Technical Implimentation

The data used for this project were acquired from several sources and have variation in spatial resolution and accuracy. Sources ranged from detailed observations made by wildlife biologists in the field to descriptions of roads and stream networks from the national data bases of the United States Geological Survey (USGS) and the Census Bureau. Additional data were provided to the research team by SANDAG, SCAG, MCB Camp Pendleton, the University of California at Santa Barbara, and others. While most source data were acquired in digital form, some data, such as the county level soils surveys prepared by the Natural Resources Conservation Service (formerly the Soils Conservation Service), were digitized from printed originals. All data were assembled, standardized to a common set of descriptive terms, and combined to produce the study's representation of the landscape.

In the GIS for this project, separate digital "layers," or maps, are used to represent the important aspects of the study area: topography, soils, vegetation, hydrology, roads, existing and planned land use, county and municipal boundaries, etc. Each separate layer is stored in "raster" form, which is a two dimensional array of "grid-cells," or "pixels." Each individual pixel represents a 30 meter x 30 meter area (approximately nine-one hundreths of a hectare, or one-quarter of an acre). Thus, each data layer of the 80km x 134km study area is represented in the GIS as a matrix of approximately 4,000 cells east-west by 3,000 cells north-south, for a total of about 12 million cells. In addition, a number of linear features, such as roads, streams, county, municipal and other legal boundaries, are maintained as a linear or "vector" data base.

The analytical models that use the base data were implemented as computer program modules using the Arc/Info GRID analysis package (Environmental Systems Research Institute, Redlands, California). Additional data re-classification and satellite data interpration was performed in IMAGINE software (ERDAS, Atlanta, Georgia). Each model combines selected layers of the base data to analyze or predict some aspect of the structure or function of the regional landscape. Some models require as an input the results of other models. This chaining process can be seen, for example, in the cougar habitat model which is partly dependent upon mule deer habitat. The alternative future scenarios were developed in Map¥Factory GIS software (Think Space, Ontario, Canada). Each scenario was represented as a land cover map with the same land use classifications as the 1990+ baseline, thus making it possible to compare present and possible future conditions.

The results of most models are represented by one or more thematic maps. A thematic map might represent a conditional state, such as land cover in 1990+, or an evaluation of species richness, or an impact such as loss of productive agricultural soil. Colors are used to identify different categories of that theme, or relative degrees of a characteristic such as density of development or soil moisture. In almost all cases, maps are rendered on shaded relief to clarify the relationships between the map theme, the physiographic terrain, and the hydrologic pattern of the study region, as shown in figure 8.

This study has been conducted using metric measures in Standard International Units (SIU). For the benefit of readers unfamiliar with the metric system, the following approximate conversion factors may be useful: