Logo Lanfrica
  • Home
  • Atlas
  • Insights
  • Docs
  • Sign in

© 2026 Lanfrica. All rights reserved. All copyrights of the resources shown on the Lanfrica website belong to the original copyright holders, unless explicitly stated otherwise.

terraiqcollective/ghana-eco-pulse

Domain:

environment and energygeospatial
Creator:
ter
Host:
Web platform for assessing forest carbon loss from ASM activities in Ghana # EcoPulse Ghana A web platform for tracking forest carbon loss from artisanal gold mining in Ghana's high forest zone, 2020–2025. ## Background Ghana's high forest zone is one of the most productive tropical forest landscapes in the world. It is also the centre of artisanal and small-scale gold mining, locally known as galamsey. Mining clears vegetation and topsoil to reach alluvial deposits, leaving abandoned pits that do not regenerate. Operational mapping of mining extent already exists. What has been missing is a clear picture of the forest carbon lost as a result, at a resolution useful to the agencies responsible for forest protection and national emissions reporting. EcoPulse Ghana addresses that gap. ## What the platform shows - Annual above-ground carbon stocks across the study area, 2020–2025 - Annual carbon loss attributable to mining conversion, with cumulative totals - Region- and district-level breakdowns of stocks and losses - Toggleable map layers for carbon stock and mining loss - A trend view comparing stock and loss over the full period The interface runs in a browser. No installation is required. ## Approach Above-ground biomass is estimated from Google Satellite Embedding predictors trained against GEDI L4A spaceborne LiDAR observations, using Random Forest regression in Google Earth Engine. Predictions are produced annually at 30 m resolution. Biomass is converted to above-ground carbon using a fraction of 0.5. Mining footprints are not generated by this project. The footprints come from the operational artisanal and small-scale mining monitoring service maintained by the Centre for Remote Sensing and Geographic Information Services (CERSGIS) at the University of Ghana, produced with a U-Net deep learning model on Planet NICFI imagery. Carbon loss is attributed by pairing each year's biomass baseline with the following year's mining footprint, identifying newly converted pixels and recording their pre-mining carbon. ## Data source …

Visit

github.com

Tasks

computer vision