The emissions tables displayed on these pages were developed using GIS procedures to spatially allocate county level emission estimates to the region within the analysis area around each Class I Area or Areas.
GIS software and appropriate surrogate data were used to spatially allocate area and mobile source emissions within the 1996 county-level WRAP inventories to the analysis area of each Class I area. Due to the relatively large number of surrogates used to allocate area and mobile source emissions, as well as the large number of Class I Areas considered, development of displays for each surrogate used would exceed the available resources for this task. Therefore, maps were developed which display the underlying land use/land cover (LULC) data and population density within the region used for the spatial allocation of county-level emissions to the analysis areas. The maps are drawn to the same scale and contain the same political and geographic features as in the overview map displays. Tables summarizing the area and mobile source emissions by county within the analysis areas were developed. Emissions presented in the tables are aggregated to general source categories.
Spatial allocation of regional or county-level emission estimates is accomplished through the use of gridding surrogates or spatial allocation factors (SAFs), for each emission source category or groups of source categories. Spatial surrogates are typically based on the proportion of a known region-wide characteristic variable that exists within the modeling domain grid cells. Traditionally, the development of spatial gridding surrogates has been performed by a variety of methods depending on the emission source category being considered, the required spatial resolution, the geographic extent of the domain, and the particular characteristics of the geospatial data available. Spatial surrogates define the percentage of county level emissions from a particular source category that is to be allocated to some spatial region, in this case the Class I Area analysis areas.
For most area sources these percentages are based on areas of a particular land use/land cover or vegetation type, while for mobile source categories the percentages are usually based on total length of a certain road type, or transportation network. Often the population is also used as a spatial surrogate for certain emission source categories.
The spatial surrogate definitions and emission source category/surrogate relationships used for the project are based on those developed by the EPA. This information is presented in full detail in the Project Report. The spatial surrogates used for the project include:
- Agricultural land
- Forest land
- Urban land
- Water
- Total population
- Urban population
- Rural population
- Inverse population
- Housing
- Railways
- Urban primary roadways
- Urban secondary roadways
- Rural primary roadways
- Rural secondary roadways
Emissions for airports (aircraft and ground support equipment) and commercial marine emissions are available in the WRAP off-road emissions data sets on a county level. These emissions were not spatially allocated in this task because of the large uncertainty that would be introduced in spatially allocated emissions using the approach discussed above.
Similarly, commercial marine emissions were estimated using activity at individual ports, but this detailed information per port is not part of the WRAP county-level commercial marine estimates. Ports with strikingly dissimilar activity rates can be found in proximity to one another. The number of ports within an area of interest is therefore not a reasonable method to allocate county commercial marine emissions.
Emissions from wildfires and prescribed wildland fires were determined for the analysis areas based on the WRAP 1996 fire databases developed by Air Sciences. These databases contain fire emissions for each fire and for each day during 1996 in terms of point locations (and are therefore treated as point sources) and, in some cases as total acres burned. Because the overall extent and shape of the burn areas are not known, and because many emission records do not include the total acres burned it was not possible to determine the areal extent of the fires within the analysis areas around each Class I Area. As an alternative, the fire emissions were treated as point sources using the coordinates given in the database.
Overlaying these data with the analysis areas allows an estimate of the fire emissions within the analysis areas and the potential to impact the Class I Areas. The Air Sciences databases do not include prescribed wildland fires for the state of California. For California, the databases maintained by the California Air Resources Board (ARB) were used for estimating fire emissions within California. The ARB databases provide only total annual emissions by type of fire and by county for 1996. Therefore, estimates of prescribed wildland fire emissions for CA were determined using procedures similar to other area sources, which are defined by county-level annual emission estimates. Estimates of the fire emissions within the analysis areas were determined by using forestland as a spatial surrogate to determine the percentage of forestland in each county within each analysis area, and allocated this percentage of the total county-level fire emissions to the appropriate Class I Areas.
Finally, as the ARB database does not include PM2.5 or NH3 emissions, these pollutants were estimated using appropriate scaling factors based on piled fires. Emissions for PM2.5 and NH3 were estimated as PM2.5=0.86*PM10 and NH3=0.0045*CO.
Data Sources
Emissions data
For all states except Alaska, WRAP 1996 emissions data are being used as the basis for characterizing the emissions around each Class I area.
The pollutants included in the emissions characterization are those pollutants estimated in the WRAP emission inventories that are important contributors to visibility impairment:
- Particulate matter (PM10 and PM2.5),
- Sulfur oxides (SOx),
- Nitrogen oxides (NOx),
- Volatile organic compounds (VOC), and
- Ammonia (NH3).
On-road mobile source emissions were obtained from the WRAP 1996 on-road emissions estimates prepared by ENVIRON. The WRAP 1996 emissions are based on EPA's MOBILE6 model, while the NET96 emissions are based on the earlier MOBILE5 model.
Off-road mobile source emissions were obtained from the updated (with NONROAD2002 model) WRAP 1996 off-road emissions estimates prepared by ENVIRON. The WRAP 1996 emissions are based on EPA's latest NONROAD model.
Area source emissions were obtained from the WRAP 1996 area source files prepared by PES. The WRAP 1996 area source emissions are largely based on the NET96, with some state updates. Emissions from wildfires and prescribed wildland fires were determined for the analysis areas based on the WRAP 1996 fire databases developed by Air Sciences. It is important to note that there are no estimates of windblown dust emissions in the WRAP area source emissions files; ENVIRON is currently developing windblown dust emissions for the WRAP region, but this work will not be completed in time for use in this project.
GIS data
A variety of geospatial data are required both for generating maps and displays as well as for spatial allocation of county-level emission estimates to the appropriate analysis areas. Geospatial and population data are available from various sources, primarily from the USGS, the EPA and the U.S. Census Bureau. The following GIS data were used for developing maps and spatial surrogates for allocation of area and mobile source emissions:
- Land Use/Land Cover (LULC) - specifies land types (urban, agricultural, forest, etc) and areal coverage for spatial allocation of area sources
- Transportation Networks - Specifies roadway types, railways, airport and ports used for spatial allocation of on-road mobile and off-road mobile emission sources. The EPA Spatial Surrogate database was used for these categories. (ftp://ftp.epa.gov/EmisInventory/emiss_shp).
- Population - Population data used for developing population density displays and for spatial allocation of county-level emission estimates were derived from a 1-km gridded global data set for 1995. These data were obtained from the Center for International Earth Science Information Network (CIESIN), Columbia University; International Food Policy Research Institute (IFPRI); and World Resources Institute (WRI), 2000. Gridded Population of the World (GPW), Version 2. Palisades, NT; CIESIN, Columbia University.
| |
Contact
the Webmaster
