The Effect of Cross-Correlation between SPEI Variables in Long-Term Drought Analysis

In this study, a comprehensive evaluation of the effect of cross-correlation coefficient between precipitation and evapotranspiration on the performance of SPEI in forecasting the long-term behavior of agricultural drought characteristics in 39 synoptic stations located in different climates and altitudes throughout Iran has been performed, and its performance has been compared with SPI. For this purpose, stochastic precipitation and evapotranspiration time series (10,000 pairs) were generated using the Lag-one Autoregressive Multiple-Site Model (Multi-AR(1)) for a wide range of correlation coefficients of the above two variables and then used to monitor and determine different drought characteristics. The results of this study showed the systematic and dependent behavior of different characteristics of drought with cross-correlation between precipitation and evapotranspiration. So that, the performance of the two SPEI and SPI indices was nonlinearly close to each other (R=0.85) by increasing the cross-correlation between the two variables (|R|⇒1), and for cross-correlation of zero, the maximum difference between the drought characteristics was obtained for the two indices. Finally, the results of this study have been used as a comprehensive guide in the accurate and realistic inference of drought events and characteristics and can be an effective aid in interpreting drought outcomes with the SPEI index.