Abstract: This description relates to a predictive model for estimating the proportions of sand, silt, and clay in a soil sample based on near-infrared spectroscopy (NIRS) performed on the soil samples. The soil texture prediction model is trained using benchmark NIRS test data and collected spectra of soil samples for which the benchmark test data has been collected. The estimates produced by the model can be used to determinate a variety of properties of the soil sample, such as hydrological properties and soil quality. Knowledge of these properties are useful for informing decisions that agronomists, producers, or farm managers make throughout the year. For example, agronomists use the determined properties to infer how the soil will perform under any given treatment or application, for example applying seeds, fertilizer, pesticide, and herbicide.

Abstract: This description relates to a predictive model for estimating the proportions of sand, silt, and clay in a soil sample based on near-infrared spectroscopy (NIRS) performed on the soil samples. The soil texture prediction model is trained using benchmark NIRS test data and collected spectra of soil samples for which the benchmark test data has been collected. The estimates produced by the model can be used to determinate a variety of properties of the soil sample, such as hydrological properties and soil quality. Knowledge of these properties are useful for informing decisions that agronomists, producers, or farm managers make throughout the year. For example, agronomists use the determined properties to infer how the soil will perform under any given treatment or application, for example applying seeds, fertilizer, pesticide, and herbicide.