Reading Species Accounts

General Information

The internet has created an explosion of information on birds and Minnesota’s avifauna is no exception. With a simple click of the mouse or the touch of a screen, a vast plethora of images, text, and data on any bird species is available. Managing and selecting the appropriate material from this glut of information was a challenge. The primary goal of the on-line version of the Minnesota Breeding Bird Atlas (MNBBA) is to present and summarize field data gathered during the Atlas survey period from 2009-2013. For each species we focused on a brief summary of its Minnesota seasonal status, North American breeding distribution and relative abundance, conservation status, and life history in terms of migration, food, and nest. This is followed by more in-depth text on the species: 1) Minnesota breeding distribution, 2) breeding habitat, 3) population abundance, and 4) conservation. Foremost in our minds, was to focus on Minnesota-centric information when it was available. The taxonomic order in which we present Minnesota breeding birds reflects the changes outlined in the 58th supplement to the American Ornithological Society’s Checklist of North American birds published July 5, 2017 (Chesser et al. 2017).  Scientific names for each species are also presented in each account.

Due to the massive amount of data available for each species, we restricted each species account to approximately 2,000 words excluding references, figures, or tables. A few species received more extensive coverage, especially those designated as endangered, threatened, or special concern, or those that are managed. The Minnesota Breeding Bird Atlas should be viewed as a living document to be expanded as well as a stimulus for further inquiry. The extensive records included in the Atlas are a reflection of the passion that Minnesotans have for birds.  This first Atlas provides a snapshot in time which we trust will be repeated at regular intervals like the second generation atlases in our neighboring states and provinces of Iowa, Manitoba, Michigan, Ontario, and Wisconsin.

If more information is desired for a species, then we urge the reader to see information from the references included in each account. We especially direct readers to the free information on Minnesota birds available at the Minnesota Ornithologists’ Union (http://moumn.org/) or at the Minnesota Department of Natural Resources (http://www.dnr.state.mn.us/nr/index.html). Additional excellent resources of general information on birds includes The Cornell Lab of Ornithology (https://www.allaboutbirds.org/) or the Birds of North America, a cooperative endeavor of the American Ornithological Society and the Cornell Lab of Ornithology (https://birdsna.org/Species-Account/bna/home). The latter is the most comprehensive summary of what is known about North American bird species,  but full access requires a library account or a minimal payment.  All of these documents were extensively used in the species accounts. We apologize to those individuals who have published research results for a species that have not been included. We simply were unable to produce a complete summary for each species, given the wealth of available information, especially for well-studied species.

We truly hope the MNBBA species accounts will motivate the collection of more quantitative data on the nesting activities, habitats, and status of Minnesota birds. For example, what are the important micro and macro characteristics that determine where species nest? How many birds are actually nesting in an area? Are nesting attempts successful, and if not, why not?  We know a fair amount about many species, especially the common ones, but there is still much to be discovered about most bird species nesting in Minnesota.

Overview

Minnesota Seasonal Status

The seasonal status is briefly described for each species in terms of its annual frequency of occurrence and season of occurrence in Minnesota. We also assumed the species is native, but note if it is a non-native species. Two species of birds are officially listed by the Minnesota Department of Natural Resources (2017a) as “invasive terrestrial animals”, the European Starling and the Mute Swan. Invasive species are those that are not native to Minnesota and cause economic or environmental harm or harm to human health.

We used the following definitions for describing the species occurrence during the breeding season:

Regular: The species is found every year in the state of Minnesota. Regular species are further classified as either:

• Permanent, year-round residents; or
• Breeding residents present primarily during the summer season.

Visitant: A non-resident species whose occurrence in Minnesota is outside the species expected distribution range; it occurs in Minnesota 5 to 10 years each decade.

Vagrant: A non-resident species whose occurrence in Minnesota is well outside the species normal distribution range; it occurs in Minnesota 1 to 4 years each decade.

Former Breeding Species:  Species that formerly bred in Minnesota prior to the year 2000.

A brief statement was also added about the species migratory status in the state and its occasional occurrence during the winter months.

Each species also was assigned to a relative abundance category during the Minnesota Breeding Bird Atlas using the total number of records included in the atlas data base. Several metrics of relative abundance could be used such as the total number of blocks where breeding evidence was detected, the total number of priority blocks where breeding evidence was detected, or the total number of detections on the randomly selected points. However, the correlation among all of these metrics was an r > 0.88 which indicated that each would provide the same answer on its relative abundance. Total records also were most useful for birds such as aquatic species, raptors, and nocturnal species that are not easily detected, except with specific observational techniques such as night surveys or searches within habitats that are difficult to access. We recognize that this is, at best, a relative index. For highly colonial species, such as many herons, egrets, gulls, terns, cormorants, and pelicans, the index is not a true reflection of abundance. Each atlas record can often represent a breeding colony that may number in the hundreds or even thousands of breeding individuals.

Categories of abundance were designated as follows (the number of species in each category is in parentheses): very abundant, > 6,000 records (9 species); abundant, 4,000 – 5,999 records (22); common, 1,000 – 3,999 records (71); uncommon, 100 – 999 records (82); rare, 10-99 records (41); and very rare, 1-9 records (24).

North American Breeding Distribution and Relative Abundance

The distribution and relative abundance maps were derived from the federal Breeding Bird Survey (BBS) because they provided knowledge about where the species was primarily found in the United States and southern Canada as well as an indication of how abundant the species is within this region (Sauer et al. 2017). Note that the BBS maps do not include Alaska nor the northern, remote roadless areas of Canada. Although BBS routes have been established in these remote regions since the early 1980s the BBS maps only display those regions of North America (the United States and southern Canada) where the data are sufficient in terms of the density of covered routes. Most general distribution maps found in field guides and other general publications just fill in areas between places where a species occurs, even though suitable habitat may not be available or even if the species is rare within a region. Unfortunately many species are not effectively detected by the BBS such as waterfowl and other waterbirds, nocturnal species like owls, or very rare species. The primary rationale for using the BBS maps, besides their public availability, was to illustrate a species relative abundance during the breeding season in the United States and southern Canada which is revealing for many species compared with standard distribution maps. Relative abundance is depicted as the average number of  birds detected per BBS route each  year. There are many widely available species distribution maps that can be found in field guides or on many websites such as the excellent All About Birds website by the Cornell Lab of Ornithology (https://www.allaboutbirds.org/).

Conservation Concern

For each species we summarized their conservation status by using rankings assigned by a variety of conservation organizations and resource agencies. A universal, global ranking for all waterfowl, waterbirds, shorebirds, and landbirds was recently developed by Partners in Flight (PIF) (2017). Data for many of the landbirds are also presented in the PIF 2016 Landbird Conservation Plan (Rosenberg et al. 2016). Although the vast majority of scores for landbirds remained unchanged in 2017 readers familiar with the 2016 Landbird Conservation Plan will notice that the 2017 scores changed for three Minnesota species: Chimney Swift (raised from 12 to 13); Nelson’s Sparrow (lowered from 14 to 12); and Great-tailed Grackle (lowered from 5 to 4). These scores are based on a combination of the species’ population size, breeding and non-breeding distribution, threats to its breeding and non-breeding range, and its overall population trend. They give an overall perspective on the North American conservation status of the species. The lowest score possible for a species is 4 (the lowest Continental Concern) and the highest score possible is 20 (the highest Continental Concern). These scores will change over time as more knowledge is gained on the population, distribution, threats, and trends for each species.

In each species account, the species Partners in Flight Continental Concern Score is presented in the context of a histogram that illustrates the distribution of scores for all the breeding species in Minnesota. The scores for Minnesota species ranged from a low of 4 (Great-tailed Grackle) to a high of 18 (Piping Plover). The delineation of “Low,” “Moderate,” and “High” scores is taken from the 2016 Landbird Conservation Plan (Rosenberg et al. 2016). The histogram includes the scores for exotic species like the Rock Pigeon and European Starling which are presented in PIF (2017), but scores for exotic species are not included in their species accounts which is consistent with the 2016 Landbird Conservation Plan (Rosenberg et al. 2016). Readers will also note that the histogram ranges from 4 to 18 but excludes the number 17; no species in Minnesota had a Continental Concern Score of 17.

We have chosen to highlight the PIF Continental Concern Scores because they have recently been developed for all birds and they reflect their conservation concern at the global level for those species that occur beyond North America. In addition to the PIF scores, for most of the waterfowl, waterbirds, and shorebirds, conservation rankings assigned by North American conservation plans developed between 2002 and 2016 are presented. Readers will note that some of these North American rankings differ from the global PIF scores. This occurs for a variety of reasons. Some may reflect the different geographical scope of the assessment, some may reflect more recent information on population trends, and some may reflect other conservation issues. The Mallard, for example, has been designated a High Continental Priority by the North American Waterfowl Management Plan because it is the most important harvested duck in North America but it has been assigned a low Continental Concern Score of 7/20 by PIF because it has such a broad distribution and large population.

We also note if a species is federally listed and/or state listed and if it has been designated a Species in Greatest Conservation Need in Minnesota (Minnesota Department of Natural Resources 2017b) and a Targeted Conservation Species by Audubon Minnesota.

Life History

For each species we provide a brief summary of its migration strategy. We identify whether it is a permanent resident, short-distance migrant (winters in the United States), medium-distance migrant (winters in Mexico, Central America, and the Caribbean), or long-distance migrant (winters in South America), and where it can be found in winter.  We also briefly summarize its food, often including how it is obtained, and nest, including nest type and where it is located.

Minnesota Breeding Distribution

To describe the breeding status of species in Minnesota, a historical path was followed which fortunately was available for Minnesota birds. For instance, a general chronological sequence included 1) P.H. Hatch (1892) in his First Report on the Notes on the Birds of Minnesota, 2) the monumental tome of T.S. Roberts Birds of Minnesota, Volumes 1 and 2 (1932, 1936), 3) Green and Janssen’s 1975 book, Minnesota Birds: Where, When, and How Many, 4) Janssen’s 1987 book on Birds of Minnesota, and 5) Hertzel and Janssen’s 1998 publication, County Nesting Records of Minnesota Birds. Note that Hatch was not always used because his observations were often very limited in detail and coverage in Minnesota. Important distributional records and compilations published in The Loon, the quarterly publication of the Minnesota Ornithologists’ Union, also were frequently cited.

These materials were generally followed with a summary of the extensive coverage of the Minnesota Biological Survey (Minnesota Department of Natural Resources 2017c) and their bird inventories of Minnesota counties. Although the focus of the Survey’s bird work was on rare birds, data were collected incidentally on all birds when point counts were conducted. Survey work was conducted from 1987 to 2014 in 83 of Minnesota’s 87 counties. The results of surveys conducted in four northern counties (Beltrami, Koochiching, Lake of the Woods, and northern St. Louis) were not available at the time the MNBBA species accounts were prepared. Finally, we present the results of the first Minnesota Breeding Bird Atlas for the data gathered from 2009 to 2013.

Each of these documents summarized in the MNBBA species accounts represents a summary on the status of breeding evidence for the primary species that were known to nest at the time of their publication in Minnesota. Each varies considerably in the level of detail presented, especially on what constitutes evidence of nesting. Many of the earlier accounts by Hatch and by Roberts could be vague on the level of detail they provided and what constituted confirmed nesting. For instance, Roberts (1932, 1936) often used the terms Cass Lake, Leech Lake, Itasca Park, and Mille Lacs. These areas are generally well-known to most Minnesotans, but they may not be specific to a county because many of these locations encompass more than one county. When the county was not clear from the document, we simply referred to the place name such as Mille Lacs. Minneapolis and St. Paul were also used as place names by Roberts, but we interpreted these as Hennepin County and Ramsey County, respectively.

Brief details on what Hatch or Roberts stated in their summary accounts regarding nesting activity are often reported for each species. Green and Janssen (1975) summarized a species’ breeding evidence as either positive (e.g., nest with eggs, or adult sitting on nest constantly, or eggshells near nest; young in nest seen or heard; downy young, or young still unable to fly, seen away from nest) or inferred nesting (nest-building or excavation of nest-hole in progress; adults seen in distraction display or feigning injury; used nest found; recently fledged young seen; adult seen carrying fecal sac, adult seen carrying food for young; adult seen entering or leaving nest site in circumstances indicating occupied nest). Two additional symbols were used to indicate former positive nesting or former inferred nesting. They incorporated breeding evidence from Roberts (1932, 1936) in their breeding evidence maps.

Janssen (1987) summarized breeding evidence since 1970. He only included confirmed nesting records for a county using the same positive definition used by Green and Janssen (1975). Hertzel and Janssen (1998) also summarized breeding evidence since 1970 and they defined a confirmed nesting record as a nest with eggs, an adult on a nest, identifiable eggshells near a nest, downy young in a nest or nearby, or flightless young. The Minnesota Biological Survey identified locations where species were recorded, but did not have a focused effort on finding nesting evidence except through incidental observations.

Breeding Evidence Codes. The MNBBA used definitions that are standard among breeding bird atlases completed in the United States and throughout many parts of the world. For instance, the definitions used by MNBBA are consistent with the atlases completed in the states and provinces surrounding Minnesota including Iowa, Manitoba, Michigan, Ontario, South Dakota, and Wisconsin. These definitions have some overlap with those used by Green and Janssen (1975), Janssen (1987), and Hertzel and Janssen (1998), but also many differences and additional categories of breeding evidence (Table 1). For example, confirmed nesting with the MNBBA included all the definitions used by Green and Janssen, Janssen, and Hertzel and Janssen, but included confirmed nesting for several behaviors that Green and Janssen (1975) classified as inferred nesting, such as nest building seen at the actual site (code NB), bird seen carrying nesting material such as sticks, grass, mud, cobwebs, etc. (code CN), and an adult seen carrying a fecal sac (FS). MNBBA also had codes for probable, possible, or simply observed within safe breeding dates. These differences in definitions require caution when making comparisons with previous confirmed breeding evidence in Minnesota. Because there are differences in definition of confirmed nesting, we added a footnote to the species account to caution the reader in making direct comparisons. For example, although the MNBBA often documented nesting in more counties than earlier accounts, the MNBBA definitions for confirmed breeding were less restrictive.

As described above, we primarily used the word “records” to describe entries by the MNBBA participants for each species. A record refers to an entry into the MNBBA data base for a specific priority block or block (see below). A record may represent a single detection of a species within a block or it may include a number of detections of a species within a block. Detections refer to both visual and aural observations of a species. Two of the breeding evidence codes (code M and S) specifically reference multiple detections of a species and are assigned as “Probable” breeding evidence. The multiple detections, however, may only constitute one record in the MNBBA data base. The word “observations” is used sparingly in this document because it is often interpreted as a visual detection only.

MNBBA Mapping. Results are presented in a series of maps and charts that includes an interactive function providing a variety of options to search atlas analyses. Major results of the atlas are summarized in each species account, such as the number of blocks or priority blocks where a species was observed and general statements about the species’ distribution or breeding evidence that was reported. Breeding evidence is summarized to the highest evidence code reported for a given block during the MNBBA. For example, a species may have been reported several times within a particular block (i.e., multiple detections) but only that record with the highest breeding evidence code is shown on the statewide breeding distribution map.

When suitable data were available for a species, a probability-based predictive model was derived using the randomly distributed point counts gathered throughout the state of Minnesota. These models used habitat at a point count, landscape information surrounding the point, and climate data associated with the point to make predictions on the probability of observing a species within its breeding range in Minnesota. Details regarding the collection of data at point counts can be found under Data and Methods – Data Collection; details of the modeling work can be found under Data and Methods – Methods of Analysis.

For the predictive species distribution maps, we used different expressions of the relative abundance of a species based on their detectability and behavior. These expressions fall into three different categories. 1) For species that were relatively common or abundant and have strong territorial behavior, we expressed densities as pairs per 40 ha (100 acres). These are largely based on singing behavior of territorial males and it is commonly assumed there is a 1:1 sex ratio of males to females (Modeling Strategy 1 in Methods of Analysis). 2) Number of individuals observed per 10-minute unlimited distance point count was used to express relative abundance for species that were wide-ranging (e.g., Chimney Swift, swallows), heard from a great distance making estimates of detection distance difficult (e.g., Ring-necked Pheasant, American Bittern, Sandhill Crane, or Pileated Woodpecker), or landbird species that are frequently detected by vocalization but generally not as territorial males (e.g., nuthatches, Ruby-throated Hummingbird, and many woodpeckers) (Modeling Strategy 2 in Methods of Analysis). 3) We used landcover suitability (low to high) for species that were rare or not amenable to quantitative sampling of their populations (e.g., raptors, waterfowl, or many landbirds that were uncommon or rare) (Modeling Strategy 3 in Methods of Analysis).

Both the breeding evidence and predictive species distribution maps could be useful for conservation planning, to identify places where a particular species may be observed, to identify places where nest sites may be found in future atlases, or to gain a better understanding of a species’ Minnesota distribution and abundance.

Surveyed blocks. MNBBA blocks were derived from Public Land Survey System (PLSS) townships. We considered blocks surveyed if they had at least 10 records during the 5-year period of the Minnesota Breeding Bird Atlas. A record is an entry of a breeding evidence category for a bird species (Confirmed, Probable, Possible, or Observation) for a block. There were a possible total of 9,774 atlas blocks and 2,352 of these were priority blocks (Table 2). Townships (6 x 6 miles) were divided into quarters to create 3 x 3 mile blocks. We randomly selected the northeast block of a township as the priority block. Theoretically, approximately one-quarter of all the blocks should be priority blocks (2,443) but many townships on the state boundary are only partial in size. As noted in the Data Collection section on Minnesota’s BBA grid system, in cases where the NE block was smaller than 8 square miles, we selected the SE block or continued clockwise to select a complete block. If the township had no blocks that met the size criteria, the township was excluded from the MNBBA. In the end, a total of 4,733 blocks (48% of possible atlas blocks) and 2,337 priority blocks (99% of priority blocks) had at least 10 records during the atlas. We used these “surveyed” block numbers to calculate the summary statistics in the pie charts and tables in the Minnesota Breeding Bird Distribution tab of each species (note that all species have a summary table, but only species with at least 10 records have pie charts). However, we included additional blocks when a species was reported from one or more blocks with less than 10 records. For example, Alder Flycatcher was reported from 11 blocks with less than 10 records. So to calculate the proportion of blocks where the species was probable, we divide 827 (the number blocks were it was probable) by 4,744 (i.e., 4,733 + 11). This is also the reason for some blocks in the block maps being coded as “not surveyed” for most species but “observed” or greater for a handful of other species. Because of this, an astute reader may notice the denominator change from species to species when calculating the proportion of blocks where a species was detected.

Table 1. Breeding evidence codes used for the Minnesota Breeding Bird Atlas project from 2009 to 2013 (Safe Dates for each species and Evidence Reporting Tips referenced below may be found under Data and Methods: Data Collection: Atlas Materials).

Category	Code	Evidence
Observed	O	Species observed in a block within safe dates, but no evidence of breeding. Use this code for species not in suitable nesting habitat. For example, this code applies to raptors or gulls flying over, herons and other colonial nesters foraging with no observed rookery, ducks summering on a pond, or pelicans loafing. Check Evidence Reporting Tips for other examples if you have questions.
Possible	X	Species encountered in suitable nesting habitat within safe dates such as a Virginia Rail in a marsh. Use this code when you hear a male singing in a likely nesting area on only one occasion. For raptors, ‘X’ is not a reliable code due to extensive home ranges; use ‘O’. Also, use caution with northern irruptive owls (Northern Hawk Owl, Great Gray Owl, Boreal Owl). Check Evidence Reporting Tips if you have questions.
Probable	M	Multiple (5 or more) singing or territorial birds of a species detected in suitable habitat within the block on the same day. Most species listed as (M) can be upgraded during a later visit. Check Evidence Reporting Tips for species specific examples.
Probable	S	Singing male in suitable habitat at same location at least 5 or more days apart. Be aware that some species do not sing to demonstrate territoriality or to attract a mate. Check Evidence Reporting Tips for species specific examples.
Probable	P	Pair observed in suitable nesting habitat. Geese, swans, ducks: ‘P’ not reliable as breeding evidence. Code as ‘X’. For more information, check the Evidence Reporting Tips.
Probable	T	Permanent territory presumed through defense of breeding territory by fighting or chasing individuals of same species. Geese, swans, ducks: ‘T’ not reliable as breeding evidence. Code as ‘X’. For more information, check the Evidence Reporting Tips.
Probable	C	Courtship or copulation behavior, includes displays or transfer of food. Geese, swan, ducks: ‘C’ not reliable as breeding evidence. Code as ‘X’. For more information, check the Evidence Reporting Tips.
Probable	N	Bird observed visiting the same likely nest site repeatedly, but insufficient behavior for upgrading to Confirmed. Especially useful for cavity nesters or for a shrub-nesting species that flies into the same thicket and disappears several times.
Probable	A	Agitated behavior or anxiety calls from adults usually indicate a nest site or young in the vicinity.
Confirmed	PE	Physiological evidence of breeding based on bird in the hand such as a highly vascularized swollen incubation (brood) patch or an egg in the oviduct. For birdbanders only using expert judgment.
Confirmed	CN	Bird seen carrying nesting material such as sticks, grass, mud, cobwebs, etc.
Confirmed	NB	Nest building seen at the actual nest site, including wrens and woodpeckers
Confirmed	DD	Distraction displays, defense of unknown nest or young, or injury feigning. Killdeers may give a “broken wing” act, a Red-winged Blackbird may dive at you near the nest site, or an Ovenbird may run about with wings fluttering.
Confirmed	UN	Used nest or eggshells found. Unless carefully identified, use this only for unmistakable eggshells and nests that were used during the Atlas period.
Confirmed	ON	Occupied nest indicated by adult entering or leaving nest site in circumstances indicating an occupied nest, including those in high trees, cliffs, cavities, and chimneys where the contents of the nest and incubating or brooding adult cannot be seen.
Confirmed	FY	Adult bird carrying food for young or feeding long-tail young. This code includes young -of-the-year birds with adults (e.g. duck broods not full grown). Be aware that some adults carry food a long distance or may be engaged in courtship feeding.
Confirmed	FL	Recently fledged short-tail young or downy young. This includes dependent young only. Be cautious of species that range widely soon after fledging.
Confirmed	FS	Adult bird seen carrying fecal sac. Many passerine adults keep their nests clean by carrying membranous, white fecal sacs away from the nest.
Confirmed	NE	Nest with eggs.
Confirmed	NY	Nest with young seen or heard.

 

Table 2. Statewide summary of blocks, effort, and number of bird species reported during the MNBBA.

	All blocks	Priority blocks
Total blocks	9,774	2,352
Blocks with at least 10 records	4,733	2,337
Species detected	249	243
Species confirmed	231	218

In our summary of the MNBBA data for a species we primarily use counties or common place names in Minnesota (e.g., North Shore of Lake Superior, Twin Cities metropolitan area) to describe distributions of the species. For convenience, we include the Minnesota County Map just below the MNBBA block map for quick reference. Many atlas blocks include two counties; a small number even include 3 counties. As a result, when the atlas data is summarized by county both counties that were covered by the block are included in the summaries. We used a rule that if 5% of the block was in a county, then the county was included for that respective breeding code. Species accounts often note which counties are included merely because blocks straddled county lines. This was particularly true for many riparian species that occurred along the Minnesota River Valley where a large number of blocks along the river’s floodplain encompass two counties. In addition, we often searched maps or satellite imagery to identify suitable habitats within a block for a species, but often suitable habitats (e.g., large wetlands) crossed county lines.

We also use the Ecological Classification System (ECS) that is commonly used by the Minnesota Department of Natural Resources (2017d). The ECS groups regions of the state into broader categories that can be more useful for management because the groupings have similar climate, geology, topography, soils, hydrology, and vegetation. In our species accounts we have confined our use of the ECS to three levels: provinces, sections, and subsections. A description of the ECS system can be found at http://www.dnr.state.mn.us/ecs/index.html. One of many options on the Interactive Map is to display the MNBBA results by ECS units.

For most species, we summarized major historical changes that have occurred in their breeding distribution in Minnesota, in the region, or in North America, plus potential reasons for these changes. The basis for this evidence was primarily derived from the historical accounts, the published literature, breeding bird atlas data available for Iowa, Manitoba, Michigan, Ontario, and Wisconsin, and especially from the Birds of North America series published by the Cornell Laboratory of Ornithology in conjunction with the American Ornithological Society (https://birdsna.org).

Breeding Habitat

Breeding habitats were broadly described and, when possible, specific descriptions from Minnesota or the surrounding region were used especially when sample sizes were large and completed in a probability based design such as using random sampling. For most species, we provide a photograph of “typical breeding habitat in Minnesota.” We only use the common name of tree or shrub species used in the text, but scientific names are provided in Appendix 1. We provide a perspective on whether the species has been reported to be sensitive to landscape issues such as habitat fragmentation of forest, grassland/prairie, or wetland habitats. For bird species that were detected with a reasonable sample size on the MNBBA point counts, a habitat profile of the types of habitat where the species was detected was included. The point counts were documented with specific geo-positioning system (gps) devices so the detection of a species could be related with overlays of their habitat from remote sensing imagery.

For these habitat profiles we used 13 different habitat types which represented the majority of major habitats found in Minnesota (Appendix 2). Many habitat classifications that exist in Minnesota, however, have 40 or more different categories. The species habitat affinity to a habitat type uses a PPI index (percent perfect indication) that expresses how strongly a species is related with a habitat type. PPI values are calculated using both how abundant a species is in the habitat category and its frequency of occurrence in the habitat. PPI values range as high as 1.0 if a species is always found in the same habitat type (Dufrene and Legendre 1997). See the Data and Methods – Methods of Analysis for more details on the calculation of PPI values.

When covering such a massive area as an entire state, documenting the habitat for rare species or rare habitats can become problematic and result in spurious or inappropriate habitat representations. We reviewed many iterations of habitat classifications with more and fewer habitat types. We concluded that the 13 habitat types used here best represented the most bird species across the entire state. Many bird species, such as those found in wetlands, lowland deciduous forests, or in upland open areas (e.g., grasslands, croplands, or shrublands) were broadly classified into a general habitat type. In these situations, we urge the reader to gain a better understanding of the species breeding habitat from the narrative statements in the text.

Population Abundance

In recent years many sophisticated models have been developed nationally to estimate the breeding populations of a majority of species that occur in North America. Several programs such as Partners in Flight (PIF) (PIF Science Committee 2013, Rosenberg et al. 2016, PIF 2017), the North American Bird Conservation Initiative (2016), Environment Canada (2011), and Environment and Climate Change Canada (Environment Canada 2014), have estimated breeding populations using the federal Breeding Bird Survey (BBS) as a major source of information for these models. We caution, however, that the estimates from Environment Canada (2011) and Environment and Climate Change Canada (Environment Canada 2014) are very conservative with broad population range estimates. Also note that we have often cited PIF (2017) in our Minnesota state population estimates, but the state estimates were originally derived by the PIF Science Committee (2013).

The BBS is a North American network of randomly selected roadside routes that are annually censused by volunteers. Routes are on secondary roads and consist of 50 stops each 0.8 km (0.5 miles) apart where all birds seen or heard in a 0.4 km radius around each point are counted over a 3-minute period (Sauer et al. 2017). The national program was initiated in 1966, but not started in Minnesota until 1967. Thousands of routes are completed each year. BBS was primarily developed to estimate breeding bird population trends in the United States and Canada. We only reported BBS population trends for those species that had sufficient sample sizes over the long term with moderate precision and moderate abundance on routes; these trends were those identified with the color blue at the BBS website (Sauer et al. 2017). The color yellow identified BBS population trends that were considered deficient primarily because the species is uncommon and/or the sample size is small.  When these trend data are used, we state so in the account and use it for a specific purpose such as when the trend was particularly convincing for Minnesota, a region, or nationally. Besides their use in estimating population trends, the BBS routes have been used to estimate breeding populations in specific U.S. states or Canadian provinces if a species occurs within that jurisdiction. These estimates are based on the proportion of a species range found within the region of interest (PIF Science Committee 2013).

Population estimates also were developed using the MNBBA data (see details in Data and Methods – Methods of Analysis). These statewide estimates frequently varied considerably from those presented by the PIF Science Committee (2013). Among the reasons for variability are 1) the MNBBA estimates were based on data gathered in Minnesota and not from the proportion of potentially suitable habitat available in the state, 2) the methods for population estimates incorporated both detectability and habitat data in the models, 3) MNBBA sampled point counts in many of the state’s roadless areas such as in the Boundary Waters Canoe Area Wilderness and in the Red Lake Peatland of northern Minnesota, and 4) MNBBA also used the extensive data gathered during the National Forest Bird Monitoring Program in the Chippewa and Superior National Forests (Niemi et al. 2016). These latter data also were used to provide mean density estimates within forested areas of Minnesota, especially in north-central Minnesota where the Chippewa National Forest is located and in northeastern Minnesota where the Superior National Forest is located.

The primary purpose of the BBS was to estimate population trends in North America (e.g., survey-wide which includes all routes completed in the United States and Canada) or for more specific areas such as Bird Conservation Regions (Figure 1), U.S. States, Canadian provinces, or for the entire United States or Canada (Sauer et al. 2017). The BBS also is a wealth of other information such as for the relative abundance maps in the United States and Canada described above, for trend maps depicting where a species is stable, increasing, or decreasing, and for specific estimates of abundance for specific areas (e.g., average species detections per route per year in Minnesota). The BBS trend figures include the 2.5% and 97.5% credible intervals for the trend estimates because the trends are based on a a Bayesian hierarchical model (see Sauer et al. 2017).  The BBS data set has been used extensively in the species accounts to present data on population estimates, population trends, and relative abundance of the species.

Figure 1. Bird Conservation Regions (BCRs) of North America (North American Bird Conservation Initiative 2017). Portions of BCRs 11 (Prairie Potholes), 12 (Boreal Hardwood Transition), 22 (Eastern Tallgrass Prairie), and 23 (Prairie Hardwood Transition) are found in Minnesota and are frequently discussed in the Minnesota Breeding Bird Atlas species accounts.

We also included density estimates within specific habitats for Minnesota, the surrounding region, the published literature, or from the Birds of North America series if they were available. One of the major problems presenting these data was the wide variety of density units that have been presented in the literature. In the text for individual species accounts, based on our own work in Minnesota, we specified whether the density estimates were for individuals, breeding pairs, or territorial males per 40 ha (100 acres). In some cases, we converted values from the literature to this commonly used expression of density or reported the densities directly from the literature source cited.

Conservation

Our final section was to provide an overall context on the status of the species in Minnesota and nationally. We cited the Continental Concern Score that was developed by Partners in Flight (2017) to provide this overall perspective on the species status. We also identify other conservation designations of the species, especially those in Minnesota. We summarize some of the major threats that are known to affect the species such as habitat change and loss, climate change, risk of migration such as collisions with towers and buildings, or other factors that appear to be affecting the species distribution or abundance. Finally, we include conservation actions that have been implemented or potentially would be useful if implemented to reduce threats to the species.