A coupled surface water storage and subsurface water dynamics model in SWAT for characterizing hydroperiod of geographically isolated wetlands

Authors: QI, J. - University Of Maryland; ZHANG, X. - University Of Maryland; LEE, S. - University Of Maryland; Moglen, Glenn; Sadeghi, Ali; McCarty, Gregory

Submitted to: Advances in Water Resources
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 7/10/2019
Publication Date: 9/1/2019
Citation: Qi, J., Zhang, X., Lee, S., Moglen, G.E., Sadeghi, A.M., McCarty, G.W. 2019. A coupled surface water storage and subsurface water dynamics model in SWAT for characterizing hydroperiod of geographically isolated wetlands. Advances in Water Resources. https://doi.org/10.1016/j.advwatres.2019.103380.
DOI: https://doi.org/10.1016/j.advwatres.2019.103380

Interpretive Summary: Wetland hydroperiod, the duration of inundation, is an important parameter driving biogeochemical cycles important for wetland ecosystem function. However, this parameter is often unevaluated in models and difficult to predict at field and watershed scales. This paper reports the development of a fully-coupled surface water storage and subsurface water dynamics module within the Soil and Water Assessment Tool (SWAT) model. The hydroperiod module of this model was calibrated and validated using high resolution temporal data that tracked surface and subsurface water dynamics on four wetlands (two restored and two natural) located on the Delmarva peninsula. The results showed that the hydroperiod module predicted the dynamics of wetland hydrology with good accuracy during both wet and dry periods. Predictions for natural wetlands tended to be better than for restored wetlands because of better wetland vegetation characterizations in the natural ecosystems. This hydroperiod module is now ready to be used at watershed scales for better representation of wetland biogeochemistry and greater understanding of the impacts of wetlands on carbon storage and sequestration in agricultural watersheds.

Technical Abstract: The hydrological modeling of wetlands is important for reliable estimation of biogeochemical processes in soils subject to periodically inundating conditions. The present study has developed a wetland module in the Richards-equation-based Soil and Water Assessment Tool (SWAT) to couple the surface water storage and soil water dynamics. The wetland module was tested using observed daily water level data from four wetlands (including restored and natural wetlands with and without impermeable soil layers) on the Delmarva Peninsula, USA. After the wetland module was calibrated, simulated daily water level and observed data were compared using two statistics: percent bias (Pbias) and Nash-Sutcliffe coefficient (NS) from 2016 to 2017. The results show that, in general, the wetland module accurately reproduced hydroperiods for both restored and natural wetlands with and without impermeable soil layers; specifically, the module was able to accurately model saturation conditions for different soil layers corresponding to wet and dry periods; the wetland module had the tendency to generate better results for natural wetlands likely because restored wetlands tended to have mixed plant types which caused difficulty for accurate estimation of evapotranspiration. The ability to describe inundation conditions for wetlands is important for biogeochemical modeling so that the newly developed wetland module has great potential in enhancing simulation of biogeochemical cycles not only at the site scale but also at the watershed scale.