Uses of Atlas Data
Atlas data are primarily collected for the purpose of assessing the geographical distribution of species in an area. The wide movements of many bird species make the location of their relatively sedentary breeding stations quite important. Therefore, bird atlases usually concentrate on breeding. However, other atlases have been developed for many other groups. In Britain, over 30 atlas projects have been completed or underway, dealing with life forms as diverse as dragonflies, mammals, and marine dinoflagellates. Therefore, in the discussion of the uses of atlas data it is important to recognize that it is possible to undertake analyses with information for a wide variety of life forms.
C.S. Robbins (pers. comm.) suggests that atlas data can be used for three general types of activities: conservation, monitoring and research. This paper will discuss these three activities while exploring six different types of uses.
Determination of Geographical Distribution
Atlas data are valuable for the original determination of the geographical distribution of a species. This fine degree of resolution, the wide geographical distribution, and the large amount of field time result in the production of a picture of the range of a species for the province or state. In Ontario, the breeding bird atlas work resulted in the addition of three new confirmed breeding species and ten possible or probable breeding species to the provincial list (Cadman, Eagles and Helleiner 1987).
The known ranges may change by hundreds of kilometers as the data reveals previously unknown populations. For example, in Ontario it was found that Lesser Yellowlegs breed over 200 kilometers further south than the range shown in Godfrey (1986). New patterns of distribution are often revealed. In Ontario, it was found that there are two distinct bands of distribution across southern Ontario for Cerulean Warbler, a fact previously unknown (Cadman, Eagles and Helleiner, 1987).
Atlas data can provide relevant and important information on the range on every species that is studied. The broad geographical coverage of atlas data allows for correlation with other ecological data bases, such as those of soil surveys, forest inventories, etc. However, it is important to recognize that atlas data can be used for much more.
Determination of Distribution Changes over Times
Atlas data for birds are usually collected within a finite period of time. These data are then coalesced into a group and published as if from one point in time. However, it is possible to break down the data to specific periods of time. For example, Dunn (1987) has shown the significant changes in the distribution of House Finches in Ontario over the 1981 to 1985 data collection period of the bird atlas. Lumsden and Smith (1987) discuss the importance of having year by year data with the interpretation of the distribution of the irruptive crossbill species. If the data are stored electronically with files for each year, it is possible to undertake analyses of these types.
In Ontario, six species of birds that had bred in Ontario were not found during the atlas period. This almost certainly means that they are now no longer present as native breeders in this area.
In some cases atlas data have been collected continually. For example, the Provisional Atlas of the Mammals of the British Isles (Arnold, 1978) includes historical as well as more current information. The maps for most species have three types of records, those before 1900, those between 1900 and 1959, and those between 1969 and 1976.
Historical information is prone to bias due to the under collection of data in some areas. However, if used judiciously, it can still be used to show changes over time. This is most certainly the case for those species that were once found in an area but were found more sparingly during more recent data collections.
C.S. Robbins (pers. comm.) has suggested that there are four important conservation questions that can be partially answered by atlas data:
Determination of Rarity
Atlas data show the geographical distribution of a species. This information can be used to give a rough idea of the relative rarity of an organism. Certainly, if a species is found in a majority of the squares in a province it cannot be considered rare. However, it is often necessary to have some abundance data for the species in a square to provide a more complete picture.
It is possible to have a widely and well distributed species considered to be rare. Loggerhead Shrike in Ontario were found in 145 squares, which is a reasonably large number, but the number of individuals in each square was quite low. It is possible to have a locally distributed species considered to be common. This can be the case where there are large numbers of individuals in the locations where the species occurs. Some colonial waterbirds could be examples of this phenomenon. The determination of rarity is therefore best done by the combination of data on distribution and on abundance.
Atlas data also can be used at local scales, such as for part of a province, or counties. This can be quite important for the understanding of local populations of breeding birds. It is often found that a species which is common across the entire province is locally rare and therefore deserving of special management in a specific area. If the atlas data are properly stored the selection of a variety of scales should be possible (Eagles, 1987).
The Ontario breeding bird atlas data are being used to assist with the assessment of status under the Ontario Endangered Species Act. As atlases elsewhere in the country come into form, the national status of breeding birds can be much more clearly defined than it is now.
Environmental Impact Assessment
In both the U.S. and Canada there are extensive provisions requiring for the proponents of development to undertake a study that outlines the environmental impact of the proposal. Atlas data can be used in two ways in EIA.
In the early stages of a proposal the EIA specialist needs to know the general environmental features of the area of study. Atlas data can provide an excellent first cut of information. Usually the atlas data are at too coarse a scale to provide specific information on the site in question, but they can still provide knowledge of the local species pool. It is very useful to know if specific important species have been recorded in the local area. The design of the field methodology can then be adapted with the information on the local area in mind. The EIA specialist must then undertake detailed field work in the specific area. These data are then compiled and summarized.
One of the important aspects of EIA is the assignment of importance to an environmental feature that is found on a site. After the field work is complete the EIA specialist can ask for the comparison of the data from the specific site to the general area. For example, it is possible to ask of the atlas data base: In how many squares was species “X” found in Halton County? Or In how many squares was species “X” found in the 50 squares surrounding the development? Dance and Fraser (1987) provide a detailed discussion of the use of atlas data in EIA.
Assessment of Landscapes Change
Landscapes change over time and wildlife populations are excellent indicators of such changes. Eagles (1984) provides a nice example of the impact of large scale agricultural change on the humble bumble bee of Great Britain. The British Bee Atlas map for Bombus humilis shows that the species has been extirpated from central England in recent decades. This is due to the removal of hedges and forests for the intensification of agricultural land use. The increased use of herbicides also probably played a role.
In the next few years we should see increasingly sophisticated analyses of the interrelationships between landscapes and wildlife populations. It is now possible to use satellite images of land, databases of soils, land use, forest potential, watersheds and other phenomena to explore the reasons for the distributions of wildlife.
The information found in atlas databases provides an unprecedented opportunity to study various aspects of basic biology, most specifically in the areas of autecological and biogeographical studies.
One of the most interesting applications deals with the analysis of species complexes. How often do certain species occur together? Are there breeding bird communities that replicate over broad geographical areas? How similar are the communities between area A and area B? How often does a particular bird species occur in association with a certain mammal species?
Calabuig (1981) has experimented with developing indices of similarity between squares in an attitudinal and latitudinal sense in central Spain. Kwak and Reyrink (undated) used a computer program, TWINSPAN, to divide the breeding bird populations of the Netherlands into 18 different districts. The program compared the atlas data among different squares and grouped those with similar populations. Taylor and Smith (1987) present an analysis of the bird communities in Ontario, using the TWINSPAN program as well. They show that clearly defined breeding bird communities do occur in Ontario and the distributions of these communities do not necessarily follow the biogeographical province boundaries that have been defined on botanical grounds.
Characteristics of Volunteers
Almost all atlas projects are based on the work of volunteer field workers. Hidden in the database are valuable clues on the behavioral characteristics of the volunteers. What was the average number of hours of participation by a volunteer? How were those hours distributed? Where did the volunteers participate? How far did the average person travel from home? All of these questions are answerable by the average atlas database.
Beyond the basic questions of coverage and participation lie questions of motivation. Why do atlasers volunteer? What do they get out of it? What did they not do to make time for atlasing? What rewards were they seeking? What rewards did they get? Why did some drop out with low levels of activity? All of these questions can only be answered by a follow-up study. Such research would be quite valuable for a more complete understanding of the volunteer aspect of atlasing.
Atlas data can be used in many ways, beyond mapping geographical distributions. It is important to recognize that there are limits to the use of this information. The research question must come first, to be followed by the application of the appropriate methods designed to answer the question. In many of the questions discussed above, atlas data are but one way to approach the problem. There are others and they should be used when appropriate.
It is important for the designers of atlas database systems to be aware that there are a wide variety of research questions that can be tackled by the data. If this fact is recognized early, then the database can be designed with this in mind. It is possible to shut out many applications with a narrowly defined database.
It is to be hoped that in the next few years we will see the continuing publication of research results from novel approaches to the study of atlas data.
Thanks to Dave Balser who read an earlier version of this document.
Arnold, H.R. 1978. Provisional Atlas of the Mammals of the British Isles. Inst. Terr. Ecol., Abbots Ripton, Huntingdon.
Cadman, M.D., P.F.J. Eagles, and F.M. Helleiner. 1987. Atlas of the Breeding Birds of Ontario. Univ. Waterloo Press, Waterloo.
Calabuig, E.L. 1981. Analisis Y Valoracion Del Metodo Del Itinerario con Estaciones de Escucha, Aplicado a la Elaboracion de un Atlas Ornitologico Regional. Pp 35-56 in Censos de Aves en el Mediterraneo (F.J. Purroy, Ed.), Dept. Zoologia, Univ. Leon, Spain.
Dance, K.W. and D.M. Fraser. 1987. Uses of Breeding Bird Atlas Data for Environmental Planning. Pp 569-471 in Atlas of the Breeding Birds of Ontario (M.D. Cadman, P.F.J. Eagles, and F.M. Helleiner, Eds.). Univ. Waterloo Press, Waterloo.
Dunn, E.H. 1987. Using Atlas Data to Monitor Changes in House Finch Distribution. Pp 574-575 in Atlas of the Breeding Birds of Ontario (M.D. Cadman, P.F.J. Eagles, and F.M. Helleiner, Eds.) Univ. Waterloo Press, Waterloo.
Eagles, P.F.J. 1984. The Planning and Management of Environmentally Sensitive Areas. Longman, London.
Eagles, P.F.J. 1987. The Role of Atlas Data in Determining the Relative Size of Ontario Breeding Bird Populations. Pp 566-568 in Atlas of the Breeding Birds of Ontario (M.D. Cadman, P.F.J. Eagles, F.M. Helleiner, Eds.). Univ. Waterloo Press, Waterloo.
Godfrey, W.E. 1986. The Birds of Canada. Natl. Mus. Can., Ottawa.
Kwak, R.G.M. and L.A.F. Reyrink. Undated. National Breeding Bird Districts and Their Relation to Landscape Features. Unpub. Manuscript.
Lumsden, H.G. and R.B.H. Smith. 1987. Crossbills, Cone Crops and Irruptions. Pp 572-573 in Atlas of the Breeding Birds of Ontario (M.D. Cadman, P.F.J. Eagles, F.M. Helleiner, Eds). Univ. Waterloo Press, Waterloo.
Taylor, P.D. and S.M. Smith. 1987. Multi-species Clusters of Birds in Southern Ontario. (M.D. Cadman, P.F.J. Eagles, F.M. Helleiner, Eds.) Univ. Waterloo Press, Waterloo.
The NORAC website is hosted and maintained by Bird Studies Canada