The Western Grebe (Aechmophorus occidentalis), a Species of Special Concern in Canada, has decreased in both distribution and abundance throughout its North American range (COSEWIC 2014). Declines in abundance have been noted in British Columbia, where the species is on the Red List as imperiled (Burger 1997, British Columbia Conservation Data Centre 2015), and in California (Robison et al. 2015), including areas that once had some of the largest concentrations of wintering Western Grebes (Burger 1997, Puget Sound Action Team 2007). The decline in wintering regions is thought to be partially a result of a southerly geographic shift of the species’ distribution due to prey availability (Wilson et al. 2013). In Alberta, the Western Grebe was recently listed as Threatened, following similar patterns of decline and due to its sensitivity to disturbance and habitat loss (AESRD and ACA 2013). Not only has Western Grebe occupancy decreased by 37% on 43 lakes in Alberta that historically supported grebes (Erickson et al. 2014), but declines in abundance have been much more pronounced—up to an estimated 63% loss of breeding adults over the past 15 years in some regions (Fig. 1). Based on the latest provincial population estimate of 9549 birds in 2012, Alberta supports 10–14% of the world’s breeding population of Western Grebes (AESRD and ACA 2013, COSEWIC 2014). Considering these data, the Alberta Western Grebe population should be an important focus for conservation.
Past studies have reported factors correlated with occupancy of the Western Grebe, including fish-bearing lakes and ice-free periods for nesting (Nuechterlein 1975, Riske 1976, Forbes 1984, Found et al. 2008), as well as correlates of persistence (constancy in occupancy over time) (Rahel 1990) probability. The relative probability of persistence of the Western Grebe was modeled relative to key habitat variables on Alberta lakes throughout its breeding range that were known to have supported the species (Erickson et al. 2014); Western Grebe persistence on a subset of once-occupied lakes was positively correlated with the proportion of shoreline bulrush (Scirpus lacustris), and was inversely related to the proportion of forested backshore. As well, Western Grebes have persisted on many lakes with extensive human development.
As a general pattern, in ecology, occupancy is correlated with abundance (Andrewartha and Birch 1954), having been documented in a diversity of taxa (Winters and Wheeler 1985, Gibbons et al. 1993, Mossman et al. 1998, Gaston et al. 1997). Indeed, for a variety of applications, occupancy surveys have been used to predict abundance (Nachman 1981, He and Gaston 2003, Royle and Nichols 2003, Boyce et al. 2016). However, in the context of a threatened species like the Western Grebe, it is important to examine occupancy and abundance separately because a high probability of persistence might be the result of a historically large Western Grebe colony rather than a reflection of current abundance. As well, systematic count surveys provide context for the creation of species recovery plans for threatened and/or endangered species because count data are used both to establish a baseline and as a target population estimate at which the species could be considered sufficiently recovered. Publicly accessible citizen-science databases such as eBird (Sullivan et al. 2009) provide an opportunity to submit opportunistic sightings but may not provide information on the number of birds or breeding status at a site. Finally, because habitat loss remains one of the top threats to the world’s endangered avian species (Wilcove et al. 1998, Stattersfield and Capper 2000, Wells 2007), exploring how habitats relate to abundance can facilitate better management and conservation of habitats on lakes that continue to support Western Grebes.
We examine the strength of the relationship between Western Grebe abundance (i.e., surveyed number of adults on a lake) and the relative probability of persistence on a suite of lakes that either historically supported or currently support grebes, and the relation between categories of Western Grebe abundance and habitat covariates. If persistence probability and abundance are not highly correlated, factors related to persistence might be different from those associated with abundance, thereby suggesting different implications for Western Grebe habitat conservation.
Data on Western Grebe abundance were collected on 43 lakes at which the species was known to occur within the past 40 years in Alberta’s Boreal Forest, Parkland, and Grassland regions (Fig. 2). In addition, each of the lakes was characterized for covariates that we hypothesized to be important environmental characteristics for the grebes. By focusing on lakes with a history of Western Grebe presence, we hoped to ascertain site suitability for continued support of the species.
The Boreal Forest region covers the northernmost part of the province and is bordered by the Foothills region to the west and Parkland region to the south. Major vegetation includes trembling aspen (Populus tremuloides) and balsam poplar (P. balsamifera), white spruce (Picea glauca), black spruce (P. mariana), and jack pine (Pinus banksiana). Some of Alberta’s largest and deepest lakes occur in this region, with 35–45% of the landscape dominated by wetlands (Natural Regions Committee 2006). The Parkland region includes the most populated areas in the province, including the cities of Edmonton, Red Deer, and Calgary. Wetlands make up between 8% and 10% of this region (Natural Regions Committee 2006). Vegetation in the region represents the transition between the northern boreal forest and southern grasslands, often with stands of aspen interspersed with grasslands. The Grassland region of southern Alberta consists of mostly agricultural lands and native grassland (e.g., Festuca spp.). Lakes comprise less than 2% of this region (Natural Regions Committee 2006).
During May through August 2008–2009, we conducted surveys for Western Grebe abundance, and characterized habitat variables on lakes in Alberta’s Boreal Forest (n = 34), Parkland (n = 6), and Grassland (n = 3) regions (Fig. 2). All surveys were conducted between 0700 and 1600 hours.
Our 43 study lakes were a subset of lakes surveyed by Alberta Environment and Parks (AEP), at which the Western Grebe historically had occurred since at least 1970, and included all known major breeding colonies since 1970. Historically, surveys for Western Grebes in Alberta were largely opportunistic, but they have become more consistent since 2000, with data reported in provincial species-at-risk reports. To remain consistent with data presented in these reports, we used AEP survey techniques to collect data on Western Grebe abundance, including nest counts, brood counts, and shoreline waterbird surveys. Survey method was dependent on the size of lake and whether there was a known Western Grebe breeding colony present. Nest counts provided an estimate of Western Grebe abundance on a particular lake, while brood counts and shoreline waterbird surveys yielded a count of Western Grebe adults. Each lake was surveyed three times within the study period; the highest estimate or count of Western Grebes was used in analysis (see Appendix 1 for abundance data and survey method).
On lakes with breeding colonies of Western Grebes, nest counts frequently are used as a proxy for adult Western Grebe abundance, with two breeding adults estimated per nest counted (Resources Inventory Committee 1999, Hanus et al. 2002). To minimize disturbance, we conducted nest counts in late July and early August, shortly after chicks hatched and left the nesting site. Observers entered the colony in chest waders or kayaks/canoes (depending on the water level). Nests were counted on straight-line transects along the length of the colony, with two to five observers within eyesight of each other. Distance between observers varied depending on the observed density of nests and density of vegetation; more nests and/or denser vegetation required narrower transects. Transect length extended a few metres beyond the edge of the colony to ensure that all nests were counted. For one lake on which a nest count was not conducted in 2008 or 2009 due to logistic factors and to minimize disturbance, the 2007 estimate was used in the analysis (see Appendix 1). All colonies were visited in either 2008 or 2009, however, to confirm breeding activity.
We conducted brood counts on a subset of breeding lakes during mid-August 2008 and 2009 to (1) confirm nest count estimates, and (2) document recruitment (Resources Inventory Committee 1999). One boat driver and one to two observers surveyed the lake using a systematic boat survey (Hanus et al. 2002), and stopped at predetermined points along transects in the open water to count all adult and juvenile birds that were visible from that location. We used brood count data for adult abundance in the data analysis if nest count data were not available for a particular breeding lake, or if the brood count data yielded a higher adult abundance estimate for that year (Appendix 1).
Shoreline waterbird surveys were used to collect Western Grebe abundance data on nonbreeding lakes, following techniques outlined by the Resources Inventory Committee (1998). Surveys were conducted from a kayak (for lakes < 5 km2), or motor-powered watercraft for medium (5–50 km2) or large (> 50 km2) lakes. From within the kayak or boat, we scanned the lake using binoculars while keeping a distance from the shoreline of 20–200 m, depending on visibility and water depth. The entire lake was surveyed to obtain a complete count of all Western Grebes (see Appendix 1 for abundance survey results).
Shoreline habitat surveys were conducted from the water (in either a kayak or boat), 20–400 m from shoreline, with distance from the shore dependent on visibility. We recorded the extent (m) along the shoreline of emergent macrophyte species known to provide nesting habitat, including cattail (Typha spp.), common reed grass (Phragmites australis), and bulrush (Wollis and Stratmoen 2010, LaPorte et al. 2013). Surveys were conducted as early as possible during the study period to record data for habitat variables shortly after spring migration and Western Grebe lake selection (throughout May/early June).
Within a geographical information system (GIS) ArcMap (Environmental Systems Research Institute 2008), we used both georeferenced aerial photography and satellite imagery (0.5–1.0 m resolution) to digitize and calculate additional variables, including shoreline perimeter and linear extent of emergent vegetation, as well as proportion of human development and proportion/type of backshore vegetation within a 500-m buffer surrounding each lake (see Found et al. 2008, Erickson et al. 2014). Anthropogenic development and emergent vegetation were digitized using ground-truthed data and digital aerial photography as a reference. We used a GIS land cover raster layer (Agriculture and Agri-Food Canada 2008) to calculate the amount and type of terrestrial vegetation and human land use in the 500-m buffer.
Western Grebes are mainly piscivorous (LaPorte et al. 2013). Fish density was a preferred metric to use in our models; however, the data available were limited to species richness. Therefore, we used the number of fish species per lake (from Alberta Environment and Parks’ Fish and Wildlife Management Information System [FWMIS]) as a metric for prey occurrence. Maximum lake depth levels were also obtained from FWMIS.
Covariates were log-transformed if needed to obtain a normal distribution (Hosmer and Lemeshow 2000, Vittinghoff et al. 2005), and were examined for multicollinearity. In the case of high correlation (|r| > 0.65) between covariates, we retained the covariate with the highest predictive ability from a univariate analysis (Hosmer and Lemeshow 2000). Covariates retained in the ordinal regression analysis are listed in Table 1.
We related Western Grebe abundance at the 27 occupied lakes to the relative probability of persistence at each lake. The relative probability of persistence was estimated using the top-ranked model from Erickson et al. (2014), which was based on Western Grebe occupancy data collected on the same suite of lakes and over the same time period as this study. The persistence model is the log-linear function:
where P(x) is proportional to the probability of persistence, β1…βn are coefficients estimated from the data, and x is a vector of predictor covariates, x1, x2,…xn (Erickson et al. 2014). We used a correlation coefficient to determine the strength of the relationship between the two parameters. Analyses were conducted in Stata 12.0 (StataCorp 2011).
We used ordered logistic regression to relate Western Grebe abundance to habitat covariates. Modeling actual abundance is a preferred method of analysis, but small sample size and skewed number of both unoccupied lakes (i.e., zeros) and lakes with high Western Grebe abundance necessitated an alternative approach. Ordinal regression models can be used to compare ranked categories of a response variable when the differences between categories are not necessarily equal (Guisan and Harrell 2000, Long and Freese 2006).
Western Grebe abundance was divided into four categories to compare unoccupied (0 Western Grebe adults), low (1–10 adults), medium (11–100 adults), and high (> 100 adults) abundance lakes, with categories selected to ensure reasonable dispersion of observations into groups for multinomial analysis. We modeled abundance categories as a function of six habitat variables (Table 1) within 12 a priori models, including a null model (Table 2). A proportional odds ratio was used to compare the highest abundance category with the lower categories for each predictor variable (UCLA 2009). We tested for the proportional odds assumption to determine if there was a difference in the coefficients between models, while a likelihood ratio test was used to compare each model to a null model without count predictors to test the significance of the model overall (UCLA 2009). Alternative models were compared using AICc for small sample size, an information-theoretic approach (Burnham and Anderson 2002).
Western Grebes were observed at 27 of the 43 study lakes during 2008–2009 (Appendix 1), with lakes classified as zero (n = 16), low (n = 13), medium (n = 7), or high (n = 7) abundance lakes. Ten lakes had evidence of breeding (presence of nests or young) during the study period, as compared to 21 historical breeding lakes, and all but two breeding lakes (Utikuma Lake and Lac la Nonne) had an established breeding colony with abundance ranging from 84 to 2716 birds (mean = 651, SE = 303). Adult grebe abundance on all occupied lakes ranged from 1 bird to 2716 birds, with a mean of 200 (SE = 103.02) and a median of 11 birds. All lakes within the "high" abundance category had established breeding colonies at the time of surveying. Although in some cases the recorded abundance differed slightly between repeated surveys on a lake within the season, no differences resulted in a change of a lake’s abundance category in the ordinal regression analysis.
The relationship between Western Grebe abundance and relative probability of persistence was significant (r = 0.44, df = 25, P < 0.05) (Fig. 3). However, abundance data accounted for only 19% of the variance in persistence.
The top ordinal logistic regression models (ΔAICc < 2) had habitat covariate combinations of Development Proportion, Bulrush Proportion, Shoreline Length, and Forest Proportion (Table 2), and were a substantial improvement (P < 0.05) over the null model according to the likelihood ratio test. Development Proportion was significantly associated with Western Grebe abundance, as was Shoreline Length, although only marginally. Bulrush Proportion and Forest Proportion exhibited positive and negative associations with grebe abundance, respectively, although these associations were not significant (Table 3).
According to the odds ratio (OR), lakes with a higher proportion of shoreline covered by bulrush were roughly 6.87 times more likely to have more than 100 grebes than were the medium, low, or zero abundance categories (Table 3). Lakes with increased development in a surrounding 500-m buffer (OR = 2.91) also favored higher ranked abundance categories, while an increased amount of forest in a 500-m buffer favored lower abundance categories or no birds at all (OR = 0.29) A greater shoreline perimeter (OR = 1.01) showed little to no association with higher abundance categories (Table 3).
The small-population paradigm states that the size of a population is a driver of persistence (Caughley 1994, Boyce 2002), with smaller populations at greater risk for extinction (MacArthur 1972). Indeed, in Alberta, the relative probability of Western Grebe persistence and abundance were significantly correlated, although the recent decrease in abundance is far greater than that in persistence, and abundance accounted for only 19% of the variance in the relative probability of persistence. This could be due to the mobile nature of migratory birds, which allows them to recolonize and continue to persist, while effects on abundance are not as easily overcome. To that end, although we surveyed only lakes with a history of Western Grebe occurrence, we were confident that we monitored the regional population because the provincial Department of Environment and Parks was monitoring other lakes in the area for waterbirds. As well, the species’ tendency toward site fidelity in Alberta—especially on lakes with established breeding colonies—coupled with provincially gathered waterbird data on additional lakes, suggests that the decline of Western Grebe on some lakes is not countered by a notable increase of the species elsewhere in the region.
Forest Proportion, Bulrush Proportion, and Development Proportion emerged as important parameters in our Western Grebe abundance models, with Bulrush Proportion and Development Proportion having positive coefficients and Forest Proportion displaying an inverse relation with higher abundance categories. These results are similar to those in the top persistence probability models from Erickson et al. (2014). However, we found that Shoreline Length was associated with abundance, whereas it had not been related to the relative probability of persistence in Erickson et al. (2014). Although we believe that the relationship between abundance and the parameters in the top models hold biological significance, we note that statistical significance remains weak given our limited sample size.
Forest Proportion was inversely associated with Western Grebe abundance, perhaps because the surveyed lakes with a high proportion of forested backshore occur primarily on the extreme northern edge of the species’ geographic range. As a result, the habitat surrounding these forested lakes is vastly different from other lakes known to support the Western Grebe, such as those within extensive marsh systems bordered by arid desert (Lindvall and Low 1982) or prairie pothole regions in the Great Plains (Allen et al. 2008). Nevertheless, substantial Western Grebe breeding colonies still occur on some of the more northerly lakes in our study area. Of these, a colony on Lac la Biche is within a designated Important Bird Area (Bird Studies Canada 2015), and the Cold Lake colony location is generally inaccessible to disturbance by motor watercraft due to the presence of a sandbar that separates the colony from the main lake body. Although these colonies have not been immune to the decline in abundance seen in other areas of the province, they so far have remained viable.
As expected, the proportion of shoreline bulrush was positively related to Western Grebe abundance. However, this variable did not include other vegetative species in which grebes have been known to nest, such as common reed grass or cattail (Nuechterlein 1975). For instance, breeding colonies on three lakes in Alberta (Wabamun Lake, Lake Isle, Lac Ste. Anne) all nested in different species of emergent vegetation during the 2008 season. Bulrush was included in the habitat models because it is preferred by Western Grebes (Riske 1976, Short 1984, LaPorte et al. 2013) and had the greatest correlation with the relative probability of grebe persistence (Erickson et al. 2014). However, other vegetation can be used for nests and cover if available, especially if bulrush is not continuous or dense enough during site selection and nest construction periods.
Although the length of shoreline was not a major predictor of grebe persistence probability (Erickson et al. 2014), it had a slight positive association with Western Grebe abundance, which suggests that a greater amount of shoreline may support larger numbers of birds. If the shoreline length is relatively large due to a complex geometry of sheltered bays or areas protected from wind and wave action, that lake might serve as an attractive site for breeding colony establishment.
Fish species richness did not contribute appreciably to any of the top models. However, we acknowledge that although the general pattern between occupancy and abundance (He and Gaston 2003) indicates that fish species richness might have some value as a crude index, obtaining estimates of abundance for species of fish that grebes use as prey could be a fruitful area for future research on patterns of grebe abundance.
Development Proportion was positively associated with Western Grebe abundance; i.e., lakes with the highest number of grebes also had high amounts of human development in a 500-m buffer surrounding the shoreline. Although this relationship was unexpected given negative impacts of human activity on waterbirds (Carney and Sydeman 1999) and the sensitivity of the Western Grebe to disturbance (AESRD and ACA 2013), it is consistent with the Erickson et al.’s (2014) persistence probability model, as well as results from Somers et al. (2015), who found that Western Grebe density was relatively high on two lakes in Saskatchewan, Canada, even in areas with increased shoreline development and recreational use. These findings suggest that the Western Grebe might have become habituated to the presence of humans on such lakes, or that the lake attributes preferred by grebes (i.e., large, deep, fish-bearing waterbodies) are the same as those selected by humans. In the latter case, these shared sites might be ecological sinks for Western Grebes, as suggested by Somers et al. (2015).
Loss of breeding season habitat is a major concern for Western Grebes in Alberta. Indeed, habitat loss and degradation is a primary threat to birds generally, affecting 85% of threatened bird species worldwide (Stattersfield and Capper 2000) and almost 87% of endangered bird species in Canada (Venter et al. 2006). Because developed lakes tend to have less emergent vegetation, especially adjacent to the shoreline (Radomski and Goeman 2001), loss of adequate habitat is a threat to Western Grebes at both breeding and nonbreeding sites. In Alberta, many known breeding colony sites already have experienced changes in habitat suitability due to activities such as snowmobiling over reed beds at Isle Lake in 2002 or high levels of boating activity at Lac Ste. Anne (AESRD and ACA 2013). Therefore, recreational lake use and development should be mitigated in a way that it does not destroy shoreline and/or emergent vegetation, especially at known breeding colony sites.
In systems where a species’ persistence and abundance are strongly correlated, it may be logistically favorable to use probability of persistence as an indicator of the viability of the population. However, abundance is an important parameter to consider with mobile species like birds that could experience large population declines while still maintaining an ability to recolonize a site. In addition—as in the case with the Western Grebe in Alberta—abundance data can be used to establish both baseline and target recovery populations of a threatened or endangered species.
Although forested lakes tended to have fewer grebes overall, a few northerly sites still support large colonies, and therefore should not be overlooked as conservation concerns. Applying protective notations or designating Important Bird Areas such as the case on Lac La Biche can be a first step to protecting specific colony locations. Specific bulrush stands within a lake frequently are used by breeding grebes year after year; therefore, it is not surprising that this variable was an important predictor in selected models of abundance. Lakes with a greater shoreline perimeter appear to have a greater effect on abundance (particularly large numbers of grebes) than on relative probability of persistence for Western Grebes. The relation between development and high abundance of Western Grebes shows that co-existence continues to be possible for this species. However, one must exercise caution in assuming this is a casual relationship, in light of several studies on the effects of disturbance on waterbirds (Carney and Sydeman 1999, Newbrey et al. 2005) and anthropogenically driven habitat loss on avian species in general (Stattersfield and Capper 2000, Gaston et al. 2003, Venter et al. 2006). Furthermore, although current abundance is high, annual surveys show that these colonies continue to decline overall (i.e., Fig. 1). Additional insight is needed into the effects of recreational lake use and/or development on patterns of Western Grebe abundance to better ascertain the success of this species over time on highly developed or recreational lakes. In the meantime, the species’ persistence and (relatively) high abundance on developed lakes is promising.
Because Alberta supports such a large proportion (10–14%) of the world’s Western Grebe breeding population, it is imperative that we mitigate threats to the birds and their habitats. Lakes with a greater shoreline perimeter might be important targets for conserving what remains of Alberta’s large colonies. Emergent vegetation (both new and old growth) should not be removed, especially on breeding lakes. Moving forward, gaining insight into the effects of habitat covariates, including anthropogenic activity, on abundance will be an important step in mitigating the current decline of Western Grebe while promoting recovery in the province.
We wish to thank many individuals from Alberta Environment and Parks (formerly Alberta Sustainable Resource Development), most notably H. Wollis, as well as staff from the area offices in Edmonton, Lac La Biche, and Red Deer, Alberta for field support, unpublished data, and aerial imagery. We thank C. Nielsen and M. Shain for technical assistance, and T. Berry, N. Martin, and J. Toth for field assistance. We appreciate the editorial comments provided by the reviewers of this manuscript, which helped improve the quality of its contents. We wish to extend a special acknowledgement to the late K. Norstrom for her expertise and dedication to Western Grebe research in Alberta, including this project. Funding was provided by the University of Alberta, the Alberta Conservation Association, the North American Waterfowl Management Plan, and the Alberta Sport, Recreation, Parks & Wildlife Foundation.
Agriculture and Agri-Food Canada. 2008. Land cover for agricultural regions of Canada, circa 2000. [online] URL: http://open.canada.ca/data/en/dataset/16d2f828-96bb-468d-9b7d-1307c81e17b8?_ga=2.1988981.515457352.1493845476-1970616929.1493845475
Alberta Environment and Sustainable Resource Development and Alberta Conservation Association (AESRD and ACA). 2013. Status of the Western Grebe (Aechmophorus occidentalis) in Alberta: update 2012. Alberta Wildlife Status Report No. 60 (Update 2012), Alberta Environment and Sustainable Resource Development, Edmonton, Alberta, Canada.
Allen, J. H., G. L. Nuechterlein, and D. Buitron. 2008. Weathering the storm: how wind and waves impact Western Grebe nest placement and success. Waterbirds 31:402–410. http://dx.doi.org/10.1675/1524-4695-31.3.402
Andrewartha, H. G., and L. C. Birch. 1954. The distribution and abundance of animals. University of Chicago Press, Chicago, Illinois, USA.
Bird Studies Canada. 2015. Important Bird Areas of Canada database. Port Rowan, Ontario, Canada. [online] URL: https://www.ibacanada.ca/site.jsp?siteID=AB097
Boyce, M. S. 2002. Reconciling the small population and declining population paradigms. Pages 41–49 in S. Beissinger and D. R. McCullough, editors. Population viability analysis. University of Chicago Press, Chicago, Illinois, USA.
Boyce, M. S., C. J. Johnson, E. H. Merrill, S. E. Nielsen, E. J. Solberg, and B. van Moorter. 2016. Can habitat selection predict abundance? Journal of Animal Ecology 85:11–20. http://dx.doi.org/10.1111/1365-2656.12359
British Columbia Conservation Data Centre. 2015. BC Species and Ecosystems Explorer. British Columbia Ministry of Environment, Victoria, British Columbia, Canada. [online] URL: http://a100.gov.bc.ca/pub/eswp/
Burger, A. E. 1997. Status of the Western Grebe in British Columbia. Wildlife Working Report No. WR-87, British Columbia Ministry of Environment, Victoria, British Columbia, Canada.
Burnham, K. P., and D. R. Anderson. 2002. Model selection and multimodel inference: a practical information-theoretic approach. Second edition. Springer, New York, USA. http://dx.doi.org/10.1007/b97636
Carney, K. M., and W. J. Sydeman. 1999. A review of human disturbance effects on nesting colonial waterbirds. Waterbirds 22:68–79. http://dx.doi.org/10.2307/1521995
Caughley, G. 1994. Directions in conservation biology. Journal of Animal Ecology 63:215–244. http://dx.doi.org/10.2307/5542
Committee on the Status of Endangered Wildlife in Canada (COSEWIC). 2014. COSEWIC assessment and status report on the Western Grebe Aechmophorus occidentalis in Canada. Ottawa, Ontario, Canada. [online] URL: http://www.sararegistry.gc.ca/virtual_sara/files/cosewic/sr_Western%20Grebe_2014_e%20.pdf
Environmental Systems Research Institute. 2008. ArcGIS: Release 9.3. Environmental Systems Research Institutes, Redlands, California, USA.
Erickson, M. E., C. Found-Jackson, and M. S. Boyce. 2014. Using latent selection difference to model persistence in a declining population. PLoS ONE 9:e98126. http://dx.doi.org/10.1371/journal.pone.0098126
Forbes, L. C. 1984. The nesting ecology of the Western Grebe in British Columbia. Canadian Wildlife Service, Delta, British Columbia, Canada.
Found, C., S. M. Webb, and M. S. Boyce. 2008. Selection of lake habitats by waterbirds in the boreal transition zone of northeastern Alberta. Canadian Journal of Zoology 86:277–285. http://dx.doi.org/10.1139/Z07-137
Gaston, K. J., T. M. Blackburn, and K. K. Goldewijk. 2003. Habitat conversion and global avian biodiversity loss. Proceedings of the Royal Society of London B: Biological Sciences. 270:1293–1300. http://dx.doi.org/10.1098/rspb.2002.2303
Gaston, K. J., T. M. Blackburn, and J. H. Lawton. 1997. Interspecific abundance-range size relationships: an appraisal of mechanisms. Journal of Animal Ecology 66:579–601. http://dx.doi.org/10.2307/5951
Gibbons, D. W., J. B. Reid, and R. A. Chapman. 1993. The new atlas of breeding birds in Britain and Ireland: 1988–1991. Poyser, London, U.K.
Guisan, A., and F. E. Harrell. 2000. Ordinal response regression models in ecology. Journal of Vegetation Science 11:617–626. http://dx.doi.org/10.2307/3236568
Hanus, S., H. Wollis, and L. Wilkinson. 2002. Western (Aechmophorus occidentalis) and Eared (Podiceps nigricollis) grebes of central Alberta: inventory, survey techniques, and management concerns. Alberta Species at Risk Report No. 41, Alberta Sustainable Resource Development, Edmonton, Alberta, Canada. http://dx.doi.org/10.5962/bhl.title.114306
He, F., and K. J. Gaston. 2003. Occupancy, spatial variance, and the abundance of species. American Naturalist 162:366–375. http://dx.doi.org/10.1086/377190
Hosmer, D. W., and S. Lemeshow. 2000. Applied logistic regression. Second edition. John Wiley and Sons, New York, New York, USA. http://dx.doi.org/10.1002/0471722146
LaPorte, N., R. W. Storer, and G. L. Nuechterlein. 2013. Western Grebe (Aechmophorus occidentalis). In A. Poole, editor. The birds of North America online. No 026a, Cornell Lab of Ornithology, Ithaca, New York, USA. [online] URL: http://bna.birds.cornell.edu/bna/species/026a
Lindvall, M. L., and J. B. Low. 1982. Nesting ecology and production of Western Grebes at Bear River Migratory Bird Refuge, Utah. Condor 84:66–70. http://dx.doi.org/10.2307/1367823
Long, J. S., and J. Freese. 2006. Regression models for categorical dependent variables using Stata. Second edition. StataCorp LP, College Station, Texas, USA.
MacArthur, R. H. 1972. Geographical ecology: patterns in the distribution of species. Harper and Row Publishers Inc., New York, New York, USA.
Mossman, M. J., L. M. Hartman, R. Hay, J. R. Sauer, and B. J. Dhuey. 1998. Monitoring long-term trends in Wisconsin frog and toad populations. Pages 169–198 in M. J. Lannoo, editor. Status and conservation of Midwestern amphibians. University of Iowa Press, Iowa City, Iowa, USA.
Nachman, G. 1981. A mathematical model of the functional relationship between density and spatial distribution of a population. Journal of Animal Ecology 50:453–460. http://dx.doi.org/10.2307/4066
Natural Regions Committee. 2006. Natural regions and subregions of Alberta. Publication No. T/852, Government of Alberta, Edmonton, Alberta, Canada.
Newbrey, J. L., M. A. Bozek, and N. D. Niemuth. 2005. Effects of lake characteristics and human disturbance on the presence of piscivorous birds in northern Wisconsin, USA. Waterbirds 28:478–486. http://dx.doi.org/10.1675/1524-4695(2005)28[478:EOLCAH]2.0.CO;2
Nuechterlein, G. L. 1975. Nesting ecology of Western Grebes on the Delta Marsh, Manitoba. Thesis. Colorado State University, Fort Collins, Colorado, USA.
Puget Sound Action Team. 2007. State of the Sound 2007. Publication No. PSAT 07-01, Olympia, Washington, USA. [online] URL: http://blog.pugetsoundinstitute.org/wp-content/uploads/2011/12/StateoftheSound2007.pdf
Radomski, P., and T. J. Goeman. 2001. Consequences of human lakeshore development on emergent and floating-leaf vegetation abundance. North American Journal of Fisheries Management 51:46–61. http://dx.doi.org/10.1577/1548-8675(2001)021<0046:COHLDO>2.0.CO;2
Rahel, F. J. 1990. The hierarchical nature of community persistence: a problem of scale. American Naturalist 136: 328–344. http://dx.doi.org/10.1086/285101
Resources Inventory Committee. 1998. Inventory methods for colonial-nesting freshwater birds: Eared Grebe, Red-necked Grebe, Western Grebe, American White Pelican, and Great Blue Heron. Standards for Components of British Columbia’s Biodiversity No. 8. British Columbia Ministry of Environment, Lands and Parks, Victoria, British Columbia, Canada. [online] URL: https://www.for.gov.bc.ca/hts/risc/pubs/tebiodiv/colonial/index.htm
Resources Inventory Committee. 1999. Inventory methods for waterfowl and allied species: loons, grebes, swans, geese, ducks, American Coot and Sandhill Crane. Standards for Components of British Columbia’s Biodiversity No. 18. British Columbia Ministry of Environment, Lands and Parks, Victoria, British Columbia, Canada. [online] URL: https://www.for.gov.bc.ca/hts/risc/pubs/tebiodiv/waterfowl/assets/waterfowl.pdf
Riske, M. E. 1976. Environmental and human impacts upon grebes breeding in central Alberta. Dissertation. University of Calgary, Calgary, Alberta, Canada.
Robison, K. M., D. W. Anderson, and R. E. Robison. 2015. Brood size and nesting phenology in Western Grebe (Aechmophorus occidentalis) and Clark’s Grebe (Aechmophorus clarkii) in Northern California. Waterbirds 1:99–105. http://dx.doi.org/10.1675/063.038.0113
Royle, J. A., and J. D. Nichols. 2003. Estimating abundance from repeated presence–absence data or point counts. Ecology 84:777–790. http://dx.doi.org/10.1890/0012-9658(2003)084[0777:EAFRPA]2.0.CO;2
Short, H. L. 1984. Habitat suitability index models: Western Grebe. Western Energy and Land Use Team, U.S. Fish and Wildlife Service, U.S. Department of the Interior, Washington, D.C., USA.
Somers, C. M., L. M. Heisler, J. L. Doucette, V. A. Kjoss, and R. M. Brigham. 2015. Lake use by three avian piscivores and humans: implications for angler perception and conservation. Open Ornithology Journal 8:10–21. http://dx.doi.org/10.2174/1874453201508010010
StataCorp. 2011. Stata statistical software. Release 12. StataCorp LP, College Station, Texas, USA.
Stattersfield, A. J., and D. R. Capper, editors. 2000. Threatened birds of the world. Lynx Edicions and BirdLife International, Barcelona, Spain and Cambridge, UK.
Sullivan, B. L., C. L. Wood, M. J. Iliff, R. E. Bonney, D. Fink, and S. Kelling. 2009. eBird: a citizen-based bird observation network in the biological sciences. Biological Conservation 142:2282–2292. http://dx.doi.org/10.1016/j.biocon.2009.05.006
UCLA: Academic Technology Services, Statistical Consulting Group. 2009. Stata annotated output: ordered logistic regression. [online] URL: http://www.ats.ucla.edu/stat/stata/output/stata_ologit_output.htm
Venter, O., N. N. Brodeur, L. Nemiroff, B. Belland, I. J. Dolinsek, and J. W. A. Grant. 2006. Threats to endangered species in Canada. BioScience 56:903–910.
Vittinghoff, E., D. V. Glidden, S. C. Shiboski, and C. E. McCulloch. 2005. Regression methods in biostatistics: linear, logistic, survival, and repeated measures models. Springer, New York, New York, USA. http://dx.doi.org/10.1007/978-1-4614-1353-0
Wells, J. V. 2007. Birder’s conservation handbook: 100 North American birds at risk. Princeton University Press, Princeton, New Jersey, USA. http://dx.doi.org/10.1515/9781400831517
Wilcove, D. S., D. Rothstein, J. Dubow, A. Phillips, and E. Losos. 1998. Quantifying threats to imperiled species in the United States: assessing the relative importance of habitat destruction, alien species, pollution, overexploitation, and disease. BioScience 48:607–615. http://dx.doi.org/10.2307/1313420
Wilson, S., E. M. Anderson, A. S. G. Wilson, D. F. Bertram, and P. Arcese. 2013. Citizen science reveals an extensive shift in the winter distribution of migratory Western Grebes. PLoS ONE 8:e65408. http://dx.doi.org/10.1371/journal.pone.0065408
Winters, G. H., and J. P. Wheeler. 1985. Interactions between stock area, stock abundance, and catchability coefficient. Canadian Journal of Fisheries and Aquatic Sciences 42:989–998. http://dx.doi.org/10.1139/f85-124
Wollis, H., and C. Stratmoen. 2010. Population study of Western Grebes in Alberta 2001–2009: implications for management and status designation. Alberta Species at Risk Report No. 138, Alberta Sustainable Resource Development, Edmonton, Alberta, Canada.