Soil water content forecasting by support vector machine in purple hilly region

Soil water distribution and variation are helpful in predicting and understanding various hydrologic processes, including weather changes, rainfall/runoff generation and irrigation scheduling. Soil water content prediction is essential to the development of advanced agriculture information systems. In this paper, we apply support vector machines to soil water content predictions and compare the results to other time series prediction methods in purple hilly area. Since support vector machines have greater generalization ability and guarantee global minima for given training data, it is believed that support vector machine will perform well for time series analysis. Predictions exhibit good agreement with actual soil water content measurements. Compared with other predictors, our results show that the SVMs predictors perform better for soil water forecasting than ANN models. We demonstrate the feasibility of applying SVMs to soil water content forecasting and prove that SVMs are applicable and perform well for soil water content data analysis.