Bivariate assessment of socioeconomic drought events based on an improved socioeconomic drought index

Abstract

Drought is a complex natural disaster that may have destructive impacts on human life and production. There are many types of droughts, among which socioeconomic drought can directly affect socioeconomics. Due to population growth and increasing water demand, socioeconomic drought has become increasingly important. The socioeconomic drought index (SEDI) provides a good framework for analyzing socioeconomic droughts, but it has some deficiencies such as ignoring watershed resilience. This study sought to improve the SEDI framework through adding water resources system resilience index (WRSRI) and run theory to the original SEDI method. Such improvement can help to identify the start and end times of socioeconomic drought events (SEDEs) more accurately. Taking the East River Basin (ERB) as the study area, socioeconomic drought characteristics from 1954 to 2014 were evaluated using the improved SEDI method and bivariate copula models. First, an improved SEDI method was used to identify SEDEs over the past 60 years, which could perform better in identifying drought than the original one. The variable distribution was more continuous after improvement, so that the copula model could be used for comprehensive drought assessment in the ERB. There was a large difference between the univariate and bivariate return periods. Comparing SEDEs with hydrological droughts and socioeconomic data, we concluded that although hydrological drought was the main cause of socioeconomic drought, the two were not completely synchronized. In the ERB, severe SEDEs could lead to economic slowdown, while minor water deficit events could hardly affect the socioeconomic development.