<link rel="stylesheet" href="/res/styles.39be4579664c3a48.css">
Artículo

Satellite-based monitoring of forage quality in grasslands of the United Kingdom using sentinel-2 data and random forest regression

Irisarri, Jorge Gonzalo Nicolás Icon; Texeira González, Marcos Alexis Icon; Harris, P. ; Pembleton, K. G.

Otros responsables

Fecha

12/2025

Editorial

Frontiers Media

Revista

Frontiers in Veterinary Science

Título del libro

Nombre del evento

Fecha del evento

Institución organizadora

ISSN

eISSN

2297-1769

ISBN

Idioma

Inglés

Tipo del evento

Clasificación temática

Resumen

Introduction: In the temperate grasslands of the UK, forage quality is a keyfactor influencing both animal performance and environmental impact. Becauseforage quality strongly affects rumen fermentation, improving it can reduceenteric methane emissions and mitigate animal nutritional stress. However,large-scale monitoring of forage quality remains limited due to the relianceon destructive, labor-intensive, and costly sampling methods. Remote sensingoffers a promising alternative for scalable monitoring.Methods: We explored an indicative approach combining optical remote sensing(Sentinel-2) with random forest regression (RFR) models to predict four criticalforage quality attributes: crude protein (CP), water-soluble carbohydrates (WSC),neutral detergent fiber (NDF), and acid detergent fiber (ADF). Calibration andvalidation were performed using >9,500 georeferenced observations collectedbetween 2020 and 2022 at the North Wyke Farm Platform in southwest UK.Forage quality was measured using near-infrared (NIR) sensors mounted onagricultural machinery across paddocks containing permanent and improvedpastures. Sentinel-2 spectral predictors included visible, NIR, and red-edgebands, and model performance was evaluated using R² and RMSE metrics.Results: Model performance was strong across all four forage quality attributes,with R² values ranging from 0.77 to 0.86 and consistently low RMSE values,indicating high predictive accuracy. Red-edge and NIR wavelengths were themost influential predictors. Improved pastures generally exhibited higher foragequality—characterized by lower ADF and higher WSC concentrations—thanpermanent pastures. Model-predicted seasonal changes were modest, whereasspatial contrasts between paddocks were much more pronounced.Discussion: The calibrated models are suitable for forage systems with speciescomposition and quality ranges similar to those represented in our datasetbut should not be directly applied to other forage types without recalibration.Overall, this work demonstrates the potential of Sentinel-2 remote sensingcombined with machine-learning approaches for tolerably accurate, large-scale forage quality monitoring. Such advancements could help improve grazingmanagement, support nutritional planning, and contribute to efforts aimed atreducing methane emissions from livestock systems.

Colecciones

Aprobación

Revisión

Complementado por

Referenciado por

Licencia Creative Commons

Excepto donde se indique lo contrario, la licencia de este ítem se describe como by-nc-sa