The data contained in the Glacier Bay Ecosystem GIS has been drawn largely from state and federal agencies. This project would not have been possible without the collaboration, support, and cooperation of these agencies, especially the National Biological Service, National Park Service, Tongass National Forest, USGS EROS Alaska Field Office, and Alaska Department of Natural Resources.
The National Biological Service (NBS) designated Glacier Bay as one of its twelve nationwide ecosystem initiative sites, in part due to the long history of research in Glacier Bay and because of resource management issues there. Although the research history is rich, the information has never been pulled together in a framework that allows information to be easily shared between researchers, used for resource management, or integrated with other scientific results or resource information. GIS technology provides the logical environment in which to integrate, manage, and disseminate this type of information.
The Glacier Bay Ecosystem GIS is an attempt to promote an ecosystem-oriented approach to understanding this area. Although many of the land and resource management agencies with jurisdictional authority for parts of the ecosystem have advanced GIS capabilities and extensive databases, no single agency had the mandate to move beyond their jurisdictional boundaries and develop a comprehensive view of the entire ecosystem. Interrain Pacific has been successful in fostering this perspective by developing cooperative data sharing alliances with these state and federal agencies.
Development of the GIS is also a component of a broader package of assistance and collaboration with the Sitka Conservation Society, Sitka, Alaska. If efforts to build understanding and foster sound stewardship, conservation, and development are to be successful, they must be locally grounded. Using the SCS as a local repository for this information resource and working to build local capacity to use GIS data and technology, we can help move towards empowering rural communities to better influence decisions regarding their local environment.
The specific objectives of the Glacier Bay Ecosystem GIS project are to:
- compile a set of "core" data themes to support ecological research, environmental monitoring, and ecosystem conservation efforts;
- provide a spatial context for organizing, understanding, and analyzing results of future research in the ecosystem; and
- develop a set of tools (techniques, programs, analytic approaches) to allow continued expansion and use of the GIS and foster innovative approaches to understanding the patterns, processes, and functions of the ecosystem.
The core data compiled into the database have been drawn largely from the databases of major state and federal agencies in Alaska. The most important contributions of information have come from the Tongass National Forest, National Park Service, USGS EROS Alaska Field Office, Alaska Department of Natural Resources, U.S. Fish and Wildlife Service, and National Oceanographic Service.
The Glacier Bay Ecosystem GIS database includes a variety of data layers to support analysis at a variety of scales from statewide to site-specific. The system also supports a variety of different GIS data structures or topologies. Arc/Info GIS software provides tools for data capture, processing, analysis, and display of both raster (cell-based) and vector (points, lines, polygons) data with a fully relational database to store information about the geographic features. Additionally, image data (scanned files, satellite data, digital aerial photos, etc.) and 2.5D surfaces are supported.
As such, the database has not been standardized to a given cell resolution or common data structure. Rather, the database includes a wide range of GIS data in a variety of formats and captured at a range of scales.
The spatial data catalog provides an overview of the data themes acquired from agencies and used as a basis for the GIS database. This table is included as a general guide to the data available. The following review of data layers provides additional information on the sources, uses, and constraints of each thematic layer. The Data Structure discussion provides a working guide to the organization of data within the computer. Finally, the Data Documentation section provides the technical information required to make full use of the Glacier Bay Ecosystem GIS database.
|US Geological Survey||Digital Elevation Model (DEM)||1:250,000||Entire ecosystem.|
|US Forest Service||watersheds||1:250,000||TNF|
|US National Park Service||watersheds||1:63,360||Glacier Bay NP|
Digital elevation model (DEM) data is extremely useful as a component of a GIS database for habitat modeling, landscape characterization, visualization, and refinement of vegetation classifications. Specific applications include development of perspective views, creation of shaded relief maps, development of slope and aspect layers, development of contour lines, viewshed analyses, habitat modeling, and others.
The 1:250,000 scale datasets were developed from the 200 foot contour. Originally developed at a resolution of 90 m, these datasets have been resampled to 50 m by USGS to match additional raster datasets used by USGS.
For DEM data, the original elevation datasets were converted to Arc/Info grid files from the DEM format. Slope and aspect grids were then derived from the elevation information. For elevation, slope, and aspect, the "raw" values (meters of elevation, degrees of slope, and compass degrees for aspect) have been maintained.
Watershed boundaries for all primary watersheds (those that empty into salt water) are complete for the Tongass National Forest with the exception of designated wilderness areas (i.e. Admiralty Island NM, West Chichagof / Yakobi Island, Tracy Arm, Russel Fjord). Watersheds for Glacier Bay National Park were digitized from USGS quads at a scale of 1:63,360.
The primary constraints with the 1:250,000 scale DEM datasets are those associated with the original data sources. That is, elevation information based on 100 foot contours at 1:250,000 scale can never be better than the original source data. A comparison between the original 100-foot contours on the USGS quads and 10- foot contours developed from the 1:250,000 scale DEM data indicated a reasonably good correspondence. Although there were some differences between the source and the processed data, the digitally-derived contours accurately portrayed subtleties in the terrain that were evident of the source maps. Coastal areas appear to have more problems in matching the coastline to the zero elevation contour. If used at scales where this mismatching is evident, the elevation data should be reprocessed using water bodies to mask these discrepancies.
Delineation of watershed boundaries in areas overlain by glacial ice was based on best professional judgment for areas within Glacier Bay National Park.
|National Oceanographic and Atmospheric Administration||Digital Nautical Charts||1:80,000||Glacier Bay, Icy Strait, Cape Spencer to Yakutat Bay|
|National Oceanographic Service||Hydrographic Surveys||varies||Intensive coverage of Glacier Bay National Park with the exception of lower Dundas Bay, Taylor Bay, and Scidmore Bay.|
Nautical charts for Glacier Bay, Icy Strait, Cape Spencer, and Cross Sound to Yakutat Bay were scanned by the National Park Service GIS, and georeferenced by Interrain Pacific. In addition, the coastline from these nautical charts was digitized by Interrain Pacific.
The National Oceanographic Service (NOS) publishes digital bathymetric data as: raw depths of all soundings from included hydrographic surveys. These datasets have large holes in the database where no digital bathymetric data exists. Interrain Pacific further processed this information by developing a surface interpolation (using Arc/Info's TIN module) that was subsequently used to create bathymetric contours. To compensate for lack of coverage in NOS hydrographic surveys, depth information was taken directly from the NOAA nautical charts when developing the bathymetric surface coverages.
The "raw" NOS data for Glacier Bay Ecosystem has not been included in the database. This data is in point format and, although not processed sufficiently to include in the core database, is an available resource with additional bathymetric information (e.g. survey number and year).
The major limitations associated with these datasets are from the paucity of data for many areas of Alaska. A future enhancement for the bathymetric database will involve incorporating the NOS data in Dundas Bay and additional information digitized from NOAA charts to extend bathymetric coverage along the outside coast to the edge of the continental shelf.
|US Forest Service||Land Status||1:63,360||Tongass National Forest|
|National Park Service||Coastline||1:63,360||Glacier Bay National Park|
|National Oceanographic and Atmospheric Administration||Nautical Chart||varies||Glacier Bay NP to Yakutat Bay|
|US Geological Survey||Coastline, State of Alaska||1:250,000||Statewide.|
The Glacier Bay Ecosystem GIS contains 9 coastline files, each with slightly different scales, extents, and features. For each district of the Tongass National Forest a coastline was developed based on aerial photo interpretation. These appear to be the most accurate, with a scale better than 1:63,360. For areas within Glacier Bay NP&P, a coastline was digitized by the National Park Service from USGS quads at a scale of 1:63,360. An ecosystem-wide coastline was derived by combining files from each National Forest District as well Glacier Bay National Park. Further, 2 more generalized coastlines include one digitized by USGS at a scale of 1:250,000, and a boundary file for the Tongass National Forest that includes coastline information. Finally, a coastline was digitized from the scanned NOAA nautical charts at scales ranging from 1:40,000 for Cape Spencer,1:80,000 for Glacier Bay and Icy Strait, and 1:300,000 for Icy Point to Yakutat Bay.
One problem associated with coastal mapping in the Glacier Bay ecosystem is that the shoreline is extremely dynamic, particularly at the mouth of tidewater glaciers, where the coastline shifts during glacial advance or retreat or as a result of rebounding following glacial retreat.
|US Forest Service||streams||1:31,680||National Forest Lands|
|National Park Service||streams||1:63,360||Glacier Bay NP&P|
|US Geological Survey||streams||1:250,000||Entire ecosystem|
|US Fish &Wildlife Service||National Wetlands Inventory||1:63,360||Selected areas within Yakutat, Juneau, Skagway, and Sitka Quads.|
|US Forest Service||Hydric soils||1:63,360||Tongass NF|
The most detailed stream and lake coverages are from the Tongass National Forest's stream database. The Tongass National Forest's stream layer include information on presence of anadramous fish species, although this information is variable in age and not documented as to source. A suggested extension of the stream database for Glacier Bay would be to apply the Alaska Department of Fish and Game's Anadramous Stream Catalog coding scheme and link anadramous fisheries data to that database, including the annually updated Anadramous Stream Catalog.
As noted above, information on presence of anadramous fish in this database is incomplete and of varying ages. This data should be reviewed and updated before use.
|US Geological Survey||Bedrock Geology||1:125,000||Glacier Bay National Park|
|AK Dept. of Natural Resources||Soils||1:2,000,000||Statewide|
|US Bureau of Mines||Mineral Locations||varies||Southeast Alaska|
Bedrock geology of Glacier Bay National Park was obtained by several USGS mapping projects, starting in the 1950's and continuing up to the present, with intensive work conducted in 1966 and 1975, 1976, and 1977. The major publications from these studies are listed in the references below. Prepared in cooperation with David A. Brew, Research Geologist emeritus, Alaskan Section, Geologic Division, U.S. Geological Survey.
Brew, D.A., Unpublished, Reconnaissance bedrock geologic map of Glacier Bay National Park, southeastern Alaska: Prepared for publication in the U.S. Geological Survey Miscellaneous Field Map Series, scale 1:125,000.
|US Forest Service||Timber Type||1:63,360||National Forest lands including Yakutat Ranger Dist., Hoonah and Sitka Ranger Dists., Juneau Ranger Dist., and Admiralty Island National Monument|
|US Geological Survey||Landsat Thematic Mapper||30 m pixel||Glacier Bay National Park and Preserve|
|US Geological Survey||AK Interim Land Cover||1:250,000||USGS Quads: Yakutat, Mount Fairweather, Skagway, Juneau|
Vegetation and land cover information is a vital component of any ecologically-oriented database. The capacity to broadly describe the amount and distribution of different vegetation types provides the capability to drive habitat models, conduct "GAP" analyses, and extrapolate information based on finer-grained vegetation descriptions to an ecosystem level.
The information sources listed above cover essentially the entire Glacier Bay ecosystem with some overlap and provide the capacity to broadly characterize vegetation patterns throughout the ecosystem. Each data source uses a different vegetation classification scheme, a result of intended use and method of data automation.
The Tongass National Forest timber type database is based on interpretation of aerial photography at approximately 1:30,000 scale and transferred to 1:31,680 scale basemaps. This is the most detailed digital vegetation information available for substantial areas within the greater ecosystem. The database contains detailed information about timbered areas including species composition, stand density, stocking levels, stand class size, and volume. Many of these timber-oriented descriptions can be useful as surrogates for important ecological variables such as structural information about a given stand. The database also contains considerably less detailed information on non-forest vegetation and cover types.
These vegetation datasets should be used with caution and only at appropriate scales, however. All of the 1:250,000 data sources listed above are suitable for regional scale analyses. The Interim Land Cover dataset is based on Landsat Multispectral Sensor (MSS) satellite imagery with an original resolution of 80 meters. The vegetation classifications used are also general and are roughly equivalent to the Viereck et al. (1992) Level 2 classification. These limitations suggest that these datasets are best suited for broad landscape characterization at 1:500,000 to 1:1,000,000 scale.
The Tongass National Forest timber type database was based on original photography from the mid 1980's. As such, some of the information may be approaching 10 years out of date. For some types of information (e.g., stable plant communities, water bodies) dated information is acceptable; vegetation subject to more rapid change is more problematic.
|Alaska Dept. of Fish & Game, US Forest Service, University of Alaska||Tongass Resource Use Cooperative Survey (TRUCS)||1:250,000||Southeast Alaska|
The Tongass Resource Use Cooperative Survey consisted of 1,467 interviews conducted in 33 Southeast Alaska communities between January 15 and March 13, 1988. The study was directed by the Institute of social and Economic Research of the University of Alaska, Anchorage. Households were selected for the study to yield statistically reliable data at the community level. Interviews were conducted by trained interviewers using structured questionnaires. The principal questionnaire used in the study consisted of three sections: Deer Hunting, Use of Other Resources, and Background Questions. A major part of the interview was devoted to mapping of resource use. Respondents mapped areas used to hunt, fish or gather: deer, salmon, other finfish, marine invertebrates, and marine mammals. This information was digitized and summarized by the North Slope Borough Geographic Information System. The Glacier Bay Ecosystem GIS contains this summarized resource use information.
|US Forest Service||Land Status||1:63,360||Tongass NF|
|US Forest Service||Tongass Land Management Plan||1:63,360||Tongass NF|
|Alaska Dept. Natural Resources||Land Status
(based on BLM records)
Maintaining accurate records of land ownership is laborious, complex, and beyond the scope and resources of this GIS project. The complexity arises in part from the recency of Alaska's statehood and the manner in which Native claims were extinguished in 1971 which both provide a tentative character to land status throughout Alaska. In addition, the different agencies that maintain digital databases on land ownership have different missions and choose to track different types and ownerships of land.
Yet land ownership information is a vital component of any system that is a useful tool for understanding broad ecological, conservation, and economic patterns. An understanding of land owners in an ecosystem — and their values, constituencies, and needs — is an integral aspect of any ecosystem-based approach to conservation, long-range planning, and development.
The readily available sources of land status information in the Glacier Bay ecosystem include the Tongass National Forest and the Alaska DNR; both sources have been used to build a composite, ecosystem-wide land ownership layer. The Tongass Forest database include owner and status and parcel boundaries for all lands within the National Forest boundary. These parcel boundaries are based on legal descriptions, plats, and other sources drafted onto a 1:63,360 scale base. The information in this database is current to 1995; the Tongass National Forest updates the information quarterly.
Land status records from Alaska DNR were used for state lands near Haines and Skagway, and are coarser in resolution than the datasets available through the Tongass National Forest. DNR bases their understanding of land status on a statewide digital (but non-geographic) database maintained by the Bureau of Land Management (BLM). These records are reasonably accurate but do not carry specific parcel boundaries. Ownership is tied to protracted township and section grids, such that if a given ownership type intersects a given section, the entire section will be indicated as being part of that ownership type. At a regional level of analysis, the coarse nature of this data is acceptable, but the user should recognize this limitation of the data source. State lands are current to January 1993; Native corporation lands are based on data from the late 1980s.
The most serious limitation with any land status layer in a GIS is the dynamic nature of the land ownership changes. Alaska is still in a process of transferring lands from federal to state and Native corporation ownership and from state to private ownership under the Alaska Statehood Act, the Alaska Native Claims Settlement Act, and various state homestead and land disposal programs. With this level of change affecting large acreages, a land status layer is seldom completely up to date. As such, frequent updates from the source agencies are recommended.
Another important limitation of this database briefly discussed above is the low level of spatial resolution of digital land records for areas outside of the Tongass National Forest. The Bureau of Land Management (BLM) is the source of land status data for the entire state. Their digital records only allow identification of land ownership to the "section" (640 acres) level. In addition, BLM doesn't track land transfers after a parcel has been transferred from federal to other ownership. For example, BLM would record the transfer of a parcel from federal to state hands, but if the state that offered that parcel to the public through a homestead program, BLM would not track the latter transaction. Applications requiring greater accuracy should use the land ownership boundaries as guides but seek additional, more accurate sources for the final word on land status.
One of the powerful aspects of a GIS is the ability to display spatial information in unique or informative ways. All visual presentations of mapped information are enhanced by visual clues and familiar features that provide context for the viewer. This type of information has been grouped together here under the heading of "base features". The information is drawn from a wide variety of sources and scales as shown below.
|Roads||US Geological Survey||1:2,000,000||Statewide|
|Land Status||Alaska Dept. of Natural
|Coastline||US Geological Survey||1:250,000||Statewide|
|Populated places||US Geological Survey||1:250,000||Statewide|
|Townships||US Geological Survey||Protracted||Statewide|
|Major Rivers||US Geological Survey||1:2,000,000||Statewide|
|1:250,000 Quads||US Geological Survey||Protracted||Statewide|
|1:63,360 Quads||US Geological Survey||Protracted||Statewide|
|Latitude/ Longitude||Alaska Dept.of Natural Resources||Protracted||Statewide|
The Glacier Bay GIS provides both a conceptual framework and formal structure for maintaining and analyzing information about the greater GLBA ecosystem. In a GIS, this framework is a rigorous system of organization that is imposed on the overall directory structure and on individual layers within the database. GIS databases have additional structure imposed by virtue of their geographic nature.
Geographic data must be maintained in a projection system, a grid of coordinates that represent locations of the surface of the earth. There are several projections commonly used in southeast Alaska, each with advantages and disadvantages. The Tongass National Forest, for example, stores their GIS data in a State Plane coordinate system, whereas the National Park Service uses Albers, NOAA nautical charts are registered in Mercator, and USGS maps are in UTM. For point, line, and polygon coverages within the Glacier Bay Ecosystem GIS we selected geographic coordinates with units in decimal degrees as a standard to insure maximum compatibility with all other coordinate systems. Moreover, all grids are stored in UTM Zone 8, corresponding with USGS quads that are used most often for navigation and data collection. For users of Arc/Info, a series of projection files have been included in the /glba_cd/util/project directory, that automate conversion among these and other common projection systems.
The GLBA GIS database is also organized as a logical system of directories and files in a computer. Without a rigorous system of organization, data would be difficult to locate (and therefore be easy to lose) and it would be impossible to write structured programs to query, process, or display the data. Typically a database has a hierarchical, branching structure that is often diagrammed as an inverted tree with "lower" levels (i.e., more specific) branching down from "higher" (more general) levels of the hierarchy.
The UNIX operating system uses such a hierarchical directory structure. A group of logically connected files, such as the Glacier Bay database, is often maintained as a discrete "filesystem", the highest level of organization, with files organized hierarchically in a series of directories and subdirectories to suit the specific needs of the project. When a pathname is referenced in UNIX, the directories are delimited with a slash (/). Throughout the following discussion, full pathnames (i.e., the series of hierarchical directories separated by slashes) will be used to discuss the organization of the database.
Arc/Info also uses a hierarchical scheme to organize information. A directory that holds Arc/Info data is called a workspace. A workspace can contain many types of Arc/Info data. A single thematic layer of information (e.g., streams or vegetation) in a workspace is called a coverage. A workspace may contain one or many coverages in addition to other files. A coverage is actually a collection of files in a directory that describe various aspects of the thematic layer. For the purposes of this discussion, a coverage can be thought of as a single discrete entity.
Although contained on 2 compact disks, the Glacier Bay GIS database is organized as a single file system named "glba_cd". Working and data areas (directories) segregate information related to the database as shown in the diagram below.
||This is the repository for the core database. It is organized into discrete databases according to the area of coverage as follows|
||This workspace contains data themes that cover Glacier Bay National Park. Data are organized into a hierarchical structure based on broad thematic content. Some coverages may contain aspects of multiple categories (e.g. land cover).|
||This workspace contains files relating to biological resources within Glacier Bay National Park and Preserve.|
||This workspace contains USGS Interim Land Cover classification for areas within Glacier Bay National Park and Preserve.|
||This workspace contains data files related to physical features within Glacier Bay National Park and Preserve, including USGS coastline, elevation, geology, hydrology, and a shaded relief grid.|
||This workspace contains data files related to nautical themes including bathymetry, slope and aspect of sub-marine surfaces, nautical contours, and a coastline digitized from NOAA charts.|
|/ charts||This workspace contains scanned and georeferenced NOAA charts.|
|/ cultural||This workspace contained data files related to land ownership, management and cultural features within Glacier Bay National Park and Preserve.|
||This workspace contains data themes that cover areas within the Chatham Area of the Tongass National Forest. Data are organized into a hierarchical structure based on broad thematic content.|
||This workspace contains data themes related to boilogical resources, primarily vegetation and land cover.|
||This workspace contains National Wetlands Inventory files organized by USGS quad at a scale of 1:63,360.|
||This workspace contains USGS Interim Land Cover classification organized by USGS quad at a scale of 1:250,000.|
||This workspace contains data files related to physical features within the Glacier Bay Ecosystem, including USGS coastline, elevation, hydrology, bathymetry of Yakutat Bay, and a shaded relief map.|
||This workspace contains data files related to land ownership, management, infrastructure and cultural features. Areas within the Tongass National Forest generally have best coverage.|
||This workspace contains subsistence use areas based on the Tongass Resource Use Community Surveys (TRUCS), organized by community.|
||The Chatham Area of the Tongass National Forest is divided into districts: Yakutat, Juneau, Hoonah, Sitka, and Admiralty Island National Monument. We used this structure within the Glacier Bay Ecosystem GIS database to store selected files on each area. These directories contain timber type, land status, elevation contours, a detailed coastline, lakes and streams.|
|/ juneau||Juneau Ranger District|
|/ admiral||Admiralty Island National Monument|
|/ sitka||Sitka Ranger District|
|/hoonah||Hoonah Ranger District|
|/ yakutat||Yakutat Ranger District|
||The "util" directory contains a variety of utility programs, files, coverages, etc. to make life a little bit easier. This directory contains the following subdirectories
||Projection files contain the parameters for moving a coverage from one projection to another. This directory contains a wide variety of projection files for the commonly used projection systems in Alaska.