H13C-1526
Realistically Predicting Saturation-Excess Runoff With El-SWAT

Monday, 14 December 2015
Poster Hall (Moscone South)
Linh Hoang1, Elliot M Schneiderman2, Tammo S Steenhuis3, Karen E Moore2 and Emmet M Owens2, (1)City University of New York/ NYC Department of Environmental Protection, Kingston, NY, United States, (2)New York City Department of Environmental Protection, Kingston, NY, United States, (3)Cornell Univ, Ithaca, NY, United States
Abstract:
Saturation excess runoff (SER) is without doubt the major runoff mechanism in the humid well vegetated areas where infiltration rates often exceed the medium rainfall intensity. Despite its preponderance, incorporating SER in the distributed models has been slow and fraught with difficulties. The short term objective of this paper to adjust the generally used Soil and Water Assessment Tool (SWAT) to include SER and test the results in the Catskill Mountains that is the source of most of New York City’s water. The long term goal is to use the adjusted distributed runoff mechanism in water quality models to aid in the design of effective management practices. The current version of SWAT uses information of soil plant characteristics and hydrologic condition to predict runoff and thus is implicitly based on infiltration-excess runoff. Previous attempts to incorporate SER mechanism in SWAT fell short because they were unable to distribute water from a Hydrological Response Unit (HRU) to another. In the current version called El-SWAT, this shortcoming has been overcome by redefining HRU to include landscape position through the topographic index, grouping the newly defined HRU into wetness classes and by introducing a perched water table with the ability to route interflow from “dryer” to “wetter” HRU wetness classes. Mathematically, the perched aquifer is a non-linear reservoir that generates rapid subsurface stormflow as the perched water table rises. The El-SWAT model was tested in the Town Brook watershed in the upper reaches of the West Branch of the Delaware in the Catskill Mountains. The results showed that El-SWAT could predict discharge well with Nash-Sutcliffe Efficiency of 0.69 and 0.84 for daily and monthly time steps. Compared to the original SWAT model, El-SWAT predicted less surface runoff and groundwater flow and a greater lateral flow component. The saturated areas in El-SWAT were concentrated in locations with high topographic index and was in agreement with field observations. With the incorporation of topography characteristics and the addition of the perched aquifer, El-SWAT gives a realistic representation of hydrological processes and will lead to better water quality models where the source of the surface runoff matters.