Please use this identifier to cite or link to this item:
|Title:||Improving the Wetland Zooplankton Index for application to Georgian Bay coastal wetlands|
|Authors:||Yantsis, Stephanie N.|
|Abstract:||<p>In 1996, the State of the Lakes Ecological Conference (SOLEC) stressed the need to create ecological indicators to monitor Great Lakes wetlands. Since then, a suite of indicators have been created by researchers throughout the Great Lakes using different environmental parameters such as water quality (Water Quality index, WQI, Chow-Fraser 2006; Agriculture PCl, Danz et al. 2007), fish (Wetland Fish Index, WFI, Seilheimer and Chow-Fraser 2006,2007), plants (Plant Index of Biotic Integrity, Rothrock and Simon 2006, 2008; Wetland Macrophyte Index, WMI, Croft and Chow-Fraser 2007) and zooplankton (Wetland Zooplankton Index, WZI, Lougheed and Chow-Fraser 2002). In a recent study, Seilheimer et aI. (2009) found that the WQI had a significant linear relationship with both the WFI and WMI, but not with the WZI. They showed that the WZI was not able to discern the pristine nature of wetlands in Georgian Bay, Lake Huron, where there is minimal human disturbance. As the first objective of my thesis, I investigate three possible reasons for this poor performance. I investigated whether the lower than expected WZI scores associated with high-quality Georgian Bay sites could have been due to 1) inadequate sampling effort 2) inclusion of highly exposed sites or 3) lack of representation of Georgian Bay sites in the development of the WZI. Using data from the Chow-Fraser database, as well as analyzing addition samples from the zooplankton archive, zooplankton abundance data was used to analyze my hypotheses. Increasing sampling effort from 1 to 5 samples per wetland did not lead to significantly higher WZI scores, even though species richness increased with sampling effort. Including Georgian Bay sites with high degree of exposure to wind and wave action did not significantly decrease WZI scores, although there was a trend towards lower overall abundance of zooplankton for exposed sites. I found strong support for the third reason, that the original development of the WZI had biased the index parameters against Georgian Bay sites. This was confirmed when I employed the same statistical approach to an expanded database that included 63 of the original 70 wetland-years along with 31 new wetland-years in Georgian Bay and 45 others in Lakes Erie and Ontario. Using the results of Partial Canonical Correspondence Analysis (pCCA), I made 5 modifications to the WZI optimum (U) and tolerance (T) values. Using an independent dataset, I found that the modified WZI (WZI09) scores were linearly related to WQI (r2=0.283; P < 0.0001; n= 50). The second objective of my thesis was to investigate whether or not aquatic macrophyte information was a stronger predictor of zooplankton community than water-quality information. I compared the percent fit of data from a co-correspondence analysis (CO-CA) of zooplankton abundance data and plant presence/absence data and a correspondence analysis (CCA) of zooplankton abundance and environmental data. Results indicated that plants were not a better predictor of zooplankton distribution than environmental variables (CO-CA: 12.8%, CCA: 13.3%, n=107). I therefore conclude that the modifications of the WZI09 have resulted in an improved indicator that can be used in tandem with other indicators to determine wetlands health throughout the Canadian shoreline of the Great Lakes, including Georgian Bay.</p>|
|Appears in Collections:||Open Access Dissertations and Theses|
Items in MacSphere are protected by copyright, with all rights reserved, unless otherwise indicated.