Mapping the Future of Soil: AI & GIS for Dynamic Land Evaluation | COSA Project Research Spotlight (6/6)

Everything in agriculture starts with the land. But how do we accurately measure the true potential of our soil in a constantly changing climate?

For our sixth and final Q1 publication highlight from the COSA project, our research team merged spatial geography with artificial intelligence to revolutionize land evaluation. We are thrilled to share this high-impact research published in the journal Agriculture.

The Science Behind the Soil

Traditional “land bonitation” (the agronomic evaluation of land quality and agricultural potential) often relies on static, historical data. Our team developed a cutting-edge GIS-Driven, Machine Learning-Enhanced Framework that makes this process adaptive. By combining Geographic Information Systems (GIS) with machine learning algorithms, the framework continuously analyzes complex spatial and environmental data to evaluate land potential dynamically.

Key Advances and Real-World Impact

• Precision Crop Matching: By understanding the exact, up-to-date health and potential of specific land parcels, farmers can perfectly match crop varieties to the optimal soil conditions.
• Dynamic Adaptation: As climate, rainfall, and soil conditions shift, the machine learning models adapt, providing stakeholders with highly accurate, future-proof land quality scores.
• Sustainable Land Management: This data-driven approach prevents land degradation and over-farming, ensuring that our agricultural practices remain highly productive and environmentally sustainable for generations to come.

Huge congratulations to the authors on this incredible contribution to digital agronomy!

Thank you to everyone who followed along with our Q1 publication series. The COSA Project is proud to continue pushing the boundaries of what is possible in Smart Agriculture!

Read the full open-access publication here:
Vaduva, B., Avram, A., Matei, O., Andreica, L., & Rusu, T. (2025). A GIS-Driven, Machine Learning-Enhanced Framework for Adaptive Land Bonitation. Agriculture 2025, 15, 1735.
https://www.mdpi.com/2077-0472/15/16/1735