Application of Stochastic Models for Rainfall and Drought Frequency Analysis in Sudan

Abstract

Sudan is one of the countries which economy depends on rain- fed agriculture with recurring cycles of natural drought. The drought phenomenon has significant widespread impacts on the community, environment and economy. The main objectives of this research are to study the characteristics of rainfall in Sudan, find suitable tools for drought characterization to be used during drought periods and propose monthly rainfall forecasting methods accuracy with inspection of the model forecasting ability. As time series analysis and forecasting have become a major tool in different applications in hydrology and environmental management fields, linear stochastic models known as ARIMA and multiplicative Seasonal Autoregressive Integrated Moving Average (SARIMA) models were used to simulate droughts based on the procedures of the models developments. The models were applied to simulate droughts using standardized precipitation index (SPI) series in many rainfall stations in the Sudan. The SPI index was used as a drought indicator for drought forecasting due to its advantages over other drought indices. These models were also used for simulating and forecasting the monthly rainfall in many rainfall stations across Sudan. The results of this research proved that the linear stochastic models (ARIMA) can be used for the rainfall stations for predicting SPI time series of multiple time scales to detect the drought severity in future. A time series model for monthly rainfall stations across Sudan, taking Gadaref station as a typical station was adjusted, processed, diagnostically checked and a typical SARIMA (0, 0, 0) (0, 1, 1)12 model was established. The model was used to forecast three years monthly rainfall values. The stochastic models developed for the stations can be employed for the development of a drought emergency management plan so as to ensure sustainable water resources management in these stations. The model was found appropriate to forecast the monthly rainfall in Gadaref station and assist decision makers to establish priorities for water demand, storage, distribution, and disaster management