Long-Lead Rainfall Prediction Based on Climate Patterns of Tele-Connection, A Case Study: Aharchay Basin

Abstract

This study aims to develop a seasonal rainfall prediction model for the Aharchay Basin, northwest of Iran. The model is based on climate patterns of tele-connection including sea level pressure (SLP) and sea surface temperature (SST) over the period from 1965 to 2005. The models cover both wet (from December to May) and dry (from June to November) seasons. For this purpose, the climatic patterns affecting the climate of the northwest of Iran were initially determined. In the second stage of the study, the correlation coefficient analysis and the Gamma Test (GT) technique were used to select the best predictors and to determine the best combination of the variables. The results revealed that the gamma test model outperformed the other model in determining the required input variables and their best combination. The seasonal rainfall in the basin was also predicted using the Support Vector Machines (SVM) and the results thus obtained were compared with those of the multivariate linear regression model as a benchmark to show the performance of the SVM model in rainfall prediction.

Keywords

• Rainfall Prediction
• Climate Patterns of Tele-Connection
• Support Vector Machines
• Gamma Test