Application of Wavelet Neural Network for Predicting Standardized Precipitation Index

Drought is one of the most important climatic phenomena which occures in all climate conditions and regions of the earth. Drought forecasting, therefore plays an important role in designing and management of natural resources and water resources systems, assessing plant evapo-transpiration. For this purpose, in this study, data from four meteorological stations nourabad, borujerd, aleshtar and doroud in Lorestan province, on time scales of 6 and 12 months were used to analyze drought by using standardized precipitation index SPI. Then, droughts were evaluated using neural network model estimation. The results showed, Boroujerd and Doroud stations have the longest drought period, and severe drought is recorded in Nourabad station. The results of the survey showed that Boroujerd Station had maximum amount ofdrought months occurred during the drought period. The results of using wavelet neural networkmodel showed best estimation of SPI for Doroud station than others in both time scales. In conclusion, the results showed more accuracy of wavelet neural network model in estimation of long-term drought, and the use of wavelet neural network model can estimate the drought effectively, Which in return facilitates the development and implementation of management strategies to avoid drought.