Authors: Yanshen Wu*, School of Geography, Faculty of Geographical Science, Beijing Normal University, Hao Guo, School of Geography, Faculty of Geographical Science, Beijing Normal University, Anyu Zhang, School of Geography, Faculty of Geographical Science, Beijing Normal University, Jing'ai Wang, School of Geography, Faculty of Geographical Science, Beijing Normal University; Key Laboratory of Environmental Change and Natural Disaster, MOE, Beijing Normal University
Topics: Hazards and Vulnerability, Hazards, Risks, and Disasters, Agricultural Geography
Keywords: Vulnerability curve, Drought, Key point, K-means clustering, Spatial variation
Session Type: Lightning Paper
Start / End Time: 8:00 AM / 9:40 AM
Room: Marshall North, Marriott, Mezzanine Level
Presentation File: No File Uploaded
As an essential component of drought risk, crop-drought vulnerability refers to the degree of adverse response of crop to a drought event. The degree of response varies with the drought intensity, and the impact of the environment. Therefore, quantifying the crop-drought vulnerability and then identifying their spatial distribution patterns will contribute to the understanding of vulnerability and the development of risk reduction strategies. This paper selects the European winter wheat growing area as the study area and the 0.5° grid as the basic assessment unit. The drought vulnerability curves of winter wheat are established based on the EPIC model. By analyzing their characteristics of key points and the shape, K-means clustering is further performed to identify areas with similar vulnerability curves and to explore the local environmental characteristics. The results show that: (1) The regional yield loss rate starts to increase rapidly from 0.13 when the drought intensity index reaches 0.18, and it turn into a relatively stable stage with the value of 0.74 when the latter reaches 0.66; (2) The stage transitions of the vulnerability curve lag in the Alpine of the Mediterranean, indicating a stronger resistance to drought in the system，which is on the contrary in the Pod Plain; (3) The vulnerability curves can be divided into five clusters. From cluster 1 to cluster 5, shapes of curves transit from concave to convex as the vulnerability increases, corresponding to the spatial distribution from low latitude to high latitude and from mountain to plain.