Computer Modeling of the Nutritional Quality of Forage Trees in Silvopastoral Systems by Remote Sensing

• Grant number: 25/20196-4
• Support Opportunities: Scholarships abroad - Research Internship - Scientific Initiation
• Start date: January 01, 2026
• End date: March 06, 2026
• Field of knowledge: Agronomical Sciences - Animal Husbandry - Pastures and Forage Crops
• Principal Investigator: Jozivaldo Prudêncio Gomes de Morais
• Grantee: Giovani Vignola Tirloni
• Supervisor: Virginia Garcia Millan
• Host Institution: Centro de Ciências Agrárias (CCA). Universidade Federal de São Carlos (UFSCAR). Araras , SP, Brazil
• Institution abroad: Universidad de Málaga (UMA), Spain
• Associated to the scholarship: 24/13746-5 - Use of digital tools to monitor redesigned grazing systems, BP.IC

Abstract

Brazilian livestock farming has expanded extensively, resulting in significant environmental impacts, such as increased deforestation and greenhouse gas emissions. Silvopastoral systems emerge as a sustainable alternative to intensify production and reduce degradation through the use of tree species integrated with grasses in an extensive environment. However, there is still a gap in knowledge about the nutritional potential of forage trees and the application of precision technologies for their management. In this context, this project aims to estimate the bromatological parameters (Crude Protein, Acid Detergent Fiber, and Neutral Detergent Fiber) in three tree species (Morus nigra, Gliricidia sepium, and Tithonia diversifolia) used as forage in silvopastoral systems. To this end, spectral data from satellite images (Planet Labs) and machine learning models (Random Forest) will be used. The research will correlate bromatological data with structural and environmental information to improve the accuracy of predictive models. The model will be trained, validated, and evaluated using metrics such as R2, MAE, and RMSE, comparing it with statistical methods. The results are expected to contribute to consolidating the use of precision tools and computational modeling in livestock farming, contributing to environmentally sustainable production intensification.