HSPF Application to Microbial Indicator Data

Trent Schade, Orin Shanks, Matthew A. Morrison and Kate Field


From March 2001 to March 2003 researchers at Oregon State University collected water quality samples at regular two-week intervals for 30 sites along five rivers in the Tillamook Basin in Northwestern Oregon. The unique dataset consists primarily of E. coli counts paired with ruminant- and human-specific source markers, that allow for the direct comparison of regulated contaminant levels with fecal source tracking information. The focus of the current study is on the five sampling stations in the Trask River Watershed, where two USGS gauging stations provide a long-term hydrologic record. Using regular time intervals in a water quality sampling protocol produces a set of data randomized in terms of the hydrological conditions for each sampling event. With this water quality data, we can produce a “dosing” type of analysis to make broad inferences about the type, source and pathways for water quality impairment. Additional analysis of the entire dataset showed that during wet periods, there is a positive correlation in some locations between flow and fecal indicators. The current study applies an Hydrological Simulation Program Fortran (HSPF) model of the Trask River Watershed portion of the Tillamook Basin in order to provide a more mechanistic understanding of the fate and transport of fecal pollutants. HSPF can distinguish at short time intervals (hours) over long period (decades) between flow originating as Surface Runoff, Interflow, and Groundwater Flow. Applied to the microbial sampling data, the separated hydrographs may yield a stronger relationship between fecal indicators and hydrologic variables. Further, this work may suggest modifications for the type and placement of best management practices for both human and animal sources of microbial contamination.

Permanent link: