Indiana Biological Survey

Division of Fishes Projects

Coastal Wetlands Habitat

Little Tail Point

Dead Horse Bay

Efforts to characterize Great Lakes biocriteria and establish regional reference conditions have generally concentrated on tributary waters within several states (Simon 1991; Lyons 1992; Thoma 1999) or have included limited numbers of sites (Wilcox et al. 1999; Burton et al. 1999). Greater portions of the Great Lakes nearshore and coastal wetlands have not been sampled, thus restricting a more global spatial depiction. Our effort was to develop sampling procedures, conduct pilot studies to determine spatial relevance of indicator development, and produce reference conditions for fish, plants, and macroinvertebrates. This effort is a preliminary study that has answered many questions regarding the quality of coastal wetlands throughout the Great Lakes.

The design strategy used was a tessellated, random-stratified sampling design that was used to obtain an unbiased selection of wetlands that could be used to estimate the range of wetland conditions in their “true” abundance. Fish assemblages within all the Great Lakes and connecting channels were used as the primary indicator to provide an unbiased validation of pending reference conditions for fish, wetland plants, and invertebrates. Site selection began with the National Wetland Inventory (NWI) electronic layers for each Great Lake for the United States shoreline.

The project consisted of two parts, a methods comparison and a regional comparison. During 2000, a pilot project was initiated in Lake Michigan, which was used to develop sampling approaches for fish, macroinvertebrates, and aquatic plant assemblages. Comparative sampling at 20 drowned river mouth wetlands utilized existing methods.

Macroinvertebrate comparative sampling included activity traps (Wilcox et al. 1999) and sweep net sampling (Burton et al.1999). Activity traps were positioned in the dominant habitat types so that two pair of traps were set for 24-hr. Sweep net sampling was conducted within the 500 m sampling reach and included 20 sampling efforts divided equally among the dominant habitat types as identified by Burton et al. (1999). Macroinvertebrate sampling showed that either method was equally adept at collecting a representative sample of the aquatic community, however, the two methods differed in the portions of the community sampled. Aquatic plant assemblage indicators included a quantitative transect approach (Albert et al. 1987) and a qualitative rapid approach (Simon et al. 2001).

During 2001, an evaluation of the status of Great Lakes coastal wetlands focused on sampling in every Great Lake and connecting channel. Data from this phase of the project was used to develop multimetric indices of biological integrity for fish assemblages. Drowned river mouth wetlands in each of the Great Lakes and connecting channels were surveyed. Lake Michigan coastal wetlands were compared to an index developed for southern Lake Michigan (Simon 1998). Fish assemblages showed significant differences between lakes, however, nonsignificant differences with respect to Ecoregion and wetland size. Only a few metrics required calibration to account for scale differences based on wetted width. Multimetric indices were developed for Lake Superior, Lake Michigan, Lake Huron, and the St. Clair system. Connecting channel indices were calibrated for the St. Clair and Detroit Rivers and the Niagara and St. Lawrence Rivers. The IBI for Lake Erie (Thoma 1999) was verified using drowned river mouth wetland data from the western and central basins.

Pumpkinseed Lepomis gibbosus