Milwaukee_Template.xml

<?xml version="1.0"?><metadata><idinfo><keywords><theme><themekt>ISO 19115 Topic Category</themekt><themekey>economy</themekey><themekey>environment</themekey><themekey>health</themekey><themekey>transportation</themekey></theme><theme><themekt>EPA GIS Keyword Thesaurus</themekt><themekey>Agriculture</themekey><themekey>Conservation</themekey><themekey>Ecosystem</themekey><themekey>Human</themekey><themekey>Land</themekey><themekey>Natural Resources</themekey></theme><place><placekt>None</placekt><placekey>Wisconsin</placekey></place><theme><themekt>EnviroAtlas</themekt></theme></keywords><citation><citeinfo><origin>University of Arkansas - CAST</origin><title>EnviroAtlas: Milwaukee Land-Cover 1 Meter</title><pubinfo><publish>US EPA Office of Research and Development (ORD) - National Exposure Research Laboratory (NERL)</publish><pubplace>Research Triangle Park, NC</pubplace></pubinfo><pubdate>20130329</pubdate></citeinfo></citation><descript><purpose>The work performed for this project is meant to advance the goals of EPA’s Sustainable and Healthy Communities Research Program (SHCRP) and is a part of SHCRP’s EnviroAtlas which attempts to visualize and analyze the benefits provided to humans by the built-up and natural environment. Using the created Land-cover data.</purpose><abstract>The type of high resolution land-cover and/or land-use maps required for urban planning and analysis do not exist for most metropolitan areas. Additionally, when such maps do exist, they typically cover only a small portion of a city’s greater metropolitan area. One reason for this dearth of urban land-cover data is that high resolution mapping of urbanized areas presents many categorical, temporal, and spatial challenges for traditional remote sensing methodologies. Categorical challenges related to verticality can be problematical for traditional pixel-based classification techniques. For example, the delineation of tree canopy and impervious surfaces is an essential aspect of this mapping project. However, many impervious surfaces are underneath the tree canopy. Therefore, if aerial imagery is the only data source for the resulting map, there cannot be an accurate portrayal of both impervious surfaces and canopy. Obviously problems such as this did not exist when urban areas were mapped at the low resolution associated with the National Land-Cover Dataset derived from 30 meter resolution Landsat imagery. Matters of scale also present interesting challenges when working with high resolution imagery. For example, features such as bare patches and water puddles in a typical suburban yard are discernible at 1 meter resolution, but should the bare patch be classified as bare soil, or a puddle as a waterbody? Clearly, such cases present problems for traditional pixel-based image processing methods.
 
The goal of the project outlined here is to develop object-based image analysis methodologies for mapping one of the 50 cities for the community component of the EnviroAtlas. It was determined that the metropolitan area of Milwaukee would serve as the proof of concept city for this work effort.We intend to address the unique problems associated with high-scale urban land-cover mapping by employing object-based image analysis (OBIA) techniques. Hay and Castilla (2006) define OBIA as, “a sub-discipline of GIScience devoted to partitioning remote sensing (RS) imagery into meaningful image-objects, and assessing their characteristics through spatial, spectral and temporal scale.” The strength of OBIA lies in its emphasis on analyzing “objects” and not individual pixels. Additionally, OBIA blends the raster and vector processing and provides tools for combining original input imagery with appropriate ancillary data sets (both vector and raster) such as road centerlines, elevation information, etc. in the classification process. This is done by, 1) creating meaningful objects from the aerial imagery, and 2) assigning all the various attributes from all the input data (both vector and raster) to the spatially corresponding object.

Because we are dealing with very large datasets, we intend to produce automated and semi-automated processes for creating the land-cover maps for the EnviroAtlas. This automation will require an image processing ruleset that works with little modification for a diversity of geographical settings, from the desert southwest to the coast of New England. Finally, because our methods must work in all areas of the U.S. we will use as our primary input dataset imagery from the USDA’s National Agricultural Imagery Program (NAIP, 2010).
</abstract></descript><timeperd><current>Date of Source Data</current><timeinfo><sngdate><caldate>June 2010</caldate></sngdate></timeinfo></timeperd><status><progress>Complete</progress><update>None Planned</update></status><accconst>None.</accconst><useconst>None</useconst><spdom><bounding><westbc>-128.02839229</westbc><eastbc>-65.20336449</eastbc><northbc>51.6773602</northbc><southbc>22.7348894</southbc></bounding></spdom><secinfo><secsys>FIPS Pub 199</secsys><secclass>No Confidentiality</secclass><sechandl>Standard Technical Controls</sechandl></secinfo><ptcontac><cntinfo><cntorgp><cntorg>US Environmental Protection Agency, Research Triangle Park</cntorg><cntper>EnviroAtlas Coordinator</cntper></cntorgp><cntpos>ORD/NERL/EnviroAtlas</cntpos><cntaddr><addrtype>mailing address</addrtype><city>Research Triangle Park</city><state>NC</state><postal>27711</postal></cntaddr><cntvoice>000-000-0000</cntvoice><cntemail>EnviroAtlas@epa.gov</cntemail></cntinfo></ptcontac></idinfo><eainfo><detailed><enttyp><enttypl>Land-Cover Definition</enttypl></enttyp><attr><attrlabl>new_attribute</attrlabl></attr></detailed><detailed><enttyp><enttypl>Landcover Definition</enttypl><enttypd>Modified Anderson Level I</enttypd></enttyp><attr><attrlabl>Trees and Forest</attrlabl><attrdef>Trees and Forest Larger than 2 square meters</attrdef></attr><attr><attrlabl>Grass</attrlabl><attrdef>Grass and Grasslands (including all herbaceous, non-woody vegetation) larger than 4 square meters</attrdef></attr><attr><attrlabl>Emergent Wetland</attrlabl><attrdef>Areas of Grass that are labeled Wetland in NWI</attrdef></attr><attr><attrlabl>Woody Wetland</attrlabl></attr><attr><attrlabl>Water</attrlabl><attrdef>Waterbodies larger than 4 square meters</attrdef></attr><attr><attrlabl>Bare Earth</attrlabl><attrdef>Sand, beaches, mining operations, excavated construction sites, etc. larger than 4 square meters</attrdef></attr><attr><attrlabl>Agriculture Fallow/Bare</attrlabl><attrdef>Agricultural land with little or no vegetative cover</attrdef></attr><attr><attrlabl>Agriculture Vegetated</attrlabl><attrdef>Agricultural land with vegetative cover</attrdef></attr><attr><attrlabl>Dark Impervious Surface</attrlabl><attrdef>Impervious areas which have a pixel value in the NAIP Green band less than 200.</attrdef></attr><attr><attrlabl>Light Impervious Surface</attrlabl><attrdef>Impervious areas which have a pixel value in the NAIP Green band greater than or equal to 200.</attrdef></attr></detailed><overview><eaover>Vector Categories:

1: Tree and Forest
2: Grass
3: Water
4: Bare Earth
5: Impervious Surface Bright
6: Impervious Surface Dark
7: Agricultural Land - Vegetated
8: Agricultural Land - Bare Soil
9. Emergent Wetlands
10. Woody Wetlands
</eaover></overview></eainfo><dataqual><logic>Tests for integrity have not been performed.</logic><complete>Complete</complete><lineage><procstep><procdate>November 2013</procdate><procdesc>Data Acquisition</procdesc></procstep><procstep><procdate>December 2013</procdate><procdesc>Data Preprocessing</procdesc></procstep><procstep><procdate>December 2013</procdate><procdesc>Ruleset Development</procdesc></procstep><procstep><procdate>January 2013</procdate><procdesc>Ruleset Testing</procdesc></procstep><procstep><procdate>January 2013</procdate><procdesc>Accuracy Assessment Tool Development</procdesc></procstep><procstep><procdate>January 2013</procdate><procdesc>Imagery Classification Testing</procdesc></procstep><procstep><procdate>February 2013</procdate><procdesc>Batch Image Classification Begins</procdesc></procstep><procstep><procdate>March 2013</procdate><procdesc>Image Classification Complete</procdesc></procstep><procstep><procdate>March 2013</procdate><procdesc>Accuracy Assessment Complete</procdesc></procstep></lineage></dataqual><distinfo><resdesc>Downloadable Data</resdesc><distliab>Although these data have been processed successfully on a computer system at the Environmental Protection Agency, no warranty expressed or implied is made regarding the accuracy or utility of the data on any other system or for general or scientific purposes, nor shall the act of distribution constitute any such warranty. It is also strongly recommended that careful attention be paid to the contents of the metadata file associated with these data to evaluate data set limitations, restrictions or intended use. The U.S. Environmental Protection Agency shall not be held liable for improper or incorrect use of the data described and/or contained herein.</distliab><distrib><cntinfo><cntperp><cntorg>US Environmental Protection Agency, Research Triangle Park</cntorg><cntper>EnviroAtlas Coordinator</cntper></cntperp><cntpos>ORD/NERL/EnviroAtlas</cntpos><cntaddr><addrtype>mailing address</addrtype><city>Research Triangle Park</city><state>NC</state><postal>27711</postal></cntaddr><cntvoice>000-000-0000</cntvoice><cntemail>EnviroAtlas@epa.gov</cntemail></cntinfo></distrib></distinfo><metainfo><metd>20130329</metd><metfrd>20170329</metfrd><metstdn>FGDC Content Standard for Digital Geospatial Metadata</metstdn><metstdv>FGDC-STD-001-1998</metstdv><metc><cntinfo><cntorgp><cntorg>US Environmental Protection Agency, Research Triangle Park</cntorg><cntper>EnviroAtlas Coordinator</cntper></cntorgp><cntpos>ORD/NERL/EnviroAtlas</cntpos><cntaddr><addrtype>mailing address</addrtype><city>Research Triangle Park</city><state>NC</state><postal>27711</postal></cntaddr><cntvoice>000-000-0000</cntvoice><cntemail>EnviroAtlas@epa.gov</cntemail></cntinfo></metc></metainfo><spref><horizsys><planar><gridsys><gridsysn>Universal Transverse Mercator</gridsysn><utm><utmzone>16</utmzone><transmer><sfctrmer>0.99960000</sfctrmer><longcm>-87</longcm><latprjo>0.00000000</latprjo><feast>500000.00000000</feast><fnorth>0.00000000</fnorth></transmer></utm></gridsys><planci><plance>coordinate pair</plance><coordrep><absres>1</absres><ordres>1</ordres></coordrep><plandu>meters</plandu></planci></planar><geodetic><horizdn>North American Datum of 1983</horizdn><ellips>Geodetic Reference System 1980</ellips><semiaxis>6378137.000000</semiaxis><denflat>298.257222</denflat></geodetic></horizsys></spref></metadata>