This is the authors' abstract. We don't add key points for this paper.
This study developed the Mangrove Loss Estimation and Adaptive Forecasting (MangroveLEAF) tool, a cloud-based platform designed to support accurate, site-specific mapping and modeling of mangrove loss for targeted conservation and restoration strategies. MangroveLEAF integrates satellite data, remote sensing indices, and machine learning to analyze historical trends and forecast future mangrove loss. It draws on Landsat data from 2010 to 2020 and environmental parameters through 2025. The approach uses 11 model configurations that incorporate vegetation, moisture, built-up, erosion, and spectral indicators, selected based on their relevance to site-specific biophysical conditions and drivers of degradation. The tool was applied to eight mangrove sites across the Philippines, with the best-performing model identified for each site, achieving prediction accuracies between 93% and 100%. These models were used as inputs to a Random Forest algorithm implemented in Google Earth Engine to generate site-specific mangrove loss projections. To support a deeper understanding of site-specific drivers, three additional analytical tools were developed: Shoreline Trend Estimation and Modeling (STEM), Built-up Regression Analysis of Coastal Habitats (BRANCH), and Survey on Traditional and Local Knowledge (STALK). Together, these tools inform a guide matrix for designing science-based interventions and policy recommendations, offering an integrated framework for mangrove conservation and management.
Answers are grounded in this paper and cite their sources.
No takes yet. Share an insight, caveat, or question.
Baloloy et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: