{61C69E94-1EB4-42B7-8C56-4D064C05243E} eng; USA utf8 dataset EnviroAtlas Coordinator U.S. Environmental Protection Agency, Office of Research and Development-Sustainable and Healthy Communities Research Program, EnviroAtlas Geospatial Data Owner (919) 541-3832 109 T.W. Alexander Drive Research Triangle Park NC 27709 EnviroAtlas@epa.gov https://www.epa.gov/enviroatlas pointOfContact 2025-02-13 ISO 19115 Geographic Information - Metadata ISO 19115:2003/Cor.1:2006(E) EnviroAtlas -- Memphis, Tennessee -- Meter-Scale Urban Land Cover (MULC) Data (2012) 2015-12-03 publication U.S. EPA Office of Research & Development (ORD) - National Exposure Research Laboratory (NERL) Research Triangle Park NC publisher US Environmental Protection Agency (US EPA) originator The Memphis, TN EnviroAtlas One Meter-scale Urban Land Cover (MULC) dataset comprises 2,733 km2 around the city of Memphis, surrounding towns, and rural areas. These leaf-on LC data and maps were derived from 1-m pixel, four-band (red, green, blue, and near-infrared) aerial photography acquired from the United States Department of Agriculture (USDA) National Agriculture Imagery Program (NAIP) on four dates in 2012: June 15, June 18, June 21 and June 23, and one date in 2013: July 12. Three separate LiDAR (Light Detection and Ranging) data sets collected on February 19, 2009 - August 2, 2010, December 1-2, 2011 and January 23-24, 2012 were integrated for Shelby Co., TN, Crittenden Co., AR, and DeSoto Co, MS. Five MULC classes were mapped directly from the NAIP and LiDAR data: Water, Impervious, Soil, Trees, and Grass/Herbaceous. Agriculture was derived from USDA Common Land Unit (CLU) data. Woody and emergent wetlands were copied from existing National Wetlands Inventory (NWI) data. Analysis of a random sampling of 612 photo-interpreted land cover reference points yielded an overall users accuracy of 86.9%. This dataset was produced by the US EPA to support research and online mapping activities related to EnviroAtlas. EnviroAtlas (https://www.epa.gov/enviroatlas) allows the user to interact with a web-based, easy-to-use, mapping application to view and analyze multiple ecosystem services for the contiguous United States. The dataset is available as downloadable data (https://edg.epa.gov/data/Public/ORD/EnviroAtlas) or as an EnviroAtlas map service. Additional descriptive information about each attribute in this dataset can be found in its associated EnviroAtlas Fact Sheet (https://www.epa.gov/enviroatlas/enviroatlas-fact-sheets). The EnviroAtlas MULC provides spatially detailed foundational data for approximately 80 derived EnviroAtlas Community information layers. Some layers are developed at the native one meter MULC resolution and others by U.S. Census block group. While the EnviroAtlas National component is based on NLCD (National Land Cover Database, www.mrlc.gov) land cover data at 30 m spatial resolution (30x30 m pixel dimensions from the Landsat sensor), the EnviroAtlas Community component is based on MULC data at 900 times that spatial resolution. This permits analysis of common landscape features in the 1-10 m range which are unresolvable using 30 m NLCD pixels. EnviroAtlas includes close to 100 data layers derived from MULC data. Three such layers are: Percent impervious area in 50m stream buffer; Value of asthma exacerbation cases avoided due to sulfur dioxide removed ($/yr); Percent of busy roadway bordered by < 25 percent tree buffer. Some of many possible applications of MULC include: • Knowing the exact location and size of vegetated and non-vegetated spaces; • Identifying highly impervious areas that would most benefit from additional trees and other vegetation to mitigate surface runoff and urban heat island effects; • Measuring the width and character of urban riparian and street buffers; • Locating potential wildlife corridors, gaps and linkages; • Planning park development for maximum benefits. The overall goal of EnviroAtlas is to employ and develop the best available science to map indicators of ecosystem services production, demand, and drivers for the nation. completed EnviroAtlas Coordinator U.S. Environmental Protection Agency, Office of Research and Development-Sustainable and Healthy Communities Research Program, EnviroAtlas Geospatial Data Owner (919) 541-3832 109 T.W. Alexander Drive Research Triangle Park NC 27709 EnviroAtlas@epa.gov https://www.epa.gov/enviroatlas pointOfContact asNeeded Ecosystem Environment Land theme EPA GIS Keyword Thesaurus Agriculture Biophysical Communities Ecosystem Services EnviroAtlas Environment Environmental Atlas Green Space Land Land Cover Surface Water Water Wetlands theme EnviroAtlas Tennessee Memphis, TN place None https://edg.epa.gov/EPA_Data_License.html otherRestrictions Use Constraints: None. Please check sources, scale, accuracy, currency and other available information. Please confirm that you are using the most recent copy of both data and metadata. Acknowledgement of the EPA would be appreciated. otherRestrictions Access Constraints: None PUBLIC https://project-open-data.github.io/schema/#accessLevel Standard Technical Controls eng; USA environment geoscientificInformation imageryBaseMapsEarthCover -90.324550 -89.624648 34.714780 35.402557 9999-01-01 true Entity and Attribute Information Entity and Attribute Overview: For easier readability, this XML metadata can be opened in a text editor (e.g., Notepad). For more information about EnviroAtlas data, go to https://enviroatlas.epa.gov/EnviroAtlas/DataFactSheets/index.html. The Memphis, TN land cover dataset includes data for the Memphis Urban Statistical Area, as defined by the US Census Bureau, plus a 1-km buffer. The total area classified was 2,733 square kilometers, with approximately 60% of the study area in Shelby County, TN, 24% in DeSoto County, MS, and 13% in Crittenden County, AR. The land cover classification was developed using 96 Digital Orthophoto Quarter Quadrangles [("quarter quads" - each quarter quad includes four bands (visible RGB plus near-infrared) with one-meter pixel size] acquired from the United States Department of Agriculture (USDA) National Agriculture Imagery Program (NAIP). The aerial photography data were collected on four dates during the summer of 2012: June 15 (39 quarter quads), June 18 (26 quarter quads), June 21 (13 quarter quads), and June 23 (5 quarter quads). There was no NAIP imagery collected for the state of Arkansas in 2012. Therefore imagery collected on July 12, 2013 was acquired for areas in Arkansas (11 quarter quads) and Tennessee. In addition to the NAIP imagery, LiDAR point cloud data were incorporated into the classification for 2,256 square kilometers, in Shelby County, TN, DeSoto County, MS, and Crittenden County, AR. This LiDAR data consisted of 3 different data sets delivered in the form of LAS tiles. The Mississippi River Delta Project LiDAR produced by Photo Science, Inc. had a nominal pulse spacing of 1m with a vertical accuracy of 15 cm, covering 226 km2 (8% of study area); The Memphis TN Project and Shelby County Project LiDAR were both produced by Woolpert, Inc., processed for intensity and height above ground together, and were treated as one complete data set. These had a nominal pulse spacing of 1 m with a vertical accuracy of 12.5 cm for Memphis and 15.9 cm for Shelby County. The horizontal accuracy of these data sets was +/- 3.8 feet at a 95% confidence level. These two datasets together covered 2030 km2 (74% of study area). LiDAR data did not exist for 18% of the total study area. Up to four returns were collected per pulse, and points were separated into the following categories by the contractor: "water," "ground," "unclassified," "medium vegetation," and "high vegetation". The "medium vegetation" and "high vegetation" were pre-classified only in the Mississippi River Delta dataset and were not used in the modeling of LiDAR for this project. Three regions of interest within the Memphis study area were identified and processed as individual study areas with separate image processing completed for each. These were as follows: 1) the Mississippi River Delta LiDAR area; 2) the Memphis TN and Shelby County LiDAR area; and 3) the area with no LiDAR data. These sub-areas were created by building an ArcGIS model to vectorize the boundary of the two LiDAR areas identified above. The portion of the original study area not intersecting the LiDAR extent was identified \as the no LiDAR data area. A shapefile with a total of 96 intersecting USGS quarter quadrangles (QQs) was overlaid on the study areas. Land cover classification was performed for each of these intersecting QQs, with each individual QQ buffered by 3 meters to prevent unclassified slivers in the final product. The grid cells with only NAIP imagery available were classified using five bands: the four original NAIP bands (red, green, blue, near-infrared) plus the normalized difference vegetation index (NDVI), which was calculated using the NAIP red and near-infrared bands. NDVI = ((band 4 - band 3) / (band 4 + band 3)). The grid cells with both NAIP imagery and LiDAR data available were classified using the five bands derived from the NAIP imagery plus two bands derived from the raw LiDAR point clouds for a total of seven bands. To derive these additional bands, the tiles in the LiDAR dataset were re-projected into the coordinate system of the NAIP imagery (NAD_1983_UTM_Zone_15N) and then clipped to the boundaries of the QQs. The LiDAR point data were then imported into ArcGIS and converted from vector to raster data. The first LiDAR-derived band was the mean intensity of LiDAR points falling within each 1m pixel. The second band was height-above-ground. A digital surface model (DSM) was created by selecting first return points and getting the maximum z-value in each 1m pixel. A bare-earth digital elevation model (DEM) was calculated by averaging the z-values of LiDAR points classified as "ground" in the dataset. NoData values in the DSM and DEM rasters were estimated using a triangulation as the interpolation type and natural neighbor as the interpolation method. The height-above-ground band was calculated by subtracting the digital surface model from the bare-earth elevation model. Height Above Ground (HAG) = DSM - DEM. A total of eight land use / land cover (LULC) classes were mapped: Water (W), Impervious (I), Soil (S), Tree (T), Grass (G), Agriculture (A), Woody wetland (WW) and Emergent wetland (EW). The five classes 1) Water, 2) Impervious, 3) Soil, 4) Tree and 5) Grass were classified using GeniePro 2.4, a machine-learning classification software program. These layers were created in GeniePro (www.observera.com) by selecting appropriate training pixels. A solution algorithm was evolved based on the training pixels by running approximately 5000 iterations to create a classification algorithm. The solution algorithms were applied to NAIP QQs images to classify land cover categories. Errors were corrected using hand-digitizing tools in GeniePro. LiDAR data was used in classifying trees, grass, soil, and impervious surfaces where available. For the portion of the study area that lacked LiDAR data, only the NAIP imagery was used in the classification. Landcover classes were classified in GeniePro in the following sequence: 1) water, 2) trees/grass/non-vegetation, 3) soil/impervious. Water was classified using the original NAIP imagery and without the use of LiDAR data. After performing a classification for an individual class for a QQ, the classification was reclassified in ArcGIS with the "Reclassify" tool so that only values for that particular class existed and other pixels were given "NoData" values. All the individual class rasters where then mosaicked together in ArcGIS using the "Mosaic to New Raster" tool to form a seamless class raster. The "Extract by Mask" tool was then used to mask the completed class out of the NAIP imagery before beginning creation of the next land cover type in the sequence. Therefore all water was masked out before classifying trees/grass; water/trees/grass were all masked out before classifying soil/impervious. The "Mosaic to New Raster" tool was again used to combine rasters for individual classes to form a seamless land cover map. Upon completion of the classification process using Genie, the agriculture class was created. 2008 Common Land Unit (CLU) data was available for the state of Tennessee; however, it was missing for the study area in Mississippi and Arkansas. CLU data (shapefile) were selected upon visual inspection. When necessary, features were edited with hand digitizing tools in ArcGIS. When producing an agriculture layer for Mississippi and Arkansas no CLU data existed; therefore, the same process of visual inspection and hand editing in ArcGIS was performed using county tax parcel data. Though this process was more tedious and time consuming than using GeniePro, it was less prone to errors. Agricultural data created through CLU and county parcel data was then merged and projected to NAD_1983_UTM_Zone_15N to create a final vector layer for the entire study area. Agricultural parcel vector data was then converted to raster to create the agriculture class for Memphis. Before the incorporation of wetland data the existing classes were mosaicked using the following hierarchy and assigned cell values: 1) Water - 10, 2) Trees - 40, 3) Impervious - 20, 4) Agriculture - 80, 5) Grass - 70, 6) Soil - 30. Wetland data for the study area was acquired from the National Wetland Inventory (NWI). Data collection for NWI ranges from 1977 to 2014; however data collected for Memphis was in the 1970s and 1980s. For NWI metadata please visit http://www.fws.gov/wetlands/Data/metadata/FWS_Wetlands.xml. For the NWI wetlands mapper please visit http://www.fws.gov/wetlands/Data/Mapper.html. Polygons corresponding to woody or emergent wetlands were extracted. Classified pixels falling within the boundaries of these polygons were extracted from the 6-class classification. Pixels initially classified as "Grass" were reclassified as "Emergent wetlands" and assigned a cell value of 92; pixels initially classified as "Tree" were reclassified as "Woody wetlands" and assigned a cell value of 91. An accuracy assessment was performed on the 6-class classification (Woody and emergent wetland classes were excluded since they were not derived from derived from an outside source) on the area falling within the Memphis Urban Statistical Area plus 1-km buffer. A random sample of 600 pixels was selected from the classification. An analyst who had not been involved in generating the land cover map performed photo interpretation for these 600 validation points. The points were labeled using a fuzzy classification approach (Gopal and Woodcock, 1994). This permits the Analyst to assign a confidence value to the photo interpreted label at reach reference point. Confidence is expressed as an integer from 1 to 5: 1: Absolutely wrong: classification value is unacceptable (Very Wrong); 2: Understandable but Wrong: classification value is not good. There is something about the site that makes the answer understandable, but there is clearly a better answer. Classification would pose a problem for users of the map. (Not Right); 3: Reasonable or Acceptable: Maybe not the best possible classification but it is acceptable; the classification does not pose a problem to users of the map. (Right); 4: Good Answer: Would be happy to find this classification given on the map (Very Right); 5: Absolutely Right: No doubt about the match. (Perfect). Using the NAIP imagery as well as ancillary imagery (e.g. Google Satellite and Street Views and Bing Aerial and Birdseye views - Google and Bing Imagery given less importance than NAIP since they differed from NAIP both temporally and in resolution), the photo-interpreter evaluated the land cover class of the random point plus the classes of its eight immediate neighbors. If the land cover class of the random point was not also the class of five of the eight neighboring pixels, then the area was considered to be insufficiently homogeneous and the random point was excluded from the accuracy assessment. A total of 63 points were excluded from the Memphis accuracy assessment. An additional 75 points were added by stratifying among classes for a total of 612 points to ensure that at least 50 points were assessed in each class. The classification values were extracted from the 6-class classification map. The classification values and interpretation values were used to create the following confusion matrices. One is more liberal ("RIGHT") view and one is more conservative ("MAX") view. The MAX interpretation is correct if the classified land cover matches the interpreter's highest score, illustrating the interpreter finds this class to be the most appropriate for that location. The RIGHT interpretation is correct if the classified land cover matches any class the interpreter has given a value of 3 or greater, illustrating the interpreter finds the classification to be acceptable but another class may be more appropriate. The following confusion matrix summarizes the accuracy assessment MAX results. For easier readability, this XML metadata can be opened in a text editor (e.g., Notepad) and the confusion matrix may be copied from text editor to an Excel spreadsheet. Confusion matrices can also be viewed in original formatting by opening in ArcCatalog. MAX_View RASTER_TXT Agricult Grass Impervious Soil Tree Water N(n) Users_Accuracy Errors_of_Commission Agricult 91 3 0 1 0 0 95 95.79 4.21 Grass 7 114 3 0 7 1 132 86.36 13.64 Impervious 0 2 67 5 1 0 75 89.33 10.67 Soil 1 29 3 27 0 0 60 45.0 55.0 Tree 1 10 1 0 183 2 197 92.89 7.11 Water 0 2 1 0 0 50 53 94.34 5.66 N(n) 100 160 75 33 191 53 612 Producers_Accuracy 91.0 71.25 89.33 81.82 95.81 94.34 Errors_of_Ommission 9.0 28.75 10.67 18.18 4.19 5.66 Kappa_Coefficient 0.835 OVERALL_ACCURACY 86.93 The overall accuracy of the MAX view was 86.9%. LiDAR data helped improve the accuracy of some classes, especially Trees, Grass, and Impervious. The following confusion matrix summarizes the accuracy assessment RIGHT results. RASTER_TXT Agricult Grass Impervious Soil Tree Water N(n) Users_Accuracy Errors_of_Commission Agricult 91 3 0 1 0 0 95 95.79 4.21 Grass 0 122 3 0 6 1 132 92.42 7.58 Impervious 0 2 68 4 1 0 75 90.67 9.33 Soil 0 28 3 29 0 0 60 48.33 51.67 Tree 0 9 1 0 185 2 197 93.91 6.09 Water 0 2 1 0 0 50 53 94.34 5.66 N(n) 91 166 76 34 192 53 612 Producers_Accuracy 100.0 73.49 89.47 85.29 96.35 94.34 Errors_of_Ommission 0.0 26.51 10.53 14.71 3.65 5.66 Kappa_Coefficient 0.861 OVERALL_ACCURACY 89.05 The overall accuracy of the RIGHT view was 86.9%. LiDAR data helped improve the accuracy of some classes, especially Trees, Grass, and Impervious. Entity and Attribute Detail Citation: https://www.epa.gov/enviroatlas/enviroatlas-fact-sheets EnviroAtlas Coordinator U.S. Environmental Protection Agency, Office of Research and Development-Sustainable and Healthy Communities Research Program, EnviroAtlas Geospatial Data Owner (919) 541-3832 109 T.W. Alexander Drive Research Triangle Park NC 27709 EnviroAtlas@epa.gov https://www.epa.gov/enviroatlas distributor https://enviroatlas.epa.gov/download/MTN_MULC_tif.zip download Download The URL providing direct access to the downloadable dataset. download https://enviroatlas.epa.gov/arcgis/rest/services/Supplemental/Landcover_AllCommunities/ImageServer ESRI:ArcGIS API The endpoint of a web service to access the dataset (REST endpoint, WMS GetCapabilities URL, or a SOAP WSDL endpoint). webMapService https://www.epa.gov/enviroatlas/enviroatlas-data information Documentation URL for accessing related documents such as data dictionary, technical information about a dataset, developer documentation, etc. information https://www.epa.gov/enviroatlas information Home Page Alternative landing page used to redirect user to a contextual, Agency-hosted "homepage" for the Dataset or API information dataset Horizontal Positional Accuracy Report Data were collected using methods that are accurate to within 2-5 meters (EPA National Geospatial Data Policy [NGDP] Accuracy Tier 2). For more information, please see EPA's NGDP at https://www.epa.gov/geospatial/geospatial-policies-and-standards) Horizontal Positional Accuracy NAIP metadata; LiDAR probably more accurate. meters 6 Features represented have not been tested for completeness Features represented have not been tested for completeness Tests for integrity have not been performed. This lineage was generated by the FGDC CSDGM to 19139 transformation See overview section for detailed NAIP and LiDAR processing workflow 2014-03-01T00:00:00 unknown 2029-02-13 This metadata was automatically generated from the FGDC Content Standard for Digital Geospatial Metadata standard (version FGDC-STD-001-1998) using the 2024-02-09T11:11:00 version of the FGDC CSDGM to ISO 19139 transform. EnviroAtlas Coordinator U.S. Environmental Protection Agency, Office of Research and Development-Sustainable and Healthy Communities Research Program, EnviroAtlas Geospatial Data Owner (919) 541-3832 109 T.W. Alexander Drive Research Triangle Park NC 27709 EnviroAtlas@epa.gov https://www.epa.gov/enviroatlas pointOfContact