The influence of extreme water levels on coastal wetland extent across the Laurentian Great Lakes

Laurentian Great Lakes coastal wetlands (GLCW) are ecological hotspots and their integrity depends upon dynamic hydrologic regimes of the Great Lakes. GLCW naturally adjust to changes in hydrologic regimes via migration, but Great Lakes water levels may be shifting faster than wetlands can manage: 2000–2015 marked an extended low water level period and was followed by record highs in 2017–2020. Our objective was to quantify how Great Lakes water levels impact GLCW linear extent (from the shoreline to open water). We calculated wetland extent and migration from 2011 to 2019 using data from 1538 vegetation transects at 342 sites across the U.S. shoreline of the Great Lakes. Mediated multiple linear regression with Bayesian hierarchical modeling investigated the relationship between water levels and wetland extent. We employed Bayesian hierarchical modeling because (1) the dataset was spatially nested, with sampling points within wetlands within Great Lakes and (2) Bayesian statistics offer flexibility for environmental modeling, such as the inclusion of mediation in models, where we can assess both direct influences of Great Lake water levels on wetland extent and indirect (i.e., mediated) influences of water levels via the presence of vegetation zones on thus wetland extent. Results showed that, overall, there was a landward migration from 2011 to 2019 (although 38 % of wetlands had lakeward migration of the wetland-upland border). Wetland length and inundation length decreased with increased water levels, as mediated by the presence of certain vegetation zones. This decrease in wetland extent is of concern because it likely relates to a decrease in wetland function and habitat. A better understanding of how GLCW migrate with shifts in water levels enables decision makers to better predict where Great Lakes coastal wetlands are at risk of being lost and thus where to prioritize management efforts.