1 Cofrin Center for Biodiversity, University of Wisconsin-Green Bay, Green Bay, WI ... of Natural Sciences, University of Wisconsin Superior, Superior, WI USA.
Avian Responses to Landscape Stressors in Great Lakes Coastal Wetlands R.W. Howe1, N.G. Walton1, E.E. Giese1, G.J. Niemi2, N.P. Danz3, T.N. Brown2, A.M. Bracey2* 1
Cofrin Center for Biodiversity, University of Wisconsin-Green Bay, Green Bay, WI USA 2 Natural Resources Research Institute, University of Minnesota Duluth, Duluth, MN USA 3 Department of Natural Sciences, University of Wisconsin Superior, Superior, WI USA
Birds respond to environmental disturbance in a variety of ways depending on the type and extent of the given stressor. Understanding how species respond to different types of stress is essential to the development of meaningful environmental indicators. We classified bird species’ responses as positive, negative, intermediate , or no response with respect to 3 stressor gradients; 1) percent agricultural land cover (%Ag), 2) percent developed land cover (%Dev), and 3) a multivariate metric (SumRel), combining these variables with population density, road density, and point source pollution data. Then we compared the proportions of species exhibiting each type of response. Responses were classified for Great Lakes coastal wetland obligate species or species strongly associated with such wetlands.
Study Area We used data gathered for 99 wetland complexes in the U.S. portion of the Great Lakes sampled for birds in 2001- 2004. These data were collected as part of the Great Lakes Environmental Indicators (GLEI) project, and analyses were confined to Ecoprovince 212, the Laurentian Mixed Forest Province (Fig. 1).
Figure 1. Location of the 99 wetland complexes used in our analyses, which only incorporated wetlands located in the Laurentian Mixed Forest Province (Ecoprovince 212).
Methods Using linear regression, we modeled the observed relationships between the abundances of coastal wetland bird species and 3 stressor gradients associated with the adjacent watershed: SumRel, %Ag, and %Dev (Fig. 2). Each stressor gradient ranges from 0 to 10, where 0 = the maximum amount of the stressor gradient ("highly degraded") and 10 = the minimum amount of the stressor gradient ("minimally degraded"). Predictor variables were modeled as polynomials (up to third order) of the original gradients, allowing for non-linear responses. Responses for 26 species were classified. A species was defined as having a response if the regression analysis was significant (p < 0.05). If the maximum value of a response occurred at greater than 8 we considered it a positive response, if less than 2, a negative response, and if the value was between these criteria it was considered an intermediate response.
Figure 2. Species response functions were modeled by quantifying the relationship between the abundance of a species and a defined stressor gradient (e.g. %Agriculture, %Development).
Results A large proportion of bird species exhibited no significant response to the stressor gradients. A greater number of species responded negatively than positively to the environmental stressor gradients (Table 1). Each type of biotic response was observed in at least one species. Negative, intermediate, and positive bird species responses to different stressor gradients are represented in Figure 3. Table 1. Summary of the types of bird responses observed for 26 Great Lakes coastal wetland species (obligate species and those strongly associated with wetlands). Numbers represent the proportions of each response for all 3 gradients (SumRel, %Ag, and %Dev). Response Neutral Negative Intermediate Positive
SumRel 0.58 0.27 0.04 0.12
Ag(%) 0.62 0.35 0.00 0.04
Dev(%) 0.81 0.12 0.00 0.08
Figure 3. Example response curves for 3 species to each of the 3 stressor gradients. a) Mallard (Anas platyrhynchos) represents a negative response to development. b) Marsh Wren (Cistothorus palustris) represents an intermediate response to the SumRel gradient. c) Alder Flycatcher (Empidonax alnorum) represents a positive response to agriculture.
Conclusions Our results show that birds respond to these environmental stressor gradients in different ways. A probability-based Index of Ecological Condition (IEC) developed by Howe et al. (2007a,b), modeled biotic responses to landscape stressors such as these. This metric has been applied to multiple-taxa within the Laurentian Great Lakes region. Our results can be used to develop a multi-taxa and multi-gradient IEC that will inform land managers about health of coastal wetlands and the landscape stressors that affect these systems.
References Howe et al. 2007a. Ecological Indicators 7:793-806 Howe et al. 2007b. Journal of Great Lakes Research 33: 93-105
Acknowledgments This research was supported by a grant from the U.S. EPA’s Science to Achieve Results Estuarine and Great Lakes program through funding to the Great Lakes Environmental Indicators project and a grant from the National Aeronautics and Space Administration . We are grateful for contributions by other scientists involved with the GLEI project. Other important contributions came from T. Hollenhorst, P. Wolter, V. Brady, J. Brazner, S. Price, D. Marks, and others, including more than 20 student field investigators. Important ideas underlying this analysis were communicated by J. Karr, D. Simberloff, P. Bertram, A. Tyre, and H. Possingham.