2014 GSA Annual Meeting in Vancouver, British Columbia (19–22 October 2014)

Paper No. 295-9
Presentation Time: 11:00 AM

MODELING THE HYDROLOGICAL EFFECTS OF WETLAND RESTORATION IN THE LE SUEUR WATERSHED WITH SWAT


MITCHELL, Nathaniel A.1, CHO, Se Jong2, DALZELL, Brent J.3, KUMARASAMY, Karthik4 and GRAN, Karen1, (1)Dept. of Earth and Environmental Sciences, University of Minnesota - Duluth, 1114 Kirby Dr, Duluth, MN 55812, (2)Dept. of Geography and Environmental Engineering, John Hopkins University, 3400 North Charles St, Baltimore, MD 21218, (3)Soil, Water, and Climate, University of Minnesota, 439 Borlaug Hall, 1991 Upper Buford Circle, St. Paul, MN 55108, (4)Department of Watershed Sciences, Utah State University, 5210 Old Main Hill, Logan, UT 84321

Documented increases in stream flows, near-channel erosion rates, and turbidity levels within the agricultural Le Sueur Watershed (LSW) in central Minnesota have motivated the discussion of potential hydrologic management options to reduce peak flows including wetland restoration. Wetlands may be able to increase the residence time of water on the landscape, increase evapotranspiration and seepage, reduce water input to the channel network, and reduce peak flows. This study uses the Soil and Water Assessment Tool (SWAT) to evaluate the extent to which peak flows entering the lower, incised reaches of the LSW can be reduced through wetland restoration scenarios with different extents and placements in the watershed. The projected flow reductions are then used to guide inputs in an erosion model for near-channel sediment sources (e.g., stream banks, bluffs, and channel incision) in the lower incised valley. This sediment model is being used in stakeholder meetings to aid discussions of regional management strategies. Wetland sites were selected based on their topography, agricultural productivity, and land use. Results to date show that the methods used in the derivation of certain SWAT parameters, such as the fraction of discharge input to wetlands (WET_FR) and the hydraulic conductivity of the wetlands’ bottoms (WET_K), are critically important to modeling results and their interpretation. While many authors have made broad assumptions regarding the WET_FR parameter, these assumptions can lead to values that are unrepresentative of the sites and their topographic surroundings. This study has developed new, physically-based methods for the derivation of WET_FR values. Utilizing a range of hydraulic conductivity values that are feasible within the study area produces large changes in wetland behavior, often with appreciable flow reductions only occurring with higher conductivity values. Hydraulic conductivity is often the limiting factor in the flow reductions offered by wetland restoration scenarios. Findings with different scenarios may help determine the potential success of implementing distributed stream flow management systems capable of reducing stream flows, erosion of near-channel features, and resulting turbidity levels.