Application of Gene Expression Programming in Drought Prediction (Case Study: Tabriz Synoptic Station)

The water body of Urmia Lake in recent years has been faced with a significant reduction in surface area and volume due to various reasons such as successive droughts. Therefore, in this study, the situation of drought was evaluated at the synoptic station of Tabriz as one of the important stations of Urmia Lake basin in different time scales using the standardized precipitation-evapotranspiration index (SPEI) and the gene expression programming (GEP) model. For this purpose, the SPEI index was used for drought monitoring at 1, 3, 6, 12, 24, and 48 months during the 53-year statistical period. The results showed that three long periods of drought related to the years 1961-1963, 1986-1992, and 1997-2009 are available during the statistical period. According to the results, the prediction accuracy is directly related to increasing the scale of SPEI and increased by increasing the scale of SPEI, so that the correlation coefficient in the test stage in the one-month scale (SPEI1) increased from 0.203 to 0.988 at 48-month scale (SPEI48) and the overall accuracy of the model increased from 57.1 in SPEI1 to 94.2 % in SPEI48.