Fixing a Flat: The Third Year of Post-Restoration Monitoring at Shiawassee National Wildlife Refuge

Abstract

The Shiawassee floodplain is a hybrid floodplain located in mid-Michigan and is home to our study site, the Shiawassee National Wildlife Refuge (SNWR). The refuge was established in 1953 to designate an area for migratory birds to seek refuge during migration. This location is topographically quite flat and is the convergence of four different rivers, thus making it very prone to flooding. These factors drove many historic farmers to dike the tributaries of these four rivers to prevent their crop fields from flooding with water. Historic diking caused SNWR to be disconnected from the Shiawassee River. Through the Great Lakes Restoration Initiative, SNWR received funding to start a restoration project to reconnect these two features. A team was formed to study the effects of restoring this wetland and reconnecting it to the river. This team includes the United States Fish and Wildlife Service (recipients of funding and managers of this refuge), the United States Geological Survey (federal researchers with expertise in this area), and a master’s team from the University of Michigan (the person power to drive this study). 2021 marks the third year of consecutive monitoring for this study and is crucial for understanding 1) the multiple life cycles of taxa present at the refuge, 2) successional changes over monitored years; while 3) establishing that funding for this project has been effectively used, and 4) providing future monitoring teams with robust data to better predict the future of SNWR. The team set out to learn how abiotic and biotic elements change across refuge wetland units of Maankiki South (MS), Maankiki North (MN), Maankiki Center (MC), Pool 1A (P1A), and Shiawassee River (SHR) and sampling months, describe the relationships between these elements, and to describe changes that have occurred across the three years of monitoring. The elements sampled included water quality, vegetation, aquatic macroinvertebrates, and fishes. Water Quality: We describe each water quality variable sampled (temperature, dissolved oxygen, conductivity, turbidity, pH, and nitrogen and phosphorous levels) and provide statistical analyses of each variable across units, seasons, and in comparison to 2019 and 2020 data. ● We found that refuge wetland units experienced harsh conditions of high average temperatures and low dissolved oxygen levels in the summer season, but refugia zones for aquatic organisms when there were unfavorable conditions elsewhere also existed since water quality conditions varied across different sampling sites. ● Future recommendations include using floodplain wetland statistical tools for analysis, fuzzy coding analyses, and sampling more evenly across vegetation zones. Fish: We describe the abundance, composition, and quality of fish communities throughout three previously studied units and two new locations within SNWR, and compare these findings to those of the 2019 and 2020 sampling seasons. We also analyzed these data in their relation to water quality, vegetation, location, and season. ● We conducted fish sampling through multiple frame and mesh size fyke nets, and electrofishing. ● We found more species than 2020 and similar amounts to 2019, though unique species were found in all years. Most unique species in 2021 came from the Shiawassee River. ● We found that community composition across all units were homogeneous, which we attribute to harsh conditions only being tolerable to a few wetland adapted species. ● We found the same abundant species as both 2019 and 2020, with the most abundant species varying by unit but with Black Bullhead, Bowfin, Pumpkinseed, Bluegill, and YOY sunfish being some of the most abundant species across all units. ● Future recommendations include continued sampling to build robust datasets, water depth analysis, and tracking of fish between the Shiawassee River and SNWR. Our research has implications for ecosystem variability and succession, as we found both increases and decreases in habitat quality, abundant species, and community compositions. This third year of sampling has shown evidence of succession following a large disturbance and provides insight into how restored wetlands vary over multiple years.