Data management, collection, and inventory approaches can be characterized along several dimensions. Two that are frequently useful to consider are scale and extent of functional/organizational integration (or coordination). The case studies that follow attempt to fill in the matrix in figure 2. Scale and coordination are highlighted at the beginning of each case study description.
| Multi-Agency Multi-Function | Multi-Agency Single Function |
Single Agency Multi-Function | Single Agency Single Function | |
| Ecoregion | . | . | . | . |
| Subregion | . | . | . | . |
| Landscape | . | . | . | . |
| Land Unit | . | . | . | . |
Figure 2. Project dimensions.
Note: Mention of trade names of software and hardware in this report, including in the case studies, is intended for information only, and is not an endorsement or recommendation by the authors or of their agencies.
Case Study 1:
Big Quilcene River Watershed Analysis Data Assembly
A. DESCRIPTION/OBJECTIVE: This case study describes the difficulties
and lessons learned from merging data sets from multiple organizations
at the watershed level. The Big Quilcene River Watershed Analysis
was a joint effort among the Olympic National Forest (USDA Forest
Service), the Washington State Department of Natural Resources,
the Jamestown S'Klallam Tribe, and the City of Port Townsend.
Other Federal agencies, Washington State agencies and private
landowners also contributed data. Information from the analysis
now supports management prescriptions for forest practices on
state and private lands, and is used as guidance for site specific
activities and long range land management planning on federal
lands.
B. SCALE: The analysis was at the Landscape (Watershed) scale. Broader scale data was used for context. The analysis examined the physical, biological, and social processes occurring in the watershed and ecosystem.
C. COORDINATION: Multi-Agency, Multi-Function
D. DESIRED OUTPUT: Joining together datasets so the entire watershed could be analyzed was the desired output of the information management process. The Big Quilcene Watershed Analysis covers 53,016 acres. Eighty percent of this land is under Federal ownership and State and private lands are the other twenty percent. Multiple data sets exist for the area. These sets are managed by various agencies and at the start of the study were in multiple and incompatible formats.
E. SUCCESSES:
Data was successfully brought together. In the process relationships were established that have fostered ever increasing cooperation in the Olympic Peninsula.
True data gaps were identified early in the process. Distinguishing data gaps from data unknowns took research and clarification time up-front. A large amount of initial time was spent updating and analyzing existing data sets. This enabled the watershed analysis team to go into the analysis stage with the best possible data, and identified true data gaps where future data gathering should be focused.
A major success was identifying and implementing a common surface water identification system. The team adopted the Washington Surface Water Identification System (WASWIS) developed by the State of Washington Department of Ecology. The WASWIS identifier acts as a backbone whereby each resource group may attach and retain their own particular surface water data. The worth of the effort became obvious when several spatial data sets were associated to a single indexed coverage. The underlying linear data did not have to be modified. Dynamic segmentation capabilities of GIS software support any mappable change in attribute data being snapped to a single set of hydrology arcs.
A vegetation layer was successfully assembled from a wide variety of source data. This layer takes the most time but is critical to almost all analysis. The team was faced with putting together vegetation information from photo interpretation, activity records, satellite mapping, stand exams, old growth surveys, ecology plots, GAP Analysis, National Wetland Inventory , plant association mapping, historical vegetation surveys, and fire history. The quality of the data varied tremendously. Sequential modifications and overlays were made using all the data to create a "best picture" final product. For example, GAP Analysis data best identified true hardwood patches, and National Wetlands Inventory data best identified nonforested wetlands.
Subsequent analyses on the Olympic Peninsula have been much more efficient. The Big Quilcene experience has led to the agencies addressing building core data sets Peninsula wide and recognizing the role of data stewardship in continuing maintenance of data sets.
F. INPUT (data, tools, skills, cost, time): Many data sources were used to generate GIS layers for the Big Quilcene Analysis. These sources included data from the USDA Forest Service (Olympic National Forest), the Washington State Departments of Natural Resources, the Washington State Department of Fish and Wildlife, and the Washington State Department of Ecology. Data from these various sources were merged together to create watershedwide layers. This merging took many forms, but whenever possible only data sets of similar data accuracy were merged together. For new data USDA Forest Service Region Six "Guidelines for Map Manuscript Preparation for Digitizing" were followed. A standard operating map scale of 1:24,000 was selected for data gathering, manuscripting, and digitizing. Whenever possible only data sets with similar data accuracy were merged together. Data accuracy was documented in a data dictionary which also included a summary of the process used to create the coverage, a map showing the extent of the coverage, and a description of the attributes associated with the line, point, or polygon.
Arc/INFO GIS Software and the ORACLE relational database system were used to manipulate the map layers and associated attribute data. This included generation of stand age and seral stage from fire history, harvest, and plant association group models.
Two employees worked full time and one employee worked halftime on information management and GIS analysis. It took six months to reach a point where all data was ready for analysis. From that point on, it was a quick trip, with the analysis completed in three more months.
G. LESSONS LEARNED: The Big Quilcene River Watershed Analysis is only one of a number of watershed analyses that have been completed in the Pacific Northwest. The first fifteen of these analyses were considered pilots, and an intensive effort was made to garner lessons learned, including lessons learned in data management (USDA, USDI November 1994). The "lessons learned" list below incorporates both Big Quilcene River Watershed Analysis lessons and the lessons garnered from the larger sample of pilot watersheds.
1. Different areas have different complexities and differing degrees of datarichness that make a standard allocation of time for data compilation difficult. What is clear is that assembling data takes the lion's share of the time necessary to complete landscape scale analysis (figure 3). The building of layers standardized across the land owners and managers beforehand could slash the time for analysis.
Keep the amount of information standardized and assembled on a broad basis minimal. The challenge is having enough but not so much the layers will never be built. For example, based on the Big Quilcene experience, a province wide successional stage layer is being built on the Olympic Peninsula. The basic attributes are fire age, managed stand age, plant association group (potential vegetation modeled from basic environmental variables), and whether existing vegetation cover is hardwoods or conifers. Successional stage is interpreted from the basic attributes.
Creation of standard layers would be better done incrementally (watershed by watershed) than not at all. If the data sets are maintained, the next round of analysis on the same land will be many times more efficient. The added difficulty (if there is any) of building a standardized layer is paid back many times.
2. By the end of the Big Quilcene Analysis resource there was thorough appreciation of the value of data stewardship. Responsibility needs to be assigned for keeping data uptodate. This is a resource area task, not a GIS task.
3. Include an information or GIS manager on the analysis team from the beginning. Until resource specialists understand what a GIS can and cannot do well, this interaction is essential.
Figure 3. Watershed analysis eight steps and percent time spent on each - Sol Duc Watershed
4. The analysis team must have a clear charter that allows identification of key questions. Managers or line officers need to articulate the issues that are within their decisionmaking space with respect to the analysis at hand. Get written signoff on the issues and key questions. If the team does not know clearly what the questions are, data collection and aggregation tangents will significantly prolong the analysis process. All parties to the analysis need to be involved in identifying the questions and the data type and quality sufficient to answer them. Different functions, organizations, and agencies often have considerably different risk levels they are willing to tolerate.
Lack of clarity resulted in the Big Quilcene Analysis collecting lots of data that was not used. Subsequent watershed analysis efforts in the Olympic Peninsula have been much more efficient. A certain amount of floundering cannot be avoided, but it certainly can be minimized.
5. The analysis team has several options if data is not available to meet needs:
Generally there is not a lack of resource information. Finding information that is relevant and in a form that can be used is the difficulty.
Be thorough in searching for available data. Data discovered later can back up analysis midcourse. Resource people on the analysis team must query their counterparts. Leaving the searching for data to GIS people does not work well.
6. Merging data sets from different agencies was a very high cost endeavor. It is very difficult to merge databases that were built for different reasons. Data quality is lost in bringing data to the least common denominator. Interagency standards need to be set at higher than the local level for basic data themes such as vegetation and hydrography. These data standards need to address how features are delineated and identified as well as attributed. It is most important to have common arcs (i.e., exact same digitized locations) between agencies, even if the data attached to the arcs are different.
Developing these data standards is no easy task, but not having some type of common standard is costing all data users in time and efficiency.
7. The ability to network agencies (both state and federal) together so information can easily be shared would greatly enhance analysis efficiency, and needs to be addressed at a scale larger than the landscape scale. However, keep in mind that data sharing goes well beyond the ability to exchange data. Data assembly, getting multiple data sets into a single useable format, is the time hog.
8. It definitely is advantageous for agencies to share common GIS software. It also helps greatly to define common word processing and spreadsheet software up-front, if possible.
9. A Forest by Forest, State Agency by State Agency, approach to developing partnership agreements (Memorandums of Understanding) would be a very time consuming and redundant process. The Olympic National forest pursued a state wide approach, but gave up when time constraints started to affect efficiency. A province wide MOU was developed and implemented. The Forest Service has since developed a statewide MOU with the Washington State Department of Natural Resources.
10. The analysis done for the Report represents only a small portion of what could be done with the data. One of the big spinoffs of these efforts is creation of a data set that can be used for more in depth followup analysis.
11. Cooperation is not easy. It has worked on the Big Quilcene and subsequent efforts on the Olympic Peninsula because partners have clear benefits from being partners. The Washington State Department of Natural Resources, for example, has greatly accelerated the rate at which it completes examination of Watershed Administrative Units mandated by State Forest Practices legislation through it's cooperation with the US Forest Service.
H. SUMMARY/CONCLUSIONS:
Different agencies have their own way of doing things, and this is reflected in data sets that are difficult to link together. For the Big Quilcene Watershed Analysis, this created a tremendous workload. High GIS and analytical skills and hard work brought the data sets together.
With proper data preparation, the time consuming data assembly part of analysis can be greatly reduced. The most efficient way for this data preparation to take place is before watershed analysis begins. If data could be developed across multiple data management groups for some common core elements, analysis level and efficiency would be greatly increased. Focus on a core, doable set.
Insist that management give you as clear a charter for your effort as possible. Lack of clarity in charter will multiply data management costs due to tangents and dead ends.
Data sets created have high value beyond their use in the immediate analysis. Data stewardship of the data must be established to assure maintenance of the data and to maximize continuing returns on a high investment.
I. REFERENCES AND/OR CONTACTS:
U.S. Department of Agriculture, Forest Service. 1990. Region 6 guidelines for map manuscript preparation for digitizing. Portland, OR: U.S. Department Of Agriculture, Forest Service, Pacific Northwest Region. Misc. pagination. Obtain from Cathy Askren, Resource Information Specialist, USDA Forest Service, Region 6, P.O. Box 3623, Portland, OR, 97208, 503-326-2507.
U.S. Department of Agriculture; Washington State Department of Natural Resources. November, 1994. Big Quilcene Watershed Analysis: An Ecological Report at the Watershed Level. Olympia, Washington. Obtain copy of executive summary and information management section from: Rodney Matye, Watershed Assessment Team Leader, USDA Forest Service, Olympic National Forest, 1835 Black Lake Blvd SW, Olympia, WA 985125623. (360) 956-2438.
U.S. Department of Agriculture; U.S. Department of the Interior. November, 1994. Summary of Proceedings of the Watershed Analysis Conference: FY94 Federal Experience. Obtain from Regional Ecosystem Office, P.O. Box 3623, Portland, OR, 97208, (503) 326-6165.
A. DESCRIPTION/OBJECTIVE: The Alaska Geographic Data Committee (AGDC) is a subcommittee of the Federal Geographic Data Committee and is composed of representatives from twenty Federal agencies and five Departments within the State Of Alaska. The AGDC facilitated the formation of a successful multi-agency partnership for the collection of digital hydrography. A cost share/work share partnership was developed among the US Geological Survey (USGS), the Bureau of Land Management (BLM), the US Fish and Wildlife Service (USFWS), and the National Park Service (NPS) for the production of updated digital hydrography over user specified federal lands in Alaska.
B. SCALE: 1,178 topographic quadrangles at 1:63,360
C. COORDINATION: Multi-Agency, Multi-Function
D. DESIRED OUTPUT: Updated hydrography in digital form (DLG-3 format). The hydrography on the base topographic mapping series (1:63,360) is updated using photorevision methods and transformed into the specified digital form. The updated digital hydrography will be used in fisheries management, recreation management, land transfer and acquisition activities, Wild and Scenic River management, oil spill contingency planning, water rights quantifications, etc.
E. RESULTS
1. Successes
AGDC provided the forum for the establishment of the Hydrography Subcommittee. The Hydrography Subcommittee provided the appropriate forum for statewide leadership and coordination for hydrography updating and related digital data collection.
2. Lessons Learned
Continue to use of the AGDC as a forum for agencies to build geographic information partnerships.
3. Promising Possibilities
The AGDC has established several other Subcommittees. These are the Land Cover Subcommittee, Clearinghouse Subcommittee, and the Government-owned/Contractor-operated Subcommittee.
4. Sharing
The AGDC plans to invite the boroughs (a.k.a. counties) to join.
F. INPUT: The production of updated digital hydrography is being done via contract. All the agencies have provided funding. BLM and USGS also provide oversight. BLM oversees the photorevision and USGS oversees the digital production.
G. SUMMARY/CONCLUSION: The AGDC has provided statewide leadership for surveying, mapping, and related spatial data coordination. For the foreseeable future, the AGDC will be a viable avenue for building geographic information partnerships among government institutions.
H. REFERENCES AND/OR CONTACTS
Contact:
Gust Panos
Chief, Branch of Mapping Science
222 West 7th Avenue #13
Anchorage, AK 99513-7599
Phone: 907-271-4549
CASE STUDY # 3:
The Gap Analysis Program
A. DESCRIPTION/OBJECTIVE: The Gap Analysis Program (GAP) is a cooperative effort to map: (a) natural land cover, (b) vertebrate species, and © the lands that are managed in ways that maintain biological diversity. The purpose of Gap Analysis is to identify the "gaps" in our network of conservation lands regarding habitat types as well as individual vertebrate species, and to build partnerships around the development and application of this information. A complete description of GAP and a 94-95 Status Report is available from the Homepage:
http://www.nr.usu.edu/gap/gaphome.html
B. SCALE: GAP produces map products with a nominal scale of 1:100,000. The current generation landcover maps have a minimum mapping unit (mmu) standard of 100 hectares, though many state projects have exceeded this and new projects are producing "pixellevel" maps from TM satellite imagery, or a mmu of 30 meters. Analyses are intended to be provided at a "mesoscale" within a broad ecoregional context, using a coarsefilter approach to identifying the elements of biodiversity that are inadequately represented in the existing network of conservation lands. The unit of analysis is the Bailey's Province or Section depending on regional needs. The intent is to identify the boundaries of each ecoregion using data from the ECOMAP project.
C. COORDINATION: Multi-Agency, Multi-Function
D. DESIRED OUTPUT:
(1) Landcover Map at the Natural Landcover Classification System (UNESCO/TNC) Alliance level;
(2) Predicted Vertebrate Distribution of all native, terrestrial vertebrate species;
(3) Public Land Ownership based on, or equivalent to the PLSS/Ownership mapping of the BLM; Management Status of all lands according to 4 levels of maintenance for biodiversity;
(4) Analysis indicating the area and percent of distribution of each element (native plant community type and vertebrate species) per ownership and management category;
(5) List of Elements having <10%, <20%, <50% of their area in status 1 & 2 managed lands.
E. RESULTS:
1. SUCCESSES:
Funded Gap Analysis projects are either under way, completed, or in early organizational stages in 46 states. There are requests for project startups in the remaining states as well as Puerto Rico, U.S. Virgin Islands. Thirtyone refereed journal articles have been written on the results of various Gap Analysis projects; fourteen of these have been published in the past two years, including one CD ROM. At least 40 more journal articles are in press at the time of this writing.
Under the MultiResolution Land Characteristics Consortium (MRLC), GAP joined with the U.S. Geological Survey (USGS), the Environmental Protection Agency (EPA), and the National Oceanographic and Atmospheric Administration (NOAA) to make the first joint purchase of Landsat Thematic Mapper satellite images covering all of the contiguous 48 states. This joint purchase saved the government millions of dollars in direct costs as well as an estimated 30 million dollars in combined program costs. The MRLC also resulted in an agreement to have the USGS provide all satellite image preprocessing, archiving, and distribution at significant additional savings to the program and increased the comparability of the final map products created by individual researchers.
Staff members participated on the Federal Geographic Data Committee's Vegetation Subcommittee to help with the development of federal standards for natural land cover types. They also worked with The Nature Conservancy and the National Park Service Mapping Program to ensure comparability of land cover classification.
Over the past two years, GAP investigators pioneered the development of airborne video for labeling land cover maps and assessment of map accuracy. Workshops on accuracy assessment of both the land cover maps and the vertebrate distribution maps were held to develop guidelines for the state projects. Using these guidelines and other methods, GAP investigators in Utah, California, Idaho, and Massachusetts conducted some of the first accuracy assessments of largearea regional land cover and animal species maps. Additionally, the first independent validation of gaps was conducted in California. Gap Analysis researchers have been fieldtesting hypotheses of vertebrate models and determining which life history characteristics are needed to improve modeling the distributions of species.
Guidelines for conducting an aquatic Gap Analysis project were developed and a draft aquatic GAP manual developed. A pilot aquatic GAP project was begun in New York to test methods and assumptions.
Two external reviews of Gap Analysis have been completed. The first, chaired by Dr. Ervin Zube of the University of Arizona, was an independent peer review. The second review was conducted for the National Council of the Paper Industry for Air and Stream Improvement, Inc. (NCASI). Both reviews provided evaluations of and guidelines for the program. Both of these critical reviews recognized GAP as an invaluable tool for solving land management and conservation problems.
GAP investigators recently edgematched the California, Idaho, Oregon, and Washington land cover maps. Other investigators are currently rendering a unified land cover map of the fourstate Mojave Basin ecoregion and of the sixstate Great Basin ecoregion. Results to date indicate minor differences relating to variability in the labels used for the original maps.
GIS data, reprints, manuals, and other GAP related information are available over the Internet for Arizona, Arkansas, California, Colorado, Idaho, Nevada, Oregon, Utah, Washington, and Wyoming at http://www.nr.usu.edu/gap. There are an average of about 16,000 queries per month to the GAP home page.
There have been more than a hundred major uses made of the GAP data sets in the last two years. Some of the categories of uses include wildlife management, county planning, land use planning by private corporations, basic research, generation of options for large-area designations, and environmental assessments.
Utah State University, in partnership with the National Biological Service (NBS), the USGS, and the Environmental Systems Research Institute, Inc. (ESRI), published the Utah GAP data set on a CDROM. Plans are under way to develop similar products for California, Arkansas, and Washington in a userfriendly, menu-driven format for use on a personal computer.
Educational outreach programs using GAP data include "NatureMapping," developed in Washington State, which has involved some 340 classroom teachers. NatureMapping provides instructional guidelines for teachers on collecting field data, specifically designates educational software for data entry, database building, and instructional video and professional support through local planning offices, Audubon Society chapters, and the Senior Environmental Corps.
Among the workshops conducted by GAP over the past two years are map accuracy assessment, aquatic GAP, standards for reporting data and project results, integration of socioeconomic data, and state-level biodiversity planning. The Gap Analysis Program also cohosted a symposium with the American Society of Photogrammetry and Remote Sensing (ASPRS).
2. LESSONS LEARNED
While GAP has learned a great deal about the technical development of biological thematic maps, the greatest lesson has been in the amount of followup required to ensure that the data is adequately distributed, integrated into users' routine applications, and the results of analyses are utilized for planning and decision making.
To that end, GAP is developing userfriendly data products for CD and Internet distribution, and surveying our cooperators about the need for workshops and new ideas about implementation. The greatest need is for the emergence in each state of a cooperative consortium that will maintain the data, conduct analyses and research, and initiate cooperative planning and management based on the results. Such groups have emerged in several states, and GAP is trying to identify ways to foster them in all others. At a minimum, followup funding is required to support a technical GAP position for some time after completion of the basic GAP data. This "extension agent" position is needed to ensure the transition to a statebased cooperative organization to install the data in cooperators systems, help them develop GIS analysis systems, train them in the use of the data, and conduct analyses for them if needed.
E. INPUT
1. Data: GAP input data varies by project, but overall, includes Landsat Thematic Mapper imagery, aerial video (analog and digital), aerial photos, previously developed landcover and vertebrate distribution maps, vertebrate occurrence records, vertebrate habitat associations, ownership maps, management plans, and ancillary GIS coverages such as DEMs, road and stream DLGs, EPA River Reach, NWI, TIGER files, U.S. Census data, precipitation, temperature, political jurisdictions, mountain ranges, and STATSGO Soils.
2. Tools: Considerable computing capability is required to carry out a project including use of 1 to several highend workstations, PCS, large format printers, scanners, and digitizers. A typical GAP project will need 1220 gigabytes of mass storage, and 8mm tape drive. Software includes ArcInfo, ERDAS, PCI, GRASS, SPECTRUM, SkyKing, and various data base packages, among others.
3. Skills needed include: remote sensing/imagery processing, plant ecology, wildlife biology, project management, cooperator coordination ,and outreach.
4. Cost and Time: GAP projects are funded at levels that vary by land area, landscape heterogeneity, opportunity for cooperative funding and inkind contributions, lab capability, etc. The average expected budget is between $400,000 $500,000 though costs have been reduced through the MRLC cooperative imagery purchase, and technological innovations. Most states acquire considerable inkind services which is a necessity. For example, the GAP Peer Review identified a target budget of approximately $1 million per state project. Project completion time also varies by the above factors, but averages approximately 4 years. We also anticipate the time requirement to come down as mapping techniques become more efficient.
F. SUMMARY/CONCLUSIONS: As GAP transitions from a research project to a program, it will have continuing challenges in securing and maintaining funding sufficient to complete the contiguous states in a timely manner, conduct updates for older projects, and continue research into new mapping and analysis methods. In particular, the recent completion of several projects now provides a data set sufficient to begin new research in biodiversity analysis, particularly representation, complimentarily of regions to represent biodiversity, surrogates for total biodiversity, and viability of habitats. The other major challenge is to integrate GAP data and analysis into the routine applications of it's cooperators and all potential data users, and to foster the emergence of cooperative biodiversity planning within each state.
G. REFERENCES AND CONTACTS:
Most information and individual state summaries and contacts can
be obtained from the Homepage at
http://www.nr.usu.edu/gap/gaphome.html,
or contact the National GAP office at 208/885-3555,
gap@uidaho.edu,
530 S. Asbury St., Ste.1, Moscow, ID 83843.
Contact:
Patrick J. Crist
Western States Coordinator
Gap Analysis Program
208/885-3901
References: (editor note: only part of this list was received)
Scott, J.M., B. Csuti, J. E. Estes, and H. Anderson. 1989. Status assessment of biodiversity protection. Conservation Biology 3(1):8587.
Scott, J.M., B. Csuti, and S. Caicco. 1991. Gap Analysis: Assessing protection needs. Pages 1526 in W. E. Hudson, editor. Landscape linkages and biodiversity. Island Press, Washington, D.C.
Scott, J.M., B. Csuti, and F. Davis. 1991. Gap Analysis: An application of geographic information systems for wildlife species. Pages 167180 in D.J. Decker, M.E. Krasny, G.R. Goff, C.R. Smith, and D.W. Gross, editors. Challenges in the conservation of biological diversity: A practitioner's guide. Westview Press, Boulder, Colorado.
Scott, J.M., B. Csuti, H. Anderson, and J.E. Estes. 1989. Letters:Biodiversity. Science 245:477480.
Scott, J.M. 1990b. Preserving life on earth: We need a new approach . Journal of Forestry 88(3):1314.
Scott, J.M., B. Csuti, D. Stoms, and F. Davis. 1992. Remote sensing for nongame wildlife habitat management. Pages 134140 in Transactions of the 56th North American Wildlife and Natural Resources Conference (1991).
Scott, J. M., B. Csuti, and E. T. LaRoe. 1992. Protecting biodiversity (Letter). Issues in Science and Technology 8(4):21.
Scott, J. M., and M .D. Jennings, editors. 1994. A handbook for GapAnalysis. Idaho Cooperative Fish and Wildlife Research Unit, University of Idaho, Moscow, Idaho.
Scott, J .M., B. Csuti, K. Smith, J.E. Estes, and S. Caicco. 1991. Gap Analysis of species richness and vegetation cover: An integrated biodiversity conservation strategy. Pages 282297 in K. Kohm, editor. Balancing on the brink of extinction. Island Press, Washington, D.C.
Scott, J .M., S. A. Temple, D. L. Harlow, and M. L. Shaffer. 1994.Restoration and management of endangered species. Pages 531539 in T.A.Bookhout, editor. Research and management techniques for wildlife and habitats. The Wildlife Society, Bethesda, Maryland.
Scott, J. M., B. Csuti, K. Smith, J. E. Estes, and S. Caicco. 1988. Beyond endangered species: An integrated conservation strategy for the preservation of biological diversity. Endangered Species UPDATE 5(10):4348.
Scott, J. M., B. Csuti, and K. Smith. 1990. Commentary: Playing Noah while paying the devil. Ecological Society Bulletin 71(3):156159.
Scott, J. M., E. T. LaRoe, and M. D. Jennings. 1995. Gap Analysis: Ageographic approach to planning for biological diversity. Page 462 in E. T. LaRoe, G. S. Farris, C. E. Puckett, P.D. Doran, and M.J. Mac, editors. Our living resources. National Biological Service, Washington, D.C.
Scott, J. M., and G. Wright. In press. The need for a hierarchical approach to conservation. In
J. M. Scott, T. H. Tear, and F. Davis, editors. Gap Analysis: A landscape approach to biodiversity planning. American Society for Photogrammetry and Remote Sensing, Bethesda, Maryland.
Scott, J. M., and B. Csuti. 1991. Gap Analysis: Blueprint for proactive conservation. FOCUS on Renewable Natural Resources 17:12.
Scott, J. M., T .H. Tear, and F. Davis, editors. In press. Gap Analysis: A landscape approach to biodiversity planning. American Society for Photogrammetry and Remote Sensing, Bethesda, Maryland.
Scott, J. M., F. Davis, B. Csuti, R. Noss, B. Butterfield, C. Groves, H. Anderson, S. Caicco, F. D'Erchia, T.C. Edwards, Jr., J. Ulliman, and R.G. Wright. 1993. Gap Analysis: A geographic approach to protection of biological diversity. Wildlife Monographs 123:141.
Scott, J. M., T. H. Tear, and L.S. Mills. 1995. Socioeconomics and the recovery of endangered species: Biological assessment in a political world. Conservation Biology (9):214216.
Scott, J. M., B. Csuti, J. D. Jacobi, and J. E. Estes. 1987. Species richness A geographic approach to protecting future biological diversity. BioScience 37:782788.
Scott, J .M., and B. Csuti. In press. Gap Analysis for biodiversity survey and maintenance. In M.L. ReakaKudla, D .E. Wilson, and E. O. Wilson, editors. Biodiversity II: Understanding and protecting our biological resources. National Academy Press, Washington, D.C.
Scott, J. M., E.D. Ables, T. C. Edwards, Jr., R. L. Eng, T. A. Gavin, L .D. Harris, J. B. Haufler, W.M. Healy, F. Knopf, O. Torgerson, and H. P. Weeks, Jr. 1991. Conservation of biological diversity: Some perspectives and a look to the future for the wildlife profession. Ad Hoc Committee on Conservation of Biological Diversity, The Wildlife Society.
Shaffer, M. In press. The conceptual basis for Gap Analysis. In J. M.Scott, T. H. Tear, and F. Davis, editors. Gap Analysis: A landscape approach to biodiversity planning. American Society for Photogrammetry and Remote Sensing, Bethesda, Maryland.
Shaw, D. M., and M. D. Jennings. 1995. Federal data bases of land characteristics. Pages 467468 in E.T. LaRoe, G. S. Farris, C. E. Puckett, P. D. Doran, and M. J. Mac, editors. Our living resources. National Biological Service, Washington, D.C.
Slaymaker, D., K. Jones, C. Griffin, and J. Finn. In press. Uses of aerial videography in Gap Analysis for deciduous forests in New England. In J.M. Scott, T.H. Tear, and F. Davis, editors. Gap Analysis: A landscape approach to biodiversity planning. American Society for Photogrammetry and Remote Sensing, Bethesda, Maryland.
Smith, C. R. 1991. Challenges in the conservation of biological resources: An epilogue. Pages 367376 in D. J. Decker, M. E. Krasny, G. R. Goff, C.R. Smith, and D. W. Gross, editors. Challenges in the conservation of biological resources: A practitioner's guide. Westview Press, Boulder, Colorado.
Smith, K., and D. Catanzaro. In press. Modeling predicted vertebrate distributions: Potential pitfalls and problems. In J. M. Scott, T. H. Tear, and F. Davis, editors. Gap Analysis: A landscape approach to biodiversity planning. American Society for Photogrammetry and Remote Sensing, Bethesda, Maryland.
Stine, P.A. 1995. A multiscale conservation assessment of plant communities in southern California. Ph.D. dissertation. University of California, Santa Barbara, California.
Stine, P.A., F. W. Davis, B. Csuti, and J. M. Scott. In press. Comparative utility of vegetation maps, prepared with different spatial resolutions for conservation planning. In R. Szaro, editor. Biodiversity in managed landscapes. Sacramento, California, July 1317, 1992. Oxford University Press, New York.
Stoms, D. M. In press. Validating largearea land cover databases with maplets. Geocarto International.
Stoms, D .M., F. W. Davis, and A .D. Hollander. In press. Hierarchical representation of species distributions for biological survey and monitoring. In Proceedings of the Second International Conference/Workshop on Integrating GIS and Environmental Modeling. Breckenridge, Colorado, September 2630, 1993.
Stoms, D. M. 1991. Mapping and monitoring regional patterns of species richness from geographic information. Ph.D. dissertation, Department of Geography, University of California, Santa Barbara.
Stoms, D. M. 1992. Effects of habitat map generalization in biodiversity assessment. Photogrammetric Engineering and Remote Sensing 58:15871591.
Stoms, D. M., and F. Davis. 1995. Biodiversity in the southwestern California region. Pages 465466 in E.T. LaRoe, G. S. Farris, C. E.Puckett, P. D. Doran, and M. J. Mac, editors. Our living resources. National Biological Service, Washington, D.C.
Stoms, D. M., F. W. Davis, C. B. Cogan, M .O. Painho, B. W. Duncan, J. Scepan, and J. M. Scott. 1993. Geographic analysis of California condor siting data. Conservation Biology 7:148159.
Stoms, D. M., F. W. Davis, C .B. Cogan, and K. Cassidy. 1994. Assessing land cover map accuracy for Gap Analysis. Pages Validation 2.12.20 in J .M. Scott and M. D. Jennings, editors. A handbook for Gap Analysis. Idaho Cooperative Fish and Wildlife Research Unit, University of Idaho, Moscow, Idaho.
Stoms, D.M. 1994. Scale dependence of species richness maps. Professional Geographer 46:346358.
Stoms, D.M., F. W. Davis, and C. B. Cogan. 1992. Sensitivity of wildlifehabitat models to uncertainties in GIS data. Photogrammetric Engineering and Remote Sensing 58:843850.
Stoms, D. M., F. W. Davis, P. A. Stine, and M. Borchert. 1992. Beyond the traditional vegetation map towards a biodiversity database. Pages 718726 in Proceedings of GIS/LIS'92, San Jose, California, November 1012, 1992.
Stoms, D.M. 1994. Actual vegetation layer. Pages Data Layers 1.11.6 in J.M. Scott and M.D. Jennings, editors. A handbook for Gap Analysis. Idaho Cooperative Fish and Wildlife Research Unit, University of Idaho, Moscow, Idaho.
Stoms, D.M., and J.E. Estes. 1993. A remote sensing research agenda for mapping and monitoring biodiversity. International Journal of Remote Sensing 14(10):18391860.
Tear, T.H., J.M. Scott, P. Hayward, and B. Griffith. 1993. Status and prospects for success of the Endangered Species Act: A look at recovery plans. Science 262:976977.
Tear, T. H., J. M. Scott, P .H. Hayward, and B. Griffith. 1995. Recovery plans and the endangered species act: Are criticisms supported by data? Conservation Biology 9(1):182195.
Thomas, K., and F. Davis. In press. Applications of Gap Analysis data in the Mojave Desert of California. In J. M. Scott, T. H. Tear, and F. Davis, editors. Gap Analysis: A landscape approach to biodiversity planning. American Society for Photogrammetry and Remote Sensing, Bethesda, Maryland.
Vickerman, S. In press. State biodiversity plans. In J .M. Scott, T. H. Tear, and F. Davis, editors. Gap Analysis: A landscape approach to biodiversity planning. American Society for Photogrammetry and Remote Sensing, Bethesda, Maryland.
Walker, R .E., D .M. Stoms, J. E. Estes, and K. D. Cayocca. 1992. Improved modeling of biological diversity with multitemporal vegetation index data. Pages 562571 in Technical papers of the 1992 annual meeting of ASPRS/ACSM. Albuquerque, New Mexico, March 38, 1992.
Walker, R .E., D. M. Stoms, F .W. Davis, and J. V. Wagtendonk. 1992. Modeling vegetation cover types froma topographic gradient in the southern Sierra Nevada. Pages 794803 in Proceedings of GIS/LIS'92, San Jose, California, November 1012, 1992.
Walsh, S .J., and F. W. Davis. 1994. Applications of remote sensing and geographic information systems in vegetation science. Journal of Vegetation Science 5:609756.
Winckler, S. 1992. Stopgap measures. The Atlantic Monthly 269(1):7481.
Winn, D. S., and T .C. Edwards, Jr. 1991. A process for evaluating, monitoring, and conservation of landscape biodiversity within the intermountain area, USDA Forest Service. USDA Forest Service, Intermountain Region, Ogden, Utah.
Wondra, P. In press. Data access: An NBS overview. In J. M. Scott, T. H. Tear, and F. Davis, editors. Gap Analysis: A landscape approach to biodiversity planning. American Society for Photogrammetry and Remote Sensing, Bethesda, Maryland.
Wright, R. G., J. G. MacCracken, and J. Hall. 1994. An ecological evaluation of proposed new conservation areas in Idaho: Evaluating proposed Idaho National Parks. Conservation Biology 8(1):207216.
Wright, G. In press. Applications for planning and expanding national parks. In J.M. Scott, T.H. Tear, and F. Davis, editors. Gap Analysis: A landscape approach to biodiversity planning. American Society for Photogrammetry and Remote Sensing, Bethesda, Maryland.
CASE STUDY 4 :
Great Lakes Assessment
A. DESCRIPTION/OBJECTIVE: Federal, state, and private land managers need various types of environmental, biological, socioeconomic, and process information for conducting natural resource assessments, planning and management. Although considerable information is either currently available or is being collected, this information often resides at different institutions, is archived at different scales and in different formats, and is typically inaccessible to government agencies or the general public except on a case by case basis. This case study describes a promising possibility for managing this information collectively in the Northern Great Lakes Region to improve its coverage, quality, and accessibility.
B. SCALE: Regional
C. COORDINATION: Multi-Agency, Multi-Function
D. DESIRED OUTPUT:
The northern Lake States comprise one of the most densely forested regions of the nation, with 41% of the total area or 50.5 million acres in forested lands. About 46 million acres of this forest land is considered commercial forest, and 52% of this commercial forest land is owned by the nonindustrial private sector. The private sector relies upon information made available through federal and state programs, whereas federal and state agencies routinely collect and manage information concerning their respective land holdings.
The second-growth forest of the Lake States is now reaching commercial maturity, and while demands for forest products are increasing within this region, lands available for timber production continue to decrease due to urban and industrial development, construction of second homes, and the emergence of conflicting land uses. Surveys indicate that urban Americans perceive a need for decreased emphasis on forest commodity production and increased emphasis on noncommodity values (e.g., recreation and aesthetics), yet rural users depend upon forests for employment and community development. These divergent expectations, and associated issues, sum to a need for obtaining and using the best information available on ecological and social conditions in the Great Lakes region for natural resource planning and management.
An interagency proposal involving the USDA Forest Service and Natural Resource Conservation Service, USDI National Biological Service, the US Environmental Protection Agency, the University of Wisconsin at Madison, the University of Minnesota at Duluth, Boise Cascade Corporation, and Mead Corporation was made to the National Performance Review Technology Innovation Fund. Funds were granted to complete a two phase project for the Great Lakes Region. The first phase will compile existing information previously managed by different federal and state organizations. This includes (1) environmental information on climatic gradients, surficial geology, ecological units, soils, hydrography, drainage patterns, and contaminants; (2) biological information on current forest conditions, the biogeography of game, nongame, and threatened and endangered species; (3) socioeconomic information on land use and ownership, human demographics, recreational demand, and road densities; and (4) ecological process information on the frequency and effects of natural and anthropogenic disturbance associated with fire, wind, and flooding events, and resource development and consumption.
Spatial information will be processed according to specified data standards in a common geographic information systems (GIS) data format. The standards will be made available to government agencies and the private sector on CDROM and over the Internet. Equipment, technologies, and expertise available at the USDA Forest Service, the USDI National Biological Service, the US Environmental Protection Agency, the USDA Natural Resource Conservation Service, the University of Wisconsin at Madison, and the University of Minnesota at Duluth will be utilized to meet these objectives.
The second phase will be the development of data visualization and decision support systems for natural resource planners and managers. It is important that the information described above be easily understood by the public, policy makers, and other nonscientific audiences. Recently a number of tools for visualizing complex environmental data have become available. These tools allow users to interactively display information on landscape features in three dimensions from digital elevation data. Users are then able to view changes in this three-dimensional landscape in terms of forest cover, urban expansion, risk from fire or floods, or prescribed management activities. The use of three-dimensional renderings of the landscape, texture mapping ("painting" a landscape with textures derived from aerial or satellite images), and realtime animation (flyovers or landscape changes over time) can provide tools by which the public can clearly see the consequences of planned and natural changes to the regional landscape.
In the area of natural resource management, these Geographic Information Systems databases and data visualization and decision support tools can be integrated with landscape simulation and forest management models (e.g., forest succession, fire spread, flood risk, insect outbreaks, management practices, or urbanization). Rather than presenting data tables or static map information, the Great Lakes project proposes to develop a decision support system which combines the maps described in phase one with computer simulation models and data visualization tools to allow a user to interact with natural resource information. Among the capabilities of this system are the abilities to I) explore a regional landscape in terms of relevant geographic data layers such as land use, land cover, roads, and rivers, in two or three dimensions, ii) run simulation models which result in landscape change (fire, forest growth, forest harvest), and iii) conduct 'what-if" scenarios on the landscape to help users better understand the results of different management plans or policies (Host et al. 1994).
A corporate information system that includes spatial information in a GIS medium, and narrative information including literature reviews, reports, and research findings, and other information for interagency and private sector applications will be developed. All data except for threatened and endangered species locations will be distributed at appropriate resolution on Internet by the Environmental Management Technical Center, National Biological Service. All data except for threatened and endangered species locations will be distributed at full resolution on CD by the Environmental Management Technical Center. Information will be accessed through a runtime version of a GIS program such as ArcView2 (Environmental Systems Research Institute) to allow 'end user' to do needs to conduct specific analyses. A data visualization environment designed to run on advanced microcomputer and workstation systems will be accessible through disk or CDROM. We will also make this software accessible as a site on a World Wide Web server. Written documentation and online training information will be provided.
(Editor Note: Dave Cleland needs to complete with successes, input, lessons learned)
CASE STUDY 5:
Interior Columbia Basin Ecosystem Management Project
Information System Development and Documentation
(editor note: case study not complete - when complete requires review of Rebecca Gravenmier, ICBEMP Spatial Analysis Team Leader)
A. DESCRIPTION/OBJECTIVE: Land use planning that is regional in scope and proportion requires information management tools and processes that go beyond the typical techniques of landscape and resource analysis. The Interior Columbia Basin Ecosystem Management Project (ICBEMP) was the largest interagency database development effort ever undertaken by the agencies involved, covering more than 144 million acres (58 million hectares). Though few projects are on this mega-scale, many of the processes and tools developed and lessons learned in supporting the ICBEMP are applicable to smaller efforts.
B. SCALE: Regional
C. COORDINATION: Multi-Agency, Multi-Function
D. DESIRED OUTPUT: Efficient information management support to the Science Integration Team and Environmental Impact Statement Teams. Managed information release processes. Transfer of processes and products to the agency implementers of direction and guidance resulting from the Project.
E. INPUT (DATA, TOOLS, SKILLS, TIME, COST):
A Spatial Analysis Team was established to manage data and support the analytical needs of the ICBEMP. This interagency group was charged with the collection of available GIS data, capture of new data, management of data, and analysis to support the Science Integration Team (SIT) and Environmental Impact Statement (EIS) Teams. The Spatial Analysis Team was also responsible for the documentation and distribution of data to project personnel, federal agencies, publics, and other interested parties.
Members of the Spatial Analysis Team were located in Walla Walla, Washington; Portland, Oregon; Missoula, Montana: Wenatchee, Washington; Boise, Idaho and Coeur d'Alene, Idaho. This dispersed approach to data collection and analysis was necessary due to the cost and inefficiency of relocating all equipment and personnel to one location.
Over 100 computers, from powerful UNIX workstations to desktop personal computes, played a role in the work of the entire ICBEMP. Dominant GIS software was ARC/INFO (Environmental Systems Research Institute, Inc.), but GRASS (Construction Engineering Research Labs, U. S. Corps of Engineers), ArcView2 (Environmental Systems Research Institute, Inc.), UTOOLS (Umatilla National Forest, Pendelton, OR) and LT4-X (Infotec Development, Inc) were also used. Analysts used Oracle, Microsoft ACCESS and EXCEL, Borland PARADOX, and QUATTRO-PRO to summarize and evaluate both non-spatial and spatial data.
Telecommunications played a key role in uniting the work and systems of the many project work sites. Transfer of large data sets required high transfer speed and capacity exceeding typical communications lines and hardware. Fractional T-1 and frame relay services were used. The wide area network (WAN) created by this infrastructure connected to local area networks (LANs) at the work sites. In this way PC'S and workstations communicated and shared data.
Over 170 different GIS data layers were compiled or created in support of the Project. The data layers were derived from source maps, photos, or transfer media ranging from 1:12,000 to 1:4,000,000 in scale. Some GIS layers mapped features continuously across the entire basin, and others were for discrete area only (e.g. subsample areas). Many of the landscape data layers were collected for an area larger than the assessment area so that ecological units could be mapped entirely.
The data with the best resolution were not necessarily the best data. The best data for the project were the data that met the analytical needs, were continuous across the Basin, and were consistent.
(describe databases - CRBLAD, Other)
(describe models - VDDT, CRBSUM)
The Project used an Oracle distributed database (Spatial Unified Data Dictionary - SPUDD) to document information about both planned and completed data sets, and to track analysis product requests. SPUDD, however, was not table to track changes to data after it was checked in. The Project designed an Automated Tracking Manager (ATM) to track GIS data and associated metadata and recreate GIS processes if needed.
F. SUCCESSES AND LESSONS LEARNED:
The information and technology plan developed for the ICBEMP was functional and fairly efficient for a project of this scope. The overall project was larger than was actually planned for in terms of the number of people involved, the amount of data collected, and the scope of analysis performed. There were several bottlenecks, problems, or items that could be improved upon for future projects of this type.
Data Development, Documentation, and Analysis
Many of the GIS data sets took over nine months to compile or create. Time could have been spent more efficiently if key base data, such as subwatersheds and broad scale vegetation (1km), had been available at the beginning of the project. Many GIS themes that should have been based on subwatersheds were created before the subwatershed layer was finalized. This caused rework of data once the subwatershed layer was available. A stepwise approach should have been followed for completion of many of the base biophyscial layers.
For a given piece of ground, mapping and classification of the environment will vary widely between different scientists, organizations, and individuals. Some of this is a function of need, and some is a function of differing interpretations, knowledge, and skills. Some of the problems encountered when trying to put together a continuous data set from different sources included disagreement in features (spatial edge mapping), classifications, and attributes. Integrated data sets, without any more specific data on where each piece of data came from, were only as accurate as the least accurate data set that went into their creation.
(add techniques used by project for integration)
Overall metadata and change tracking using SPUDD and ATM worked very well. Metadata could have been better reviewed, edited, and verified against the actual GIS data layers early in the project, however. Much time was spent after the fact, correcting inaccurate information or gathering more from the data source.
A formal quality control process for analysis and data layers was in place in Walla Walla and Portland, the hubs for data assembly, integration, organization, and management. Both the GIS analyst and the appropriate Science Integration Team member reviewed GIS data layers before they were used for analysis. However, in several instances data from the individual Forests or BLM Districts was missing, and this was not caught until later and required reprocessing of data and analysis. Under projects with tight time lines, it is often difficult to perform quality review of information when individuals are simultaneously performing a variety of high priority tasks. The team tried to keep the spatial and attribute features of the data as reliable as possible, as well as ensuring that no logic errors were introduced in the analysis or summarization routines.
Standard map templates were developed and shared among analysis sites for large size plots (1:1,000,000 and 1:2,000,000) early in the project. Page-size templates to ensure all maps were the same extent and could be overlayed were not developed until much later in the project. This caused frustration for managers when they tried to use maps from a variety of sources as overheads and they did not match one another.
The ICBEMP data release process was very successful. By having a specialized group handle and process data release, analysts were able to concentrate on GIS analysis. The acquisition, release, and distribution of data via the Internet did not materialize until the end of the project. This was unfortunate from a number of aspects, including interagency connection at the federal level, coordination with other agencies and organizations (data collection and distribution), problem solving, and information sharing. An effective information sharing route would be the use of World Wide Web/anonymous FTP to distribute data and share information with agencies and organizations with Internet connections. This is currently underway and ICBEMP data is expected to be made available in the spring or summer of 1996.
Many of the SIT members were unable to determine up front what their analysis needs were going to be for a project of this scope. A formal information needs assessment was not completed for the ICBEMP. This would have been helpful in managing the scope of the project, as more data and analysis were continually added throughout the life of the project. However, since much of the data and analysis had not been developed before, the creation of an information needs assessment plan was not possible at the onset of the project.
Distributed Work Sites and Systems
The telecommunications network between the distributed analysis sites allowed resources that would not have been available otherwise, human and machine, to be accessed for the project. It would have been difficult to physically move some of these resources as they were also dedicated for other activities and duties outside the ICBEMP. Additionally, costs of relocating these personnel and equipment were reduced. On the other hand, the lack of face-to-face communication hindered understanding of both data and analysis needs. It also took more time to solve problems and answer questions concerning analysis requests between the various offices.
The Spatial Analysis Team was ultimately responsible for all the GIS data and associated databases collected or created for the ICBEMP. The Spatial Team Lead, however, was not in direct control of the data collection or analysis at sites other than Walla Walla or Portland. the Landscape Ecology and Aquatic Teams were directing the data collection and analysis at the other locations. This created problems with data standards and documentation, quality control of products, and the setting of work priorities. It may have been more efficient to have all GIS analysis reporting to one Spatial Team Lead. Less time would have been spent trying to correct data and documentation or to understand analysis that had been completed at other sites.
G. SUMMARY/CONCLUSIONS:
The ICBEMP experience shows that information management is possible and critical in an interagency environment. The difficulties encountered also highlight the importance of a long term information resource management strategy for the implementation of ecosystem management. A key element for ecosystem management is the need for consistent, current, and accurate information concerning the ecological and biophysical environments across the landscape. The collection and management of this data and information between federal, state, and local agencies needs to be effectively coordinated and shared. An ongoing commitment to ecosystem management and modeling makes it absolutely necessary for federal agencies to reassess their individual resource inventory programs and strive for a set of interagency standards tied to ecosystem management.
H. REFERENCES AND/OR CONTACTS:
Reference:
(anticipate above publication will be published summer, 1996)
CASE STUDY 6 :
Development of Interagency Vegetation Standards in the Pacific Northwest
A. DESCRIPTION/OBJECTIVE: The Northwest (President's) Forest Plan directs the building of an interagency Geographic Information System. This case study describes the development of standards for the vegetation themes.
B. SCALE: Regional
C. COORDINATION: Multi-Agency, Multi-Function
D. DESIRED OUTPUT:
E. SUCCESSES:
Final standards were approved by the Intergovernmental Advisory Committee in April of 1996. The set of elements is small and doable total tree canopy cover, forest canopy structure, tree overstory size class , species, stand year of origin, land cover class, plant series, plant association group, and plot location.
Implementation of these standards requires the data producing agencies to examine how their individual efforts can be focused to produce unified products. This process is now underway the team membership now being those individuals with program responsibility for vegetation data production and management in the resource management agencies, the GAP program, and the private sector.
F. INPUT:
A small interagency team, the Vegetation Strike Team, was assembled in January 1995 to identify information needs, formulate data standards, and make implementation recommendations.
Needs were identified through a formal information needs assessment survey of the PNW pilot watershed analysis efforts (see Big Quilcene River Case Study) and personal interviews of resource specialists from throughout the region. Criteria were developed to focus the wide array of information currently being used in watershed analysis into a set of basic, highly used, core elements.
Proposed core elements had to pass through a feasibility sieve where the ability to actually assemble the data across the full range of the President's Forest Plan was evaluated. This evaluation considered the present available agency data and the ability to capture the elements by remote sensing.
The results of the Strike Team findings were summarized in a report which, along with a voting form, was distributed to a larger interagency team, the Data Coordination Team, comprised of one representative from each agency and Northwest Forest Plan provinces for comment and review. The report was also distributed to other Northwest Forest Plan work groups, the Federal Geographic Data Committee vegetation subcommittee members, individuals interviewed by the strike team and numerous others.
Entire elapsed time from initial team meeting to approved standards was fifteen months. The co-leads for the team devoted a major share of their work time to the effort during that period.
G. LESSONS LEARNED
Factors making for success of this effort included tight scope control, the review mechanisms, the decision on an analysis scale the standards would serve (watershed level), and the ability to do backwards needs analysis by looking at pilot watershed analyses information use. Direction from a body make up of agency heads (Intergovernmental Advisory Committee) was also key.
The need to move immediately to implementation is recognized. There is ability to tier the effort down to finer level standards efforts with fewer partners. This is currently being done as standards are developed for project level vegetation information between the Oregon State Office of the Bureau of Land Management and Region 6 of the US Forest Service.
Private land confidentiality must be respected. The team assured coverage expected over the entire range of the Plan was obtainable with remote sensing.
It will be easier when the Internet is in more common use to assure all parties have the opportunity to see a standards proposal but Internet distribution will never substitute for directed interviews and surveys. These are very time consuming but this effort would have failed without them.
The test of real success in this effort will be in implementation.
H. SUMMARY/CONCLUSIONS
Seek to be effective in standards efforts and not absolutely right. Effectiveness depends on dedicated resources, wide directed participation, scope control, and interagency management review and decision authority.
I. REFERENCES AND/OR CONTACTS:
Contacts:
Catherine Askren
USDA Forest Service, Region 6
P.O. Box 3623
Portland, OR 97208
Phone: 503-326-2507
Chris Cadwell
USDI Bureau of Land Management
Eugene District
P.O. Box 10226
Eugene, OR 97408
Phone: 541-683-6120
Reference:
CASE STUDY 7:
Forest Plan Monitoring:
Role of Remote Sensing and GIS on the Mark Twain National Forest
A. DESCRIPTION/OBJECTIVE: To apply change detection procedures using Landsat Thematic Mapper (TM) imagery and a geographic information system (GIS) to monitor the Forest Land and Resource Management Plan (FLRMP) accomplishments in conjunction with the midpoint review.
B. SCALE: Target scale is Landscape and Land Unit
C. DESIRED OUTPUT: Maps showing the amount of change or disturbance during the plan period versus the amount desired under the plan. These maps were one measure of how well the forest was progressing towards the vegetation conditions specified under the desired future conditions in the plan. A secondary set of maps were needed to display management alternatives in easily understandable visual formats to the public.
D. RESULTS
1. SUCCESSES
Using Landsat TM for classification of land cover change proved to be an excellent source of information for forest plan monitoring. The methods were easily understandable and readily applicable to larger areas of land.
2. LESSONS LEARNED
Landsat TM imagery provided a comprehensive view of the forest across ownerships that allowed resource specialists to evaluate the landscape as a whole, not just the pieces owned by the Forest Service.
3. PROMISING POSSIBILITIES
The initial change detection study was conducted on a small test area on forest. Procedures developed on the pilot area have since been applied forest wide by resource specialists and have become a part of the forest plan review process.
4. SHARING
Resource specialists have made presentations about this project to the private sector.
E. INPUT: ERDAS and ARC/INFO software were used to perform change detection on the Landsat TM imagery. The land cover change detection was combined in a GIS with other forest data (i.e., stand inventories, ownership, etc) to produce the desired output. The total cost of implementing the change detection process across the entire forest was less than $0.20 per acre.
F. SUMMARY/CONCLUSION: The results of the Landsat TM change detection classification and the GIS overlay analysis performed by the forest suggest that forest canopy changes can be accurately detected and classified in a straightforward procedure by using multitemporal Landsat TM images. The derived classifications can aid in the monitoring of forestrelated activities over time and results can be readily integrated into a GIS.
G. REFERENCES AND/OR CONTACTS:
Mark Twain National Forest:
Mike Schanta (m.schanta:r09f05a)
401 Fairgrounds Rd.
Rolla, MO 65401
314-364-4621
Remote Sensing Applications Center
Henry Lachowski (h.lachowski:w03a)
2222 West 2300 South
Salt Lake City, UT 84119
801-975-3662
CASE STUDY 8:
Range Allotment Mapping
A. DESCRIPTION/OBJECTIVE: To evaluate Landsat Thematic Mapper (TM) imagery and airborne video for mapping range vegetation on an allotment and to familiarize range resource managers with current remote sensing technology.
B. SCALE: Target scale is Landscape and Land Unit
C. DESIRED OUTPUT: A map containing 12 classes of vegetation and digital files of the allotment boundary, existing pastures, and proposed pastures. These products, stored in a GIS, were used to illustrate pasture management alternatives as input to an allotment analysis.
D. RESULTS
1. SUCCESSES
Using airborne video data tagged with GPS proved to be a successful means of collecting more detailed vegetation information about a particular site. The information from the airborne video was then used to improve the Landsat TM vegetation classification.
2. LESSONS LEARNED
Airborne video and Landsat TM provide a useful complement of both closeup and large area perspectives on the resources being mapped, though if only one type of data could be used for large area mapping, it should be the Landsat TM.
3. PROMISING POSSIBILITIES
Use of this data for other purposes including calculation of acreage for vegetation types, development of field sampling schemes for ground data collection, and development of management alternatives when used with ancillary data.
4. SHARING
E. INPUT: ERDAS software was used to produce the vegetation map. The vegetation map was combined with the other digital files in ARC/INFO to produce the desired output.
F. SUMMARY/CONCLUSION: The level of accuracy (67%) of the vegetation map from Landsat TM was adequate for performing many functions in an allotment analysis, particularly for stratifying the area for field sampling.
G. REFERENCES AND/OR CONTACTS:
Santa Fe National Forest (Coyote RD):
Roberto Martinez (r.martinez:r03f10d01a)
Coyote, NM 87012
505-638-5526
Remote Sensing Applications Center
Henry Lachowski (h.lachowski:w03a)
2222 West 2300 South
Salt Lake City, UT 84119
801-975-3662
CASE STUDY 9:
Developing a Wild and Scenic River Management Plan
A. DESCRIPTION/OBJECTIVE: Support the development of a Wild and Scenic River Management plan by creating several digital and paper visual products to display management alternatives and their possible implications and to help in presenting the plan to the public.
B. SCALE: Target scale is Landscape and Land Unit
C. DESIRED OUTPUT: Various maps depicting an assortment of existing and proposed recreational and management activities along the East Fork of the Jemez River.
D. RESULTS
1. SUCCESSES
Remote sensing and GIS proved to be useful tools for this project, providing managers with uptodate information about the resources they manage. By combining remote sensing and GIS data sets in different ways, new applicationspecific maps were quickly created and modified as needed
for scenic quality assessment, management planning, and recreational opportunity promotion.
2. LESSONS LEARNED
The maps were intended to be used at public meetings as an interesting sideline. Instead they became the focal point of the meeting (see SHARING below). The public "connected" with the information portrayed in a visual (or map) format faster and more easily than they had in the past.
3. PROMISING POSSIBILITIES
Three dimensional scenic quality assessment maps providing an overview of the recreation area from a particular vantage point were one of the products generated for this project. To take this information further, data visualization software could be used to examine the projected outcome
of various management alternatives for scenic quality impacts prior to implementation.
4. SHARING
Maps produced for the Wild and Scenic River Management Plan were used by resource specialists at public meetings to present visual descriptions of existing and proposed changes in management practices, existing and proposed recreational uses, and vegetation diversity. The maps became a focal point for the meetings, quickly getting people involved by stimulating conversation and facilitating the flow of information between the public and land managers.
E. INPUT: The data sources for the visual products included satellite imagery, airborne videography, GPS, DEM's, and CFF's. Software used included ERDAS for processing of the satellite imagery and ARC/INFO for combining the satellite imagery with other data.
F. SUMMARY/CONCLUSIONS: Remote sensing and GIS were important tools for preparing the management plan, providing basic spatial information in a visual, easytounderstand format. Since the data were in digital format, maps could be modified to suit a particular application or reflect recent changes in the data.
G. REFERENCES AND/OR CONTACTS:
Santa Fe National Forest
(Carl Linderman has left the forest and is now in Region 6)
Remote Sensing Applications Center
Henry Lachowski (h.lachowski:w03a)
2222 West 2300 South
Salt Lake City, UT 84119
801-975-3662
CASE STUDY 10:
Use of satellite data for resource management on Kootenai
A. DESCRIPTION/OBJECTIVE: Use of satellite digital data and "photo composites" for: updating stand inventory maps, capturing current road information, used as backdrops on 615 equipment, and classifying the data for use in wildlife models, supplement vegetation information, and producing a snap shot in time.
B. SCALE: Data was registered and geocoded to the 1:24,000 scale. It is used from the project level to forest wide applications.
C. DESIRED OUTPUT: 1:24,000 timber compartment images produced by merging TM and PAN data, windowed independent of quad lines. 1:24,000 "orthophoto" like SPOT images registered and windowed by quad. 1:126,720 scale merged TM/SPOT false natural color forest wide image. Data from both data sets for classification purposes and backdrops used on the 615 platform.
D. RESULTS
1. SUCCESSES
Use of the hardcopy products was immediate and wide ranging. Updates of timber compartment maps and capturing current roads data was especially useful. The forest wide image was well accepted and provided our specialists, managers, and customers a common base and time to monitor changes into the future. We are only now beginning to derive benefits from the classified TM data. It is becoming useful for modeling sensitive and endangered species habitat and as supplemental vegetation information.
2. LESSONS LEARNED
Classification of the TM data, for multiple applications is time consuming. Mangers need to think beyond a single data set and determine what skills are needed, where within the organization, and how often or if we will have a recurring need for classified data. Issues of resolution and scale are complex and need to be addressed to gain efficiencies.
3. PROMISING POSSIBILITIES
As availability to the 615 hardware increases, use of satellite and other digital imagery is moving closer the ground and the resource specialists. The combination of the technology, current data, and field going subject matter expertise offers us a better chance at understanding and implementing management practices that will consider process, function, and structure, over a broader landscape perspective.
4. SHARING
Sharing of this technology has been accomplished through use. The first use was through the hardcopy "orthophoto" like quads and the compartment maps. Secondly the forest wide image was and continues to be a great way to share with the public. And third, the use of the classified TM data as it is applied to species modeling and vegetation classification. Bottom line... make the technology and data useful for work on the ground and sharing will happen.
E. INPUT: ERDAS/IDIRISI/ARCINFO were used for the classification effort. The SPOT data is used as is for backdrops and "orthophoto" like quads. The TM and Pan were merged to improve contrast in the compartment maps and forest wide image.
F. SUMMARY/CONCLUSION: The panchromatic data was useful immediately. It enabled the Forest to have fresh information that could be put to use quickly; hardcopy products to take to the field and images to use, as backdrops, on our GIS platform. The TM data will have many more long term uses but has required more time and expertise to make it valuable. We have no doubt that the future of this technology is bright, however, program management (for multispectral data) needs to be addressed. What skills, are need where, for future and recurring applications.
G. REFERENCES AND/OR CONTACTS:
Kootenai National Forest:
Dale Hawley (d.hawley:r01f14a)
506 US HWY 2 West
Libby, Mt. 59923
406-293-6211
Remote Sensing Applications Center
Henry Lachowski (h.lachowski:w03a)
2222 West 2300 South
Salt Lake City, UT 84119
801-975-3662
CASE STUDY 11:
An Investigation on the Use of Digital Orthophoto Quadrangles in a GIS
A. DESCRIPTION/OBJECTIVE: To examine procedures for loading and manipulating digital orthophoto quadrangle (DOQ) data, to determine possible uses for the data in a GIS, and to examine the data characteristics of file size and storage requirements, radiometric differences, and registration to other layers in GIS.
B. SCALE: Target scale is Land Unit
C. DESIRED OUTPUT: A set of procedures for loading, viewing, and manipulating DOQ data, a list of possible uses of DOQ data as well as estimates of file size and storage requirement, and maps showing radiometric differences and registration of DOQ data with other GIS data layers.
D. RESULTS
1. SUCCESSES
Procedures were developed to load and manipulate DOQ data. Possible uses of the data are listed under PROMISING POSSIBILITIES. File size and storage requirements were determined and maps were generated showing radiometric differences and registration to other GIS data layers.
2. LESSONS LEARNED
The accuracy and standards of any GIS data used with the DOQs will affect the apparent registration with DOQ image data. For example, at 1:24,000 scale the accuracy standards for CFFs, which are a high quality and highly detailed data set, allow for 90% of the data to be within 12 meters of their true location (and for 10% to be more than 12 meters from true). So, when overlaying the CFFs on the DOQs, the data is within accuracy standards even if it is up to 12 pixels off (in the DOQs, 1 pixel = 1 meter on the ground).
3. PROMISING POSSIBILITIES
Some possible uses of DOQ data may include: revising or updating digital vector layer files, updating/verifying vegetation inventories, habitat analysis, fire activity analysis (fuel types, flight plans, etc), crown closure determinations, environmental impact assessment, change detection studies, flood analysis, soil erosion assessment, watershed analysis, and perspective view analysis.
4. SHARING
E. INPUT: ERDAS software was used to import and manipulating the DOQ data. ARC/INFO software was used to examine the registration of the DOQ data to other GIS data layers.
F. SUMMARY/CONCLUSIONS: A DOQ could be used in a project anytime a high resolution, orthorectified, black and white image was needed. When working with a DOQ data, the factors of registration to other data layers, radiometric differences, and file size constraints in a project must be considered. These uses and considerations will need to be explored more thoroughly with detailed research and in specific, task oriented projects.
G. REFERENCES AND/OR CONTACTS:
Superior National Forest
Edward Lindquist (e.lindquist:r09f09a)
515 W. First Street
Duluth, MN 55801
218-720-5483
Remote Sensing Applications Center
Henry Lachowski (h.lachowski:w03a)
2222 West 2300 South
Salt Lake City, UT 84119
801-975-3662
CASE STUDY 12:
Fire Analysis for Ecosystem Management using Landsat Thematic
Mapper (TM) Satellite Imagery on the Payette National Forest
A. DESCRIPTION/OBJECTIVE: The project had three objectives: to test the utility of Landsat TM imagery to derive an existing vegetation layer; to develop a pre and postfire vegetation classification scheme that could be used with Landsat TM data for a variety of resource analyses; and to map fire intensity and severity and use this data to analyze firecaused change in landscape patterns.
B. SCALE: The target scale for this project is Landscape Unit.
C. DESIRED OUTPUT: Creation of pre and postburn classification maps to support several ongoing tasks such as wildlife habitat resulting from diverse landscape mosaics created by the fire, identification of tree mortality, and identification of potential prescription burn areas and facilitating fire suppression.
D. RESULTS
1. SUCCESSES
Landsat TM imagery proved to be a very useful tool for quickly and inexpensively producing a general burn intensity map.
2. LESSONS LEARNED
In a comparison of visually interpreted high altitude photography and digital classification of Landsat TM imagery for evaluating forest fire burn intensity it was found that the Landsat TM imagery was capable of accurately stratifying the landscape into high intensity burn, other burn, and nonburned areas. High altitude photography was still needed to separate the medium and low intensity burn areas.
3. PROMISING POSSIBILITIES
If a good quality Landsat TM scene is available, it may be the most cost effective method for getting a quick burn classification map. The best approach may be to classify the burn intensities into 3 classes high intensity, moderate/low intensity, and unburned. While not providing a highly detailed burn intensity map, it would provide a solid basis for stratification and a guide for field study over large areas which could lead to a more detailed and accurate map of burn intensities. This is important because the analysis of fire caused landscape patterns over large areas will greatly assist efforts to implement a landscape approach to management on National Forests.
4. SHARING
E. INPUT: ERDAS Imagine software was used to classify the Landsat TM imagery. The derived vegetation classification was combined with other GIS data layers from Cartographic Features Files (CFFs) and other sources using ARC/INFO software.
F. SUMMARY/CONCLUSIONS: The use of remote sensing technologies is an important part of fire management and fire ecology. Having the capability of analyzing broad regions and assessing change is a vital part of resource management. The Payette National Forest has mapped burn intensity for the Corral/Blackwell fire complex and is developing a preburn vegetation layer of the same area for analysis. The burn intensity layer for the postburn classification has been completed and converted into polygon format for resource analysis.
G. REFERENCES AND/OR CONTACTS:
Payette National Forest
Susan Boudreau (s.boudreau:r04f12a)
P.O. Box 1026
McCall, ID 83638
208-634-0745
Remote Sensing Applications Center
Henry Lachowski (h.lachowski:w03a)
2222 West 2300 South
Salt Lake City, UT 84119
801-975-3662
CASE STUDY 13:
Monitoring Aspen Decline Using Remote Sensing and GIS-- Gravelly
Mountain Landscape, Southwestern Montana
A. DESCRIPTION/OBJECTIVE: To map the current and historical distribution of aspen within the Gravelly Mountain Range landscape unit in southwestern Montana using Landsat Thematic Mapper (TM) satellite imagery, and to map the extent of aspen decline between 1947 and 1992 with the aid of historical photographs.
B. SCALE: Target scale is Landscape unit.
C. DESIRED OUTPUT: Maps developed from Landsat TM and historical photographs showing the distribution and amount of aspen in 1992 and in 1947 as well as a map showing the change between the two years.
D. RESULTS
1. SUCCESSES
Aspen could be detected and decline shown using a combination of Landsat TM imagery and historical photography. The maps produced were more detailed than the forest personnel could have produced otherwise.
2. LESSONS LEARNED
Aspen was easy to detect using Landsat TM imagery but it was often confused with willow and other wet vegetation types. In some areas this resulted in an overestimation of aspen but this was considered acceptable due to the time constraints of the project. Additional photo evaluation, field work, and editing would reduce the confusion and improve the estimate of aspen amount. The change detection also showed that in areas of aspen decline, conifers were the most common successional species.
3. PROMISING POSSIBILITIES
The comprehensive view of the forest provided by Landsat TM imagery will allow resource specialists to make recommendations and decisions based on landscape evaluations across all ownerships, not just Forest Service. The techniques developed were simple and could be further developed for a more indepth analysis of the causal elements of aspen decline.
4. SHARING
E. INPUT:
F. SUMMARY/CONCLUSIONS: This study demonstrated that aspen stands can be successfully mapped using Landsat TM imagery. Use of digital satellite imagery and historical photos also proved to be a useful means of determining the location and amount of aspen decline.
G. REFERENCES AND/OR CONTACTS:
Beaverhead Deerlodge National Forest
Jim McNamara (j.mcnamara:r01f02a)
420 Barrett St.
Dillon, MT 59725
406-683-3900
Remote Sensing Applications Center
Henry Lachowski (h.lachowski:w03a)
2222 West 2300 South
Salt Lake City, UT 84119
801-975-3662
CASE STUDY 14:
Aquatic Biodiversity Assessment of the Platte/Niobrara River Basin
Using ECOMAP Products
A. DESCRIPTION/OBJECTIVE: Conduct initial stages of an aquatic biodiversity assessment based on combined terrestrial and aquatic ecological units and on fish distribution data. Describe original ranges of native fish and explain spatial patterns of species and stocks.
B. SCALE: Combine terrestrial provinces and sections with aquatic river basins to create subbasins, refined by hydrologic unit boundaries within river basins.
C. DESIRED OUTPUT: Map of PlatteNiobrara River Basins subdivided into subbasins and major watersheds that reflect the influence of physiography and hydrographic boundaries on fish distributions.
D. RESULTS:
1. SUCCESSES: Physiography and hydrography explain the original ranges of native fish species and stocks extremely well. Electronic Digital Elevation Model shadedrelief data improve the accuracy of physiographic lines. GIS data enabled rapid development of multiscaled maps.
2. LESSONS LEARNED: Both physiographic and hydrographic boundaries are often needed to delineate meaningful subbasins. Some fish species live only in certain physiographic settings and others are more adaptable.
3. PROMISING POSSIBILITIES: Distributions of native and naturalized fish species and stocks can be explained and mapped by existing physiographic and hydrologic units. Maps can be created at multiple scales to display these relationships and improve assessment efficiency and effectiveness.
4. SHARING: Coordination with The Nature Conservancy, State Natural Heritage Programs, and State Departments of Wildlife is encouraged.
E. INPUT [data, tools, skills, cost, time): GIS software was used to refine physiographic boundaries and combine physiographic and hydrographic lines into one composite map of subbasins and major watersheds in less than one day. The data were refined by consulting existing reports within five days. Fisheries biologist, hydrologist, and GIS skills were required.
F. SUMMARY/CONCLUSIONS: GIS manipulation of ECOMAP products, combined with existing information on fish distributions, provides a rapid method of initial aquatic biodiversity assessment at coarse scales. This information can then be refined using field data on distributions of fish species and stocks.
G. REFERENCES AND/OR CONTACTS:
National Hierarchical Framework
of Ecological Units; A Hierarchical Framework of Aquatic Ecological
Units in North America; Jim Maxwell (303-275-5096); Gordon Sloane
(303-275-5010); Jerry Freeouf (303-275-5095).
CASE STUDY 15:
Ecological Unit Inventory Using Satellite Imagery and Digital
Elevation Models on the Bridger-Teton National Forest
A. DESCRIPTION/OBJECTIVE: Conduct an ecological unit inventory using a geographic information system (GIS), Landsat Thematic Mapper (TM), and digital elevation models (DEM) to expedite mapping, add consistency, and stratify at various levels the study area according to the National Hierarchical Framework of Ecological Units for sampling into "premapped" ecological units. Survey meets USDA National Cooperative Soil Survey (NCSS) and USDA Forest Service Handbook (FSH 2090.11) standards and guidelines.
B. SCALE: Target scale is Landscape and Land Unit. This scale corresponds to Order 3 and Order 4 intensity (NCSS). Ecoregion, Subregion, and Landscape scales were initially created using satellite imagery.
C. DESIRED OUTPUT: Ecological unit inventory maps/coverages at the Landtype level. Output included a data base for vegetation and soil information. Subregion and Landscape maps were created by aggregating units, using the "bottom up" approach.
D. RESULTS
(successes, lessons learned, promising possibilities, sharing):
1. SUCCESSES:
AVHRR satellite imagery (1.1 km pixels) proved to be an appropriate source of imagery for observing and refining units at the Ecoregion scale.
Landsat TM (30 pixels) proved to be an appropriate source for displaying, mapping, and refining ecological units (Sections and Subsections) at the Subregion scale. Landsat TM is more appropriate over these large areas than aerial photography. Subsections were delineated directly into the GIS through the use of a Landsat TM image as a back coverage.
Landsat TM was also used to delineate ecological units at the Landscape scale (Landtype Associations). Additional information from other sources (state geological maps, historic maps, literature reviews, and input from resource professionals familiar with the area) was used to refine the boundaries.
2. LESSONS LEARNED
Landsat TM imagery, DEMs, aerial photographs, and ground collected data were used to assist in delineation of ecological units at the Land Unit scale (Landtype and Landtype Phase). Satellite imagery by itself was not adequate for developing units at this scale. However, the use of remote sensing and GIS technologies allowed the production of a field ready "premap" in about half the time it would have taken to create a similar type map from aerial photos alone.
Ecological unit lines do not always necessarily coincide with vegetation/slope/aspect lines derived from the digital data. The satellite imagery was just one of a number of useful tools in a complex, interactive process dependent upon input from many data sources.
The usefulness of the technologies to aid the mapper in defining premap delineations at the Land Unit level varied greatly with the complexity and nature of the landscape being examined. While some products reduced the time to adequately premap one area, they were of little or no assistance in contrasting areas. Therefore, the development of a single analysis process for the entire survey area was inappropriate.
The survey area required stratification into subsection or landtype association units prior to the computer analysis.
The stereo view provided by overlapping aerial photography is useful for evaluating the premap ecological units. Premap units generated by the computer must be viewed in stereo for refinement of boundary placements.
E. INPUT (data, tools, skills, cost, time):
1. ARC/INFO and ERDAS software were used to analyze DEMs and Landsat TM for ecological unit mapping. Paradox DBase was used for the soil and vegetation data base that accompanied the maps.
2. Automated processing techniques were used to generate maps of the study area in approximately five weeks. Additional time was required to manually refine the "premapped" ecological units (Landtype level) using aerial photos, orthophotoquads, field data, and other ancillary data, and map them onto aerial photographs.
F. SUMMARY/CONCLUSIONS: Satellite remote sensing provides a unique and comprehensive perspective that complements aerial photos, field sampling, and other data sources in mapping ecological units. Satellite imagery is the only feasible way of obtaining the synoptic view necessary for large area assessments which may cover many ecological, political, and jurisdictional boundaries. The key to success in using remote sensing for delineating ecological units is to match the appropriate source of information with the appropriate task.
It would be difficult if not impossible to fully automate the premapping process. The flexibility of interaction at each step is important to allow scientists the ability to input important parameters and interrelationships that may be unique to a given unit.
G. REFERENCES AND/OR CONTACTS:
Wirth, T., P. Maus, H. Lachowski, and D. Fallon. 1996. Mapping Ecosystems. In Earth Observation Magazine, pp. 1418.
CASE STUDY 16:
Integrated Resource Inventory of Land, Water, and Vegetation
in Preparation for a Geographic Information System, Rocky Mountain
Region, Forest Service.
A. DESCRIPTION/OBJECTIVE: Provide a framework to coordinate and integrate functional resource inventories; streamline inventories for time, cost and personnel efficiencies; provide consistent and reliable spatial and tabular information for use with geographic information systems; provide a corporate framework and standards for natural resource information in the Rocky Mountain Region; define uniform classification systems needed to support common units where no standards exist. The IRI consists of three integrated themes of information: existing vegetation (Common Vegetation Unit), terrestrial ecological unit (Common Land Unit), and aquatic ecological unit (Common Water Unit).
B. SCALE: All themes are developed at 1:24,000 and registered to the cartographic features files base map. The Common Vegetation Unit and Common Land Unit are developed (mapped) at the Land Unit scale. The Common Water Unit is developed at the Subwatershed through Valley Section scales. Common Land Units are used to develop and/or refine ecological units at the Landscape scale (Landtype Associations).
C. DESIRED OUTPUT: Three integrated themes (layers and the associated tabular information) of basic resource information developed according to a common, consistent corporate standards. Themes are vertically integrated to recognize (and therefore encourage recognition of) coincident lines between themes. All ground collected data is registered to a point. Map unit or polygon data may be associated with a polygon, or a line segment such as a stream reach when the area is too small to be represented by a polygon at the map scale.
D. RESULTS:
1. SUCCESSES
Resolution of coincident lines between themes. Reduction of electronic editing and maintenance costs through integration and coincident line resolution. Reduce costs associated with traditional functional inventories through a design that provides common units that support multiple resource needs.
Aerial photography proved to be the best source of imagery for providing the desired level of detail; required little or no additional training; and had not special software or hardware requirements. Landsat TM, small scale color infrared photography, and other sources of remotely sensed imagery were aids to mapping.
2. LESSONS LEARNED
3. SHARING
Consistent information provides a better basis for comparisons of resources between administrative units, allows sharing of data sets, and sharing of applications among and between administrative units without the need for adjustments to accommodate individual data sets.
E. INPUT [data, tools, skills, cost, time):
Minimum data input dependent on the specific theme of information. Much of the aquatic information is derived from GIS. The minimum level of information for existing vegetation and terrestrial information is photo interpretation and existing information, supplemented by ground data collection where necessary. Cost and time is highly dependent on skills and quality of existing information.
F. SUMMARY/CONCLUSIONS: Standards provide the ability to consolidate and integrate basic resource information into three distinct, integrated maps that will be consistent across a five state Region. Information standards also allow information to be reliably aggregated upward to address broad scale planning, ecosystem and landscape ecology issues across administrative boundaries.
G. REFERENCES AND/OR CONTACTS:
