Simulating and predicting soil water dynamics using three models for the Taihu Lake region of China

Abstract Drought stress under a changing climate can significantly affect agricultural production. Simulation of soil water dynamics in field conditions becomes necessary to understand changes develop irrigation guidelines. In this study, three models including Auto-Regressive Integrated Moving Average (ARIMA), Back-Propagation Artificial Neural Network (BP-ANN), and Least Squares Support Vector Machine (LS-SVM) were used simulate the content 0–14 cm 14–33 layers across Taihu Lake region China. Rainfall, evaporation, temperature, humidity wind speed that considered BP-ANN LS-SVM, but not ARIMA. The results showed variability layer was greater than cm. Correlation coefficients (r) between simulations observations highest (0.9827) using LS-SVM layer, while they lowest (0.7019) ARIMA layer; no significant difference r values observed two with model. Compared other models, model seems be more accurate for forecasting moisture. suggested agro-climatic data predict severity drought provide guidance increase crop production