Joint Deficit Index (JDI) is computed based on combination of the 12 modified Standardized Precipitation Index (SPImod) corresponding to the 12 time scales 1-12 month using the empirical and theoretical copula functions. Researchers suggest calculating the JDI as empirically due to difficulty and time-consuming in computing its theoretical form. The aim of this paper is to compare the empirical and theoretical copula-based joint deficit indices at 42 weather stations in Iran for the common period 1966-2010. For calculating the theoretical JDI (TJDI), we chosen the best copula function from the four candidate functions including Student’s t, Clayton, Gumbel and Frank using two information criteria. In contrast, calculation of the empirical JDI (EJDI) not need fitting any copula function to data, and therefore it can be computed more simply than TJDI. Results showed that the Student’s t copula is the best function at all selected stations; therefore, the theoretical JDI was computed based on this copula function. Although both indices, i.e. EJDI and TJDI showed a similar behavior of drought severity, the EJDI faced two problems in all selected stations: 1) The lowest value in the EJDI time series is frequently repeated in different parts of its time series, 2) The EJDI was not able to identify the peak of drought severity during the critical drought periods. In contrast, the TJDI did not face the mentioned issues and therefore that is suggested for monitoring overall status of droughts in the study area.
Manuscript profile
