Assessment of water budgets and the hydrologic performance of a created mitigation wetland—A modeling approach

The study used a water balance model (DRAINMOD) to compute water budgets of a mitigation wetland created in the Piedmont region of Virginia. The calibration of the model was conducted with automated well data collected during the 17 month monitoring period along with precipitation, temperature, soil physical properties (soil water characteristic curve and saturated hydraulic conductivity) and estimated site characteristics (surface roughness and surface storage). The model was tested for two areas, one nondisturbed and the other disturbed, by construction practices commonly adopted for a mitigation wetland created in the region. A third model was created to represent the disturbed boundary conditions (wetland design), but substituted soil data observed at the nondisturbed study area. DRAINMOD successfully predicted the hydrologic regimes of both nondisturbed and disturbed areas. The model of the nondisturbed area could not accurately predict hydrology of the disturbed area. More importantly, the model of the disturbed area with the soils data from the nondisturbed area could not accurately predict the hydrology of the disturbed area. The models were used to evaluate a set of performance criteria across a 60-year (1952 to 2011) simulation period. Ponding for longer than 60 consecutive days during the growing season occurred at the disturbed study area in 39 out of 60 years and these conditions lasted the entire growing season (219 days) in multiple years. Prolonged inundation of the surface for longer than 100 consecutive days took place in at least 15 of the years simulated compared to two years in the nondisturbed model. The modified disturbed model (using nondisturbed soil data) satisfied jurisdictional hydrology more frequently compared to the disturbed model (33 years versus 22 years, respectively) and prolonged inundation was limited to 8 years during the simulation period with the longest single event lasting 168 consecutive days. The differences were attributed to the reduced drainable porosity and vertical saturated hydraulic conductivity in the disturbed wetland area which translated to a demand for surface storage in order to achieve accurate model calibration and jurisdictional wetland hydrology. The study shows that disturbance to key soil properties will require surface storage to achieve jurisdictional hydrology, and that construction practices can result in longer durations of ponding during the growing season, thus potentially altering the habitat type for the wetland from what was originally designed (e.g., from a forested wetland to open water or emergent habitats).