US Environmental Protection Agency EnviroAtlas Meter-Scale Urban Land Cover (MULC) data, machine learning and ecosystem goods and services (EGS): information for healthy communities
Abstract
High resolution l and cover depict the physical material at the earth's surface and provide the foundation to support science-based decision - making for healthy communities and landscapes . Here we present the Meter-scale Urban Land Cover (MULC) dataset , the machine learning methods used for its development and the derived ecosystem goods and services (EGS) maps and metrics that support human and ecosystem health analysis and decision making. We describe this unique, high-resolution (1 x1 meter pixel size ) land cover data set developed for 30 U.S. communities by the United States Environmental Protection Agency (EPA) for the EnviroAtlas online geospatial tool (https://www.epa.gov/enviroatlas) . MULC data supports approximately 100 derived EnviroAtlas maps and metrics that quantify nature's benefits ( EGS ) . We highlight our experience using machine learning in supervised classification of combined aerial photography and LiDAR raster data. To date , we have produced and adapted from other sources , more than 98,000 km 2 of 1 - meter resolution land cover . The MULC dataset series is comprised of seven predominant land cover classes with an 80 % average overall accuracy and 0. 75 mean Kappa coefficient . MULC data, fact sheets and metadata are freely available via web browser in EnviroAtlas , via web map service (WMS), and via download at edg.epa.gov. MULC data in EnviroAtlas support applications such as green space and urban habitat mapping, assessing linkages between socioeconomic information and environment al quality and community conditions , optimizing locations for street -side or urban-zone tree planting, mosquito habitat analysis to inform arbovirus control , urban heat island mitigation, education and as training data for other future land cover classification endeavors.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2019
- Bibcode:
- 2019AGUFMGC31N1385P
- Keywords:
-
- 6309 Decision making under uncertainty;
- POLICY SCIENCES & PUBLIC ISSUES