2. Background on Shapefiles and Surrogates
2.1 Shapefiles
Shapefiles contain geographic data objects in the form of points, lines, or polygons with associated informational attributes that describe the geometric features. Shapefiles are a GIS industry standard format developed by the Environmental Systems Research Institute (ESRI) (see http://www.esri.com/). Shapefiles contain geospatial vector data and are the primary type of input data for the SA Vector Tools.
Each shapefile stores data for a single feature class (i.e., a shapefile
contains information for points, lines, or polygons). Point shapes are
single-coordinate features such as smoke stacks, schools, or ports. Line shapes
can be continuous lines, such as roads, and they can also be polylines, like
the branches or tributaries of a river. Polygon shapes can be simple areas or
multipart areas like states and countries (e.g., th state of
A shapefile consists of a group of files that contain the coordinates of the shapes, associated attributes, and optionally, map projection information. Each shapefile contains three mandatory files:
· .shp – coordinates of shapes or features (points, lines, or polygons)
· .shx – shape positional index data, to support quickly searching through the data contained in the shapefile
· .dbf – attribute data (e.g., population counts, road classes, airport capacities) for each shape in dBASE III format.
A shapefile can also contain many optional files, the details of which are not important to this discussion.
2.2 Spatial Surrogates
Spatial surrogates are most commonly used to map county-level emissions data onto the rectangular grid cells used by an Eulerian air quality model such as the Community Multiscale Air Quality (CMAQ) modeling system (see www.cmascenter.org). A spatial surrogate is a value greater than zero and less than or equal to one that specifies the fraction of the emissions in an area (usually a county) that should be allocated to a particular model grid cell. Many types of emission inventories contain data summed by county. Because the area of a given county may fall into several model grid cells, spatial surrogates must be used to indicate the fraction of the emissions for that county that is to be assigned to each grid cell that overlaps the county. Currently, EPA uses about 65 different types of surrogates that are computed from about 24 different shapefiles. Spatial surrogates typically must be computed for each grid to be modeled, unless the grid used is a fully aligned subset of a grid for which spatial surrogates are already available.
When surrogates are created, usually there is some type of geographic
attribute that is used to "weight" the county emissions into grid
cells in a manner that is more specific than a simple uniform spread over the
county, which would not reflect the differences in emissions between cities,
rural areas, ports, etc. For example, weights
could be information about points that represent ports (e.g., berths); or the
vehicle miles traveled on lines representing the locations of roads or
railroads; or polygon-based weight attributes such as geographic area,
population, number of households and land use categories. You can create spatial surrogates with the
Spatial Allocator from point-, line-, or polygon-based shapefiles that contain
weight attribute information. A single surrogate (srg) value for
where srg = surrogate, C = county, and GC = grid cell. Note that a surrogate does not have to be for a county; it could be for some other geographic region such as a state, province, nation, or census tract. These polygons are known in the program as the "data polygons."
Three types of surrogates can be created with this program: polygon-based, line-based, and point-based. Polygon-based surrogates use attribute information that is based on area (e.g., population in a census tract). The surrogate value is calculated as the ratio of the attribute value in the intersection of the county and the grid cell to the total value of the attribute in a specific "data polygon" (e.g., county, state, census tract). Examples of polygon-based weight attributes are area, population, number of households, and land use. The numerator (i.e., the value of the weight attribute in the intersection of the county C and grid cell GC) and the denominator (i.e., the total attribute value in the county or other data polygon) are calculated according to the following equations:
where srg=surrogate, C = county (or other data polygon), GC = grid cell, WP = weight polygon, int. = intersection of, i = weight polygon, and n = the number of weight polygons.
For line-based surrogates, the length of the linear weight feature (e.g., railroad, river, road) replaces area in the above equations. For point-based surrogates, instead of using area, the software will allocate a value of 1 if the weight point falls within the region of interest or a value of 0 (zero) if it does not. In some cases, no special weight attribute is desired. Instead, the surrogate is based purely on the area of the polygon, the length of the polyline, or the count of the points. This mode is supported by the Spatial Allocator. In this case, the above equations simplify to the following:
To Section 3.1: Download and Installation