Prediction of arable land change with artificial neural network model for Chongqing Municipality in China

Arable land has been decreasing due to rapid population growth and economic development as well as urban expansion. To obtain a better understanding of controlling land use and to design mechanisms to ensure sustainable land management, an accurate prediction of arable land is a key issue fundamentally. In this study, artificial neural network (ANN) model is applied to estimate the arable land change for Chongqing Municipality in China. The prediction is implemented using a feed-forward neural network, trained by back-propagation algorithm. In order to investigate the socioeconomic influences on arable land reduction, the ANN model is trained based on population, gross domestic production and fixed assets investment. Further, the prediction results from ANN model and linear regression model are compared to test the performance of model validation. Consequently, artificial neural network shows the ability to catch non-linear relationships between arable land change and socioeconomic factors and to predict arable land change with a high degree of accuracy. In conclusion, the ANN model is applicable of predicting arable land change and some suggestions would be provide for researchers and decision makers.