<p>Agricultural futures contracts are increasingly traded in agricultural markets, providing an effective risk management tool for agricultural producers, traders, and investors. However, risk management in agricultural futures markets has become increasingly complex, as agricultural commodity prices are affected by various factors, including climate change, supply chain issues, and changes in policies and regulations. Therefore, it is crucial to accurately assess and manage the risk of agricultural futures, but there are few studies on calculating Value at Risk (VaR) and Expected Shortfall (ES) in the agricultural futures market. This paper uses a Monte Carlo simulation based on Gaussian Mixture Model to calculate VaR and ES of agricultural futures. Firstly, we introduce the theory of Gaussian Mixture Model and EM algorithms and establish the Monte Carlo simulation model of VaR and ES based on a method of volatility weight adjustment, then we select nine kinds of agricultural futures data to carry out the empirical analyses based on Python, including the goodness of fit test of simulated return distribution and the comparative analyses of VaR and ES calculated by different models. Finally, based on the results of the backtesting, it is found that the Gaussian Mixture Model is the most effective in calculating the risk of agricultural futures compared to the traditional method.</p>