|Population and Habitat
Viability Assessments for Golden-cheeked Warblers and Black-capped Vireos: Usefulness to
Partners in Flight Conservation Planning
Carol J. Beardmore1 and Jeff S. Hatfield 2
FUNCTION AND STRUCTURE OF PHVAs
A PHVA is a tool to compile, evaluate, and synthesize data and build a framework for conservation actions. It provides an in-depth examination and synthesis of what is known of a species' life history, ecology, management, and other factors to determine courses of action to manage for viable populations. Assessments include consideration of model analysis, habitat management, captive breeding (if appropriate), genetic factors (if appropriate), life history, status, threats, geographic distribution, education and information, other conservation efforts, human demography, research, and any other component deemed necessary. In contrast, Population Viability Analysis (PVA) refers to computer simulation modeling of biological processes that produce probabilities of extinction or survival under different scenarios. That is, a PVA is often only one step of a PHVA.
Why were PHVAs conducted?
Recovery plans for both species list viability model development as a task in the Narrative Outline for Recovery Actions (U. S. Fish and Wildlife Service 1991, U. S. Fish and Wildlife Service 1992). The intent of USFWS was to initiate a process that would result in information that could be used in determining the population sizes, and the habitat patch size and distribution needed, to ensure viable populations (i. e., one of the recovery criteria). Other results that were anticipated included recommendations for habitat management, research needs, and public outreach strategies.
How were the PHVAs conducted?
A PHVA workshop, hosted by the USFWS, was held for each species. The format of the workshop was similar to PHVA workshops conducted by the IUCN's Conservation Breeding Specialist Group (CBSG 1993). The consensus-building, group discussion format was assisted by a trained facilitator. A group of biologists representing federal, state, county, and local governments; universities; consulting companies; and nongovernmental organizations was invited to participate. These biologists had either direct research experience with the warbler or vireo in the field, direct association with managing the species' habitat, or responsibility for education programs that would benefit recovery. The group was kept small to facilitate group discussion, but included a representative from all groups that had conducted research or management on these species. CBSG often includes many stakeholders in their workshops, including private landowners involved in implementation. However, because of the complexity of developing biological objectives for these species, our workshops focused only on the species' biological needs and not on implementation. Future workshops are planned that will involve landowners and land managers, and that will encourage their participation in developing and implementing recovery solutions using the biological information developed in the first workshops.
The workshops had several steps. Introductory material was presented on the general intent of the workshop, an overview of small population biology, new biological data, the importance of informing the public of what is needed to recover endangered species, and a demonstration of the simulation model. Then, the group developed a set of workshop goals. The Black-capped Vireo group agreed to the following set of goals, which were similar to those developed in the Golden-cheeked Warbler workshop (U. S. Fish and Wildlife Service 1996a).
* Update and describe range-wide distribution and status (breeding, wintering)
* Update information on habitat characteristics
* Identify and rank threats
The above goals were divided among four subgroups to address: (1) Population biology and modeling, (2) Habitat management strategies, (3) Outreach and partnerships, and (4) Distribution, status, and threats. All subgroups identified future research needs for their topics. Each subgroup drafted a report that summarized information, discussions, and recommendations. After the workshop, draft reports were compiled into chapters of a draft document and sent out for participant review. Each species group asked to have a subsequent one-day meeting during which comments on the draft document were discussed and chapters modified. The revised chapters from the one-day meetings were incorporated into final reports (U. S. Fish and Wildlife Service 1996a, U. S. Fish and Wildlife Service 1996b).
The reports from the PHVA workshops are intended as a framework from which managers can develop habitat in such a way that it will help meet the report recommendations. We were cognizant of possible misinterpretations, and were careful to include our assumptions and caveats. The results represent the best direction we have with the current state of knowledge, recognizing that we used minimal data for some values. Further data collection and analysis and the results of recommended conservation efforts will change the scope of recovery needs. That is, we intend the PHVAs to be part of an iterative process. The workshops' participants will meet again, both to target strategies for specific situations and to refine recommendations with new data. To implement recommendations found in the final reports, local workshops will be needed to involve landowners and land managers in developing specific strategies for action at the site level. Also, when research provides new information about processes, threats, and/or species biology, a reassessment of the targets and recommendations may be needed.
What resources were needed to conduct a PHVA workshop?
The Golden-cheeked Warbler and Black-capped Vireo PHVAs were supported by a National Biological Service (NBS) grant and the participants' agencies and organizations. Total expenses for one workshop were about $7,000, including a computer simulation model, biostatistician support, and partial costs for travel and meeting facilities. This figure does not include the field work necessary for data gathering, but does include data synthesis.
What data were needed for the PHVA?
By working through the available data in the population biology and modeling subgroups, the participants realized that the complexity of the simulation model that could be used for the analysis relates directly to the adequacy of the data available. If a substantial amount of the necessary data is available, a customized model can be developed that simulates important aspects of a species' life history; if data are limited or not appropriate for the model, a less sophisticated model can be used. For the Black-capped Vireo and Golden-cheeked Warbler, we did not have sufficient data to develop a customized simulation model. Instead we used a generalized, commercial model, RAMAS/metapop (AkŠakaya 1994). RAMAS/metapop, like all PVA models, required an estimate of survivorship (from mark-recapture banding) at the same location for more than three consecutive years (to provide temporal variance), fecundity (number of fledglings per adult), and dispersal rates and distances (anecdotal information was used for the warbler and vireo from banded individuals).
PHVA WORKSHOP RESULTS
Each workshop had valuable tangible and intangible results. The final reports have population targets and habitat recommendations that can be used by agencies, organizations, and individuals to plan, justify, and guide conservation actions. Intangible values were produced by the interactions among participants. Each workshop capitalized on the dynamics of a group that was brought together for a specific purpose. The participants built consensus on actions needed to achieve viable populations. Information, including personal knowledge and experience which is not found in the literature, was assembled by the participants. Through the shared workshop experience, participants gained appreciation for each other's perspectives, each others' work, and for the workshop product. Workshop participants realized that new and ongoing research should focus on certain aspects of warbler and vireo biology that previously had been addressed insufficiently. Researchers understood how a better organized, cooperative approach to research was needed that would generate the specific data that could be applied to improve the simulation models and meet other needs. Participants realized how their work contributes, or could be modified to contribute, to the bigger picture. In general, the workshops motivated participants to take action when they returned to their own work stations.
HOW CAN PHVAs BE USEFUL TO PIF?
The PHVA process includes the basic steps of data collection and synthesis, and building consensus about recommendations to conserve a species. The PIF process for developing population objectives, habitat strategies, and research and outreach recommendations should be essentially the same as the process for developing a PHVA. Often, PHVAs are not seen as applicable to the PIF planning process, primarily because of the perception that PHVAs are only computer modeling exercises, not the consensus-building group process described above.
Regarding data collection and synthesis, several important recommendations are obtained from the PHVA process. The exercise of compiling information, both for modeling and making habitat management recommendations, is useful. During this step, two realizations usually emerge. One is that more information exists than is originally suspected. Often, an important source of information is the experience of local experts, which surfaces during group interactions and is incorporated into recommendations. This information is especially important because local variations can dictate different needs and approaches even for the same species. The other realization is identifying what data are missing or insufficient; this is important for directing future research priorities. In the warbler example, we learned that we had only two banding projects that could be used to estimate survivorship. We recommended that future banding projects should commit to more than three years, because multiple years of banding are needed to provide the survivorship estimate. For the vireo, many workers were observing nesting success, but the methodology varied among projects. In this case, we recommended a standardized methodology to produce the necessary fecundity estimate.
Although simulation model analysis was used in the warbler and vireo PHVAs, it is not a requirement of PHVAs. A PIF group may decide whether a computer simulation is needed. Because of the expense and length of time required to collect survivorship, fecundity, and dispersal data, population viability analyses should be undertaken in special cases. Factors such as vulnerability of the species to risks and documented population declines can be considered to determine if a computer simulation would be useful. However, the exercise of gathering the available data for high priority PIF specieseven if limitedand determining if they could be applied to an existing model can be useful, and quickly focuses research needs. If MAPS or BBIRD stations are in the area, survivorship estimates (DeSante 1992) and fecundity estimates (Martin et al. 1996) from these programs can be considered. If data do not exist for simulation modeling, or the PIF group does not want to include modeling in their plans, then the process of "closeting" experts, examining what is known, and making recommendations based on the other components of PHVAs is valuable.
In the Golden-cheeked Warbler example, even with limited data we learned that relatively large, viable populations (1,000 3,000 breeding pairs per population, depending on rates of habitat loss) were needed to prevent an unacceptable probability of extinction (U. S. Fish and Wildlife Service 1996a). The Black-capped Vireo modeling indicated that smaller populations (200 1,000 breeding females) would be viable, but only with cowbird removal (U. S. Fish and Wildlife Service 1996b). Because multiple, viable populations are recovery criteria for both species, the species' ranges were subdivided based on ecology and geography, thus giving the number of populations needed for recovery. PIF biological objectives could also be derived from such an approach, such as:
* a viable population should have x acres of optimal habitat that is managed to produce y pairs (e.g., for the Golden-cheeked Warbler y = 1,000-3,000 pairs, and x acres of optimal habitat is figured by taking a local density, x = y pairs, X number of acres needed per pair).
In conclusion, PHVAs should be considered as a planning tool in bird conservation. In this process data and knowledge can be examined and organized, deficiencies can be found, and recommendations can be made. Computer simulations may or may not be a part of the assessment.
1 Partners in Flight, c/o Arizona Game and Fish Dept.
2 Patuxent Wildlife Research Center