Authors: Roger Auch*, United States Geological Survey, EROS Center, Kristi Sayler, United States Geological Survey, EROS Center, Bruce Pengra, KBRwyle, Contractor to USGS EROS Center, Josephine Horton, Innovate!, Contractor to USGS EROS Center, Janis Taylor, KBRwyle, Contractor to USGS EROS Center
Topics: Land Use and Land Cover Change, Remote Sensing, Regional Geography
Keywords: LCMAP, land-cover mapping, Piedmont ecoregion, land change reference dataset, forestry, urbanization
Session Type: Poster
Start / End Time: 9:55 AM / 11:35 AM
Room: Lincoln 2, Marriott, Exhibition Level
Presentation File: No File Uploaded
Although the Landsat satellite system has been acquiring images of the United States for decades, the ability to produce annual land-cover maps has not been easily feasible until time-series data stacks and advanced algorithms were created. The U.S. Geological Survey’s LCMAP (Land Cover Monitoring, Assessment, and Projection) initiative is currently producing the first systematic annual land-cover mapping effort spanning the Landsat 30-meter resolution era. Along with 10 different map products, LCMAP is creating a robust 25,000-point multi-faceted reference dataset. Production of version 1 map products has just begun but prototype data for some regions are internally available. Such data for the Piedmont ecoregion in the southeastern U.S. allowed for the first testing of annual land-cover classification map products in conjunction with the reference data. Results indicate that LCMAP captures the region’s annual land-cover composition with good to excellent accuracy and tells the overall leading stories of change in the Piedmont; forestry land-use (timber cutting and tree regrowth) and urbanization. The accuracy of specific annual land-cover changes in the Piedmont, however, had much lower statistical measures for several reasons, including how to capture change in a “yearly” fashion in a time continuum where change can happen any time. In theory, LCMAP can detect change at any arbitrary point in time but operationally such custom products are not yet available. Other current LCMAP products show promise in detecting change that may not cause a land-cover classification change and may offer additional information about ongoing land processes and thus advancing land-cover mapping.