Geospatial analysis of agricultural drought vulnerability using a composite index based on exposure, sensitivity and adaptive capacity

Vulnerability of agriculture to drought conditions is closely related to the challenges of sustainable agriculture and food security. Information on agricultural drought vulnerability status at different scales is crucial for prescription, development and implementation of long-term drought management measures. This study is on developing a composite index for measuring crop-generic agricultural drought vulnerability at disaggregated level, in the erstwhile Andhra Pradesh state of India. By analyzing the data on soils, irrigation, farm holdings, rainfall, rainy days, cropping pattern and satellite-based crop condition, 22 input indicators for 1038 spatial units called Mandals (within-the-district administrative units) were generated and segregated into exposure, sensitivity and adaptive capacity components of vulnerability. The composite index was computed for each component by generating the weights for inputs through variance approach. Agricultural drought vulnerability index was generated for each Mandal by combining the three component indices and beta distribution was fitted to it to generate vulnerability classes. Results indicate that the vulnerable group of Mandals represents 57% of agricultural area and 67% of rain-fed crop area of the state. The geographical pattern of vulnerability is well captured by the composite index. Districts dominant with vulnerable Manclals showed frequent drought occurrence and higher crop yield variability. The vulnerability map can be instrumental in prioritizing the areas for drought management interventions. Since the methodology was applied over a large geographic area with diverse conditions of weather, soil and crop, its utility over other geographies is also established. Indicators selection and weights generation are the critical aspects for further research. (C) 2015 Elsevier Ltd. All rights reserved.