关键词: 脆弱性, 风险评价, 旱灾, EPIC模型, 小麦产量, 中国

Abstract: Food security is the base of human survival and national stability, of which food supplydemand balance variation and its induced food price fluctuation are the two major concerns. Global food production has been kept on increase since 2000, though in a slow pace, which should be able to meet the basic diet requirement. According to the 4th IPCC report, however, from the end of 2007, the worldwide food supply problem has been paid great attention to due to a sky rocket of food price. The earth is getting warmer and the frequency of drought is increased. Consequently, the intensification of water stress to crops, resulted from both the long term climate change and increase of extreme weather events, may lead to the reduction of crop yield and the increase of drought disasters. To understand the mechanism of crop drought disaster process is the key to the understanding of food security and hence a possible solution to the problem. In the past researches, the concepts of physical vulnerability and social vulnerability are not always yet explicitly distinguished or separated regarding to agricultural drought disaster risk assessment, which restricts the understanding of processbased mechanism of disaster risk dramatically. The major focuses of studies on drought disaster risk assessment range from statistical analysis and probability analysis to prediction of drought hazard or disaster losses, whilst few attached importance to the quantitative separation of physical and social vulnerabilities. 
The goals of this paper are to explore and quantify the physical vulnerability and drought hazard and risk. The detailed process is as follows: Firstly, agricultural drought disaster risk assessment model is developed based on the physical vulnerability of crops. This model covers agricultural drought hazard possibility distribution estimation, physical vulnerability curve estimation, land usebased exposure assessment, coping capacity integration and risk assessment. Secondly, a case study on two typical wheat species in China is carried out, and a quantitative physical vulnerability to agricultural drought hazards is analyzed using a physicalprocess based crop growth model and Erosion Productivity Impact Calculator (EPIC). At the beginning, the onedimension (point scale) crop growth model EPIC with a capability of depicting water stress and temperature stress is extended to a twodimension (area scale) model, and the Spatial Erosion Productivity Impact Calculator (S-EPIC) can be used to simulate crop yield under different intensity scenarios of drought. Two extreme scenarios for crop growth and yield simulation were investigated. Finally, the maps of drought risk occurring every 2, 5, 10 and 20 years were charted separately based on the calculated probable distribution of drought hazard and the vulnerability curve. The results reveal that the wheat yield loss caused by drought is decreased from the northwestern part to the southeastern part of China.

Key words: vulnerability, risk assessment, drought disaster, EPIC model, wheat yield, China