Bird AtlasesA Chance for Birders to Make an Important Contribution
William E. Davis, Jr.*
*Division of Science and Mathematics, College of General Studies,
Boston University, 871 Commonwealth Avenue, Boston, Massachusetts 02215
published in Birding
In the beginning...
Maps showing the geographic distributions of bird species date back to the early part of this
century and have become a standard feature of field guides, state and
provincial bird books, and monographs on birds. These distribution maps are
generally based on published information and nest record compilations.
Grid-based atlases (bound maps usually supplemented with text) are a more
recent invention and have proliferated into a broad spectrum of projects and
publications with an equally broad spectrum of goals.
The pioneering grid-based atlasing projects began in Great Britain in the 1950s with a
botanical atlas that prompted a pilot project on mapping the breeding bird
distribution of England's West Midlands. This in turn led to a broader
project culminating in the publication of the Atlas of Breeding Birds in
Britain and Ireland (Sharrock 1976). The European Ornithological Atlas
Committee was established in 1971, and a flurry of activity on the continent
resulted in the publication of similar atlases in Denmark, France, West
Germany, the Netherlands, and Switzerland by 1980, as well as several
covering more restricted geographic areas. These atlases were based on
5-kilometer (km) or 10-km per-side grid squares that provide small enough
scale to make even fairly small changes in distribution of birds apparent in
the future. These atlases differ from previous bird distribution atlases in
representing a "snap-shot" in time, all-be-it a snap-shot with a usual
exposure time of five years, rather than a compilation of historical
records, and in the use of virtual armies of birders to provide the database
from which the atlases are compiled. These atlases provide a baseline of
known distributions against which future snap-shots can be compared. The
atlasing fever rapidly spread world-wide, and atlases, at a variety of
scales have been published in Africa, Australia, Europe, New Zealand, and
North America (Robbins 1982).
In North America, the first grid-based atlas projects to reach fruition
began when several western states, including Montana and Colorado, compiled
bird distribution atlases, but these relied heavily on existing information,
and their scale was large--the compilers used a grid system in which each
"block" (the basic geographical unit in which data are gathered, presented,
and analyzed) was one degree of latitude by one degree of longitude (a "latilong").
Most recent atlasing projects have used approximately 5-kilometer square
blocks, of which approximately 400 fit a one degree latilong block! Hence
some people argue that their large scale and lack of time constraints make
it difficult for latilong atlases to perform one of the major functions of a
breeding bird atlas--to monitor changes in distribution and abundance
through time. At the latilong scale there would have to be a catastrophic
plummet in a species' numbers and distribution for such a change to be
observed in subsequent atlasing efforts. The rather unique grid-based atlas,
Breeding Birds of North Dakota (Stewart 1975), was the product of an
extended, largely one-man project, in which the grid block was the township
(six miles per side). It was based primarily on field work from 1961-1970,
but historical records from a variety of sources were included. The breeding status of a species is generally ranked as "observed,"
"possible", "probable", or "confirmed", with criteria--often hotly debated
and not uniformly used--for each category. For example, if an individual
bird is seen in a block during the breeding season but not in suitable
breeding habitat, it is listed as observed; if in suitable habitat it is
listed as possible. A pair of birds in suitable breeding habitat would be
listed as probable, and a nest with eggs or young as confirmed.
The "true" atlas projects begin
North American atlas projects based on small-scale grids began in Maryland in the 1970s where a
county-by-county project was started. The first publication of results was
The Breeding Bird Atlas of Montgomery and Howard Counties, Maryland
which appeared in the local bird journal Maryland Birdlife (Klimkiewicz and
Solem 1978). This was a harbinger. With it began a deluge of county,
regional, province, and state publications that continue to pour across the
counters of book dealers.
In 1980 the North American Ornithological Atlas Committee (NORAC) was organized, and in 1981 a
Northeastern Breeding Bird Atlas Conference was held at the Vermont
Institute of Natural Science in Woodstock, Vermont. The outcome was a
published Proceedings volume (Laughlin 1982) loaded with informative papers
including an overview of international atlasing by Chandler S. Robbins, a
major contributor to the initiation of atlasing projects in North America. A
second conference and Proceedings followed in 1986, sponsored by the Cornell
Laboratory of Ornithology in Ithaca, New York (Sutcliffe et al. 1986). This
burst of activity led to the publication of several major papers on atlasing
breeding birds (e.g., Udvardy 1981, Laughlin et al. 1982, Butcher and Smith
1986), and eventually NORAC published Handbook for Atlasing American
Breeding Birds in English and Spanish editions (Smith 1990). It was reported
at the 1986 conference that more than 30 states and provinces had atlas
projects under way. After periods of data collection, which usually involved
approximately five or six (often an "extra" year is added) years, the long,
expensive, and often painful process of compilation and publication of the
results was undertaken. This process has resulted in the production a wide
variety of formats, styles, and supplemental information. Some atlases have
been hardback, some soft cover, some contain species accounts, some are
copiously illustrated, and some have fancy supplemental materials such as
overlay transparencies that facilitate comparing breeding bird distribution
with a variety of habitat and climatic phenomena. A few examples here (Not
included here) of this broad spectrum of publications show the diversity of
format and project design, and illustrate the problems, strengths, and
weaknesses inherent in atlasing projects.
Atlas projects reach fruition
The Atlas of Breeding Birds of Vermont (Laughlin and Kibbe 1985) was the first state or
provincial atlas published in North America. It became a model for many
other atlases and typifies many of the positive aspects and pitfalls common
to atlas projects. The grid scale chosen was approximately 5-km per-side
blocks that were established by subdividing a standard U.S. Geological
Survey (U.S.G.S.) 7.5-minute quadrangle map into six blocks. Because of a
shortage of manpower (a chronic problem in atlasing projects) only one block
(a randomly selected "priority block") per quadrangle was targeted for
intensive coverage, although information for the other blocks was accepted
when available. In addition, 24 blocks containing "unique and fragile
habitat" were also targeted for intensive coverage. Six years of data
collection (1976-1981) were used to construct a map for each species,
showing by the size of dots the location of blocks where the species was a
possible, probable, or confirmed breeder.
A species account for each confirmed breeding species was printed on facing pages with a
pen-and-ink drawing of the bird and the distribution map. Possible and
probable breeding species had smaller maps and shorter natural history
accounts. Also included were eight plastic overlay maps that facilitate
comparison of a species' breeding distribution with physiographic regions,
county boundaries, elevations, agricultural lands, precipitation, and July
temperatures. This excellent hardback book also illustrated some of the
major problems inherent in most atlasing projects. Perhaps the most serious
problem is a lack of information on species abundance. A block with one
breeding pair cannot be distinguished from a block with 1000 pairs. The
usual biases involving detection of rare, secretive, or nocturnal birds and
early nesting species, uneven coverage of priority blocks, and differences
in observer experience and ability are discussed and cautionary
pronouncements included. All things considered, this is an excellent and
The New York state atlasing project, and its publication The Atlas of Breeding Birds in New
York State (Andrle and Carroll 1988), was a daunting project. More than
4300 volunteers contributed more than 200,000 hours in the field. There was
at least some field coverage for nearly all of the 5335 25-km2
blocks! About a quarter of the blocks were visited by "block busting" teams
consisting of two people who spent a day (and night) in each block "busted".
The grid differed from most state projects in that a metric-based system was
employed rather than subdivisions of U.S Geological Survey maps. This
system, the New York Transverse Mercator Grid, divided the state into 5-km
per-side squares. Species accounts together with a drawing of the bird
occupy a single page of text (often less) and present historical information
about distribution as well as breeding biology information. The distribution
map is on the facing page. The maps are large and each block where the
species was observed is pictured with a solid black square if breeding was
confirmed, or other symbols for possible or probable nesting. Maps depicting
vegetation, elevation, etc. are included, and plastic overlays of most of
these features are available at extra cost. The latter are helpful because
the species maps are, of necessity, virtually featureless. The drawings
occupy nearly a quarter of the text page but are generally excellent in
quality. Appendices describe the ecological zones of the state (e.g.,
Appalachian Plateau), natural and cultural (e.g., mosquito ditch) ecological
communities, and tables of breeding-season data (e.g., egg dates) for each
species. The book is wider than it is tall (landscape format), presumably an
adaptation for the shape of New York state. Like most of the published
atlases there is an attempt to provide a great deal more information about
the birds than just what is depicted in the distribution maps.
The Atlas of Breeding Birds of Michigan (Brewer et al. 1991) is a large (8.5 x 11 inches),
handsome book, that employs the same basic design for species accounts as
the New York atlas. The species distribution maps are large and easy to
read, with large blue squares representing confirmed breeding, circles
probable breeding, and tiny squares possible breeding. The grid base is,
however, the 36-square-mile township, each of which is subdivided into four
9-square-mile blocks. All the blocks in the lower half of Michigan's Lower
Peninsula received coverage, but only two blocks out of four received
coverage in the northern half, and one block per township in the Upper
Peninsula. This "stratified sampling" effort resulted from a number of
considerations including uniformity of habitat and concentration of
volunteers, and resulted in a total of 4301 "priority blocks" targeted for
intensive coverage. To the usual biases was added the inclusion of gamebird
and raptor data from other studies. The book includes a chapter on land,
climate, and vegetation, supplemented with maps including fascinating ones
depicting presettlement vegetation and current vegetation, a chapter on the
pre-European settlement avifauna, an analysis of post-settlement changes,
and chapters on biogeography and ecology and conservation perspectives. The
distribution maps are supplemented when feasible with small abundance maps
from the National Biological Service's Breeding Bird Survey data, which
attempt to address the census data deficiency common to most atlas projects.
As with most state and provincial atlases, the references section is a
veritable gold mine of information about local birds, much of it tucked away
in local bird journals. This book is much, much more than a compendium of
Atlas of Breeding Birds in Pennsylvania (Brauning 1992) is another attractive and informative atlas. This project,
as have most in the United States, used the U.S.G.S. topographic maps of 7.5
minutes of latitude and longitude subdivided into six blocks. The coverage
was interesting, however, in that the southeastern block in each quadrangle
was designated as a priority block, but coverage was targeted for all
blocks. This priority block system provided initial intense coverage for one
block that could subsequently act as a measure of quality control for the
other five blocks during the seven years of field work. In this very
ambitious project all 4928 blocks received some coverage and produced nearly
32,000 records. The book contains a brief history of atlasing projects and a
history of Pennsylvania ornithology, along with the obligatory methods
section, and is augmented by eight plastic overlays.
The Atlas of Breeding Birds in Rhode Island (1982-1987) (Enser 1992) and Atlas of Breeding
Birds in Maine, 1978-1983 (Adamus n.d.) are examples of atlases compiled
and published on a low budget. The Rhode Island atlas presents one species
per page, with a short paragraph of text at most, whereas the Maine atlas
has no species accounts accompanying the distribution maps, but in many
cases provides short annotations and references. The latter has a useful
bibliography of Maine ornithology published since 1949, and it presents all
of the atlas data in tabular form, which would be useful for some retrieval
purposes. The Maine atlas differs from those of most eastern states in using
a larger scale with blocks consisting of entire U.S.G.S. 7.5-minute
Remarkably comprehensive atlases, full of ecological information, have been produced for several
counties in California (e.g., Marin County [Shuford 1993]) and more than a
dozen county projects are underway or complete in California alone. Atlas
of the Breeding Birds of Monterey County (Roberson and Tenney 1993) used
a priority block system to accommodate a limited supply of volunteers for a
large area (Monterey County is larger than the state of Rhode Island). One
interesting feature of this atlas is the inclusion of abundance estimates
based on an exponential scale of 1 pair only, 2-10, 11-100, or 101-1000
pairs. These data provide very rough population estimates for the county.
How useful these estimates are is the subject of some debate, but it
represents an attempt to correct one of the main deficiencies in most
atlasing projects. For many species two maps are presented, one showing
recorded breeding categories of confirmed, probable, and possible, and the
other showing abundance estimates. Population estimates for many species are
included in the text of the species accounts. This is a beautifully
illustrated and comprehensive book.
Another county atlas of note is the Atlas of the Breeding Birds of Los Alamos County, New Mexico
(Travis 1992). Los Alamos County is small (111 square miles), and each of
the sixty 2,5-km- per-side blocks received coverage during the five years of
field work. Interesting features of this atlas include tables of bird
species/habitat associations (e.g., 21 species are associated with pinyon/juniper
woodlands), and numerous other ecological analyses.
The Canadians have done a remarkable job in producing high- quality atlases under the difficult
circumstances of vast regions with sparse population and difficult or
limited access. The Ontario atlas project used 10-kilometer-per-side blocks
for the more populated southern section but 100-km-per-side blocks for the
sparsely settled northern regions of the province. The final product,
Atlas of the Breeding Birds of Ontario (Cadman et al. 1987) was the
first of the Canadian projects to be published. The atlas includes
histograms of abundance estimates along with the usual breeding distribution
maps. Unfortunately the confirmed, probable, and possible categories were
represented on the maps with squares of three intensities of the same color,
which make them hard to distinguish.
In Alberta, where there are more than 6000 10-km-per-side squares, one of every block of
four squares was designated a priority block in southern portions of the
province and one out of 100 for remote northern areas. Atlasers were
encouraged to organize "square bashing parties" in order to achieve the
desired coverage. The project received more than 40,000 hours of field work
from nearly a thousand atlasers. The final product, The Atlas of Breeding
Birds of Alberta (Semenchuk 1993), is large (9 x 12 inches) and
beautifully illustrated with color photographs of each species and
distribution maps with colored backgrounds showing the natural regions
(e.g., boreal forest, grasslands), making it a truly elegant book. Other
excellent provincial atlases include those for the Maritime Provinces (Erskine
1992) and most recently Quebec (Cyr 1995).
What does the future hold?
Atlasing projects continue to come to fruition in the United States. The atlas for Ohio
appeared in 1991 (Peterjohn and Rice) but is already out of print. This
atlas featured a priority block system supplemented by "special areas",
including unique or localized habitats such as hemlock ravine and wetlands.
In 1994 atlases for West Virginia (Buckelew and Hall), Connecticut (Bevier), and New Hampshire (Foss)
were published. The West Virginia atlas presented one species per page but
included seven plastic overlays of useful features such as percentage of
forest cover, annual precipitation, and topographic features. The
Connecticut atlas was published in landscape format with each species
account occupying two (facing) pages with a map and line drawing for each
species. It also has an interesting eleven page section on interpreting
distributions of breeding birds. The New Hampshire atlas is a large and
handsome book with huge, easily read maps, and extensive text for each
In 1995 the atlas for South Dakota (Peterson) appeared and in 1996 the atlas for Kentucky
(Palmer-Ball). The former features a spiral wire binding with large maps
with information presented in tabular form on habitat element types (e.g.,
natural, manmade), early and late dates, and abundance estimates. Each
species is covered in a single page with text restricted to a few sentences.
The Kentucky atlas presents separate maps for breeding status and abundance,
and includes a photograph of every species. It also features data analysis
by physiographic province and a tabular summary of each species' breeding
Several other publications are expected shortly (e.g., for Colorado, Delaware, Iowa,
Maryland and the District of Columbia, New Jersey, Saskatchewan, Tennessee,
Publication of atlasing results often takes years after the completion of the atlas field work. The
most extreme example of this is the atlas for Massachusetts. This was the
first state-wide project to be implemented, with field work conducted from
1974-1979. A variety of problems and some indecision concerning publishing a
separate atlas or including the maps with the recently published Birds of
Massachusetts (Veit and Peterson 1993) led to a series of delays.
Publication is currently scheduled for 1996 or 1997. Interestingly, from a
strictly scientific point of view it really does not matter when an atlas
gets published, as long as it gets published, because it represents a datum
plane of breeding bird distribution. The data never become obsolete and are
as valid and important 25 years after they were collected as they were at
the time of collection. In addition, the data from atlas projects are often
available long before they are published in atlas form. For example, in the
Maryland atlas project, the data were entered into computers each year and
made available for land management decisions by state and county agencies
(C. Robbins, pers. comm.). The burgeoning technologies of Internet and
cyberspace will surely further facilitate public access to atlas data in the
Atlas projects that quantify breeding bird distribution in North America have
provided, and continue to provide, birders with an increasingly important
role in scientific ornithology and in habitat and avian conservation.
Birders have contributed literally hundreds of thousands of hours to
atlasing projects in the past two decades. The importance of these projects
continues to grow. They provide detailed information on the distribution of
breeding birds for birders wishing to see them. But they accomplish much
more than that. They provide quantitative information for land managers and
conservationists that can be used to make informed land management
decisions, including critical habitat assessment, setting priorities for
land acquisition, and assessing the effects of habitat fragmentation.
Quantitative information on breeding birds can be powerful ammunition in a
conservation organization's arsenal when critical habitat is threatened by
development or funds need to be raised for land acquisition. When you become
involved with an atlasing project you are contributing to a long-term
project the value of which will increase as future snap-shots-in-time make
possible assessment of population trends and a broad spectrum of other
ecological perspectives. In Europe, the first of the second generation
snapshots has already been taken. The New Atlas of Breeding Birds in
Britain and Ireland: 1988-1991 (Gibbons et al. 1993) is the first of the
second generation atlases to be published. For each species it includes maps
depicting abundance, distribution, and status changes that have occurred in
the twenty years since the first generation atlas was published.
Participation in atlasing projects provides an opportunity for those for
whom birding is primarily a recreational pastime to make important
contributions to scientific ornithology and conservation initiatives. And
it's fun! If your state or province has already completed its first round of
atlasing, why not try another state--most atlasing projects would welcome
help, including "block busting." And some states and provinces are nearing
the time when the second round of atlasing is about to begin.
I thank Sarah B. Laughlin, Wayne R. Petersen, Chandler S. Robbins, and Charles R.
Smith for their helpful suggestions on previous drafts of the manuscript.
Adamus, P. R. n.d. Atlas of Breeding Birds in Maine, 1978-1983. Maine Department of Inland
Fisheries and Wildlife, Augusta.
Andrle, R. F., and J. R. Carroll. 1988. The Atlas of Breeding Birds in New York State.
Cornell University Press, Ithaca.
Bevier, L. R. 1994. The Atlas of Breeding Birds of Connecticut. Bulletin 113, State
Geological and Natural History Survey of Connecticut, Hartford.
Brauning, D. W. 1992. Atlas of Breeding Birds in Pennsylvania. University of Pittsburgh
Brewer, R., G. A. McPeek, and R. J. Adams, Jr. 1991. The Atlas of Breeding Birds of
Michigan. Michigan State University Press, East Lansing.
Buckelew, A. R., Jr., and G. A. Hall. 1994. The West Virginia Breeding Bird Atlas.
University of Pittsburgh Press, Pittsburgh.
Butcher, G. S., and C. R. Smith. 1986. Breeding bird atlases add zip to summer birding.
American Birds 40: 419-428.
Cadman, D., P. F. J. Eagles, and F. M. Helleiner. 1987. Atlas of Breeding Birds of Ontario.
University of Waterloo Press, Waterloo.
Cyr, A., and J. Larivee. 1995. Atlas Saisonnier des Oieaux du Quebec.
Presses de l'Universite de Sherbrooke and Societe de Loisir Ornithologique
de l'Estrie, Inc., Sherbrooke.
Enser, R. W. 1992. The Atlas of Breeding Birds in Rhode Island. Rhode Island Department
of Environmental Management.
Erskine, A. J. 1992. Atlas of Breeding Birds of the Maritime Provinces. Nova Scotia Museum,
Foss, C. R. 1994. Atlas of Breeding Birds in New Hampshire. Audubon Society of New
Gibbons, D. W., J. B. Reid, and R. A. Chapman. 1993. The New Atlas of Breeding Birds in
Britain and Ireland: 1988-1991. T & AD Poyser, London.
Klimkiewicz, M. K., and J. K. Solem. 1979. The breeding bird atlas of Montgomery and Howard
Counties, Maryland. Maryland Birdlife 34: 3-39.
Laughlin, S. B. 1982. Proceedings of the Northeastern Breeding Bird Atlas Conference.
Vermont Institute of Natural Science, Woodstock.
Laughlin, S. B., and D. P. Kibbe. 1985. The Atlas of Breeding Birds of Vermont.
University Press of New England, Hanover, for the Vermont Institute of
Laughlin, S. B., D. P. Kibbe, and P. F. J. Eagles. 1982. Atlasing the
distribution of the breeding birds of North America. American Birds 36:
Palmer-Ball, B. L., Jr. 1996. The Kentucky Breeding Bird Atlas. University of Kentucky Press,
Peterjohn, B. G., and D. L. Rice. 1991. The Ohio Breeding Bird Atlas. Ohio Department of
Natural Resources, Columbus.
Peterson, R. A. 1995. The South Dakota Breeding Bird Atlas. South Dakota Ornithologists'
Robbins, C. S. 1982. Overview of international atlasing. In S. B. Laughlin (ed.).
Proceedings of the Northeastern Breeding Bird Atlas Conference. Vermont
Institute of Natural Science, Woodstock.
Roberson, D., and C. Tenney. 1993. Atlas of the Breeding Birds of Monterey County
California. Monterey Peninsula Audubon Society, Monterey.
Semenchuk, G. P. 1992. The Atlas of Breeding Birds of Alberta. Federation of Alberta
Sharrock, J. T. R. 1976. The Atlas of Breeding Birds in Britian and Ireland. T & A. D.
Poyser, Calton, England.
Smith, C. R. 1990. Handbook for Atlasing North American Breeding Birds. Vermont Institute
of Natural Science, Woodstock, for the North American Ornithological Atlas
Shuford, W. D. 1993. The Marin County Breeding Bird Atlas. Bushtit Books, Bolinas,
Stewart, R. E. 1975. Breeding Birds of North Dakota. Tri-College Center for Environmental
Sutcliffe, S. M., R. E. Bonney, and J. D. Lowe. 1986. Proceedings of the Northeastern
Breeding Bird Atlas Conference. Cornell Laboratory of Ornithology, Ithaca.
Travis, J. R. 1992. Atlas of Breeding Birds of Los Alamos County, New Mexico. Los Alamos
National Laboratory, Los Alamos.
Udvardy, M. D. F. 1981. An overview of grid-based atlas works in ornithology. In: Studies
in Avian Biology No. 6:103-109.
Veit, R., and W. Petersen. 1993. Birds of Massachusetts. Massachusetts Audubon Society,