Assessing the effects of contaminants on restored wetland macroinvertebrate communities and a comparison of two wetland quality indices

Abstract

Inland wetlands are vital ecosystems that provide numerous services to humans and wildlife, including habitat, flood control, sediment trapping, and nutrient cycling. In the Midwest, the primary motivation for historic wetland drainage was agricultural production. In recent decades, conservationists have made considerable efforts to restore and create lost wetlands; however, the presence of these wetlands in agriculturally dominated landscapes means they are subject to anthropogenic stressors like contamination by nutrients and pesticides. This study examined the factors influencing macroinvertebrate and aquatic vegetation community structure in restored agricultural wetlands. I also compared the ability of two biotic indices to track anthropogenic impact and examined drivers of waterfowl site occupancy. In 2021 and 2022, wetland community and habitat structure were surveyed in 41 ponds on 27 conservation easements throughout central Wisconsin. Our findings indicate that pond depth affected vegetation and macroinvertebrate community composition. However, these effects may not be important for conservation as there was no relationship between pond depth and species richness, diversity, or biotic index scores. Additionally, I found that neither the Wisconsin Wetland Macroinvertebrate Biotic Index (WWMBI) nor the Wisconsin Floristic Quality Index (wFQI) correlated with a measure of human disturbance known as the human disturbance gradient. Results indicate that invertebrate communities may be more strongly affected by phosphorus concentration and fish presence than any other factor. Waterfowl site occupancy was modeled against several habitat characteristics and quality metrics for Mallards (Anas platyrhynchos), Blue-winged teal (Spatula discors), and Wood Ducks (Aix sponsa). Results show that survey date most heavily influenced detection, and that wFQI score and pond size may influence occupancy estimates.