USGS:Biology Biological Informatics Program Vegetation Characterization Program Program Reviews 1998 National Park Service Review

Early in 1998, the Inventory and Monitoring Advisory Council and the Science Advisory Council of the National Park Service requested a review of the NPS/USGS-BRD Vegetation Mapping Program. Accordingly, a group of primarily NPS staff met in the Boston Support Office on 1-2 Sept 1998 for the review. The objective was to provide NPS and USGS-BRD with clear recommendations for the future direction of the program.

The agenda covered the background of the program, a description of the process (sampling, classification, mapping, time frame, etc.), an update on the progress of the program, and a review of the products and website. Representatives from ASIS, VOYA, PORE, ACAD, ROCR, JOTR, and the Prairie Cluster LTEM Program also attended the meeting and discussed the program, methods, and data related to their experience. The presentations and discussions assisted the group with gaining a better understanding of the program and the products.

A large part of the meeting was set aside for discussing the various aspects of the Vegetation Mapping Program, notably:

(1) whether the program is providing the proper products to NPS and, if not, what products are needed

(2) whether the program is the most cost effective method for generating the products and what alternatives are available.

Proper Products

Tom Owens (BRD-OBIO-CBI) presented two demonstrations of the data developed by the program. One was based in ARCView and focused on the deliverables for Devils Tower National Monument. Tom demonstrated the links between the final vegetation layer, the field data, descriptions, and accuracy assessment and used the TUZI project to demonstrate the data that are presently available on the Vegetation Mapping Program web site. The park representatives related their current use and their projected future use of the data. Although many of the parks will need more detailed data for park-specific issues, the data that are provided by the program are at a reasonable level of detail. The cost of the program and the products were not considered in these discussions. The focus was on the products themselves.

The consensus of the group was that the program provided the proper products for the National Park Service. The participants agreed that (1) the quality of the data and the variety of products are a level of detail that is seldom available but necessary for parks and (2) the methods for producing these products are valid and acceptable. The major concern was not with the scope or quality of the products but with the speed with which the program can be implemented at the current funding levels and availability of personnel.

The Vegetation Mapping Program was initially envisioned as a 10-year program to provide baseline data for the parks that conduct long-term ecological monitoring, reflecting vegetation in the year 2000 time frame. A graphic comparison by Tom Owens of the expected or planned budget with the actual budget of the program revealed that at the present funding level, completion of the program requires 50-60 years. Of course, the need for vegetation data by most parks is immediate and a 50-60 year time frame is unacceptable.

The extended timeframes, in some cases were confounded by an unavailability of contract personnel for some of the field work in the parks. A resolution of the problem has been the training of personnel from NPS, BRD, and the US Fish and Wildlife Service, thus broadening the base of personnel that is qualified to support the program.

Two additional none-park-specific products were suggested by the group:

A need for a system (GIS/decision support, data management) that integrates the data from all of the I&M programs in the parks, so that all park personnel can use the data in an integrated environment. Such a system is outside the scope of the Vegetation Mapping Program but has been taken up by the Data Managers Group of the I&M Program.

The participants also expressed need for more outreach, specifically, demonstrations at conferences (George Wright, Superintendent's, GIS, etc). Such demonstrations should not only present the program products but demonstrate the applications of data in parks, inform people about the status of the program, and help generate more support for the program.

Cost-Effectiveness-Effectiveness

Addressing the cost-effectiveness of the Vegetation Mapping Program was difficult without an itemized list of the cost of the various program steps. Such a list, which would allow reviewers to better understand the cost of specific products and the basic start-up cost of the program, will be developed for future discussions.

Three cost-related alternatives were identified: (1) scaling back the program, (2) doing more with the available funds, and (3) obtaining more funds. Some of the options were not fully discussed because the reviewers could not identify the associated cost savings. Many of the options, therefore, were not recommended at this time but recommended for future discussions when more information is available on specific costs of the program components.

Scaling Back the Program of the Number of Products

The several available options for scaling back the program include a reduction of the level of detail in the mapping and descriptions such as the minimum mapping unit (currently 0.5 ha) and the level of descriptive detail (mapping at the Alliance rather than at the Association). This would reduce the field work, analysis, photo interpretation, automation, and accuracy assessment. It would reduce the cost of the program in each park but would also limit the usefulness of the data. Although this option was not comprehensively discussed, the participants felt that the present level of products is appropriate as long as adequate funding can be made available.

The option of providing two levels of data was also identified. In place of a comprehensively detailed level of data, a reduced level of detail may be sufficient for some parks. If so, parks that require comprehensive detail and parks that require lesser detail must be identified. However, such a strategy may raise rather than lower costs in the long-term because parks that require less detail now may need comprehensive detail in the future. Savings from providing either of two levels of details could not be determined because of lack of pertinent cost data.

For the same reasons, the group did not recommend a reduction in the number of parks serviced by the program but suggested that such an option be re-examined when data are available.

Scaling back the accuracy assessment of the program may represent a cost saving. Although the time spent at the various field sites for the accuracy assessment is minimal, the time investment of accessing the sites and the large number of sites required for the assessment is significant. Options include the removal of some easy classes from the assessment, assessment of the accuracy at the Alliance level rather than the Association level of the classification, use of fewer replicates for each class (currently 25-30 sites/class), and assessment of regional products rather than products for each park (assuming that the accuracy for products in a region are consistent).

Doing More with the Available Funds with the Available Funds

The suggested alternatives ranged from more cost-efficient personnel to mapping of clusters of parks with similar environs. Several of these options had been explored in the past, and these explorations should be continued and expanded.

The program initially relied on contract personnel for the full range of services for the photography, field data collection, analysis, photo interpretation, development of classification and descriptions, automation, and accuracy assessment. Because of the lack of availability of contract personnel for some of these activities (scheduling conflicts) and the high cost of some contract personnel, the program initiated the use of two additional groups of personnel (BOR and BRD-EMTC). Additionally, the program often funds seasonal positions in various parks for field data collection and GPS operation. Because the estimated cost of the field operations is one third or more of the project costs, significant savings can be realized by using additional local or regional personnel. Such personnel would be responsible for sampling the vegetation in the parks, entering the field data into an automated system (Plots database), and performing the accuracy assessment. The group felt that developing regional teams for this type of work rather than hiring personnel by park would raise cost-effectiveness as well as economies of scale. Once photo interpreters are familiar with interpretable features and vegetation classes in a region, continuation of their work in another park in the same region rather than in a dissimilar geographic area would expedite completion of the task.

Several parks (ROCR, PORE, ACAD, etc) are currently providing all or part of the field work for their vegetation mapping programs, and one park (JOTR) is assisting the photo interpreters by training one of the staff in this area. When a park can provide this type of assistance, the program benefits from cost savings and the park benefits from a better understanding of the process and the data and by its personnel gaining skills.

The participants suggested initiatives to encourage creative local augmentation of funding or assistance to the program. The model that funded the I&M Bird Inventories in 1998 was suggested for the Vegetation Mapping Program. The parks from the top of the existing priority list would be invited to develop and submit proposals for initiating the program in their parks. Such an invitation may encourage inter- and intra-agency partnerships that are often unavailable at a national level.

Obtaining More Funds

At least two options for obtaining more funds were discussed. One focused on additional funding through the budget process of NPS and BRD (individually and jointly). The BRD representatives related the difficulty of negotiating the USGS budget process for these types of programs unless they are approached as joint initiatives.

The second option focused on other programs and agencies that will benefit from the program. Other programs include FirePro and NPS Maintenance. The program representatives have met with FirePro representatives in the past and discussed the similarities between the vegetation data collected for the program and fire fuel data that are collected for the FirePro Program. The data collection of the two programs overlaps. The two representatives of the Fire Community at the meeting supported joint efforts and suggested the coordination of joint efforts at a local or regional level rather than at a national level. Two parks in the Intermountain Region were specifically discussed as possible tests. At least two other parks are already working between these programs (YOSE and PORE).

Other agencies that benefit from the program include federal and state agencies with managed lands adjacent to parks. Because the program offers the parks an opportunity to map beyond their boundaries, there is an opportunity to involve the neighboring agencies for possible inclusion of their lands in the mapping. The US Fish and Wildlife Service and the US Corps of Engineers already expressed interest in such an arrangement. Full exploitation of these kinds of partnerships requires early identification.

Finally, there may be funding sources outside the government such as Canon, non-profits, and others. This has yet to be investigated.

Other Recommendations

The visibility of the program must be raised as examples of the use of the products by parks become available. High visibility will buy more support and partners. Someone will have to act as a cheerleader for not only the Vegetation Mapping Program but for the entire I&M Program.

The initial NPS contract for vegetation mapping with ESRI expires in March 1999. At present, this is the only vehicle NPS has for direct funding. YOSE and PORE are using this contract to develop products on their own. When the contract expires, NPS will not have this option. It is therefore recommended that NPS ensures other contracts or agreements for parks that are able to fund programs on their own.

Endnote:

All participants are concerned with the timeframe implied by the current budget. Everyone hoped for the best possible funding in order to continue the present level of detail within the program. This, however, may not be realistic. The participants hesitate to recommend changes of the techniques and products unless they are given a timeframe (e.g., 10-year program with existing funding, 20-year program, etc). The participants are willing to continue these discussions when more information becomes available or if the Science Advisory Council provides more explicit direction.

For more information contact:

Mike Story NPS-NRID
Remote Sensing Specialist
Voice: 303-202-4236
FAX: 303-202-4219
mike_story@nps.gov

Participants

Sarah Allen NPS PAB Cluster/PORE
Gillian Bowser NPS JOTR
Eden Crane NPS ROCR
David Delsorso NPS PWRO
George Dickison NPS AKSO
Mark Duffy NPS ASIS
Mary Foley NPS Boston SO
Linda Gregory NPS ACAD
John Karish NPS Phila SO
Sam Lammie NPS VOYA
Chris Lea NPS ACAD
David Manski NPS ACAD
John Mosesso USGS-BRD OBIO (BRD Review)
Maury Nyquist USGS-BRD CBI
Thomas Owens USGS-BRD CBI
Paul Reeberg NPS PWRO
Nancy Roeper FWS Division of Refuges (BRD Review)
Nigel Shaw NPS Boston SO
Mike Story NPS NRID
Lisa Thomas NPS Prairie Cluster LTEM
Sara Wesser NPS AKSO