Association of Fish&Wildlife Agencies

Recognition that a cooperative, nonadversarial conservation approach was required to address bird and habitat issues led to formation in 1990 of Partners in Flight (PIF). The PIF mission is expressed through three related concepts:

Helping species at risk. Species exhibiting warning signs today must be conserved before they become imperiled. Allowing species to become threatened or endangered results in long-term and costly recovery efforts whose success often is not guaranteed.

Keeping common birds common. Native birds, both resident and migratory, must be retained in healthy numbers throughout their natural ranges.

Voluntary partnerships for birds, habitat, and people. A central premise of PIF is that the resources of public and private organizations throughout the Americas must be combined, coordinated, and increased in order to achieve success in conserving bird populations in this hemisphere.

Over the last seven years, PIF has engaged in a comprehensive planning effort, resulting in several dozen regional bird conservation plans covering all states or physiographic areas in the U.S. (Pashley et al. 2000). Similar regional efforts are underway in Canada and Mexico. These regional and state PIF plans (see www.partnersinflight.org) identify priority species and habitats, set goals and objectives, discuss local issues and opportunities, and outline strategies for local or regional partners to implement bird conservation objectives. In 2004, PIF published its North American Landbird Conservation Plan (Rich et al. 2004), which synthesized landbird priorities and objectives at a continental scale and set forth a coordinated approach to landbird conservation among nations and regions of North America. The North American Landbird Conservation Plan is a blueprint for continental landbird conservation and, as such, is not intended to replace existing or developing regional and state PIF plans. The conservation and management strategies required for several hundred landbird species are far too complex and variable across North America to be treated only at a continental scale. Implementation of on-theground bird conservation strategies must take place at state, provincial, and local levels, guided by regional and continental planning.

Throughout the development of regional and continental bird conservation plans, PIF has followed a stepwise planning approach, based on the best available scientific data and judgments from a broad spectrum of bird conservation experts. Originally described as the PIF "Flight Plan" (Pashley et al. 2000), these steps include

assessing conservation vulnerability among all native landbird species,

identifying species most in need of conservation attention at continental and regional levels,

setting quantitative population objectives for species of conservation importance,

identifying conservation needs and recommended actions for priority species and their habitats,

outlining an implementation strategy for meeting species and habitat objectives, and

evaluating success, making revisions, and setting updated objectives for the future.

An unprecedented opportunity to implement bird and habitat objectives identified by PIF and the other bird-conservation initiatives exists with the State Wildlife Grants program. In order to make the best use of the State Wildlife Grants program, Congress charged each state and territory with developing a statewide Comprehensive Wildlife Conservation Strategy (Strategies). These Strategies will provide an essential foundation for the future of wildlife conservation and a stimulus to engage the states, federal agencies, and other conservation partners to strategically think about their individual and coordinate roles in prioritizing conservation efforts across the nation. In developing the Strategies, the needs of bird species and their habitats will be considered along with needs for all other taxa.

Our intent in this report is to summarize the detailed information from PIF bird conservation plans that is most relevant to state planners and biologists as they develop the Strategies. The report is broken into two parts—Part I is a user’s guide that explains the standardized set of procedures and assumptions used to develop the data and information for each state. The data and information is presented in Part II. This information includes

priority species for each state, based on the PIF continental and physiographic area planning process;

population estimates for each state, broken into portions of Bird Conservation Regions (BCRs) within each state;

population objectives and numerical targets for each priority species, based on methods defined in the PIF North American Landbird Conservation Plan; and

a cross referenced list of where relevant conservation issues, management recommendations, research and monitoring needs are found in the physiographic area and state PIF plans.

The priority species in each state, along with their state and BCR population estimates and targets, are presented separately for each major habitat type within each state. These suites of species are similar to the habitat-species suites presented in many physiographic area and state PIF plans and represent groups of focal species that will benefit most from conservation actions within a given habitat type. The priority species are listed within the species-habitat suite that is their primary breeding habitat.

The first table within each habitat-species suite contains the priority species that breed primarily in that habitat, a population estimate within the portion of each BCR found within the states for each species, the continental population objectives from the PIF North American Landbird Conservation Plan for each species, and the state numerical population target for each species that is needed to meet the continental objectives. The first table also contains two columns indicating the species priority tier scores for breeding (B_Tier) and non breeding or wintering (N_Tier) species. The second table within each habitat-species suite presents the numerical statewide population objectives for the priority species using the information from the first table. The last section within each habitat-species suite is a cross referenced list of where information on goals, objectives, strategies and individual species accounts are located in the relevant PIF physiographic area and state plans.

The information described above is provided for every state, following a standardized set of procedures and assumptions (detailed in Part I of this report), recognizing that states will vary in their approach to developing the strategies and in their need for specific types of information from PIF plans. The methods described below are adapted from several key sources, which should be consulted for greater detail if needed. These sources include: Carter et al. (2000), for (now slightly out-of-date) descriptions of the PIF species prioritization process; Pashley et al. (2000), for a summary of the PIF physiographic area plans; Panjabi et al. (2001), for a handbook to using the PIF species assessment database; the Rocky Mountain Bird Observatory web site (www.rmbo.org/pif/pifdb.html), for viewing PIF species assessment scores and accompanying data; Rosenberg and Wells (in press) for an example of a regional PIF synthesis; and Rosenberg and Blancher (in press), for a description of how PIF sets population objectives. These references are currently, or will soon be, available on the PIF web site: www.partnersinflight.org. In addition, the following sections of the North American Landbird Conservation Plan (Rich et al. 2004) are most relevant to understanding the process for developing state and local population estimates and objectives:

Table 1, Species of continental importance for the U.S. and Canada;

Appendix A, Assessment scores and estimated population size of North American landbirds; and

Appendix B, Methods used to estimate population sizes and percents. Finally, the author is available for questions and consultation at kvr2@cornell.edu or (607) 254-2412.

PIF priority species

Bird species may be of conservation importance in a state for several reasons, including rarity or high threats to populations within the state, representation in rare or unique habitats within the state, or because of larger regional or even global concerns for the species. PIF bird conservation plans provide lists of species considered priorities for conservation action at continental, physiographic area, or (in some cases) state levels. Providing PIF priority species lists to all states provides planners with an assessment of which species may require coordinated regional (i.e. inter-state) actions, as well as for which species their state can make the most significant regional or national conservation contribution. In Part II of this report, we provide PIF priority lists for each state, partitioned according to the status of each species in portions of BCRs overlapping each state.

The first step in PIF’s planning process is to identify species most in need of attention, based on a conservation status assessment of each species throughout its range and annual cycle. PIF has developed a process that evaluates several components of species vulnerability and provides an overall conservation assessment of the species (Hunter et al. 1993, Carter et al. 2000, Panjabi et al. 2001). This process has been tested, reviewed, and updated, and its scientific credibility acknowledged by the American Ornithologists’ Union (Beissinger et al. 2000).

Species assessment is based on the PIF North American Species Assessment Database, which contains standardized data on the status of North American landbirds at the continental scale (www.rmbo.org/pif/pifdb.html). During the development of the North American Landbird Conservation Plan, the PIF Science Committee reviewed the data and consulted other appropriate experts on all factors in the database to ensure that our assessment reflects the current state of knowledge.

Each species was assigned "global" scores for six factors that assess distinct aspects of vulnerability across the species' entire range: Population Size (PS), Breeding Distribution (BD), Nonbreeding Distribution (ND), Threats to Breeding (TB), Threats to Nonbreeding (TN), and Population Trend (PT). Scores for each factor reflect the degree of each species’ vulnerability (i.e., risk of significant population decline or rangewide extinction) as a result of that factor. Scores ranged from “1” for low vulnerability to “5” for high vulnerability. Complete descriptions, justifications, scoring criteria, and definitions for each factor can be found in Panjabi et al. (2001), available at the Rocky Mountain Bird Observatory web site (www.rmbo.org/pif/pifdb.html).

To determine species of conservation importance at the continental scale, we calculated a Combined Score, which is a single metric of a species’ relative conservation importance. The Combined Score is calculated as (highest of TB or TN scores) + (highest of BD or ND scores) + PT + PS. This score can range from 4 for a widespread, relatively secure species for which we have few concerns, to 20 for a species of the very highest concern. The most vulnerable species are those with a combination of small and declining populations, limited distributions, and deteriorating habitats. Species were placed on PIF's Continental Watch List if they had a Combined Score ≥ 14, or a Combined Score = 13 with Population Trend score = 5 (the latter representing a 50% decline over 30 years).

In addition to species on the Continental Watch List, we identified species of regional importance in each BCR, following methods used in many PIF physiographic area plans (e.g. Rosenberg and Wells, in press, www.partnersinflight.org). Species of regional importance were identified based on four “global” factors (PS, BD, ND, TN), as well as threats to breeding populations (TB), population trend (PT), and area importance (AI) scores, which are specific to each BCR. Area Importance is scored according to the relative abundance of a species in a BCR (based on BBS abundance or equivalent), relative to the maximum abundance that species achieves across all BCRs. Categories of priority status are determined by examining combinations of factor scores, as well as the total rank score, which is a measure of overall conservation priority. For more information about how scores are assigned see the PIF Handbook of Species Prioritization at http://www.rmbo.org/pubs/downoads/Handbook.pdf.

Species of conservation importance identified for each state, therefore, represent a combination of species on the PIF Continental Watch List, and additional species that meet criteria for regional importance in BCRs that overlap each state. The resulting "priority tiers" listed for each state are defined specifically as follow:

Tier I. High Continental Importance. — Species on the PIF Continental Watch List, which are typically of conservation concern throughout their range. These are species showing high vulnerability in a number of factors, expressed as any combination of high global parameter scores, with AI ≥ 2 (so that species without manageable populations in the region are omitted). High level conservation attention warranted.

Tier II. High Regional Priority. Species that are of moderate continental priority (not on Continental Watch List), but are important to consider for conservation within a region because of various combinations of high parameter scores, as defined below; total of 7 parameter scores = ≥ 19.

Tier IIA. High Regional Concern. Species that are experiencing declines in the core of their range and that require conservation action to reverse or stabilize trends. These are species with a combination of high area importance and declining (or unknown) population trend; total of 7 parameters ≥ 19, with AI + PT ≥ 8.

Tier IIB. High Regional Responsibility. Species for which this region shares in the responsibility for long-term conservation, even if they are not currently declining or threatened. These are species of moderate overall priority with a disproportionately high percentage of their total population in the region; total of 7 parameters ≥ 19, with AI = 5 or % population > threshold (see http://www.rmbo.org/pubs/downoads/Handbook.pdf).

Tier IIC. High Regional Threats. Species of moderate overall priority that are uncommon in a region and whose remaining populations are threatened, usually because of extreme threats to sensitive habitats. These are species with high breeding threats scores within the region (or in combination with high nonbreeding threats outside the region); total of 7 parameters ≥ 19 with TB + TN > 6, or local TB or TN = 5.

Scores for all breeding and wintering species in each BCR, as well as priority tiers assigned to each species, may be found at: http://www.rmbo.org/pif/pifdb.html.

Population estimates

As part of the development of the North American Landbird Conservation Plan (Rich et al 2004), PIF estimated the “current global” population size for each of 448 landbird species. These estimates give an impression of the size of the landbird resource, and more importantly they emphasize the magnitude of the task of attaining landbird population objectives. Continental (U.S. and Canada) population estimates provide a starting point for estimating population sizes in states, provinces or BCRs, and an understanding of the magnitude of attaining objectives regionally. In Part II of this report we provide population estimates for all continental and regionally important bird species presented by habitat-species suite in each state, listed by the portions of BCRs within that state.

The following methodology is excerpted from Rich et al (2004); Appendix B, which provides additional details. We used Breeding Bird Survey (BBS) data from the 1990s as the basis for population estimates across the U.S. and across Canada south of the arctic (i.e., excluding BCR 3, see next section). BBS-based estimates of abundance were calculated according to the following steps:

1) For each BBS route run within acceptable weather conditions, counts were averaged across years to give a single average count for the 1990s for each species recorded on each route.

2) Species counts were averaged across all BBS routes in each geo-political polygon defined by the intersection of a BCR and a state.

3) Indices of abundance were calculated for each geo-political polygon by multiplying average counts per BBS route times area of the geo-political polygon, and dividing by the theoretical area covered by a BBS route (25.1 km2, assuming 400-m radius around each of the 50 count circles). For example, the index of abundance for Wood Thrushes in the BCR 28 portion of New York equals 10.71 birds/route (29 routes sampled in 1990s) x 37,872 km2 (area of New York) / 25.1 km2 (area per BBS route) equals approximately 16,000.

4) State-wide indices of abundance were calculated by simple addition across all polygons making up each state, thus giving a population index for Wood Thrushes in all of New York of approximately 52,000. BCR-wide indices of abundance were calculated in the same manner.

5) State-wide indices of abundance were converted to population estimates by applying three correction factors (see Rosenberg and Blancher, in press, for more detail on these correction factors):

Pair correction: Indices were multiplied by two on the assumption that typically a single member of a breeding pair is observed during BBS tallies;

Detection area correction: Each species was placed into one of five detection distance categories, based on presumed effective detection during 3-minute BBS counts: 80m, 125m, 200m, 400m and 800m. Because area of detection increases as the square of detection distance, the detection area correction is then simply the square of the ratio between 400m (theoretical BBS count circle) and species-specific effective distance. For example for Wood Thrush, placed in the 200m class, the population index is multiplied by a detection area correction of 4 (square of 400/200). Note that effective detection distances are intended to incorporate not only the distance at which a species is normally heard and seen, but also the distance the species moves during a 3-min count period – this is why some wide-ranging species have been assigned an 800-m detection distance despite being counted within a 400-m BBS circle.

Time of day correction: Almost all species show a temporal change in detection across the 50 BBS stops, some declining from a dawn chorus, others peaking after sunrise or later in the morning. A time of day correction is applied to the population index to adjust counts to the maximum time of detection. This adjusts for birds not detected at other times of the morning. The correction factor is the ratio of counts at the peak of detection (calculated using a polynomial curve fit to smooth out stop-by-stop variance) relative to the average count over whole BBS routes. Time of day correction factors were calculated from survey-wide BBS stop-by-stop data. For Wood Thrush, whose detectability declines from a peak at BBS stop 1, the time of day correction is 2.30.

For Wood Thrushes, the population estimate for New York = 52,000 (index from step 4) x 2 (pair correction) x 4 (detection area correction) x 2.30 (time of day correction) = roughly 960,000 breeding individuals.

For a variety of reasons, the population estimates presented in Part II are rough estimates and will need to be improved over time, especially for use at smaller scales. Two main assumptions of the approach are mentioned here (see Rosenberg and Blancher, in press).

Assumption: Habitats are sampled in approximate proportion to their occurrence in the regional landscape. Although BBS is designed to provide a random sample of the landscape, limitations of a road-based survey mean that the landscape sampled is a biased representation of available habitat – for example species characteristic of high elevation habitats are likely to be undersampled by BBS simply because roads tend to follow valley bottoms in mountainous regions. In northern BCRs, there is a geographic bias, with most BBS data available from the southern portions of those BCRs. Checklist and Breeding Bird Census sites are determined by individual scientists and volunteers, so are not a random sample of arctic regions. We have not accounted for habitat bias in our continental estimates, in part because it will differ from region to region, and because the magnitude of bias has not yet been estimated in many regions or at a continental scale. Correction for habitat bias should be considered when using the methods described above at smaller scales.

Assumption: Birds present but not detected during BBS counts are accounted for, on balance, by one or more of the three density corrections applied above (pair, detection area, and time of day corrections). Species that have a peak of detection outside of the BBS sampling window (e.g., early-season breeders, most nocturnal species) are likely to have been underestimated. Pair corrections may result in over-estimation of population size, if a high proportion of counts involve either both members of a pair, or unmated birds.

In addition to stating assumptions behind this population estimation procedure, PIF has assessed the accuracy and precision of population estimates for each species. Overall, about two-thirds of the “global” population estimates presented in the North American Landbird Conservation Plan are rated as having fair to moderate accuracy, expected to be within and usually well within an order of magnitude of the correct breeding population. Estimates are least accurate for wideranging species with large populations in northern Canada or south of the U.S.; for many of these species our estimates for the U.S. population and that of many states, will be more accurate than our “global” estimate. For species estimates based largely on BBS abundance, we also assessed the precision of these estimates, a measure of how repeatable the estimate is given the variance among counts. Results show that most estimates are repeatable within 10% or 20%; i.e., repeatability of the estimate is generally high or very high, even when accuracy was rated as fair to moderate (summarized in Rich et al, 2004; Appendix B).

Population objectives

For each state, we list the population objective and numerical population targets for the continental and regionally important species by habitat-species suite, within the portions of BCRs that overlap that state (See Part II). Population objectives are based on rangewide population trends of species, as described below, and targets are based on population estimates described above.

Setting population objectives requires knowledge of population size and trends, as well as agreement on historic baselines to which present-day populations can be compared. As a starting point, the target for PIF priority species is to maintain current populations, or to return declining populations at least to their numbers in the late 1960s. This date was selected because we believe that target is achievable and realistic for most species of conservation importance. Acceptance of this baseline recognizes that the extensive losses and modifications of habitat since the European settlement of North America are historical realities that are not likely to be reversed to a significant extent at the continental level. It also recognizes that prior to 1966 and the start of the Breeding Bird Survey, there were no consistent data for most landbird species upon which to base measurable population objectives.

Population objectives were determined for each species based on degree of population change since 1966, according to the trend data used in the species assessment process. However, we recognize that trend estimates are not exact. Rather than proposing population objectives that represent estimates of the actual number of birds in 1966 (which would generate a different target for each species), we assigned each species to one of four population objective categories, as described below. For now, these objectives are based on trends at the continental level, and defined for each state to help meet continental targets. Refinement of this process and comparison with local and regional data and targets, may dictate more or less aggressive objectives than for the species continental targets. For regionally important species not on the Continental Watch List, continental estimates of trends are used to determine objectives and targets, as with Watch List species described in Rich et al (2004). For species that are the subject of legally mandated Recovery Plans, we defer to the objectives of those plans.

Double Population: For all species that have undergone severe declines of 50% or more over 30 years (i.e., those with Population Trend scores of 5), the objective is to double the current population over the next 30 years. Reversing declines and doubling present-day populations is warranted for nearly a third of the 100 species on PIF's Continental Watch List. Increase Population by 50%: For species that have undergone moderate declines (15-50% over 30 years, as indicated by Population Trend scores of 4), the objective is to increase the population by 50% over the next 30 years. This objective is warranted for 23 Continental Watch List Species.

Maintain/Increase Population: Watch List Species with uncertain or unknown past trend (Population Trend scores of 3) may be seriously declining without our knowledge. Our conservative objective for these species, therefore, is to maintain or increase current populations in the next 30 years while simultaneously improving our knowledge of population status. This is the objective for 33 Continental Watch List Species.

Maintain Population: For species with stable or increasing populations, PIF’s objective is to at least maintain current populations.

By combining the suggested population objectives with our initial estimates of population size, a first approximation of a numerical population target for each species at the continental, regional and state levels can be determined. These are listed separately for each species in each BCR polygon within a state. For example, the North American Landbird Conservation Plan calls for a doubling of present-day Brewer’s Sparrow populations over the next 30 years to restore a rangewide population of roughly 32 million breeding individuals. The portion of that target suggested for Idaho, therefore would be to double the present-day population of 1.2 million Brewer's Sparrows, with 83% of that target being met in the BCR-9 portion of the state.

Comparing and refining estimates

Numerous experts and many rounds of review have helped to make the process of estimating populations as accurate as possible, and we hope that it will become a valuable tool in landbird conservation. But we realize that these estimates provide a starting point, not a final answer. Significant discussions have already taken place and current efforts are underway to refine the process and assumptions of our methodology for calculating population estimates and assigning population objectives. For example, our correction factors are being reviewed and revised by regional PIF groups, and additional correction factors are being proposed and developed (e.g., a habitat bias correction factor). Regional and local population objectives are being proposed based on data and knowledge at that scale to cumulatively support the continental population objective. Additionally, a Science Review Team not involved in the process of developing this methodology has been created and will be meeting soon to provide an independent review of the population estimation process with recommendations. Please consult with your state PIF representative and/or your BCR or JV Coordinator to determine the status of these efforts and how they may apply to the State Wildlife planning process. We look forward to the outcome of all these efforts and anticipate substantial improvements with every North American Landbird Conservation Plan update as data, analysis and concepts improve.

Therefore, it is important for users of these estimates to understand the following:

All species assessment scores have a degree of uncertainty in the underlying information and professional judgments were made in setting each score. See Carter et al. 2000 and Panjabi et al. 2001 for details.

The “global” population size estimates rely on several assumptions and have a level of error that can only be approximated. Estimates will be revised as data improve and as the estimation process is refined. Revised estimates will be posted annually on the PIF web site (www.partnersinflight.com).

Rule sets were used to select Species of Continental Importance and to assign those species to categories for Conservation Action and Monitoring Need. Different rule sets would produce different lists, but the ones used here are the result of exhaustive discussion and analysis by landbird experts.

Population objectives are based on past population trend and are independent of population size estimates. Changes in population size estimates will have no effect on objectives, but improved trend information could have large effects. Objectives will be revised as appropriate.

At the Population Objectives workshop in Texas (February, 2004), and at several regional workshops, the process of comparing our "top-down" BBS-derived population estimates (see Part II) with locally derived "bottom-up" estimates was begun. We urge that additional comparisons be made wherever local data for priority bird species exist. These comparisons have produced "mixed" results to date, with some species well within expected ranges and other estimates based on different approaches not yet very close.

Disparity or similarity between estimates derived from the process described in this report vs. a bottom-up (local/regional) approach does not necessarily mean that one or both estimates are wrong. In all cases, the assumptions of each approach need to be carefully evaluated. Initial comparisons have suggested that one or more correction factors for some species may need revision, resulting in comparisons that are much closer. In some cases, re-thinking of local density information with respect to what constitutes suitable habitat has also resulted in closer comparisons. Ultimately, population estimates based on accurate habitat-suitability and - availability models will be needed to evaluate the accuracy and utility of the process described in this report.

Issues and recommendations by habitat-species suites

For each habitat-species suite, we provide a cross referenced list where information on conservation issues, management recommendations, research and monitoring needs, and individual species accounts most relevant to that habitat or group of species are located. The list will direct the reader to specific pages within the relevant sections of PIF physiographic area and state plans. Our aim is simply to provide a link to information from the bird conservation plans. We therefore have not added new information or filled in gaps where information for a particular species or habitat is lacking at present.

Next Step: Deriving habitat objectives

In most (but not all) cases, conservation actions aimed at maintaining or restoring healthy populations of landbirds will be directed at habitat. It is desirable, therefore, to set explicit habitat objectives and targets for species or suites of priority bird species. Deriving quantitative habitat objectives for birds requires an additional set of assumptions (for example, that habitat is limiting for a given species), knowledge of local or habitat-specific densities, knowledge of habitat-related limiting factors, and often a complex GIS-based modeling approach. Such analyses are beyond the scope of this report. A standardized five-step approach to modeling habitat requirements for meeting PIF landbird objectives was discussed at the recent Population Objectives Workshop in Port Aransas, TX (February, 2004). After further discussions a detailed description of the five-step, long term process will be disseminated.

A simple, first-cut estimate of habitat requirements for bird populations can be attempted, however, using population estimates and habitat-specific density information. These estimates may be derived using the equation: Habitat Area = Population Estimate X Density (birds/unit area). These habitat objectives have been used in some PIF physiographic area plans (primarily in Northeast Region), but their acceptance is far from widespread. Nevertheless, these numerical estimates have proven extremely valuable for conveying region-wide habitat conservation needs, such as in the Wildlife Management Institute's Farm Bill report, How Much is Enough? (WMI 2001). For example, we estimated that roughly 4.5 million acres of suitable grassland habitat are required across the Northeast U.S. to support 1.6 million pairs of nesting Bobolinks -- these numbers were translated into specific goals for Farm Bill programs.

These simple habitat estimates may be most reasonable for birds with simple habitat requirements, such as grassland species. For many forest birds, however, local density-derived habitat estimates are not considered reasonable, because species are patchily distributed within a matrix of potentially suitable habitat. For these birds, defining "suitable" habitat is critical, and more complex habitat-modeling techniques are required.

References

Carter, M. F., W. C. Hunter, D. N. Pashley, and K. V. Rosenberg. 2000. Setting conservation priorities for landbirds in the United States: The Partners in Flight Approach. Auk 117:541– 548.

Beissinger, S.R., J.M. Reed, J.M. Wunderle, Jr., S.K Robinson, and D.M. Finch. 2000. Report of the AOU Conservation Committee on the Partners in Flight species prioritization plan. The Auk. 117(2): 549-561.

Hunter, W. C., M.F. Carter, D.N. Pashley, and K. Barker. 1993. The Partners in Flight prioritization scheme. Pages 109-119 in Status and Management of Neotropical Migratory Birds (D. Finch and P. Stangel, Eds.). USDA Forest Service General Technical Report RM- 229.USDA Forest Service, Fort Collins, Colorado.

Panjabi, A., C. Beardmore, P. Blancher, G. Butcher, M. Carter, D. Demarest, E. Dunn, C. Hunter, D. Pashley, K. Rosenberg, T. Rich, and T. Will. 2001. The Partners in Flight Handbook on Species Assessment and Prioritization. Version 1.1. Rocky Mountain Bird Observatory, Brighton, Colorado.

Pashley, D. N., C. J. Beardmore, J. A. Fitzgerald, R. P. Ford, W. C. Hunter, M. S. Morrison, and PIF Priorities and Objectives Defined at the State and BCR Level: Users’ Guide 11 March 2004

K. V. Rosenberg. 2000. Partners in Flight: Conservation of the Land Birds of the United States. American Bird Conservancy. The Plains, Virginia.

Rich, T.D., C.J. Beardmore, H. Berlanga, P.J. Blancher, M.S.W. Bradstreet, G.S. Butcher, D.W. Demarest, E.H. Dunn, W.C. Hunter, E.E. Inigo-Elias, J.A. Kennedy, A.M. Martell, A.O. Panjabi, D.N. Pashley, K.V. Rosenberg, C.M. Rustay, J.S. Wendt, T.C. Will. 2004. Partners in Flight North American Landbird Conservation Plan. Cornell Lab of Ornithology. Ithaca, NY.

Rosenberg, K. V., and P. J. Blancher. In press. Setting numerical population objectives for priority landbird species. Pages xx–xx in Proceedings of the Third International Partners in Flight Conference (C. J. Ralph and T. D. Rich, Eds.). USDA Forest Service General Technical Report PSW-GTR-191. Albany, California.

Rosenberg, K. V., and J.V. Wells. In press. Conservation priorities for terrestrial birds in the northeastern United States. Pages xx–xx in Proceedings of the Third International Partners in Flight Conference (C. J. Ralph and T. D. Rich, Eds.). USDA Forest Service General Technical Report PSW-GTR-191. Albany, California.