By Reagen Tibbs
I recently had the opportunity to attend the 2026 CANVAS Conference in Portland, Oregon. This conference is the joint annual meeting of the American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America. It brings together researchers, Extension educators, crop advisers, and industry professionals from across the country and the world. The conference featured presentations on crop production, soil health, precision agriculture, nutrient management, emerging technologies, and much more. While the topics covered a wide range of disciplines, several presentations highlighted research trends that may influence future management decisions on Illinois farms. The following are several research projects that stood out because of their relevance to crop production, technology adoption, and farm management in the Midwest.
Drones and Remote Sensing for Tar Spot Detection
Tar spot remains one of the most significant foliar diseases facing corn producers in the Corn Belt. Researchers from Michigan State University evaluated whether drone-based multispectral imagery could be used to detect tar spot severity, track disease progression, and assess fungicide efficacy. Using drone imagery collected over two growing seasons, researchers examined several vegetation indices and machine learning models to determine whether disease severity could be identified remotely. Disease pressure varied considerably among locations, with only one site in one year experiencing severe infections. While NDVI values generally declined as tar spot severity increased, researchers noted that NDVI also declines naturally as corn reaches maturity, making it difficult to separate disease effects from crop development. Machine learning models were unable to consistently predict disease presence when trained on data from one year and applied to another. For Illinois producers, these findings suggest that remote sensing technologies may eventually become useful scouting tools, but they are not yet a replacement for boots-on-the-ground field observations. Hybrid resistance, regular scouting, and timely fungicide decisions remain the foundation of effective tar spot management.
Can Artificial Intelligence Improve Nitrogen Recommendations?
Artificial intelligence continues to receive considerable attention in agriculture, particularly for its potential to improve nutrient management decisions. Researchers from the University of Minnesota evaluated a system that combined process-based crop models with machine learning to predict site-specific optimum nitrogen rates within individual areas of a field. Using data from on-farm nitrogen response trials, including weather conditions, field operations, and yield information, researchers developed models capable of estimating economically optimal nitrogen rates across field grids. The most successful models were those that incorporated both agronomic knowledge and machine learning techniques rather than relying solely on artificial intelligence. Weather variables, especially precipitation, had the greatest influence on prediction accuracy, while field topography also played an important role. For Illinois producers, this research highlights the potential for future precision nitrogen management tools to improve efficiency and profitability. However, the findings also reinforce that agronomic expertise and an understanding of local growing conditions remain essential when interpreting recommendations generated by artificial intelligence systems.
Source : illinois.edu