|Impact of Best Management Practices on Water Quality of Two Small Watersheds in Indiana: Role of Spatial Scale (EPA/600/R-05/080) July 2004
Transport and fate of sediments and nutrients within watersheds have important implications for water quality and water resources. Water quality issues often arise because sediments are carriers of various pollutants, such as pathogens, toxic substances, and excessive nutrients.
Section 303(d) of the Clean Water Act requires all states to develop and implement a total maximum daily load for their impaired water resources. Implementation of best management practices (BMPs) is a conventional approach for controlling nonpoint sources of sediments and nutrients. However, implementation of BMPs has rarely been monitored long-term to study the effectiveness of the BMPs. Long-term data on flow and water quality within watersheds, before and after implementation of BMPs, are not generally available.
Mathematical models provide an effective and powerful tool for evaluating long-term performance of BMPs, especially new ones that have had little or no history of use. In this study, a process-based modeling framework was developed to evaluate the effectiveness of parallel terraces, field borders, grassed waterways, and grade stabilization structures in reducing sediment and nutrient yields in two small agricultural watersheds (less than 10 square kilometers) in Indiana. The Soil and Water Assessment Tool (SWAT) served as the watershed model.
Based on the functionality of each BMP, appropriate model parameters were selected and altered to represent the effect of the BMP on hydrologic and water quality processes. An analysis was performed to evaluate the sensitivity of model computations to selected parameters. Results indicated that parallel terraces and field borders were effective at a field scale, while grassed waterways and grade stabilization structures were more effective at a watershed scale.
Distributed-parameter models partitioned the watershed into subunits (subwatersheds, hyrologic response units, and grids) during computations to represent heterogeneity within the watershed. Homogeneous properties were assumed for each computational unit. Identification of the stream network and partitioning of the study area into subunits may significantly affect hydrologic and water quality simulations of a distributed-parameter model.
Because model outputs are influenced by geomorphologic resolution, the evaluation of performance of BMPs based on model predictions will be influenced as well. Thus, examination of the efficacy of BMPs must be conducted in conjunction with studies performed at multiple spatial scales.
In this study, sediment and nutrient outputs from the calibrated SWAT model are compared at various watershed discretization levels, both with and without implementation of these BMPs. Results indicated that evaluation of the impacts of these BMPs on sediment and nutrient yields at the outlet of the two agricultural watersheds was very sensitive to the level of discretization applied for modeling. An optimal watershed discretization level for representation of the BMPs was identified through numerical simulations. It appears that the average subwatershed area corresponding to approximately 4 percent of the total watershed area is needed to represent the influence of BMPs in a modeling effort.
The results of this study depend on location and on the type of BMPs. However, the methodology can be used for similar studies in other watersheds with different BMPs.
You will need Adobe Reader to view some of the files on this page.