1998 USGS Biological Resources Division Review

Report of

Vegetation Mapping Review Team

on the

USGS Biological Resources Division

Vegetation Mapping Program

December 1998

Team Members
John Moeller, USGS, Federal Geographic Data Committee
Denice Shaw, Enviromental Protection Agency
Gyde Lund, U.S. Forest Service(Retired)
Nancy Roper, Fish & Wildlife Service
Tom Loveland, USGS, National Mapping Division

I. Executive Summary:

The National Park Service (NPS) initiated a Vegetation Mapping Program in 1994 to classify and map the vegetation of over more than 250 National Park units. The program is a partnership between the U.S. Geological Survey Biological Resources Division (USGS) and the National Park Service that supports NPS Managers, program specialists and scientists with accurate, scientifically valid and consistent vegetation classification and spatial data that can be used for a wide variety of purposes.

In July 1998, a small team with expertise and experience with research and geospatial technology was formed to review the program. The goal of the review was to determine whether the program is meeting current NPS and USGS needs and is maximizing its contribution to national, regional, and local vegetation data collection activities.

The Review Team believes that the NPS and USGS should be commended for implementing this pioneering effort. The Vegetation Mapping Program is founded on innovation in classifying vegetation, in bringing together experts to design the program and in developing implementation standards and protocols. The program has contributed greatly to the Federal Geographic Data Committee (FGDC) approved National Vegetation Classification Standard and provides a solid foundation for future contributions to the National Spatial Data Infrastructure and the National Biological Information Infrastructure (NSDI and NBII). The Review Team also believes that the NPS and USGS can improve the program as it continues to develop, to better support agency missions and to fully include interagency partnerships to achieve national, regional, and local objectives.

In summary, the team found that the program is providing positive results, is producing quality products, is well designed, scientifically credible, and is contributing to the implementation of the National Spatial Data Infrastructure and the National Biological Information Infrastructure. The program, however, is not meeting NPS scheduling needs and is not funded at the level desired by NPS managers and users or by USGS program officials.

Provided below are findings and recommendations in seven topic areas that address concerns identified in the review. Recommendations are intended to help the National Park Vegetation Mapping Program become a model for implementation within the U.S.A. and elsewhere.

II. Background:

This program review was conducted by an independent team of geospatial technology and applications specialists, to assess whether the Vegetation Mapping Program, as currently conducted, is meeting NPS and USGS needs. Also, by taking a close look at program operations and management, the review team sought to provide guidance for improving future work and maximizing program contributions to other national, regional, and local vegetation data collection activities.

III. Purpose:

The Vegetation Mapping Program was initiated in 1994 by the NPS as part of its Inventory and Monitoring Program. It involves a long-term effort to map the vegetative cover of more than 200 units of the national park system. The program was transferred to the National Biological Service in 1994 and is now administered by the USGS Biological Resources Division. The program employs a uniform hierarchical vegetation methodology to characterize existing vegetation. The vegetation data along with a large amount of ancillary data is used to create a geographical information system for each park mapped. Data products are used by park personnel to support resource management decisions and add to the information base used for park interpretative, compliance, and other operations. At the current program appropriation of $1.167 million per year, the estimated time for completion of all interested park units stretches past the midpoint of the next century.

The complex process followed for each mapping project is documented in peer reviewed protocols such as: a national vegetation classification system, field methods, accuracy assessment procedures, and others (see protocols at biology.usgs.gov/npsveg/). Although much of the actual field work, including vegetation classification, mapping, and accuracy assessments are conducted by non-government contractors, the entire process is closely monitored by BRD and NPS staff located at the BRD Center for Biological Informatics in Colorado. The oversight of this group provides a baseline of experience and helps assure consistency in products developed for a wide range of parks and monuments that vary in size, vegetation type, and management needs. The process followed for each park project can be summarized by grouping essential activities as follows: planning meetings and discussions, collection and analysis of existing data, development of a vegetation classification, development of a sampling strategy, field work, data input and analysis, photo interpretation, cartography, validation and accuracy assessment. The twenty eight final products prepared for each unit mapped are provided digitally on the BRD web page cited above. Generally, products include: classification reports, keys and descriptions, aerial photography and overlays, spatial data, field data, and accuracy assessment data.

Currently, mapping projects have been completed at five NPS units with fifteen others in progress. Because each unit requires a 2-3 year effort, and because available resources limit new starts to 3-5 per year, the large number of "priority one " units will not, under present conditions, be addressed until the middle of the next century. An alternative strategy to significantly increase the capability of the program seems appropriate.

IV. Review Process:

In July of 1998, the USGS invited Mr. John Moeller, FGDC Staff Director, to serve as team leader for the review. Mr. Moeller selected four highly qualified technical experts to assist in the review including: Mr. Tom Loveland, USGS EROS Data Center, Ms. Denice Shaw, U.S. Environmental Protection Agency, Mr. Gyde Lund - USDA Forest Service retired, previous FGDC leader for the vegetation classification standard, and Ms. Nancy Roeper - U.S. Fish and Wildlife Service representative to the FGDC Vegetation Committee.

On August 5, 1998, the review team met to discuss USGS expectations of the review, to develop a review plan, a list of interview questions and interviewees. Over the next two and a half months, the team contacted more than 32 individuals with a extensive experience and knowledge of the program. A list of these individuals can be found at the end of this report. In addition to conducting interviews, members of the team reviewed a variety of documents defining or produced by the vegetation mapping program, and participated in a NPS Review Meeting held in Boston on September 1 -2. The team met a final time to, interview program staff, synthesize comments and develop this report in Denver, Colorado from October 19-21.

V. Observations, and Recommendations:

The review process focused on the health and viability of the program and on elements of the program raised as issues during interviews. As a result, the clear consensus of the team is that the program is basically solid, important, and necessary. This program was founded on innovation and continuing innovation is an essential component of a long-term successful project. Program leadership should continually evaluate and consider new ideas and technologies that will improve the scientific integrity and application of program results. Continuing innovation means that external experts should be routinely consulted regarding new ways of improving the program. Recognizing that refinement and improvement is continuous, we offer the following observations (O) and recommendations (R) for future activities. All comments offered in a constructive spirit and are intended to strengthen and extend the life of the program

Philosophy and Organization:

O- The overall design and philosophy of the program is technically strong and is well-suited to meet the needs of the various park units.

O- The program clearly holds a potential benefit to other organizations beyond NPS. The potential role and benefit to other Departmental programs, other agencies and States has not been explored or exploited. The benefit of doing so may include increased funding opportunities.

O- A strong point of the program is the use of outside experts from groups such as The Nature Conservancy (TNC) and State Natural Heritage Programs However, this at times has created bottlenecks because of scheduling conflicts. It may be possible to replace some of the contract expertise with USGS and other staff, but this may weaken the overall effort due to the strong credentials of the contractors. It may be possible to use TNC and Heritage staff as trainers and supervisors of a larger core of field staff (i.e., permanent staff, seasonal employees, university personnel, volunteers). The challenge, if such a strategy is adopted, will be to ensure the continued consistency and quality of products..

O- The use of competitive contracts has been successful as it provided a sound start-up framework. Development activities completed under this mechanism were valuable and successful. Contracting also provided a mechanism to bring exceptional expertise into the design process and has provide a core of experts that are being used in innovative complementary projects such as those at Yosemite and Joshua Tree. However, this implementation model has been expensive and given the limited funds available to the program, the use of contract staff at the Bureau of Reclamation and BRD's Environmental Management Technical Center (EMTC) appears to be more cost effective.

R- The overall vegetation mapping program is strong and should be continued. It is recognized as a pioneering venture within the NPS due to its scientific integrity and comprehensive design.

R- Standards and protocols should be considered for broad implementation and use in other land management programs of other agencies.

R- The USGS and NPS, in cooperation with the U.S. Forest Service, should promote acceptance of the standards and protocols developed for the vegetation mapping program through the FGDC.

R- The program should continue to use a variety of contracting mechanisms and NPS personnel rather than a single central contractor. This will provide a flexible means of assembling the strongest possible team for all components of the program.

R- USGS and NPS should increase the training done by The Nature Conservancy and State Heritage program staff for classification system development and field work.

Standards:

O- The real strength of the program is the collective set of standards used throughout the program. For applications within each park, the standards were found to be appropriate; for application of the data at national or regional levels, the suitability of these standards is not known.

O- There was considerable debate concerning changing to a coarser minimum mapping unit. While this may have merit, there is no clear indication that there will be a cost savings associated with such a change. Therefore, the committee agreed that maintaining the 0.5 hectare minimum mapping unit is warranted.

O- Accuracy assessment standards produce quality results but at a significant and perhaps unaffordable cost. It is important to explore options for reducing the field costs. If the standard is lowered the cost will go down as less field work will be needed in the mapping phase. Other options include doing all field work at the onset of the project rather than in two separate seasons, implementing the assessment as an add-on to other park field exercises, reducing accuracy targets, either by reducing the number of targets or the scale of information required, generalizing classes with lower accuracy, and others. Specific solutions may differ based on the unique needs of an individual park.

O- The vegetation classification system is being developed at the state level as projects are executed. The classification system was not well-tested prior to the start of the project program. While this system is quite appropriate for park management, implementation is made difficult because of the emerging nature of the definitions. In addition, for technical reasons such as the relationship between the expression of vegetation characteristics using aerial photography, there are complex cases where it is not possible to map at the association level. Even with such complexities, however, the system is being implemented as consistently as possible.

O- A spin-off benefit of the NPS vegetation project is the acceleration of the floristic components of the National Vegetation Classification Standard ( NVCS). This will have long-term benefit to vegetation science and resource managers across the nation.

R- The program should continue to carefully adhere to FGDC and NSDI standards and practices and there should be an effort to transition the NVCS terminology to that used by the FGDC.

R- The USGS and NPS should articulate the need for nationally consistent standards and protocols in the vegetation mapping program. While there is clear evidence of the utility of the standards at the park unit level, the need for national consistency has not been clearly expressed.

R- The accuracy standard should be reevaluated by the technical leadership of the program. While the accuracy assessment standard and protocol is a critical part of the program the specific standards criteria and all elements of the accuracy assessment protocol may not be necessary considering the planned applications of the data set. Thus, accuracy standards and the assessment protocol should be reevaluated with a goal of reducing cost and accelerating the completion of the program.

R- The formation level of the NVCS is appropriate for national consistency. Mapping at the alliance level should be justified, based on need on a park by park basis.

R- The USGS and NPS should review the NVCS version being used for the vegetation mapping program and it should be modified, as needed, to conform to the FGDC vegetation classification standard.

R- Minimum program standards should remain at 0.5 ha minimum mapping units, mapping at the formation level, and 80% accuracy per class. Park management should identify and justify areas or classes where greater spatial and thematic detail and/or additional attribute information is required to meet management needs.

Protocols/Methods:

O- Protocols are flexible and therefore enable additional detail as needed from park to park; this is clearly a strength of the technical strategy.

O- Protocols could form building blocks for FGDC vegetation classification implementation procedures. Exposure and review of the protocols within FGDC and other appropriate communities should be pursued.

O- There is a continuing need to assess the role of new technology in the project.

O- The relevance of National Technical Means data to the program should be explored.

R- The scoping meetings held at the beginning of each new start should focus on the anticipated uses of project results so that the appropriate level of spatial and thematic detail is understood and the need for additional attributes identified. A proper mix of participants, including managers, ecologists, and technicians, and participants from other sectors if practical, is essential for this meeting. Further, a written declaration of the NPS contribution to the project should be executed prior to initiation of the project.

R- The USGS and NPS need to better plan and sequence important events, funds, and other inputs associated with this program.

R- The NPS should share the criteria for park unit priority rankings with all parks.

R- The USGS should take the lead in bringing major vegetation mapping programs, land management agencies, and other relevant organizations together to share information, coordinate activities, and identify priorities for mapping by geographic areas.

R- The NPS should strongly consider grouping parks on a regional or state basis so that the USGS can implement a group of similar projects and benefit from consolidated mapping and vegetation characterization.

R- A team approach to each mapping project, involving NPS and USGS, as well as other federal and state participants, and mapping and field contractors is essential and should be continued.

R- The USGS should track and summarize all project costs by and functional task so that a complete picture of project costs can be documented. Cost estimates and final costs should be provided to NPS park managers and others associated with the project.

Implementation Procedures:

O- There is a need to improve the sequencing of project events (i.e., arrangement of local resources such as field staff, vehicles, housing, etc.) and funding streams, so that there are fewer starts and stops. Basically, there is a strong need for careful up-front planning at the beginning of every project and thorough communication between all affected parties to ensure consistent expectations. Further, USGS should assure that project monies flow smoothly and quickly from Headquarter accounts to individual project accounts as soon as is the annual appropriation process is completed.

O- Scoping meetings are essential to the success of each project. Effective meetings will ensure that the requirements of each park manager are clearly defined and understood so that the results of the project are done at a level of detail sufficient to support park decisions. Scoping meetings have been useful for identifying unique information requirements of each park mapped. At these meetings, there must be a proper mix of participants including ecologists, planners, and especially managers and decision makers.

O- Data acquisition and field missions should be planned with careful consideration of the phenology of the vegetation resources of each park. In some cases, the windows of opportunity may be so small, that higher levels of precision in task execution is necessary.

O- It is critical that photo interpreters and ecologists continue to participate together in field data collection missions.

O- The way in which mapping priorities were established is not clear. There does not seem to be a mechanism for reconsidering priorities which could alter the timing and cost of the program.

O- There appear to be strengths associated with a regional approach to project implementation. By initiating all parks in a given biogeographic region, it may be possible to reduce costs through the use of a single air photo acquisition coordinated field studies, and a dedicated focus on a common set of ecological conditions.

O- The link to, or awareness of, other ongoing work is not evident. There may be missed opportunities particularly with similar programs within the USGS.

O- It is highly desirable that the program take full advantage of available NPS staff, both permanent and temporary, for field data collection and verification. Point Reyes is example of a park that makes excellent use of NPS field staff in various aspects of their mapping project.

O- Use of contract mechanisms such as that used with ESRI, versus use of in-house staff may be a sensitive issue, but it appears that the later is more cost-effective. However, there appear to be appropriate roles for both depending on whether the focus is program development or implementation and production.

R- The USGS and NPS should revisit the protocol documentation and update elements that have changed as the program has evolved. Protocols should be periodically revisited thereafter. Continuation of strict adherence to the BRD metadata standards is highly encouraged.

R- The USGS and NPS should review emerging technologies and data sources, including National Technical Means, for use in the program.

Relevancy and Use of Output:

O- The program has produced a wealth of data and documentation that have value to specific parks. The applications of project results is an evolving story. Because so few parks have been completed, compelling examples of the utility of the results are just emerging. Because the reason for the program is to contribute to the short- and long-term management of NPS lands, a great deal of attention must be given to documenting those case studies in which the data sets have proven to be beneficial.

O- Data sets provide a wonderful baseline. However, many potential uses of the data have not yet been explored or expressed (i.e., bur oak distributions, spotted owl issues, exotics).

O- One challenging question is that of product use. There is clear benefit inherent in maps, digital data, site descriptions, aerial and field photos, etc. within each park and this alone may justify the effort. Another question is whether the same products have relevancy in national and regional applications. The contribution of the work and data to larger national issues is not being recognized and credited. There is a strong need to articulate the value of this program in a national context.

R- The USGS and NPS should illustrate how the data sets being developed are or will contribute to national needs and mandates (i.e., classification system development, exotics management, threatened and endangered species, biodiversity, NSDI and NBII, climate change, etc.). This information is important when advocating for national standards and consistency.

R- The NPS with the support of the USGS should develop a capability to use vegetation mapping project results in a wide range of park applications. Specific functions that need to be completed will vary by park unit. However, access to appropriately configured computers and training in the use of the data sets are minimum requirements.

Coordination and Outreach:

O- The long-term viability of the program may be enhanced through efforts devoted to the publication of information on all aspects of the program. This could include resource management forums, articles in the scientific literature, and popular publications.

O- Outreach and training is essential so that proper use and understanding of the role of the data is needed.

O- There are good procedures, innovative work, and novel ways in which some parks have gotten work done. However, doing good work does not ensure that the community knows about it.

O- It appears that other agencies generally do not know about the program.

O- There are many programs doing vegetation mapping around and in parks, but most of it is not coordinated.

O- There needs to be greater awareness by park staff of the program resources that can help planning and preparing for mapping projects.

R- USGS and NPS leadership should recognize and publicize the value of the data in national initiatives (i.e. NSDI vegetation standards).

R- USGS and NPS should develop a fact sheet or brochure that summarizes the goals, objectives, and key elements of the vegetation mapping program for distribution outside of the two bureaus.

R- USGS and NPS should develop guidelines outlining the steps that can quickly be initiated to prepare park staff to use the vegetation classification prepared for their park and organize for the implementation of a project.

R- The USGS should initiate the production and circulation of a newsletter that shares ideas, successes, innovations, lessons, and schedules associated with the vegetation mapping program.

R- The USGS and NPS should publish standards, protocols, results, and innovative applications in the open literature so that the larger resource management and scientific community may benefit from the results of this program.

R- The FGDC should be asked to facilitate outreach and education to the broader community on the forms of the protocols and strategies for implementing classification standards.

R- The USGS Gap Analysis Program and the Vegetation Mapping Program staff should continue to explore areas of synergy between the two programs, such as classification system development, exchange of data sets, and collaboration on accuracy assessment strategies.

Schedule and Funding:

O- Funding does not permit the timely completion of a one-time inventory of all priority NPS units. The potential for monitoring and change analysis is impossible at this time. This schedule is not meeting the needs of many parks, and especially those affected by dramatic changes such as fire, hurricanes, exotic invasions, etc.

O- Funding, quite simply, is inadequate for a program of this scope. As a result, there is an incentive for individual park managers to find alternative mechanisms for developing a vegetation baseline. While this may meet local needs, the opportunity to contribute to service-wide or national issues or to maintain consistency among parks could be lost.

R- A new 2001 budget initiative for an estimated $4-5 million per year, supported jointly by USGS, NPS and the FGDC, is highly recommended to allow program objectives to be met in a more reasonable time frame. The USGS and NPS should take the action to urge FGDC to propose and support a vegetation data collection initiative for FY2001 building on principles of Community/Federal Information Partnership.

R- There should be a minimum in-kind park contribution requirement established before a new mapping project is initiated. This may be financial, the provision of staff resources, vehicles, housing, or other contributions that reduces the overall cost to the national program. Of particular value would be the provision of staff to conduct field data collection and accuracy assessment tasks.

R- A more aggressive effort should be undertaken to identify new funding sources. For example, fire, global change, and other programs may be appropriate sources of funding augmentation. It may also help assure the involvement, assessment and identification of other projects that collect vegetation data and help establish collaboration that contributes to the national program. Identification of other federal agencies, States and local governments with a common need for information of this type may result in additional funding options. The feasibility of reducing funds through a regional implementation strategy should be assessed.

R- The NPS should establish a policy requiring park units that secure vegetation mapping from sources other than the USGS program, follow the standards established for program.

Individuals Interviewed For This Review

National Park Service - Mike Soukup, Abbey Miller, Mike Story, John Karish, Mary Foley, John Dennis, Sue Salmons, Sam Lammie, Linda Gregory, Sarah Allen

USGS, Gladys Cotter, Mark Fornwall, James Getter, Maury Nyquist, Tom Owens, Ralph Root, Gary Waggoner, Mike Jennings, Nick VanDriel, Patrick Crist, Sara Lubinski Frank D'Ercia,

The Nature Conservancy - Mark Shaffer, Denny Grossman, Pat Comer

Bureau of Land Management - Tom Costello, Fred Batson

U.S. Forest Service - Chuck Dull, Tom Bobbe

Others - Deb Southworth, U.S. FWS, Randy Vaughn, ESRI, Tony Curtis, Glatting Jackson Community Planning