Continental and local declines in numerous bird populations have led to concern for the future of migratory and resident bird species. The reasons for declines are complex. Loss, modification and fragmentation of breeding habitat, loss of wintering and migratory habitat, and brood parasitism have been implicated. Scientists and concerned citizens agreed that a coordinated, cooperative conservation initiative focusing on non-game land birds was needed. In late 1990, the National Fish and Wildlife Foundation brought together representatives from federal, state, and local government agencies, foundations, conservation groups, industry, and the academic community to form a program to address the problem. Thus, Partners in Flight (PIF) was conceived as a voluntary, international coalition of governmental agencies, conservation groups, academic institutions, private businesses, and everyday citizens dedicated to "keeping common birds common" and reversing the downward trends in bird populations. Partners in Flight's primary goal is to direct resources to the conservation of non-game land birds and their habitats through cooperative efforts in monitoring, research, management, education, and international cooperation.

Colorado Partners in Flight, whose mission is to promote and enhance conservation and management efforts for Colorado birds, officially came into being in 1991. Participants include the Colorado Division of Wildlife, Colorado Bird Observatory, Bureau of Land Management, U.S. Fish and Wildlife Service, U.S. Forest Service, National Park Service, Department of Defense, USGS Biological Resources Division, Colorado State Parks, municipal government parks and open space departments, Audubon of Colorado and its chapters, The Nature Conservancy, Colorado Natural Heritage Program, American Birding Association, Denver Museum of Natural History, Denver Field Ornithologists, Colorado Field Ornithologists, university researchers, dedicated individuals, and many others.

The Flight Plan: Bird Conservation Planning

The "Flight Plan" is a national PIF document that defines the strategy for coordinating, developing and writing Partners in Flight Bird Conservation Plans. Much as the North American Waterfowl Management Plan does for waterfowl, PIF's geographically based plans will direct efforts and prioritize funding for conserving non-game land birds.

Effective and efficient ecological management requires determining which species and habitats are most in need of conservation. The PIF plans identify priority species and habitats, and establish objectives for bird populations in physiographic areas (PIF planning units defined by biotic communities and bird distribution). The plans identify the type, quantity, and quality of habitats required by priority species on the local level and at the landscape scale. The plans also identify conservation opportunities and partnerships needed to accomplish their objectives. PIF bird conservation plans complement the long-established North American Waterfowl Management Plan and the more recently initiated Shorebird Conservation Plan and North American Colonial Waterbird Conservation Plan.

The first official step in developing a Bird Conservation Plan (BCP) in Colorado was to apply for a National Fish and Wildlife Foundation (NFWF) grant. This grant provided support for a Colorado PIF Bird Conservation Facilitator to coordinate the development of the Colorado Bird Conservation Plan. The grant agreement for $32,000 in NFWF funds was signed in February 1998, requiring a match of $48,000 in challenge funds. Prior to the grant's approval, the Colorado Division of Wildlife pledged a match of $38,000, which demonstrated the Division's strong commitment to non-game bird conservation. These matching funds became available in May 1998, and work on the plan began. The remaining $10,000 in challenge funds to complete the grant requirements were raised later.

Dozens of people have since participated in planning meetings, drafted language, and reviewed documents. State working groups and habitat working groups, functioning under the umbrella of the Western Working Group of the U.S. Partners in Flight program, have held numerous planning meetings. The meetings were open to anyone with an interest in bird conservation and were designed to solicit, discuss, and compile information into a document that forms the core of the plan. These meetings revealed significant data and observations unavailable in the scientific literature, information that highlighted the need for specific approaches and objectives to manage local variations in habitat use and ecosystem function.

The goals of the Colorado Bird Conservation Plan are to conserve and monitor all bird species in Colorado, monitor the quantity and quality of bird habitat, conserve unique representatives of and/or core areas in each major habitat, protect local sites that are important for conservation of priority species, promote management practices that benefit birds on all lands, conserve wintering grounds and migration habitat, develop outreach and educational programs, identify and promote research priorities, and evaluate the success of the Colorado PIF bird conservation implementation program on an annual basis.

Partners in Flight recognizes that significant gaps in our knowledge of Colorado's birds still exist. However, it intends to assemble the best and most current scientific information into a format that land managers and landowners can use to put ideas into action. As new information becomes available through the research and monitoring recommended in this plan and from the discovery of previously overlooked information, it will be incorporated into updated versions of the plan. Thus, the Colorado Bird Conservation Plan is to be considered a dynamic document.

The Prioritization Process

Rather than presenting information about all species, the Colorado Partners in Flight Bird Conservation Plan identifies species that have conservation priority in each of its planning units; it assumes that conservation measures focused on these priority species and their habitats will benefit the other species using these habitats as well.

The Partners in Flight Species Prioritization Process, first developed in 1991, has been continually reviewed and refined (Carter et al. 1999). This process ranks each species of North American breeding bird based upon seven measures of conservation "vulnerability." The rankings include global measures (breeding distribution, non-breeding distribution, and relative abundance) and measures specific to each physiographic area [area of importance (AI), population trend (PT), and threats to breeding (TB)]. For each category a species is given a score of 1-5, with 1 indicating the lowest vulnerability and 5 the greatest vulnerability. The sum of the scores produces a composite score ranging from 7-35. Species with high overall scores are most vulnerable to extinction (although they often are not listed as endangered at present) and need conservation measures or at least need to be carefully monitored through their ranges. The PIF prioritization process can be applied at the global, national, state, or physiographic area level.

Partners in Flight has generated scores for all breeding species in Colorado by physiographic area. Selected species scores are presented in Appendix B. A further analysis of these scores was used to identify priority species, by physiographic area, for Colorado. The criteria which define priority species are categorized into several "tiers," entry levels into the priority species pool. No tier is more important than any other, and although each priority species may receive a single tier designation, it may also qualify for inclusion in other tiers. Furthermore, some species peripheral to Colorado are not included as priority species even though they may meet tier criteria. The tiers are defined as follows:

I. High overall (global) priority--species scoring 22 in the PIF prioritization system. Indicates high vulnerability of populations throughout the species range, irrespective of specific status in the physiographic area. Peripheral species are omitted.

II. High physiographic area priority--species scoring 19-21 in the PIF system, with AI + PT 8. Indicates a species of moderately high global vulnerability and with both relatively high abundance and a declining or uncertain population trend in the physiographic area.

III. Additional Watch List--species on PIF's national Watch List that did not already meet criteria I or II. Watch List species score 20 (global scores only ), or 18-19 with PT = 5.

IV. Abundant yet declining--any additional species for which the score for AI = 5 and the score for PT = 5. May identify species or a habitat type in need of monitoring.

V. Area responsibility--additional species with relatively high proportion of global population in the physiographic area [>5% for areas < 200,000 km² (77,200 mi²); 10% for areas > 200,000 km²]. Signifies that the area shares in responsibility for long-term conservation of species, even if not currently threatened.

VI. Additional listed--species on federal or state endangered, threatened or special concern lists that did not meet any of the above criteria. These are often rare or peripheral populations.

VII. Local concern--species of justifiable local concern or interest. May represent geographically variable population or be representative of specific habitat of conservation concern.

In addition to the PIF scores, Appendix B includes scores from COVERS, the Colorado Vertebrate Ranking System developed for the Colorado Division of Wildlife. The COVERS categories listed for each species include: Stage1 Biology (B1, indicating degree of biological imperilment); Stage 2 Biology (B2, sum of Stage 2 Biology variable scores); Stage 2 Importance of Colorado Populations (IMPORT, evaluates the "value" of individuals to the health of the global population and to the biotic community of the state of Colorado); and Total Biology (T.Bio, sum of all Stage 1 and Stage 2 Biology variable scores). Appendix B also indicates species having a special status designated by an agency within the State of Colorado.

Sorting the species in the priority pool by habitat identifies the highest priority habitats and associated species. A high priority habitat may have many priority species using it or a few priority species with high concern/vulnerability scores. Priority habitats are listed by physiographic area and are discussed in detail in the plan.

The highest priority species are listed by habitat within the appropriate physiographic area and a species account is provided. Priority species discussed in the plan are identified by tiers in Appendix B. A species may be discussed for any of the following reasons: 1) it represents a group or suite of species that will likely react similarly to management recommendations; 2) it utilizes a biotic component that is essential to a functioning habitat (e.g., a species using forest canopy or requiring a certain percentage of ground cover for nesting); 3) it is an area sensitive species or requires another landscape feature, such as a large territory; 4) it requires a certain successional stage; 5) the habitat is its primary or only breeding habitat; 6) monitoring it will permit ready assessment of the success of conservation efforts; 7) it is on either the federal endangered and threatened species list, or a watch list/species of concern list from a state agency or national conservation organization.

Defining Status and Setting Biological Objectives

Important steps in developing a bird conservation plan for the state include determining the current status of each priority species identified in the "tier process" and setting biological objectives that will enable us to evaluate our conservation efforts.

In defining the status of and setting biological objectives for each priority species in Colorado, many sources of information have been explored. In many cases the best source of information is the national Breeding Bird Survey (BBS). For this reason, it is important to define how we are using these BBS data, and to recognize the limitations of these data. The BBS employs trained volunteer observers who conduct roadside bird surveys along randomly distributed routes. The data obtained since its establishment in 1966 provide valuable information about bird distribution, population trends, and abundance across the country. However, for a number of reasons related to survey design or species life history (including geographic areas with limited road coverage, a limited source of trained observers, species with clumped populations, nocturnal species, or habitats not easily surveyed from a roadside) BBS data for some species are limited and inadequate to estimate distribution and population trends. These limitations especially affect data from the sparsely populated West. In many cases, species inadequately sampled by BBS methods are not satisfactorily surveyed by alternative methods either. This issue is being addressed in Colorado by such programs as Monitoring Colorado's Birds (MCB), and the plan identifies a number of specific needs for improved population surveys. The data available from BBS and any other sources will be presented in the "Status and Reasons for Concern" sections of the bird species accounts.

For this plan, we have adopted interpretations of BBS data that are consistent with BBS data analysis protocols and PIF uses (e.g., the prioritization process) of the data. Population trends (percent/year) are estimated by the BBS using the route-regression method described by Geissler and Sauer (1990). Regional trends are estimated as a weighted average of trends on individual routes. For further information about the analysis methods used, see the BBS website http://www.mbr-pwrc.usgs.gov/bbs.

For BBS population trend data, we will consider as statistically significant any trend with a P-value 0.10 and a sample size 14. Based on this standard, a P-value larger than 0.10 indicates no statistical evidence of a trend, positive or negative (i.e., there is no evidence that the regression line is significantly different from zero). P-values larger than 0.10 may mean that there really is no trend, or that there is a trend that can't be detected due to low sample size, high variances, very small magnitude trends, or a combination of factors. For trend information, whether significant or not, we will provide the P-value and sample size to permit the readers to reach their own conclusions.

In addition, we frequently use BBS data to define the distribution of a species in the physiographic area. For this purpose we have calculated the following values for that portion of the physiographic area that lies within Colorado (i.e., not for the entire physiographic area) using data from 1988-1997: 1) mean percentage of routes run in the physiographic area on which the species was detected; 2) mean number of individuals detected per route; and 3) number of routes.

From the status and distribution information, biological objectives were developed. Biological objectives must be specific and measurable so that we can evaluate our progress toward the goals of keeping populations well-distributed throughout their natural range and reasonably common, stable, and self-sustaining.

Distribution is generally defined as the area in which a species is "present," and is distinguished from abundance. Due to lack of information about most species that would allow us to define or measure a biological objective as the number of X km² in which the species is present, we have chosen a surrogate. For common and widespread species we set distribution objectives as the desired proportion of BBS routes (or other equivalent units) on which the species is to be recorded. For rare species, or ones that occur in only a few areas (e.g., colonial species), the number of sites supporting the species is more meaningful and is used instead of BBS routes.

For many species, meeting a distribution objective will also ensure that the population is reasonably abundant, stable, and self-sustaining, and no additional objectives are needed. However, this is not always the case (e.g., for species with clumped distributions). In such cases, the total population could decline seriously, with each site still supporting some birds. Thus, an abundance component should be included in the objectives for these species. In most cases these biological abundance objectives are based on the BBS population trend data and call for maintaining a positive trend of X%, reversing a negative trend, or improving a current flat-line trend estimate until it is a positive trend. Another sort of abundance objective requires a mean of X individuals recorded per route.