Migratory and Breeding Bird Survey

Of the Big Lake Natural Area, Alberta



Christina Elliott

Valerie Nelson

Miles Constable







January, 2004


Photograph of Tundra Swans at Big Lake courtesy of Dr. R.K. Lane.

All rights reserved.


Executive Summary


The Big Lake Natural Area (BLNA) northwest of Edmonton and just outside of St. Albert is a group of habitats that is being affected by large-scale development and proposals for alternate land use. It is unknown whether this change in habitat utilization has had a substantial effect on the use of the area by waterfowl and upland avian species for either migratory or breeding purposes. Baseline data is necessary to determine species diversity and abundance in the area and to implement a more strategic timeline for monitoring habitat change and creating an integrated conservation plan. The Big Lake Natural Area is a globally recognized Important Bird Area (IBA) and measures are being taken to determine whether the notable congregatory species and nesting colonies of Franklin's Gulls will continue to be present at the lake or whether recent decreases in water levels and development have caused a long-term decline.


Recent drought conditions appear to have contributed to the level of diversity in the Big Lake Natural Area, reducing the surface area of surrounding wetlands forcing many birds to use the BLNA. Similar Shannon-Weaver diversity indices were calculated for many of the habitats in the BLNA, partly owing to their close proximity. Lower than expected values were calculated for what should be productive areas i.e. the white spruce forest, possibly due to habitat fragmentation and changing environmental factors. The previously present colony of Franklin's Gulls on the west side of the lake was no longer productive, despite the presence of the species in the BLNA; however, such colonies may remain viable even after years of absences.




As part of the North American Waterfowl Management Plan (NAWMP) monitoring of waterfowl and upland avian species is deemed a necessary part of conservation (Prescott, 1995).


Big Lake (53o 36' N, 113o 43' W) is a wetland ecosystem located in central Alberta, Canada, bordered by the municipalities of Edmonton, Parkland County, Sturgeon County and St. Albert (please see map, Figure 1). The lake supports extensive stands of emergent vegetation. The depth in the central portions of its two basins is generally less than 4 meters, with minimum depths near the margins of 0.3 meters. The area determined for the designation of Big Lake to the Alberta Special Places program is 21.4 km2.


Big Lake fits into the category of medium freshwater wetland, covering at least 2.14 hectares of land with shorelines and emergent vegetation extending even farther, a high priority for NAWMP conservation in aspen parkland regions (Prescott, 1995). The NAWMP also recognizes a higher diversity of native vegetation than tame or developed land, and that the best method of maintaining biodiversity lies in habitat preservation and census to determine management practices (Prescott, 1995). In Canada, the importance of grass and parklands in the prairie region are noted as having the largest breeding duck populations in the country (Caswell and Dickson, 1997). As well, migratory populations have been known to make up nearly 75% of the present species in Canada owing largely to its high latitudes. This increases the importance of surveys in the spring and fall to monitor populations largely affected by weather alterations (Hussell, 1997).


In addition to the absence of systematic records of populations on Big Lake, there has been a recent influx of land development projects in the immediate surrounding area, making a monitoring project to note both species presence and abundance a necessity. Big Lake has been promoted for its great diversity and abundance of species, including large numbers of nesting Franklin’s Gulls and migrating Tundra Swans, along with a variety of other species year-round. The desire for a management plan for the Big Lake Natural Area to take into account development pressures, habitat sites important to migration, nesting and species richness and the importance of wetlands in maintaining such diversity makes a comprehensive biological survey of the BLNA necessary.


It has been suggested that since the 1960s songbirds alone have declined up to 50% due to habitat fragmentation (Rowell and Rice, 1998), an unfortunate loss that makes species presence important to monitor, especially in an area as biologically diverse as the Big Lake Natural Area. Johnston (1990) emphasizes long term census projects to observe changes in avian populations that correspond to changes in habitat in order to alert organizations concerned with biodiversity and allow for implementation of proper strategy to compensate.


Big Lake as a natural area has been globally recognized as an important Bird Area (IBA), international recognition given to areas important to conservation efforts. The program was developed by Bird Life International and is supported through Bird Studies Canada and the Canadian Nature Federation (Kirk, 2001). This status was originally awarded to the BLNA for its congregatory species and waterfowl concentration on the lake and surrounding land used for agriculture, hunting, pasture, recreation and industrial purposes (IBA, 2003).




Surveying of all habitat types in the BLNA was conducted during spring migration and summer breeding, systematically from May 9, 2003 until July 10, 2003 with opportunistic surveys occurring during early migration in late April and expected to continue through September. Habitats were surveyed from one-half hour before sunrise until 11:00h on days without significant precipitation, and an attempt was made to repeat surveys conducted on days with winds estimated to be above 2 on the Beaufort scale (Bilyk et al., 1998). Spring migration surveys occurred from May 9 through May 31, 2003, and summer breeding surveys from June 2 to July 10. 2003, similar intervals to those performed in previously conducted studies (Bilyk et al., 1998).

Study area


The Big Lake Natural Area is located northwest of Edmonton, Alberta and west of the city of St. Albert in the Aspen Parkland ecoregion. The area of interest to this study encompasses both the main wetland itself (21.4 km2) and surrounding areas of small deciduous and coniferous woodlands, seasonal and semi-permanent wetlands. farmland, grassland and more developed landscapes along the shoreline and in close proximity to the lake and designated natural area (Figure 1).


Surveys were based on delineated habitat types that were determined though a grouping of similar vegetative features, along with a monitoring of seasonality of wetlands and comparison to previously conducted studies in this or similar mosaic landscapes. Defining features of each habitat are described in Table 1. These were in part delineated at the inception of the project, but also became dependent on rate of drying throughout the duration of the study, especially in areas of marshy vegetation.


Figure 1 Map of the Big Lake Natural Area (BLNA).

(outlined in red)



Survey Techniques


Surveys were conducted either along the Natural Area boundary, where the shoreline makeup permitted, or at prescribed intervals where access to the lake was permissible and water levels beyond the lake allowed for decent visibility. Kayaking was also used as a method of examining mid-water reedbeds for nests and viewing areas of shoreline inaccessible on foot. Each site was assessed for landmarks to determine location, vegetation type and ground saturation to place it in a habitat grouping. At each visit, species seen and heard were identified with the aid of binoculars and spotting scope and abundance of each species was recorded. Playback recordings were not implemented. Surveys were completed by scanning the lake with binoculars to find closer groupings of waterfowl, then scanning with a spotting scope to identify those at greater distance. Birds that could not be identified using either of these tools were determined to be farther than the designated area allotted for the specific site being surveyed. Care was taken not to double count by noting the direction of species moving along the water and in the air and to remain far enough from the subjects to not cause a disturbance. Distance between shoreline sites was determined by both landmarks and diminished possibility of overestimation of abundance. Although the practice in some studies has been to double the number of "breeding territories" during breeding surveys to provide an estimation of bird numbers, this has also been found to over estimate populations and therefore was not used in this study (Johnston, 1990; Dunn and Bradstreet, 1997).


During breeding surveys, locations of nests were noted, and an attempt was made to identify species by the eggs or female seen leaving the nest. Eggs were counted, although due to the secretive nature of most nests and numerous instances of egg mortality these numbers were not included in population estimates or analysis. Nestlings and juveniles were included, with the assumption being made that these exceptions would compensate for each other. Data was recorded for migration data between May 9 and May 31, 2003 and breeding recorded between June 2 and July 10, 2003 as per Bilyk et al., 1998. Data was then transferred onto a spreadsheet separating sites by habitat classification, location and date and noting any relevant precipitation or wind level for the site.



Each habitat type was analyzed in terms of two measures of biodiversity, using similar methods to Prescott et al. (1995). The Shannon-Weaver diversity index was calculated for each of the habitat types, for migration, breeding and as an overall mean. As the value of H increases, it implies a higher level of diversity. Simple species richness averages were also noted for each of the nine habitats.


The Shannon-Weaver diversity index takes into account both species richness and evenness of individual abundance between species, allowing for a more accurate measure of biodiversity than a simple account of number of species present in a given habitat (Beals et al., 2000). However, despite this consideration, the index is only relevant when comparing species within a community and their relative abundance (Prescott et al., I995).


1. Shannon-Weaver diversity index

Using the formula H' = where pi is the proportion of the total number of species in the community made up of the ith species (Beals el al.. 2000). The values for each species were summed for each day each site was surveyed, and then averaged for each site and subsequently for each habitat.


 2. Species richness

The total numbers of species seen in all surveys across each habitat type within the census time frame were summed and recorded on a per-site basis to account for differences in sampling. It does not take into account the proportion and distribution of each species within the local community.


Each habitat was also analyzed for numbers of species present in each Order grouping; Gaviiformes, Ciconiiformes, Podicipediformes, Anseriformes, Falconiformes, Gruiformes, Charadriiformes, Piciformes, and the Order Passeriformes and abundance of individuals in those Orders in each habitat. Organized by date of the census, if more than one site was surveyed in a habitat, the bar graphs (Appendix 1) denote average numbers and date groupings if all sites were not surveyed on the same date. A list of birds heard or observed during migration as well as one during breeding season is included for each habitat. Bird species of special status were also noted, as well as those species for which nests were observed and plausibly or positively identified (Appendix 2).


Table 1. Descriptions of habitat types identified in the Big Lake Natural Area. 

(Adapted from Russell and Spiers, City of Edmonton and Bilyk et al., 1998)

Habitat Type



Developed Mosaic

Primarily golf courses, consisting of carefully mowed grassland with interspersed manufactured wetland and treed areas. Remains a suitable habitat for wildlife despite anthropocentric influence.

Flooded Farmland

Previously managed farmland not currently growing crops but saturated for at least May and June. Growing natural but stunted vegetation.

Lake Margin Complex

Reeds and cattails (Scirpus sp. and Typha sp.) close to the lakeshore, followed by willow and shrub-like vegetation and forest, either deciduous or coniferous forest, usually associated with an upward slope in terrain. If these habitats are relatively narrow in width, the habitat is defined as lake margin complex.

Small Permanent Freshwater Wetland

Wetland under 1 hectare in area, in this case defining irrigation canals and storm water collection habitats, defined by extra organic materials and precise rather than progressive edge habitat.

Medium Permanent Freshwater Wetland

Wetlands between 1 and 8 hectares, Big Lake proper.

Seasonal Wetland/Native Grassland

Grassland not disturbed by cultivation or recent grooming. Often located near areas leading up to marsh-type wetland habitats. Habitat containing abundant Typha and Scirpus species, requiring at least seasonal saturation of soils and/or shallow flooding until the beginning of July.

Semi-permanent Freshwater Wetland

Wetlands that retain water until autumn for 7 out of 10 years.

Upland Deciduous Forest

Primarily deciduous forest, either behind a long shoreline or a tree-stand away from the water but within the Big Lake natural area. Farther from the lake than those areas considered lake margin complex.

Upland White Spruce Forest

Primarily coniferous forest, either behind a long shoreline or a tree-stand away from the water but within the Big Lake Natural Area. Farther from the lake than those areas considered lake margin complex.



Maps of study sites and graphs of species numbers and individual abundance are included for each habitat in Appendix 1, along with the analyses and ranking of each habitat. A total of 196 surveys were done in the 9 habitat types (Table 1) in the Big Lake Natural Area, 81 during spring migration and 115 during breeding season. All surveys were conducted between May 9 and July 10, 2003 with opportunistic species observations from April migration being taken into account in the cumulative species list (Appendix 2). In total, 119 species were observed over the 3 month survey period and incidental observation reports.


As much of the avian habitat included in the Big Lake Natural Area is an amalgamation of the described types; those including a higher proportion of wetlands had more diversity than upland-only habitats, although these in turn had higher diversity indices than seasonal types of habitats. In terms of both per-site species richness and the Shannon-Weaver diversity index (Table 3), the small permanent wetlands had the greatest variety and even spread of species of all nine Big Lake habitats. These wetlands tend to have a high ratio of edge to area, and provide a lot of different species with needed habitat for feeding, hiding, and breeding.


The developed mosaic habitat (largely golf courses), despite its groomed nature also offers a very favourable environment for both migratory and breeding avian species with a high diversity index (H'= 2.10) as well as species richness of 24 and 30 species during migratory and breeding surveys, respectively. This habitat is much like the small permanent wetlands in having a high “edge effect” where many different species find niches.


The lake margin complex took up by far the largest portion of the natural area and being a habitat conglomerate had a high diversity rating with very high total species richness. Again, this habitat is characterized by a lot of “edge”, being defined as the border between the emergent reeds and grasses and open water. Many species in the Big Lake Natural Area use this habitat for breeding, nesting, feeding, and loafing.


The flooded farmland habitat had a higher diversity index during migration when its water supply was abundant. Once the migratory birds left and the water receded the flooded farmland habitat was not very attractive to many breeding species. This is evidenced by both it’s relatively low species richness value and low Shannon-Weaver diversity index (1.69) largely due to an uneven species presence.


Most of the species found in the medium permanent freshwater sites were waterfowl, not surprising considering the quantity of water available. There was a high diversity of waterfowl and some edge species that gave it a high species richness index although a relatively low diversity index.


The seasonal wetland/native grassland is more diverse in its grassland period, with the breeding season having 37 of the 45 total species found in that habitat. The Shannon-Weaver Diversity Index for this habitat is the lowest in the BLNA owing to poor evenness of species abundance.


Semi-permanent wetlands had medium diversity indices and per-site species richness, but collectively have the third highest species richness.


Both upland forest habitats have reduced species richness due to their specialized environment for Passeriformes species. Despite their lower species richness, these habitats have medium Shannon-Weaver diversity indices (2.01 and 1.93) indicating relatively few individuals of many species present.


Table 2. The number of species recorded in each habitat type

Habitat Type

Total Migratory Species

Total Breeding Species

Developed Mosaic



Flooded Farmland



Lake Margin Complex



Small Permanent Wetland



Medium Permanent  Freshwater Wetland



Seasonal Wetland/Native  Grassland



Semi-permanent Wetland



Upland Deciduous Forest



Upland White Spruce Forest




Table 3. Measures of diversity of each of the habitats*

Mean Species Richness/ Site;

Total Species Richness

Shannon-Weaver Diversity Index (H')

Small Permanent Wetland (39.0); (53)

Small Permanent Wetland (2.17)

Developed Mosaic (39.0); (39)

Developed Mosaic (2.10)

Medium Permanent Freshwater Wetland (36.5); (59)

Lake Margin Complex (2.04)

Semi-permanent Wetland (35.3); (55)

Upland White Spruce Forest (2.01)

Lake Margin Complex (34.1); (78)

Semi-permanent Wetland (1.94)

Seasonal Wetland/Native Grassland (28.0); (45)

Upland Deciduous Forest (1.93)

Upland Deciduous Forest (25.0); (25)

Medium Permanent Freshwater Wetland (1.92)

Flooded Farmland (25.0); (25)

Flooded Farmland (1.69)

Upland White Spruce Forest (16.0); (16)

Seasonal Wetland/Native Grassland (1.68)

* - Ranked in order from most to least diverse according to per-site and total species richness and the Shannon-Weaver Diversity Index.


Discussion and Recommendations


Out of the 115 species observed or heard in the Big Lake Natural Area during the spring and summer of 2003, 78 of those species were found in the lake margin complex habitat alone. Because of the composite nature of this habitat, its diversity illustrates the importance of maintaining proper edge habitats around wetlands to enable a greater diversity of species to utilize resources in the area. In addition to species diversity, this habitat also housed the largest number of species considered by Alberta Environment as "sensitive"' in the province. The medium freshwater permanent wetland, seasonal wetland/native grassland, semi-permanent wetland and small permanent wetland habitats also require habitat protection for their sensitive species as well as species richness.


In general, wetland size has been noted to contribute to greater diversity of species (Bilyk et at. 1998: Prescott et al., 1995) but in the 2003 Big Lake Natural Area the small permanent wetlands proved to house a greater diversity of species by Shannon-Weaver's definition. This diversity index takes into account both species richness and species evenness by measuring the proportion of individuals within species in a defined habitat (Beals et al., 2000, Kinzig et al., 2001). The relatively shallow nature of Big Lake, especially prevalent in 2003 due to low precipitation in the area in 2002 (Environment Canada, 2003), may have influenced species observations and contributed to the lower diversity index experienced in the medium as compared to other habitats like the small permanent wetland habitats. Duck nesting habitats are greatly affected by changes in surface area of wetlands, and especially nest density is affected by habitat degradation that may be caused by a decrease in water levels (Laperle, 1974). Water levels have also been found to affect composition of species in wetland areas, so a monitoring of both species composition and abundance over a period of years is an essential to protect from unnecessary declines in species numbers (Laperle, 1974: Johnston. 1990).


Despite the fact that Shannon-Weaver indices are of the most used in biodiversity ranking (Beals et al., 2000), Kinzig et al. (2001) note that there are many factors, including ecosystem productivity, area disturbance, and species nesting that are not taken into account by any known index. Biodiversity is known to be affected by changes in habitat, especially habitat fragmentation which is thought to cause a decline in species diversity (Rowell and Rice, 1998). This result can be seen in the two upland forest habitats, which did not yield the diversity expected, but whose presence is integral to biodiversity in the BLNA because of the specialized and unique habitats they offer. As an example, the white-breasted nuthatch was found only in the upland white spruce forest habitat, and the downy woodpecker and yellow-bellied sapsucker were observed only in the upland deciduous forest habitat.


It is assumed that stable and increasing populations of avian species denote healthy populations and preserved habitats (Caswell and Dickson, 1997) although to adequately determine the success of the Big Lake Natural Area a continuing program of monitoring is necessary.


Similar to the Beaverhill Lake study done by Bilyk et al. (1998), many of the habitats observed at Big Lake are in extremely close proximity. This explains the similar H' values for all of the observed habitats. A difference in sampling density affects the number of species observed, but is also representative of the proportion of area covered by each habitat type. The accuracy of this type of sampling is dependent on environmental factors, including water levels, conditions while sampling and general weather over the season in aspen parkland and surrounding areas may affect avian preferences toward a certain area for migration stops or breeding.


The colony nests of Franklin's Gulls were not present in the BLNA in the summer of 2003. The region west of the lake known at one time to host the Franklin's Gulls was quite dry during the breeding season this year (pers. obs.), even though precipitation in the Alberta aspen parkland increased in May, June, July and August 2003 from 2002 (Environment Canada, 2003). This may be a result of a depleted water table due to years of below average precipitation, and/or due to attempts to drain this area for farming/industrial purposes. Colonial nesting species are known to be quite vulnerable to habitat change and disturbances, which is an issue in the BLNA, but they also have been found to choose habitats uniquely, contributing to the possibility that the gulls may return to the site if habitat conditions are restored over the next few years (Hanneman and Heckbert, 2001). As nesting colonies of Black Terns were observed in the semi-permanent wetland habitat of Big Lake, there still remains suitable habitat for colonial nesters in the immediate area. A study to determine if drainage patterns in the wetlands just to the west of the BLNA have been affected by construction of drainage canals should be considered. Continuation of observations and correlation with habitat change and maintenance of the edge habitats to prevent habitat fragmentation and loss remain the most important and significant steps toward conservation and maintenance of diversity in the BLNA.



This work was funded by grants from the Alberta Community Development Fund, the Alberta STEP program, PetroCanada and TransCanada Pipelines through the Canadian Nature Federation’s Important Bird Area Community Action Fund.


Mike Barr of Ducks Unlimited Canada provided in-kind support in the field design of the project and the review of this paper.


This project was developed and produced by Miles Constable of the Big Lake Environmental Support Society (BLESS), the designated stewards of the Big Lake Natural Area. Louise Horstman, President of BLESS also contributed significant time to the project.


Thanks go out to George Newton of the Federation of Alberta Naturalists for advice on funding applications, and to the members of BLESS who kindly donated equipment to enable the survey to be completed.



Literature Cited

Alberta Community Development, 2001. Big Lake Natural Area, the Management Plan. Pamphlet. Edmonton.


Alberta Environment, 2000. The General Status of Alberta Wild Species 2000. Alberta Sustainable Resource Development, Edmonton.


Beals. M.. L. Gross and S. Harrell. 2000. Diversity indices: Shannon's H and E.



Bilyk. J.N, M. Barr and G.L. Holyroyd. 1998. Migratory and summer bird-habitat associations of the Beaverhill Lake landscape. Beaverhill Bird Observatory, Edmonton, AB. 62 pp.


Caswelt F.D. and K.M. Dickson. 1997. Evaluating the status of waterfowl populations in Canada. In: Monitoring bird populations: the Canadian experience. In: Dunn, E.H., M.D.


Cadman and J.B. Falls (Eds.) Occasional Paper #95. Canadian Wildlife Service, Environment Canada, Edmonton. pp. 8-15.


Environment Canada. 2003. Climate and Weather Archives, http://www.climate.weatheroffice.ec.ge.ca/advanceSearch/searchHistoricDataStations_e.html


Dunn. E.H. and M.S.W. Bradstreet. 1997. Using breeding season surveys to monitor changes in Canadian landbird populations. In: Dunn, E.H., M.D. Cadman and J.B. Falls (Eds.)


Hanneman. M.P. and M.D. Heckbert. 2001. Colonial nesting waterbird survey in the northwest boreal region, 2000. Alberta Species at Risk. Report #7. AB Sustainable Resource and Development. 21pp.


Hussell DJ.T. 1997. Monitoring migrants to detect changes in populations of birds breeding in Canada: present status and future prospects. In: Survey designs and statistical methods for the estimation of avian population trends. Sauer. J.R. and S. Droege (Eds.) U.S. Fish and Wildlife Service, Biological Report 90(1): 33.


IBA. 2003. Canadian online Important Bird Area directory website http://www.bse-eoc.org/iba/site.ctin?sitelD=ABO68&lang-en


Johnston, D.W. 1990. Descriptions of surveys breeding bird censuses. In: Survey designs and statistical methods for the estimation of avian population trends. Sauer, J.R. and S. Droege (Eds.) U.S. Fish ad Wildlife Service, Biological Report 90(1): 33.


Kinzig, A.P.. S.W. Pacala and D. Tilman. Eds. 2001. The Functional Consequences of Biodiversity: Empirical Progress and Theoretical Extensions. Princeton University Press, Princeton and Oxford. pp. 11-14.


Kirk. D. 2001. A decision support tool to aid in evaluating significance of adverse effects on birds fbr environmental assessment. Canadian Environment Assessment Agency Research and Development monograph Series. Government of Canada. 69 pp.


Laperle. M. 1974. Effects of water level fluctuation on duck breeding success. In. Boyd. H. (ed.).1974. Waterfowl studies, CWS report series #29 in eastern Canada, 1969-1973. pp. 18-30.


Prescott. D.R.C., A..T. Murphy and F. Ewaschuk. 1995. An avian community approach to determining biodiversity values of NAWMP habitats in the aspen parkland of Alberta.


Alberta NAWMP Centre. NAWMP-012. Edmonton, Alberta. 58 pp.


Rowell. P.A. and C. Rice. 1998. Manning diversified forest products research trust fund MDFP 13/96 Beaverhill Bird Observatory Neotropical Bird Monitoring 1997;8. Tofield, AB. 5 pp.


Russell. W.B. and G.A. Spiers. 198-. Vegetation pretyping of the annexed lands of Edmonton and detailed environmental assessment of the Big Lake area. Prepared for the City of Edmonton. 66 pp.