Abstract In this study, Standardized Effective Precipitation Index SEPI (for 3 and 6-month time scales) derived from Standardized Precipitation Index (SPI) methodology has been employed for agricultural drought frequency identification of two meteorological stations in northern Iraq. Since drought is mainly characterized by severity and duration, copula methodology is adopted to analyze the bivariate frequency of these two characteristics. Three Archimedean copulas were used and compared using root mean square error (RMSE), Bayesian information criteria (BIC), and Nash–Sutcliffe efficiency (NSE) performance criteria to select the most suitable copula model. The results showed that Generalized Pareto and Lognormal were the best qualified marginal distributions for drought duration and severity, respectively. Gumbel copula was the most suitable function to joint modeling of drought characteristics. The novelty in this study is the use of the meteorological variable (effective precipitation) in the bivariate frequency analysis of agricultural drought characteristics based on copula. Since the SEPI requires only effective precipitation data, it can be used to overcome the limitations of unavailability or lack of soil moisture data as in the study area.
展开▼
