Conservation of biodiversity depends on identification and preservation of habitat conditions that support healthy populations of coexisting species. Healthy populations are those that have sufficient breeding productivity to offset mortality and thereby sustain themselves and sometimes less-healthy populations as well. Data on habitat-specific breeding productivity and survival of individual species and groups of coexisting species are needed to identify land-use and habitat management practices that will support healthy populations, avoid crisis management situations, and promote biodiversity. Yet, this critical and basic information is lacking for many species, including most nongame birds.
Currently, our primary information on population status of nongame birds comes from survey programs designed to detect relatively large changes in population size over broad geographic regions. Such coarse population changes typically reflect problems that have developed over large areas and long time periods and clearly are important, but may be difficult to correct. Moreover, correction is hindered because survey programs do not provide needed information on habitat conditions and sites that support healthy populations. Indeed, population change is a relatively insensitive measure of population health, particularly at local scales, because population size can be maintained in unhealthy local populations by immigration of recruits from healthy populations. Detection of population problems earlier in their development when problems are more localized, and identification of population health by specific localities and habitats, can allow more effective targeting and execution of management solutions.
Study of demographic characteristics (breeding productivity and survival) that determine population health can allow early detection of local population problems and identification of their proximate causes (e.g., low breeding productivity, low adult survival, high nest predation rates). More importantly, studies of habitat-specific demography can identify habitat conditions that support healthy populations and potentially identify the ultimate (environmental) causes of unhealthy populations by relating nest survival and productivity to environmental conditions. In this way, demographic studies potentially can identify environmental problems, habitat conditions, and geographic locations for targeted management action. Yet, the intensive nature and small scale of most demographic studies have limited their scope of inference.
The Breeding Biology Research and Monitoring Database (BBIRD) program was initiated in 1992 to establish a national data-sharing network to increase the geographic scope of inference of demographic studies. Growing recognition of the importance and necessity of demographic information, along with improving information for locating nests (e.g., Martin and Geupel 1993), has led to increasing numbers of demographic studies. The BBIRD program harnesses this burgeoning awareness and activity by establishing a network for collaboration and data-sharing among independent scientists and natural resource agency personnel studying avian breeding productivity at sites across North America. BBIRD participants adopt standard field protocols for monitoring nesting productivity, measuring vegetation, and counting birds to allow comparisons across local sites. Cooperators submit copies of their data to a central repository at the Montana Cooperative Wildlife Research Unit, where overview analyses are conducted to identify regional and continental trends in the data. Collection and analysis of data from each local site are overseen by independent investigators, ensuring high data quality and allowing rapid identification of important local results. Information in the shared database is available to all program participants (with some restriction to protect publishing rights). Access to the shared data allows participants to place their results in a regional or national context and to conduct larger scale analyses that go beyond what is possible for single site studies. For example, the program includes studies in large blocks of minimally-disturbed habitat that can act as controls to compare with sites that are impacted by various land uses.
The BBIRD protocol is adaptable to a wide range of study objectives. Current BBIRD projects include investigations of forest fragmentation, grazing, forestry practices, breeding biology, sexual selection, and life history evolution. The program now includes 32 sites in 23 states including Puerto Rico, and is growing rapidly. Over 17,000 nests of more than 150 bird species were monitored during the first 3 years of the program and data on up to an additional 7,000 nests are anticipated for 1995, the fourth year of the program.
The BBIRD protocol includes the three following elements. Participants in the program are encouraged to follow the entire protocol, but the first element is the only requirement for participation.
(1) Nest location and monitoring
We have had excellent success training individuals with little or no prior field experience to locate and monitor nests. At most BBIRD sites 4-5 person field crews locate and monitor more than 250 nests per year, and at some sites over 700 nests. The basics of this training are outlined in Martin and Geupel (1993).
(2) Measuring vegetation at various spatial scales
Vegetation characteristics are measured in forested habitats using a modified and enhanced version of the James and Shugart (1970) method, and details are provided in a protocol manual that is available upon request. Other sampling methods are being developed for shrub and grassland systems. Vegetation is measured at: 1) nest sites and associated non-use sites, 2) a stratified random sample of points within each nest search plot, and 3) four sites surrounding each census point. The ability to measure landscape level habitat characteristics of the area surrounding all BBIRD plots is currently being developed by the national program.
(3) Point counts on nest search plots
Nest search plots are censused 3 times per season using 10-minute, 50m fixed-radius point counts. Censuses are used to provide a yearly index of population size.
Participants meet annually, generally in late February or early March, to allow: 1) presentation of recent results by participants, 2) increased collaboration and interaction among participants with relevant data, and 3) discussion of any suggested modifications to the program. In 1996 there will be a meeting in late October in Fort Collins, CO. In short, this is a constantly-evolving program.
The BBIRD approach documents population health and causes of reproductive failure and relates these measures to habitat and landscape conditions. By replicating intensive local studies at sites across North America, BBIRD allows local researchers to compare their results with other sites with differing land use regimes (including relatively undisturbed sites) and facilitates identification of regional and continental trends. Ultimately, these data can be used to develop land management practices that will maintain healthy populations of non-game birds.
James, F. C., and H. H. Shugart, Jr. 1970. A quantitative method of habitat description. Audubon Field Notes 24:727-736.
Martin, T. E., and G. R. Geupel. 1993. Nest-monitoring plots: Methods for locating nests and monitoring success. Journal of Field Ornithology 64: 507-519.
Thomas E. Martin is Assistant Leader of Montana Cooperative Wildlife Research Unit, University of Montana, Missoula, MT 59812. Charles R. Paine is with the Max McGraw Wildlife Foundation. Wes Hochachka is a Research Associate with the Montana Cooperative Wildlife Research Unit. Anyone desiring more information or interesting in joining BBIRD should contact Tom Martin at the above address or by email at tmartin@selway.umt.edu. BBIRD also maintains a homepage at http://www.wru.umt.edu/bbird.