Prediction of meteorological drought in arid and semi-arid regions using PDSI and SDSM: a case study in Fars Province, Iran

Abstract

Drought is one of the most significant environmental disasters, especially in arid and semi-arid regions. Drought indices as a tool for management practices seeking to deal with the drought phenomenon are widely used around the world. One of these indicators is the Palmer drought severity index (PDSI), which is used in many parts of the world to assess the drought situation and continuation. In this study, the drought state of Fars Province in Iran was evaluated by using the PDSI over 1995–2014 according to meteorological data from six weather stations in the province. A statistical downscaling model (SDSM) was used to apply the output results of the general circulation model in Fars Province. To implement data processing and prediction of climate data, a statistical period 1995–2014 was considered as the monitoring period, and a statistical period 2019–2048 was for the prediction period. The results revealed that there is a good agreement between the simulated precipitation (R2>0.63; R2, determination coefficient; MAE<0.52; MAE, mean absolute error; RMSE<0.56; RMSE, Root Mean Squared Error) and temperature (R2>0.95, MAE<1.74, and RMSE<1.78) with the observed data from the stations. The results of the drought monitoring model presented that dry periods would increase over the next three decades as compared to the historical data. The studies showed the highest drought in the meteorological stations Abadeh and Lar during the prediction period under two future scenarios representative concentration pathways (RCP4.5 and RCP8.5). According to the results of the validation periods and efficiency criteria, we suggest that the SDSM is a proper tool for predicting drought in arid and semi-arid regions.