- IN33A-01: Large area 2 m land cover mapping and change analysis of Amhara Ethiopia using convolutional neural network ensembles on WorldView imagery for 2009 – 2024
-
NOLA CC
Author(s):Generic 'disconnected' Message
Woubet Alemu, NASA Goddard Space Flight Center, Biospheric Sciences Laboratory (First Author, Presenting Author)
Christopher Neigh, NASA Goddard Space Flight Center
Jordan Caraballo-Vega, NASA Goddard Space Flight Center
Margaret Wooten, NASA Goddard Space Flight Center
Ejigu Muluken, Amhara National Regional State Bureau of Agriculture and Rural Development
Gebre Michael Maru, Amhara National Regional State Bureau of Rural Land Administration and Use
Molly Brown, University of Maryland College Park
Konrad Wessels, George Mason University Fairfax
Tracking how land is used and changed over time is important for protecting the environment and supporting rural communities—but it's hard to do in places like Ethiopia, where farms are small and the land is complex. In this study, we used very detailed satellite images (from Maxar Technology's commercial WorldView data) and advanced computer models to map land use changes in Ethiopia’s Amhara Region between 2009 and 2024. We found two major trends. First, many farmers are switching from growing crops to planting eucalyptus trees, which provide quick income but may harm water, soil, and biodiversity in the long run. Second, farmland is spreading into steeper hillsides—areas that are more likely to suffer from soil erosion—because flat land is running out. Our study shows that using high-resolution commercial satellite data gives a much clearer picture of what’s happening on the ground than coarser global datasets. This kind of detailed information can help governments and communities make better decisions about how to manage land in a way that protects nature while supporting people’s livelihoods.
Scientific DisciplineSuggested ItinerariesNeighborhoodType
Enter Note
Go to previous page in this tab
Session


