Developing a Bacteria Total Maximum Daily Load for the Big Sioux River

A bacteria Total Maximum Daily Load (TMDL) assessment was completed on four reaches of the Big Sioux River from Dell Rapids to Brandon, South Dakota. The project area was unique in that it encompassed a mixture of rich agricultural land as well as the densely populated area within and surrounding the city of Sioux Falls, a Phase I National Pollutant Discharge Elimination System (NPDES) community. The challenge of the project was to separate the impact the city of Sioux Falls is having on the overall bacteria loading from that originating outside the city limits. Furthermore, quantification of the relative loadings from each of the land uses within the city (industrial, commercial, residential) and below existing Best Management Practices (BMPs) needed to be assessed to understand the effectiveness of the city's current stormwater treatment capability and to properly prioritize future BMP implementation. Implementing BMPs within an NPDES community is different than those for a nonpermitted source-driven load in that adoption is regulated rather than voluntary, making the accurate quantification of the water-quality impacts and associated BMP recommendations imperative.The assessment was completed in three phases: planning, monitoring, and TMDL development. During the planning phase, available water-quality and-quantity data, along with relevant background watershed data, were compiled to gain a basic understanding of the impairment, waterbodies, and watershed characteristics affecting impairment. The subsequent monitoring phase involved implementing an adaptive monitoring plan that filled in any gaps in the historical data as identified in Phase I. The final phase of the project involved calibrating and validating a watershed model application, using the data gathered in the first two phases, to develop the TMDL document and identify BMP scenarios that will bring the Big Sioux River into compliance with water-quality standards.For the monitoring phase, 20 sampling sites were selected to characterize the water quality and loading sources in the city of Sioux Falls. Water-quality sites fell into three general categories: mainstem sites, tributary, and stormwater outfall sites. Mainstem sites were sites located on the mainstem Big Sioux River beginning upstream of Sioux Falls and continuing downstream to near Brandon, South Dakota. Three main tributaries were sampled: Skunk Creek, Silver Creek, and Slip-Up Creek. At each of the mainstem and tributary sites, weekly grabs samples were collected as well as storm event samples using automatic samplers equipped with cell–phone modems that allowed sampling to be remotely triggered. There were nine stormwater outfall sites with three automatic samplers that rotated between sites for a total of two samples at each site. Additional stage-only sites were added with continuous-stage recording equipment installed to better characterize the hydrology through the city. Extensive quality assurance/quality control (QA/QC) procedures were in place for this project, including specific procedures designed to measure the quality of bacteria data collected using ISCO samplers. To further support TMDL development, a watershed model was developed using the Hydrologic Simulation Program - FORTRAN (HSPF) to better characterize the critical conditions, perform the source allocation, and understand the impact alternative management/BMPs would have in achieving the reductions prescribed by the TMDL.The focus of this paper is to discuss the results of the monitoring and watershed modeling and how this information was used in developing the TMDL and prioritizing future BMPs.