The Southern Appalachian Assessment (SAA), conducted from November, 1994 through April, 1996, relied heavily on geographic information system (GIS) technology and information. The multiagency effort presented a challenge to cooperators on how to effectively work together in achieving the goals and objectives of the assessment within the constraint of a tight timeline. This paper focuses on how the GIS component of the SAA met that challenge, with respect to, organization, development, and implementation of an interagency GIS.
In October, 1994, the cooperating partners of the multiagency Southern Appalachian Man and the Biosphere Program (SAMAB) decided to collaborate on an assessment of the status and condition of the ecological resources in the Southern Appalachian Region. In addition, the regional assessment was to identify and prioritize areas in need of additional protection or ecological restoration. SAMAB is a consortium of federal and state agencies working in partnership to promote ecosystem management and sustainable development in Southern Appalachia. The membership includes: the National Biological Service, the US Forest Service, the US Environmental Protection Agency, the Tennessee Valley Authority, the National Park Service, the US Fish and Wildlife Service, the US Geological Survey, the Department of Energy's Oak Ridge National Laboratory, the Army Corps of Engineers, the Appalachian Regional Commission, the Economic Development Administration, and the States of Tennessee, Georgia, and North Carolina.
It was recognized early in the SAA planning stage that GIS technology and information would play a vital role in the assessment and reporting processes. Thus the primary objective of the SAA GIS was to coordinate and provide the SAA cooperators with an appropriate multiagency GIS infrastructure and analytical support in the multiteam organization of the assessment project. The GIS effort included the identification, compilation, integration, and analysis of ecological and supporting data for the assessment activities. The effort also included providing the cooperators, other researchers, and the public with appropriate access to the information. Given that there were no proven designs or true success stories for multiagency regional ecological assessments, the processes and accomplishments of the SAA and its GIS component are important for other potential multiagency efforts to consider.
The Southern Appalachian Assessment (SAA), conducted from November, 1994 through April, 1996, relied heavily on geographic information system (GIS) technology and information. The multiagency effort presented a challenge to cooperators on how to effectively work together in achieving the goals and objectives of the assessment within the constraint of a tight timeline. The effort to organize, develop, and implement an interagency GIS to support the SAA was a large undertaking. While the effort was called a success, it was not without problems and personal sacrifice.
In October, 1994, the cooperating partners of the multiagency Southern Appalachian Man and the Biosphere Program (SAMAB) decided to collaborate on an assessment of the status and condition of the ecological resources in the Southern Appalachian Region. In addition, the regional assessment was to identify and prioritize areas in need of additional protection or ecological restoration. The study area was defined as the Southern Appalachian Mountains, ridges and valleys from the Northern tip of Virginia well into Alabama. The area, defined with county boundaries, encompasses seven states in its 37.5 million acres.
The SAA Study Area
SAMAB is a consortium of federal and state agencies working in partnership to promote ecosystem management and sustainable development in Southern Appalachia. The membership includes: the National Biological Service, the US Forest Service, the US Environmental Protection Agency, the Tennessee Valley Authority, the National Park Service, the US Fish and Wildlife Service, the US Geological Survey, the Department of Energy's Oak Ridge National Laboratory, the Army Corps of Engineers, the Appalachian Regional Commission, the Economic Development Administration, and the States of Tennessee, Georgia, and North Carolina. The Natural Resources Conservation Service joined the cooperative in February, 1996.
The primary objective of the SAA GIS was to coordinate and provide the SAA cooperators with an appropriate multiagency GIS infrastructure and analytical support in the multiteam organization of the assessment project.
It was recognized early in the SAA planning stage that GIS technology and information would play a vital role in the assessment and reporting processes. In fact, GIS became one of the primary assessment tools. Thus the success of the SAA GIS component was critical to the success of the overall assessment.
The GIS effort included the identification, compilation, integration, and analysis of ecological and supporting data for the assessment activities. The effort also included providing the cooperators, other researchers, and the public with appropriate access to the information. Given that there were no proven designs or true success stories for multiagency regional ecological assessments, the processes and accomplishments of the SAA and its GIS component are important for other potential multiagency efforts to consider.
The SAA GIS was a non centralized organization. There were twelve GIS operation sites coordinated through one site. The sites represented the active participation of the National Biological Service, the US Forest Service, the Environmental Protection Agency, the Tennessee Valley Authority, the National Park Service, US Geological Survey, and the Bureau of Land Management.
The organization idea was to place data and GIS analytical expertise and support closest to the scientists. This style of organization had advantages in that it aligned GIS support directly with the other organizational components of the SAA, but it also proved difficult in terms of coordination. However, given any project with over 150 active particpants at dozens of locations in different agencies, coordination is a challenge.
The overall assessment organization included the four main resource teams of atmospheric, aquatic, terrestrial, and social, economic, and cultural. Both the terrestrial and social, economic, and cultural teams were subdivided into specialty areas. The terrestrial team had a forest health subteam and a plant and animal subteam. The social, economic, and cultural team was divided into a human dimensions subteam, a recreation subteam, a forest products subteam, and a roadless/wilderness subteam. In addition to the resource teams, there was a policy and oversight team, a report writing team, a public involvement team, an integration team, and the GIS/data base team.
The GIS team determined a liaison person for each of the teams or subteams. These individuals were thus members of two teams, the GIS team and their assigned team. Their responsibility was to assist in determining the GIS data and resource needs that the resource team required in their assessment effort. The liaisons were also responsible for monitoring the development of the required data sets for their resource team and determining critical timelines. They also coordinated production of the team's GIS graphics for the technical and summary reports.
ARCView training was provided to SAA scientists so that they could employ GIS directly in their research. Several SAA scientists became very competent on the use of ARCView and it played an important part in their analysis.
Assignments were also made among the GIS team on responsibilities for data set development. That responsibility included source data acquisition, data conversion, enhancements to the desired use requirements, quality assurance, and metadata documentation. Hundreds of data sets were developed and processed during the course of the assessment.
The interaction among the GIS team members was accomplished in several ways. There were some face to face meetings early on in the effort, only one of which had the bulk of the GIS participants in a attendance. Regular team-wide communication contact was accomplished via weekly conference calls and team-wide Emails. Next day mailings were frequent, especially towards the end of the project, as timelines became more critical. Finally, the one-on-one phone communication or face to face encounters at other resource group team meetings were important types of GIS team interactions.
There were problems with the GIS team organization and participation. These problems were largely due to the differences in how different agencies operate and handle their work assignments. However, some of the problems were common to any project. These include personality conflicts, opinion differences, varying skill levels, and time management.
The effort to identify the information needs of the assessment scientists proved to be difficult. The resource teams were being organized at the same time that the GIS team was trying to determine what data sets to compile for whom and when. Furthermore, very few of the scientists had ever worked on a regional assessment prior to the SAA. Many of the scientists were on a learning curve on how to conduct their team's business, so a classical user needs assessment fell short of what was required. The understanding of what the SAA scientists needed was primarily gained through an interactive dialog of GIS members stating what was available and the scientists voicing their developing understanding of what they could do in the assessment.
Some data needs were intuitive and quickly identified. These included, among others, land cover for the SAA and the state heritage program data.
One enormous advantage in the data base compilation process was the fact that all of the cooperators active in the SAA used ArcInfo. An early version problem was encountered, since the US EPA site was still operating version 6.1.1. However, they moved to version 7.0.3 and we became fully compatible among the agencies. Thus all data were converted into a ArcInfo data base compatible with the primary users.
The data base had a standard projection and coordinate system. Since the study area spanned two UTM zones, a decision was made to define a unique SAA projection and coordinate system. The definition was a localized Albers Equal Area Conic projection employing the North American Datum of 1983 (NAD83).
The desire to share data with the agencies and the public drove the objective to use publicly available data as much as possible. It was an overall SAA objective to use existing data in the assessment. Therefore, the data base development effort largely consisted of converting existing data into ArcInfo with the standard projection. Two examples of the data base development were the classification of land cover and the conversion of USGS digital line graph (DLG) data to ARC with additional attributes.
The land cover classification effort, performed by Pacific Meridian Resources, Inc. under contract, was the largest of the data set compilation efforts. The data set, derived from Landsat Thematic Mapper data, was classified into 16 land cover types with a two acre minimum mapping unit.
SAA Land Cover
Traffic count attributes were assigned to the ARC version of the 1:100,000 USGS DLG class 1 and class 2 road segments. That information allowed for some of the modeling efforts to characterize road corridors.
Data was complied at several scales. These include 1:2,000,000, 1:250,000, 1:100,000, and 1:24,000. The 1:100,000 scale proved to be the focus of much of the data analysis. Data layer themes included the following:
Anonymous ftp sites, that anyone could write to, proved to be very important in coverage and file exchange. However, the poor Internet accessibility of the US Forest Service required that tape media be exchange via next day mailings.
The data sets that were complied were sent onto the coordination site where the master data base was complied. As the master data base was complied 8mm tapes were produced and distributed. Additionally, two separate beta versions of the data base (SAA version 1.0 and SAA version 2.0) were written onto CD-ROMs at the US Forest Service site in Columbia, SC. These CD-ROMs were individually written so the number of CDs that were distributed was small.
The major product of the data base development effort was the publication of version 3.0 of the data base. Approximately 1,000 copies of the SAA GIS CD-ROM set were published in March, 1996. The SAA GIS data base CD-ROM set consists of five CD-ROMs. The set is being distributed to libraries, among the participant agencies, and the public on a first come, first serve basis. As of this date, no decision has been made about additional publication copies being produced, however, it is possible that some number of additional copies will be made available on a cost recovery basis.
The metadata documentation of the published CD-ROM set was far from being at the optimal state of release. However, the improved metadata and data sets are accessible through the SAMAB Southern Appalachian Home Page at:
The SAA summary and technical reports are also retrievable from the SAMAB home page.
The SAA GIS team members provided valuable analytical support to the assessment. Often this was performed in collaboration with particular resource team members. Landscape analysis for potential habitat models and recreation based settings was an important example of the SAA GIS contribution.
Analysis of the element of occurrence record (EOR) data, containing information on threatened and endangered species, was performed at only one GIS operation site. This was primarily done in order to limit access to the detailed data which was provided by the state heritage programs with the understanding that the detailed locational level of data was to remain confidential.
The reporting support was initially underestimated. Hundreds of graphics were produced for the technical and summary reports. Standards could have been improved, however, even the basic standards were not always followed through by some GIS team members. Towards the end of the project a great deal of extra effort was required to put GIS graphic projects into a more acceptable form. One of the original concepts of scientists producing their own final version graphics in ARCView fell short from a quality standpoint.
The SAA GIS made a vital contribution to the Southern Appalachian Assessment. GIS graphics and analytical results are a major part of the summary and technical reports. The SAA GIS Data Base CD-ROM set is an important contribution for future analysis by the agencies, special interest groups, and the public. The information represents a huge savings to others who might have otherwise attempted to compile similar data sets.
In terms of what was accomplished, the SAA GIS was a success. The operations could have been improved, however, a foundation for future cooperation among agencies in the region has been well established. While the current cooperation is not perfect, cooperation is occuring in the region.
Karl A. Hermann, GIS Coordinator
Southern Appalachian Ecosystem Initiative
National Biological Service Cooperative
University of Tennessee
17 Ridgeway Road
Norris, TN 37828
samab@utk.edu