94th ESA Annual Meeting (August 2 -- 7, 2009)

PS 24-33 - Constructed ponds as breeding habitat for the horned grebe in the Peace Parkland, Canada

Tuesday, August 4, 2009
Exhibit Hall NE & SE, Albuquerque Convention Center
Cynthia A. Paszkowski1, Eva C. Kuczynski1 and Beverly A. Gingras2, (1)Department of Biological Sciences, University of Alberta, Edmonton, AB, Canada, (2)Environment Canada, Edmonton, AB, Canada
Background/Question/Methods

The destruction and degradation of wetlands have resulted in significant declines in populations of aquatic birds throughout North America. The horned grebe (Podiceps auritus) is an example of a species that was considered “of high concern” by the Northern Prairie and Parkland Waterbird Conservation Plan (2004). Current conservation strategies for aquatic birds in western Canada include not only preserving, but also restoring and constructing wetland habitat. We examined use by breeding horned grebes of borrow-pits (0.1 – 2.6 ha ponds created during highway construction) in the Peace Parkland, northwest Alberta, Canada. Our goals were to 1) document patterns of occurrence and breeding success of grebes on borrow-pit ponds and 2) to determine if occupied and unoccupied ponds, and ponds with pairs that did or did not produce chicks differed in water chemistry, local habitat features, or landscape characteristics. In May 2003 and 2007 we surveyed 330 constructed ponds for grebes. In 2007 we chose 100 occupied and 100 unoccupied ponds for additional surveys in June, July and August 2007, and again in June and August 2008. We collected habitat and landscape data for these 200 ponds, and limnological data for a subset of 46 ponds. Generalized linear mixed-model regression analysis, coupled with Akaike’s Information Criterion, was used to determine which environmental variables were most effective in predicting occupancy of ponds by horned grebes. 

Results/Conclusions

The best model included all local habitat and landscape features. Grebes occupied larger ponds with more emergent and riparian vegetation. Grebes tended to avoid ponds surrounded by greater proportions of forest within 1 km and that supported beaver or human activities along their shorelines. Models that incorporated water chemistry parameters were not effective in predicting pond occupancy. Of ponds inhabited by adult grebes in June, 78.8% (of 85 ponds) in 2007 and 86.5% (of 96 ponds) in 2008 produced between one and six chicks. Our models based on habitat and landscape features had little power to predict which occupied ponds would produce chicks. We conclude that the construction of small wetlands offers a viable means of increasing breeding habitat for the horned grebe.