National and subnational boundaries are commonly available for use in Geographic Information Systems (GIS) as vector model data. The vector or map model of data in GIS represents objects as discrete phenomena with precisely defined x, y locations (points), strings of locations that define paths (lines), and closed paths that represent extents or boundaries (polygons). Continuously varying phenomena, such as elevation or population, and remotely sensed data are commonly modeled as regular array of cells or pixels with numeric or categorical values in each that combine to make a grid that represents a portion of the earth's surface. For more information on GIS and the raster and vector data models, see the core curriculum in GIS from the National Center for Geographic Information and Analysis.
Integration of vector and raster data requires conversion to the same data model. To calculate zonal statistics based on political units, e.g., to calculate the average elevation of a country, gridded boundaries are necessary. A global grid of national boundaries that is useful for these types of analysis is available from the Gridded Population of the World project (select 'National Identifier Grid' from the 'Select data' dropdown to download). Additionally, a number of useful statistics have been pre-calculated using national-level gridded boundaries and biophysical data sets as part of the Population, Landscape, and Climate Estimates data collection. These statistics allow for statistical analysis of national-level variables with biophysical parameters without using GIS.