Meeting the food demands of a growing global population with limited agricultural resources is one of the greatest challenges of our time. Crop yields are highly sensitive to weather variability and extreme conditions, which are becoming increasingly common due to climate change. Accurate regional yield forecasts are therefore crucial for helping farmers adapt, ensuring food security, and strengthening the resilience of agriculture.
In her doctoral thesis at Lund University, physical geographer Xueying Li has focused on refining both short- and long-term yield predictions in southern Sweden. Her research integrates AI, satellite data, and climate information into both data-driven and process-based models to better understand ecosystem dynamics. The work resulted in four scientific articles, each addressing different aspects of Swedish agriculture—including yield forecasts for spring barley and winter wheat.
“My findings provide a robust framework for improving agricultural management in southern Sweden. These forecasts enable farmers to optimize irrigation and harvest timing. Agricultural companies and policymakers can also anticipate fluctuations and adjust import and export strategies in advance, which strengthens the resilience of the food system,” says Xueying Li.
Higher Yields in the Future
Using the ecosystem model LPJ-GUESS (Lund-Potsdam-Jena General Ecosystem Simulator), Li was also able to predict crop yields under future climate scenarios. LPJ-GUESS simulates interactions between vegetation, soil, and climate on regional to global scales, making it possible to forecast long-term yields far into the future.
“What surprised me most was that rising carbon dioxide concentrations and higher temperatures during the growing season are expected to result in increased yields for both spring barley and winter wheat by the end of this century,” says Xueying Li.
Supporting Farmers in Decision-Making
Southern Sweden has the largest cultivated area and the greatest crop diversity in the country. Yield prediction in this region is of utmost importance for understanding how crop production may be affected by climate change. The results offer powerful guidance for improving agricultural management and safeguarding food security in a changing climate.
“This work will enable farmers and policymakers to make more informed decisions. I hope that, in the not-too-distant future, these forecasts can be used operationally by Swedish farmers. While the model is primarily designed to predict yields in Skåne, it has the potential to be applied at a European level,” says Li.
Xueying Li’s doctoral thesis: Improving crop yield prediction in Sweden using satellite remote sensing and the ecosystem model LPJ-GUESS
