Drought analysis using bivariate copulas in the Indian secondary cities

Droughts are considered essential concerns in water resource planning and management. Observed and projected changes in summer monsoon precipitation might influence the drought occurrence in India; however not well known. Using precipitation from ERA5 reanalysis and CMIP6 models, we analyze the drought frequency in Indian secondary cities (Bhopal, Bhuj and Kozhikode) in the observed (1951-2020) and the future warming climate. A copula-based methodology will be used to construct a joint drought duration and severity distribution. First, the paired drought duration and severity data are abstracted from the observed drought events, defined as the SPI (Standardized Precipitation Index). Based on the goodness of fit test (GFT), the found relevant distributions are used to model the drought duration and severity, respectively. Secondly, copulas are employed to construct the dependence structure for drought duration and severity. The bivariate probabilistic properties of droughts, such as joint probabilities and bivariate return periods, will also be investigated to demonstrate comprehensive drought assessments. The analysis will be beneficial in determining a given water supply system's capabilities and required auxiliary water resources during severe droughts; hence this information is useful in providing continuous water supply to the residents.