GIS For Environmental Management: A Hierarchical Database Structure

R. Douglas Ramsey, Kimberly Patraw, Brian Biggs, Allan Falconer, Tom Van Neil, Merland Halisky, Richard Spencer
Department of Geography and Earth Resources, Utah State University, Logan, UT 84322-5240. voice (801)-797-3783, Fax (801)797-4048
E-mail: doug@nr.usu.edu

ABSTRACT

Ecosystems function across many scales. However, resource managers must often use data at whatever scale is available. To provide the resource manager with data that can be used to evaluate these systems at various scales and to place local information in a regional context, a database has been generated at Utah State University, Department of Geography and Earth Resources. The database can be used to address environmental management problems at multiple scales. It brings together information gathered by the U.S. Geologic Survey, the National Biological Service, and the Utah National Guard to place ground based information in a local, regional, state, and national context. A graphical user interface has been developed using Arc Macro Language (AML) that allows a user to display and query landcover and landuse, historic and current climatic conditions, topography, plant phenological temporal profiles, critical wildlife habitat, land ownership, field data, and remotely sensed data from 1km AVHRR, to 30 meter Landsat TM, to 1m airborne videography and scanned aerial photos.

Introduction:

Environmental management of local areas often requires data at varying scales, either for placing local data in a regional context, or for observing larger systems that affect local areas. Gathering data for this type of analysis is normally done on a case by case basis. The database created at Utah State University provides data at multiple scales using a graphical user interface that not only provides easy access for the more sophisticated user, but also protects the novice user from missusing the data such as by comparing data at different spatial scales. The Arc Macro Language has been used to design a menu system, and a data structure has been set up to meet the above goals while allowing for easy update and management of both the interface and the database.

Data:

The database is designed to incorporate all types of data at varying scales. Four scales have been chosen as base categories within which data may reside. The chosen scales are: 1)national, 2)ecoregion, 3)state, 4)local. The four scales allow the environmental manager the ability to study and make decisions upon multiple variables at a resolution where the variables are applicable. For example, a state level manager deciding on where to locate a resevoir will require: national scale maps to identify impacts on existing interstate roads and waterways; ecoregion scale maps to determine effects upon ecosystem vitality; state scale maps to identify impact on cities, roads, land ownership, and land cover and use; and finally local scale maps to identify localized vegetation, riparian area, and threatened and endangered species impacts, as well as site geological and anthropogenic influences. In the past, managers have had to compile a variety of data at varying scales on a case by case basis. This entails not only duplication in amassing the data, but also duplication in the integration of the data set into a usable database. A single database of commonly used, defensible data frees the manager from the arduous task of finding and collecting data, and the time can be used for higher quality decision making.

Scale Considerations:
The importance of spatial scale in ecological studies, and the effect scale has on the phenomenon being observed has been widely discussed (Allen, et.al. 1982, O'Neill et.al. 1986, Meentemyer 1989). Multiple scales may be necessary for observing a complete system over time and space. Also, scale has become a questionable item with the advent of GIS, since layers can be zoomed in or out, so it is extremely important to keep data of similar scale grouped apart from data of lessor or greater scale. Otherwise, a person with little understanding of GIS may overlay and analyze data of non-common scale, and obtain completely erroneous results without recognizing any inconsistency. The grouping of compatible data reduces this possibility.

The database provides data in four separate spatial scale groups. The data is amassed and depending upon its scale is placed in one of four data sets of similar scale: national, ecoregion, state, local. The scale range for each category is the following:


National	1:1,000,000 - 1:2,000,000
Ecoregion	1:250,000   - 1:500,000
State		1:100,000
Local		1:24,000    - 1:50,000
Populating the database is done by obtaining the most accurate, quality controlled data from throughout the nation. Much of the data collected for this paper was obtained from four data sources: 1)ArcUSA(tm), 2)USGS Conterminous Land Cover Characteristics Database, 3)NBS Gap Analysis, and 4)the Camp W. G. Williams ecosystem management project. Each of these data sets has been created by peer-reviewed scientists, and has been quality tested to ensure defensibility. The four data sets are also consistent with the four scales mentioned previously, and are selected for their variety of information ranging from anthropogenic to geological to biological. The database is continually changing as higher quality data is created and becomes available.

National Level
ArcUSA(tm), the USGS Conterminous U.S. Land Cover Characteristics Database, and data from the National Climatic Data Center were used for the national level database. The ArcUSA(tm) data was produced at a scale of 1:2,000,000, consistent with the national level, and contains a broad range of data including cartographic features (state and county boundaries, roads, railroads, rivers, lakes, federal land areas, county seats); indexes (latitude/longitude grids, USGS topographic maps, Landsat scenes); and statistical attributes for states and counties (population by age and race, income, hospitals and doctor, local government spending, major soil types, agricultural products raised and sold). The ArcUSA data is dated circa 1992. (Esri, 1992)

The U.S. Geological Survey's EROS Data Center, as part of the U.S. Global Change Research Program, has developed prototype 1-km spatial data sets for global environmental research. The data set relies heavily upon information gathered from Advanced Very High Resolution Radiometer (AVHRR) satellite images. The data set contains a prototype land cover characteristics database for the conterminous United States, bi-weekly NDVI data, ecoregion boundaries as defined by Omernik (Omernik, 1987), 1km DEM derived from the Defense Mapping Agency's 30-arc second data, USGS Land Use Land Cover data, frost-free data, and various derived data sets (Simple Biosphere Model, Biosphere-Atmosphere Transfer Scheme, Onset of Greenness, Peak of Greenness, and Duration of Greenness). The scale of this data is consistent with the national and ecoregion level. These data are circa 1990 (Loveland et.al., 1991). Refer to Figures 1 and 2 bellow for examples of these data.

Figure 1 Figure 2

Data from the National Climatic Data Center was collected from a Unidata system located at Utah State University. Point data were collected and interpolated over a national 10km grid. The result is a database consisting of monthly 30 average minimum, mean, and maximum temperatures and precipitation. In addition, Arc Macro Language code was written to collect current Unidata climate information and produce daily maps of the above variables and project climatic conditions 48 hours into the future. This weather machine is currently in operation at USU and provides 10km historic, current, and future condition maps for inclusion to the national database. Figures 3 and 4 show examples of these grids.

Figure 3 Figure 4

State Level:
The National Biological Service is conducting a nationwide mapping effort on a state by state basis named Gap Analysis. The Gap Analysis program is an effort to map vegetation, predicted vertebrate distribution, and land ownership at a scale of 1:100,000, with a minimum mapping unit of 100 hectares. Utah State University has been a leader in the program, producing the first complete Gap Analysis data set at this scale, and includes a vegetation map consisting of 38 classes, a predicted vertebrate distribution map with 535 species, and a detailed map of land ownership attributed with biodiversity management status. Gap Analysis data is derived from recent Landsat Thematic Mapper imagery, DEM, and ancillary data. The data is usable at the state level scale, and more states can be incorporated as they become available. Utah Gap Analysis data is dated circa 1994 (Edwards et.al., 1995). State level data also includes monthly NOAA-AVHRR NDVI imagery and climatic information as well (Figure 5).

Figure 5

Local Level:
Where local data sets have been created, they are easily incorporated into the database. The Camp Williams National Guard Base contracted with Utah State University to develop multiple data layers at a 1:24,000 scale, consistent with the local level scale for ecosystem management. The Camp Williams data was created over a two year period from 1992 to 1994, and was dervived using a combination of Landsat Thematic Mapper imagery, aerial photography, aerial videography, and ground surveying. The data set consists of physiographic data (soils, elevation, aspect); cartographic data (roads, rivers, buildings); floristic data (vegetation distribution and trend over time, plot transects); historical data (fires, land use); faunal data (vertebrate distribution, threatened and endangered species locations); and remotely sensed data (Landsat TM and MSS, aerial photography and videography) (Shultz, 1993). Figure 6 shows a soil adjusted vegetation index image produced from Landsat TM data over the 25,000 acre National Guard Camp.

Figure 6

Other Data Sets:
Other data sets may be useful in making certain management decisions, and are easily incorporated into the database at the appropriate scale. Examples of these data include: digital elevation data from the Defense Mapping Agency at the 3-arc second and 1-arc second scales, STATSGO soils maps for the nation, TIGER socioeconomic and political boundary data, and 1:100,000 digital line graph data from the USGS.

Graphical User Interface:

A graphical user interface has been designed using Arc Macro Language (AML) to provide querying and analysis functions on the database. Menu systems are used to guide the user to different spatial scales, overlay functions, and analysis capabilities. Metadata is also provided as an option within the menu system. While the interface is designed for the unsophisticated ArcInfo user, insulating him from the commands and disallowing missuse of the data, it also provides the freedom for the expert user to input commands and perform his own analyses. These menus include point and click functions that display images and graphics as well as metadata information and provide analysis functions. Figures 7 through 11 provide examples of menus used to access the database.

Figure 7 Figure 8 Figure 9 Figure 10 Figure 11

Conclusion

The database offers a method of providing data overlays and analysis capabilities to both the sophisticated and unsophisticated user at four levels of spatial scale. It incorporates national, regional, state, and local scales of important geographic, ecologic, geologic, and anthropogenic information to be used by researchers, managers, and policy makers. The database brings together existing data sets into an integrated, easily usable package.

References

Allen, T.F.H. and Thomas B. Starr, 1982. Hierarchy: Perspectives for Ecological Complexity. University of Chicago Press.

Edwards, Thomas C. Jr., Collin G. Homer, Scott D. Bassett, Allan Falconer, R. Douglas Ramsey, and Doug W. Wight. 1995. Utah Gap Analysis: An Environmental Information System, Report. Utah Cooperative Fish and Wildlife Research Unit, Logan, Utah.

Environmental Systems Research Institute. ArcUSA, edition 1. Esri, Redlands, CA. 1992.

Loveland, T.R., Merchant, J., Ohlen, D.O., and Brown, J., 1991, Development of a land cover characteristics data base for the conterminous United States: Photogrammetric Engineering and Remote Sensing, v. 57, no. 11, p. 1453-1463.

Meentemeyer, Vernon 1989. Geographical Perspectives of Space, Time, and Scale. Landscape Ecology. Vol. 3, No. 314, pp. 163-173.

O'Neill, R.V., D.L. DeAngelis, J.B. Waide, and T.F.H. Allen, 1986. A Hierarchical Concept of Ecosystems. Princeton University Press.

Shultz, L., 1993. The Camp Williams project: Ecosystem based management and research on a military reservation: Report of the collaborative research team: Nonpublished paper, 5 pp.