The level of cultivated land security risk in China based on the PNN network

Currently, cultivated land areas throughout China are differentially susceptible to a variety of natural hazards, social risks, and economic challenges. A risk evaluation model is developed in order to capture the degree of risk to the quality of cultivated land and cultivated land areas in China. Based on the Probabilistic Neural Network (PNN), the model seeks to provide insights/recommendations on the use of cultivated land in China, by employing a set of environmental indicators. Threshold risk levels are established in order to reduce the vulnerability of cultivated lands. Five cultivated land area risk categories are defined. It is shown that more resources should be dedicated to protecting the environment and cultivated land areas in China. By reducing the risk of natural hazards and socio-economic pressures, it is expected that the quality of cultivated land in China can be improved. Finally, regions primarily dedicated to food production should be given additional protection from natural and anthropogenic risks.