1 w wz 9«4y pp. 0 ~ 0 지역회귀분석을이용한홍수피해위험도산정 Flood Risk Estimation Using Regional Regression Analysis 장옥재 * 김영오 ** Jang, Ock-Jae ½ Kim, Young-Oh Abstract Although desire for living without hazardous damages grows these days, threats from natural disasters which we are currently exposed to are quiet different from what we have experienced. To cope with this changing situation, it is necessary to assess the characteristics of the natural disasters. Therefore, the main purpose of this research is to suggest a methodology to estimate the potential property loss and assess the flood risk using a regional regression analysis. Since the flood damage mainly consists of loss of lives and property damages, it is reasonable to express the results of a flood risk assessment with the loss of lives and the property damages that are vulnerable to flood. The regional regression analysis has been commonly used to find relationships between regional characteristics of a watershed and parameters of rainfall-runoff models or probability distribution models. In our research, however, this model is applied to estimate the potential flood damage as follows; 1) a nonlinear model between the flood damage and the hourly rainfall is found in gauged regions which have sufficient damage and rainfall data, and 2) a regression model is developed from the relationship between the coefficients of the nonlinear models and socio-economic indicators in the gauged regions. This method enables us to quantitatively analyze the impact of the regional indicators on the flood damage and to estimate the damage through the application of the regional regression model to ungauged regions which do not have sufficient data. Moreover the flood risk map is developed by Flood Vulnerability Index (FVI) which is equal to the ratio of the estimated flood damage to the total regional property. Comparing the results of this research with Potential Flood Damage (PFD) reported in the Long-term Korea National Water Resources Plan, the exports' mistaken opinions could affect the weighting procedure of PFD, but the proposed approach based on the regional regression would overcome the drawback of PFD. It was found that FVI is highly correlated with the past damage, while PFD does not reflect the regional vulnerabilities. Key wordsg: regional regression analysis, flood risk map, flood damage estimation ú w x l w š w ƒwš»z y y x ƒ w w l x w. yw y w» w ƒ w p sƒw w w. z w ƒ vw wš, mw ƒ y x sƒw w. y w vw ù vwƒ» y x sƒ y w ù tx w š q. z - x ù y s x p txw» w w( Œª) y vw w. z d ( y vw ƒ w ) y vw k x z w š, z w Á z w txw. w y vw e w w ª d ( ƒ w ) mw p ƒ w y vw w. w y vw Flood Vulnerability Index (FVI) w mw p w ü vw wš, y x ùkü.» w z y x sƒ w y vw (Potential Flood Damage; PFD) w PFD ƒ ƒ e ƒ q ù z w w w. w FVI wvw ùkþ PFD w w ùkû. w : z, y x, y vw * w œ w y œw ( trial10@snu.ac.kr) ** z Á w œ w y œw 1

2 1. zƒ w w w x l w š w w ƒwš. w wš w»z y»z w» xw w w. w y w z w w w w x wš, œ w w w ùƒ w., w x» (risk-based design) v w (National Research Council, 1995; w NRC) w w w l w w» w x sƒ(risk assessment) x (risk management) ywš (Ikeda, 2006).» x sƒ w (hazard identification), (vulnerability assessment), x ³ (risk characterization) w w w w ³ j» w š, w ù š y ƒ ƒ w w ù wƒ sƒw w w vw j» j. g w w w x wš» ù xk ùkü. x sƒ y x j» w ƒ œwš, ƒ z š w z ü w x (NRC, 1983). y w x w» w w x sƒ w w» ¾ w x sƒ w v w w ƒ mw (export-opinions-based method) l mw (historical-data-based method) ƒ w (Zou, 2009). ƒ w vwƒ ù p w sƒ w. w vw k w (pressure) w (response) š w ew vw ù k(state) w w (indicators) wš, w ƒ e x sƒw. ƒ w ƒ ƒ e w wš, mw wù t mww Munich Re (2003) w s³ w vw (Annual Average Loss, ALL) 80% ƒ e, š ƒ vw(probable Maximum Loss, PML) 20% ƒ e w. w SCEMD (2005) w k w š sƒ ƒ e w yw» ƒ e w š, Schmidt-Thome (2005) ƒ w w ƒ e ƒ e ¾ w delphi method w w. ƒ w tx ù y wvw s ƒ sw k ù ƒ e ƒ e ƒ. w l w w x sƒw l w x sƒw ƒ q k. Disaster Risk Index (UNDP, 2004) ƒ ƒ w x w w w ³ w. w w wš š w ƒ z z w ƒ w w wš mw w x sƒw. w DEFRA (2006) w y e, š w w ƒ y x sƒ w š. l w w x w wz» ƒ» w. ü» w z y x sƒ š y vw (Potential Flood Damage; w PFD) wš, y vw y w y vw x sƒw wš w. w y vw š w Á z, w z w w z w. ƒ w w w vw wš y ùkü. 2. y x sƒ 2.1 y vw ü» w z( m, 2006a) wš PFD ( Œ) zá ¾ s w p e y w ùkü e p q wš n w ù, ³ e w z w» w v. PFD (2001) 150 w 2001» w z( m, 2001) 2 w wz «y

3 t 1. PFD ( m, 2006a) x, F RI, F PO ƒ w ù (F PD, β 1 = 0.4) ù (F PF, β 2 = 0.3) ù y (F RUB, β 3 =0.2) ù z» (F SOC, β 4 = 0.1) ù y vw (F PD, β 5 = 0.3) ù y (F PD, β 6 = 0.2) ù (F CI, β 7 = 0.3) ù (F DAM, β 8 = 0.1) ù ü (F PUMP, β 9 = 0.1) PFD PFD = = α 1 F PO α 2 FRI α 1 [ ] β 1 F PD + β 2 F PF + β 3 F RUB + β 4 F SOC β 5 F FE + β 6 F PR + β 7 F CI + β 8 F DAM + β 9 F α 2 [ ] PUMP 117 w w. t 1 ùküš. ƒ PFD α 1 α š, w ƒ β i ƒ q w» w z w w ù ew β i w ƒ w w. š PFD w w ü e ƒ sƒ ƒ w wš, x k l PFD y te w (½ z, 2004). w PFD w y x sƒ wš w. ü tƒ w» y vw w wz vw m te e w sƒ» w. 2.2 y x w y x» 1 w ùký. w y x w w y ƒ ù w ƒ? y ƒ w y ƒ w ƒ w w. w y ü. w y ƒ w ü ù kü vw w w ƒ, Á z» w. w ù rv 1. y x» (DEFRA (2006) w ) š y ƒ w vw x w. p w w y vw w» w 2 ww.» r y vw v w ƒ w d, w d š w. d (1) y vw z w. (y vw ) = f 1 (,,...) (1) z f 1 a, b, c š w d z w. a, b, c = f 2 ( d Á z w ) (2) d z w d w. d w - x p z w š w w w. d p (2) w (1) š mw d z f 1 w. y d y vw z wš, d z mw z w ƒ y vw w. mw ƒ y w q w. 2.3 y vw y vw w» w w w w. y vw m y ƒ w w»ƒ j w ƒ w z w y w. w w z ( y w v wƒ w ú w œwš. z wy vw x 3

4 vwd 22 (1985 ~2006 ),,, w vw w w vw 2006» y w. ƒ vwƒ w» ƒ swš d 1) z wš, Thiessen w ƒ e w. e ƒ 1985 z wù e w w» vw 15 d w. 15 yw z w» w ƒ w š w w» q. x w ƒ wvw w k t t w vw w vw wš kt w vw w w q w w w w q. p w ew t w w ù kû kt št 20 m/sec ƒ». Saffir-Simpson tropical cyclones w št 17 m/sec st(tropical storm) š 33 m/sec l hurricane, typhoon (UNDP, 2004) kt ƒ w 9 11 l 9 12 ¾ 2 ü 100 mm ü w w vw švw w. št 26.1 m/sec w w w w w 2. y x e w w q. w w y x š» št 17 m/sec w 2) w w w xk v w. w w y w vw ü w» w w.,, 10 mm, 15 mm, 3, 5 wš vw w. ƒ ùkù» w 15 mm ( 15 mm w w) 5 3. y vw ( p ) 1), ½s, ù,, ù, š, ûw, d ƒ,, d ƒ, y d ƒ d ƒ 2)» št w w» l» w (26.8 m/sec), ~ (21.7 m/sec), ~ (25.7 m/sec), ~ (43.0 m/sec), ~ (22.5 m/sec), ~ (18.7 m/sec), ~ (17.5 m/sec), ~ (26.1 m/sec) 4 w wz «y

5 t 2. w vw x z e c a b 1 b 2 R 2 1,787, ,016, ,363, ,775, Ÿ + 3,807, p + 8,397, (11 ) 1,581, ,201, ,822, s 2,016, , ,755, ,479, û + 2,077, Ÿ (7 + 1,121, ) 1,384, , w 2,723, Ÿ 3,439, Ÿ û + 20,238, ) 1985 z e w + wù e t 3. z t t / ƒ / k m ( /km 2 )» y 15 /65 d ƒ (, w, k ) ƒ s³ k xk ( ƒ,, ) (%) ( ) ( ) m xy (,, ) ( n d, s³ ) m (1985 ~2005 ) ( ƒ m ƒ w yr ( y (ha), l (HP), (m 2 ) yr ( (» 5 ) ùkü. w r 15 mm en w (excessive rainfall) ùküš, 5 w ƒ j ùkü š w. w w xk growth function w (3) xk ùkü. c y = exp[ a+ b 1 ( P ) > 15 + b 2 ( P )] peak 5hr» y y vw ( ), P >15 15 mm, P peak 5hr 5 Growth function w 3 w ù exponential w w ùký». w w d (3) z wy vw x 5

6 w ùký d k» w x w j w. w 3 15 mm 5 ƒ e w y w. ƒ ü w š, y w z w w. ƒ e growth function w» w d» w š, SPSS 12(SPSS INC., 2003) v w. w RMSEƒ ƒ kw x z Levenberg-Marquart (Marquart, 1963) w. x z t 2 4, 20 e w ƒ 0.9 z ƒ w q. 2.4 z w vw x z p w w w w. t 3 vw / k, txw vw ƒ k m y ƒƒ w. k w k w ( m, 2006b;, 2004; m, 2006). w y vw ùkü 22 wù tx» p mww s³ w k w š q. t 3 ùkü m xy k ü m,, w., œ,, sw w sww w. w w. š 65 ù 0~1¾ tx ù, 0~ ¾ t xw» w y w (UNDP, 2004). V i» V i = V i V i y, V i z ƒ w š w z w. z ƒ 20 xz w w ƒ sww sww 0.1 ƒw sww (stepwise selection method) w. (4) k ƒ f q (Variation Index Factor; VIF)ƒ 10 š w ƒ w. ƒ z w (5)~(9) w e wù w. q ƒ 10 z w. w e we z w w p p Leverage, Cook's distance, DFFITS š w. c y vw ew ùkû ƒ y w j» w q. c = 9.520ÁSP ÁPH ÁAHA+15,533Á UA+36.15ÁUH+5,607,598 (N = 17, R 2 = 0.965) (5)» SP 65, PH ƒ ( ƒ ), AHA ƒ s³, UA, UH wƒ (5) ƒ t y r 65 ƒ 0.036, ƒ ƒ 0.727, ƒ s³ , wƒ ƒ 0.185, ùküš ƒ ƒ c ƒ w eš ù ƒ ƒeƒ» ƒ s ³ y ƒ q. ew a, b 1, b 2 y vw ƒ j p ùkü p z ƒ y vw k y ƒ š, ƒ k y ƒ. w w α j vw ƒw 3 d w š, b 1 b 2 t 2 ƒ j» ƒ w w q w. a= ÁTP ÁPD ÁSR ÁTax ÁUH (N=16, R 2 =0.906) (6) b 1 =0.807ÁSR ÁTH ÁPH UH 0.010ÁAHA+0.005ÁPump 0.025ÁIA (N=17, R 2 =0.863) (7) b 2 = ÁTP ÁPP+2.777ÁSR ÁUH 0.195ÁUR+0.015ÁPR 2.228ÁIR ÁPump (N=17, R 2 =0.912) (8)» TP, PD ( /km 2 ), SR 65 6 w wz «y

7 s t 4. p ( :» d ) 19~24 hr 25~36 hr sx GEV Gumbel location parameter scale parameter shape parameter (mm) mm (mm) (mm) , Tax û ( ), UH wƒ, TH ƒ, PH ƒ, AHA ƒ s³, Pump rv, IA, PP» y, UR, PR, IR (6) t y , ƒ ƒw j ƒ j ùkû û t y ƒ ƒw t w y w ùk û. ù b 1, b 2 w œm w ƒ ƒ y ƒ j ƒ g y z ùk û. z ùk üš d» d n. ƒ e ƒ 3~4 d w ù d w leave-one-out cross validation w. c 16 l z wš, ù wù l w d w ww. 4ƒ w 0.760~0.850 ù kü d ew š d w q. 2.5 ƒ p w ƒ w y vw w w w w.» š m (2000) Huff sš w. Huff(1967) Illinois ew 400 mi 2¾ w w w y w y» ü s w s x w 1~4 y w. ƒ w y ƒ» y w w. PT i ( ) PR i ( ) Ti ( ) = % T 0 Ri ( ) = % R 0 (9) (10)» PT(i) T 0 w T(i) (%) PR(i) R 0 w T(i)¾ ƒ R(i) (%) w w w w. 19~24 25~36 ƒ w mw ƒ 100 w. w v w z w w vw w»» p ù p. Huff sš ƒ y 10%~90% 9ƒ xkƒ w w ew y 50% w. 3. z w y v w xz š z k p 25 e w z sƒw. y vw lƒ w d 14 e w 1985 z 3 e wù w y vw xz ƒ vw w. ù 11 d z wy vw x 7

8 c ( 10 6 ) t 5. ƒ y vw 19~24 ƒ vw FVI ( ) a b 1 b 2 t œ ƒ ( /m 2 ) 25~36 ƒ vw FVI ( ) d 1, , E E , E E-07 4, , E E-07 19, , E E-06 Ÿ + 3, , E E , , E E-06 1, , E E-07 4, , E E-07 8, , E E-07 s 2, , E E , E E-09 d û 2, , E E-16 1, E E-05 3, , E E-05 3, , E E-05 s 2, , E E-09 1, , E E-14 q 2, , E E-12 2, , E E-05 s 2, , E E-09 1, , E E-06 3, , E E-07 ) FVI E 10 d z k xz w š, k ƒ vw w. t ~24 j vw ƒ w ù y x w w ƒ vw j ù. m w p w w p ƒ y vw w y wƒ d w 8 w wz «y

9 . y (Flood Vulnerability Index, FVI) w. ƒ y vw ƒ y vw FVI= = (t œ ƒ/m ) ( e 2, m 2 ) (11) FVI ƒ e p w ùküš wz mw t q w 4 safe area ùkü FVI k y ƒ w w ùkü. 4(a) 19~24 s, s, q, û, ƒ y w q ƒ ƒ w. ù 4(b) r w w,,, ƒ y w ùkû.» w z wš PFD w PFD e ƒ 1018 «wù sw PFD w y w. ù w FVI l 15 ¾ 24 ü w e ƒ (, 2002) r FVI y w ùkù w. w w y x ( m, 2006b) e ƒ x ƒ vw ƒ ù ƒ. y w ƒ e y vw w mw w š, wz n w w q. 4.» ƒ w w x s ƒ w w w» w w w w. y vw vwƒ» y w vw wš sƒ ƒ vw ùkü. vw w xz wš, z p z w z w. mw y vw ƒ w p z wš, l vw w. w PFD ƒ q ƒ e w e ƒ w w. p w w v w w z vw w y w w. w p w ƒ e y l w w k q. w p 25 e w 100 ƒ w ƒ e vw y w. ƒ w y vw»» w ùkü ù y w y x mw ƒ y w e w. e ƒ wù «sw x ƒ sƒ PFD { w x w š w. w» w wz z w ƒ? w y vw w wƒ? w ƒ ƒ v w. w FVI t œ ƒ y w š ù ƒe š w» šƒ ù š ƒe w» ƒ. wz Apel (2004) w ù šƒ w x ƒe wì w š w v w. z w w ƒ t û wš, w t j w g ù / w. w w t œw» zá š w w» œw,, ƒ sw m w wz û¼. mw ƒ wš w m» sƒ ww 2008 ( y:07 B04) w œw mw.. š x m (2000) s. 1999» š - 2«Õw ƒ Ö, 2«. m (2001)» w z (2001~2020) š. š. m (2006a)» w z (2006~2020) š. z wy vw x 9

10 š. m (2006b) y x vw (II).» š. m (2006) y vwp y vw t w. š. ½ z (2004) e w. w, w. (2002) e l ,,, ½³y (2001) y vw (PFD) s ƒ. w wz w t z, w wz, pp (2004) q x w w w š. w, w. Apel, H., Thieken, A.H., Merz, B., and Bloschl, G. (2004) Flood risk assessment and associated uncertainty. Natural Hazards and Earth System Sciences, European Geoscience Union, Vol. 4, pp. 295~308. DEFRA (Department of the Environment, Food and Rural Affairs) (2006) Flood risks to people: Guidance document. FD2321/ TR2, Environment Agency, London. Huff, F.A. (1967) Time distribution of rainfall in heavy storms. Water Resources Research, AGU, Vol. 3, No. 4, pp Ikeda, S. (2006) An integrated risk analysis framework for emerging disaster risks: Toward a better risk management of flood disaster in urban communities. Terrapub, Tokyo. Marquart, D. (1963) An algorithm for least-squares estimation of nonlinear parameters. SIAM Journal of Applied Mathematics, Vol. 11, pp Munich Re (2003) Topics-Annual review: Natural catastrophes Munich Reinsurance Company, Germany. National Research Council (1983) Risk assessment in the federal government: Managing the process. National Academy Press, Washington D.C. National Research Council (1995) Flood risk management and the American river basin: An evaluation. National Academy Press, Washington D.C. SCEMD (2002) State of South Carolina hazards assessment. South Carolina Emergency Management Division, South Carolina. Schmidt-Thome, P. (Ed.) (2005) The spatial effects and management of natural and technological hazards in general and in relation to climate change. ESPON Project 1.3.1, Geological Survey of Finland, Finland. SPSS INC. (2003) SPSS base 12.0 user guide. Chicago. UNDP (2004) Reducing disaster risk: A challenge for development. United Nations Development Programme, Bureau for Crisis Prevention and Recovery, New York. Zou, L., and Wei, Y. (2009) Impact assessment using DEA of coastal hazards on social-economy in Southeast Asia. Natural Hazards, Springer, Vol. 48, pp ú : ú : ú : w wz «y