Deep learning model maps fourfold increase in Brazil's small reservoirs

Researchers have developed a deep learning model to map small reservoirs in Brazil, which are often overlooked in larger datasets. The model, trained on Landsat imagery from 1984 to 2025, accurately identifies and segments these reservoirs, even those smaller than 1 km². This effort has revealed a nearly fourfold increase in the number of reservoirs, from over 260,000 to nearly one million, with a corresponding rise in total surface area. The resulting dataset provides a comprehensive view of the evolution of these small impoundments and their potential impacts on freshwater ecosystems. AI

IMPACT Provides a novel dataset for environmental monitoring and resource management, demonstrating AI's utility in ecological studies.

RANK_REASON The cluster contains an academic paper detailing a new methodology and dataset. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• Brazil
• deep learning
• Landsat 5-9

• paper
• product

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →