132-2022 - HD Grid Soil Sampling and Analysis for Advancing Agronomy through HyperLayer Data Collection and Analytics

The overarching goal of this project is to develop predictive algorithms based on machine learning (ML) for estimating the concentrations of plant available nutrients in the soil, instant identification of yield limiting factors, calculating best achievable yield goals, and automated calculation of site-specific fertilizer prescriptions. Multiple layers of geospatial data, such as multi- and hyperspectral imagery, thermal imagery, microwave soil moisture measurements, ground penetrating radar data, high density soil tests, soil compaction, elevation data and field surface analysis, gamma-ray soil spectrometry analysis, soil electrical conductivity, and yield monitoring data, will be used to build new ML predictive algorithms, which will allow for automated analysis of various agronomic field characteristics including plant available soil nutrients and water, instant identification of yield-limiting factors, automated calculation of optimal fertilizer rates, and site-specific estimation of best achievable yield potential.