Development of Land Use Mapping Methods Based on Multisensor Data and Machine Learning Algorithms

• Grant number: 26/06628-1
• Support Opportunities: Scholarships abroad - Research Internship - Post-doctor
• Start date: August 01, 2026
• End date: July 31, 2027
• Field of knowledge: Physical Sciences and Mathematics - Geosciences
• Principal Investigator: Édson Luis Bolfe
• Grantee: Danielle Elis Garcia Furuya
• Supervisor: Wolfram Schlenker
• Host Institution: Embrapa Agricultura Digital. Empresa Brasileira de Pesquisa Agropecuária (EMBRAPA). Campinas , SP, Brazil
• Institution abroad: Harvard University, United States
• Associated to the scholarship: 24/05205-4 - Development of land use mapping methods based on multi-sensor data and machine learning algorithms, BP.PD

Abstract

Several agricultural mapping initiatives have been conducted to support rural planning and crop forecasting. However, due to the complex dynamics and scale of Brazilian agriculture, there remains a growing need to develop methodologies with detail, frequency, and accuracy to discriminate between crops, pastures, and forestry areas. The increasing availability of satellite imagery with high spatial, temporal, and spectral resolutions, combined with advances in artificial intelligence and machine learning algorithms, has significantly boosted research in agricultural remote sensing for classification, monitoring, and management. This research fellowship proposal is part of the Center of Science for Development in Digital Agriculture (CCD-AD/SemeAr) and focuses on activities in the Agrotechnological Districts (DATs). The objectives include: (1) analyzing land use and land cover (LULC) based on field-collected data; (2) conducting a detailed assessment of apple orchards, including quantifying areas adopting anti-hail nets; and (3) investigating potential climate-related interferences, especially hailstorms affecting crop production. The project will integrate multisensor data, including satellite imagery, with machine learning algorithms to enhance agricultural mapping. Expected outcomes include the development of new methodologies for agricultural monitoring, the generation of qualified scientific publications providing relevant information about LULC and apple orchard areas, and techniques applicable to agricultural producers in Brazil and other countries. (AU)