Monthly Precipitation Forecasting in Ponel Raingauge Station Using Linear Regression and Singular Spectrum Analysis

Rainfall/ precipitation, as one of the most important inputs to hydrological systems, is one of the most significant parameters in many hydrological models. In the recent decades, different types of forecasting methods are employed for forecasting and analyzing monthly precipitation rates. Linear regression is one of the methods are being used for this purpose. Recently, the use of singular spectrum analysis in water resources studies for removing random components of hydrological series has extensively increased. The main objective of this study is to investigate the use of linear regression coupled with singular spectrum analysis for monthly precipitation forecasting. The monthly data of Ponel raingauge station which span the period from 1991 to 2010 (i.e. 20 years) were used to develop the proposed model. The proposed model was compared with regular linear regression and the results indicated the superiority of combined linear regression and singular spectrum analysis models.