Variations and Sensitivity Analysis on Aridity Index (AI) in Some Climate Samples in Iran

As the ratio of evapotranspiration to precipitation, Aridity Index (AI) is used to quantify the value of aridity. In this study, sensitivity analysis was performed on AI using the partial derivative method based on 7 meteorological variables including Pre, TMax, TMin, ea, U2 and RS in 5 different climates based on De Martonne’s classification method during the period 1990-2019 on annual time scale. Mann-Kendall test and Sen’s slope estimation method were employed for the purpose of trend detection of AI variations. CanESM2 atmospheric circulation method under two scenarios including RCP4.5 and RCP8.5 and Statistical Downscaling Model (SDSM) were used to predict AI variations in the future period of 2020-2050. The Results of AI trend analysis indicated increased aridity in all stations excluding Mashhad. The Results of sensitivity coefficient showed that precipitation, among all other climate variables, contributed the most to AI variations, as 10% increase in precipitation resulted in a decrease in AI in Rasht, Ilam, Yasouj, Mashhad and Kerman by 10.72%, 11.81%, 12.48%, 11.14% and 12.82%, respectively. AI had then the highest sensitivity to TMax and TMin in Ilam and Rasht stations, to ea in Kerman and Mashhad stations, and to Rs in Yasouj station. Throughout the next 30-year period, it is expected that AI values increase in all observed stations except Yasouj. The highest and lowest annual AI increase rates throughout this period would take place in Rasht station under scenario RCP8.5, and in Mashhad station under scenario RCP4.5.