Climate Change Impact on Agricultural Land Suitability: An Interpretable Machine Learning-Based Eurasia Case Study

Abstract

Climate change is an issue for the food we eat everywhere in the world. It is changing the land where we grow our food. The way we get water to our crops is also being affected by climate change. We need to know what is going on with climate change so we can keep growing food in a way that’s good for the planet. In some places people are already having a time with money and social problems. Climate change is especially important for these places. This study is, about how climate change’s affecting the land where we grow our food in Central Eurasia. We are talking about climate change. How it affects our food. Climate change is a problem that we need to deal with so we can keep eating food. This region is really sensitive to weather changes so we want to know how it affects our food. This is a problem for the land where we grow food especially in Central Eurasia. We need to think about what this means for the food we eat. We are working to meet the United Nations goals to reduce hunger. People not getting enough nutrients. Our goal with this research is to figure out the risks of the land not being good, for farming and how we will get water to our crops when the weather is different. We use machine learning methods to see how good the land is for farming under different carbon emission situations. The model we made can figure out what kind of land is suitable for what. It is right about 86 percent of the time. It also gets the details right 72 percent of the time which is really good. The model is easy to understand so we can see how things like weather affect the land and what we can grow on it. When we looked at the results we found that some areas are in trouble like Eastern Europe and Northern Asia. In these places the changing climate is making the land less good for farming which’s a big problem, for growing food. The findings offer valuable guidance for policymakers and stakeholders by supporting informed decision-making on resource allocation, including water management and fertilizer distribution. Overall, this study demonstrates the effectiveness of interpretable machine learning as a decision-support tool for anticipating and mitigating the impacts of climate change on agricultural land suitability and long-term food security.