Can Laboratory Column Studies Really Predict Field-Scale BMP Performance? A Comparison of 10 Years of Field-Scale Performance Data With Original Bench-Scale Test Results

Date / Time:
Tuesday, Sep 15 11:00am to 11:30am
Track / Session:
Track: Stormwater Infrastructure and Natural Waterways / Session 1

There is much published research describing laboratory tests on the performance and selection of filtration media to target pollutants in stormwater controls, however, comparisons of laboratory test results to long-term field performance of full-scale stormwater controls are uncommon. Similarly, there is uncertainty about the long-term performance and maintenance needs of these stormwater controls since most laboratory are short-term and do not operate until reaching clogging or breakthrough under actual site loading rates, and since most field-scale studies have been limited to a season or two of performance monitoring.

To begin filling these critical knowledge gaps, this presentation compares the laboratory performance of a Sand-Zeolite-Granular Activated Carbon (GAC) treatment media blend with the results of long-term monitoring of full-scale stormwater controls using the same media at The Boeing Company’s Santa Susana Field Laboratory (SSFL). Initial column studies were performed at a Penn State - Harrisburg laboratory to identify an optimal media mix and contact time that would remove site constituents of concern (primarily metals and dioxins) and also provide a longer “time to clogging” to extend the lifetime of the media in field stormwater controls. The preferred media mix was selected and implemented in biofilter and media filter stormwater controls distributed throughout a NPDES-regulated watershed at the SSFL where centralized end-of-pipe treatment was not feasible. To confirm the continued performance of implemented controls, routine sampling has occurred since 2010, providing rare long-term insight into control performance.

This presentation compares the laboratory and long-term field performance through statistical evaluations of influent and effluent concentrations for TSS, dioxins, copper, lead, cadmium, and chromium. This comparison highlights the importance of using representative influent samples in the laboratory to best represent the wide range of expected site stormwater conditions (a challenge when using stored water and synthetic stormwater), both in general chemical composition and in targeted chemical influent concentrations, in order to best predict full-scale implementation performance.

This presentation also summarizes the long-term trends to assess signs of chemical breakthrough and clogging, which is a rare opportunity that is afforded by highly robust performance monitoring data. Influent sediment load estimates were calculated for each control using measured influent TSS concentrations and rainfall, and modeling estimates of runoff volume and percent capture. Time to clogging was estimated using the cumulative TSS loading per unit area of media. Remarkably, the lab-based estimates accurately predicted the clogging of one stormwater control that had its media replaced in the following year. Additionally, this analysis correctly predicted that there has been no sign of breakthrough, and showed the remaining time that each stormwater control has until they are expected to clog (a range of 2 to 20+ years).

Finally, this presentation addresses the conference theme of sustainability through lessons learned from the laboratory study and subsequent long-term implementation of stormwater controls. This information can be used to guide the design and selection of stormwater control media at other locations to better estimate long-term pollutant removal, to improve lab column study design and results extrapolation, to improve estimates of useful life of stormwater controls before major maintenance, to inform stormwater control sizing based on recommended contact times, and to balance the cost of pretreatment and maintenance frequency based on anticipated sediment loading. These planning and evaluation methods can be applied by audience members who will also be invited to share some of their tricky or surprising stormwater control testing or performance experiences.

Primary Speaker:
Bob Pitt, University of Alabama
Bob Pitt is the Emeritus Cudworth Professor of Urban Water Systems at the University of Alabama. He has more than 50 years of industrial and academic experience researching the effects, sources, and control of urban runoff. He and his graduate students have conducted research on integrating green infrastructure controls in combined sewer areas; construction site erosion characterization and control; stormwater treatment using media filtration; urban PAH sources and fates; heavy metal releases from materials; groundwater impacts from stormwater infiltration; beneficial uses of stormwater; sources of indicator bacteria; and continued work on enhancements to the Source Loading and Management Model (WinSLAMM).
Supporting Speaker 1:
Maia Colyar, Geosyntec Consultants
Maia Colyar is a water resources engineer at Geosyntec Consultants in Santa Barbara. She is experienced in leveraging hydrologic modeling, data analysis, and field visits to support National Pollutant Discharge Elimination System (NPDES) and Industrial General Permit (IGP) regulatory compliance including clean water act litigation support, as well as Best Management Practice (BMP) planning and design. She applies her technical background in chemistry, hydrology, and engineering to find solutions for complex water quality, stormwater, and watershed management problems.