The hydrology of seasonally inundated depression wetlands can be highly sensitive to climatic fluctuations. Hydroperiod—the number of days per year that a wetland is inundated—is often of primary ecological importance in these systems and can vary interannually depending on climate conditions. In this study we re-examined an existing hydrologic model to simulate daily water levels in Sinking Pond, a 35-hectare seasonally inundated karst-depression wetland in Tennessee, USA. We recalibrated the model using 22 years of climate and water-level observations and used the recalibrated model to reconstruct (hindcast) daily water levels over a 165-year period from 1855 to 2019. A trend analysis of the climatic data and reconstructed water levels over the hindcasting period indicated substantial increases in pond hydroperiod over time, apparently related to increasing regional precipitation. Wetland hydroperiod increased on average by 5.9 days per decade between 1920 and 2019, with a breakpoint around the year 1970. Hydroperiod changes of this magnitude may have profound consequences for wetland ecology, such as a transition from a forested wetland to a mostly open-water pond at the Sinking Pond site. More broadly, this study illustrates the needs for robust hydrologic models of depression wetlands and for consideration of model transferability in time (i.e., hindcasting and forecasting) under non-stationary hydroclimatic conditions. As climate change is expected to influence water cycles, hydrologic processes, and wetland ecohydrology in the coming decades, hydrologic model projections may become increasingly important to detect, anticipate, and potentially mitigate ecological impacts in depression wetland ecosystems.
|Title||Increasing hydroperiod in a karst-depression wetland based on 165 years of simulated daily water levels|
|Authors||Jennifer M. Cartwright, William J. Wolfe|
|Publication Subtype||Journal Article|
|Record Source||USGS Publications Warehouse|
|USGS Organization||Lower Mississippi-Gulf Water Science Center|