Digital Mapping of Plant Available Water Using PlanetScope Imagery and Pedotransfer Functions

Soil texture is a fundamental physical property that strongly influences other soil properties critical to agricultural productivity. Comprehensive information on soil properties and their spatial variability is essential for the implementation of effective soil management strategies. Digital soil mapping typically relies on data collected from discrete sampling points. Therefore, estimating topsoil properties from satellite data at a high resolution remains a significant challenge. Nevertheless, estimative models can be developed by relating soil physical parameters to remotely sensed and topographic variables.

In this study, pedotransfer functions were developed using multiple linear regression (MLR) to derive soil texture maps for sand silt clay and organic matter (OM). Functions were based on PlanetScope imagery and topographic data derived from a digital elevation model (DEM) at 3 m resolution. The models established statistical relationships between soil properties and selected predictor variables. Model performance was assessed using the coefficient of determination R², showing satisfactory results: 0.64 for clay, 0.64 for OM, and 0.82 for sand.

Finally, management zones based on plant-available water derived from the updated European hydraulic pedotransfer functions (PTFs) were delineated using spatial fuzzy C-means clustering (SFCM), providing a practical framework for precision agriculture and sustainable land management.

Index Terms—precision agriculture, soil texture, multiple linear regression, spatial fuzzy C-means clustering

Acknowledgements: The authors acknowledge the support of SANTO, from Universidad Politécnica de Madrid (RP220220C024) and to NetLIFE-CODES from Agencia Estatal de Investigación (PID2024-157869NB-I00)