Research proposal for seasonal rainfall forecasting method in order to make irrigation plans for Ca river basin

Abstract

     In recent years, the climate change has been one of the hot issues that need a lot of attention of researchers, particularly those related to rainfall for planning for irrigation in order to raise the management efficiency of the irrigation systems' operation. The change in seasonal rainfall which directly affects the irrigation regime and the water source is the basic data for planning of irrigation systems. The question is whether the changes in the amount of seasonal rainfall can be predicted with accuracy at acceptable levels. In this article, the Adaptive Neuro-fuzzy Inference System (ANFIS) model has been proposed to develop a precipitation model for the Ca river basin. Data for calculations were obtained at four representative meteorological stations in the Ca river basin from 1975 to 2014. Different seasonal rainfall forecast models were constructed with different input rainfall parameters, the predictive performance of these models is compared through statistical parameters to identify and propose models with the best prediction. The results show that the M4 model gives the best and most reliable results for 3-month and 6-month seasonal rainfall forecasts for the study area.