The Department of Defense (DoD) is under increasing pressure from regulators and local communities to reduce the amount of toxic pollutants being discharged with storm water into harbors, bays, lakes, and streams. Successful completion of this project will provide the DoD with a method of removing toxic contaminants from storm water runoff, thereby avoiding Notices of Violation from regulating agencies and improving public perception of DoD environmental stewardship.

Storm water runoff from DoD industrial sites is not easily treated by current commercial off-the-shelf (COTS) technology. Most COTS technologies for storm water treatment are designed for municipal applications such as trash, nutrient, and sediment removal. Also, many storm water treatment technologies require large areas of land for detention basins and similar structures. This type of space requirement is often at a premium at DoD sites, and is especially unavailable at many industrial locations.

The dual media storm water filtration system is inherently simple. It is based on a standard sand filter design used for treating storm water runoff. The original filter medium (sand) is replaced with inexpensive adsorbent materials. The system operates without manual intervention, has no pumps, controls, or other sophisticated mechanical or electrical components. Annual maintenance consists of removing and replacing top layer of pea gravel and geo-fabric, which strains sediment from storm water runoff as it enters the media bed.

The objective of this project was to validate the dual media storm water filtration system’s capability to remove metals, organic compounds, and other toxic pollutants in storm water runoff. The specific objective of the Navy Regional Recycling Center (NRRC) demonstration was to validate a technology that passes a 96-hour bioassay (toxicity) test as required by the California Regional Water Quality Control Board (CRWQCB), San Diego Region by removing copper and zinc from the runoff water. The demonstration performance goal was to reduce copper to less than 63.6 μg/L and zinc to less than 117 μg/L. The specific objective of the Anniston Army Depot (ANAD) demonstration was to comply with their storm water permit requiring that “The discharge shall have no sheen, and there shall be no discharge of visible oil, floating solids, or visible foam in other than trace amounts.” In addition, a secondary objective was to further validate the filters’ capability to remove zinc and copper similarly to NRRC even though it is not specifically required in ANAD’s storm water permit.

The technology utilized in these demonstrations was the combination of simple, low cost storm water treatment technology (sand filters) reconfigured with engineered materials (adsorbents) specifically selected to remove targeted industrial pollutants that makes the demonstrated technology innovative. Sand filters mimic natural sediment traps to trap particles of contaminating materials. Sand filters have been used in the past to treat storm water runoff from shopping center parking lots, residential areas, and other non-industrial applications. The main application has been to remove suspended solids. In these demonstrations, the design was modified to allow the use of special adsorbent materials to increase the efficiency of removal of metals and organic compounds.

For the 11 storm events that occurred during the demonstration period (March 2006 to April 2007), the dual media storm water filter system passed the 90% survival requirement 64% of the time, and passed the 70% survival requirement 82% of the time. All of the acute toxicity test requirements were met 100% of the time for the last five storm events after modifications were made to the configuration of the top fabric layer of the media bed. The average removal efficiencies (efficiency ratio) for the last five storm events of the demonstration period met the permit requirements for aluminum, copper, and lead, and was within 4% of the 117 μg/L limit for zinc. The effluent sampling results from the ANAD demonstration showed excellent pollutant removal efficiency. No sheen, visible oil, floating solids, or visible foam was observed in the effluent throughout the demonstration period. Naval Facilities Engineering Service Center (NAVFAC ESC) also believes that the performance of the system at ANAD is similar to the NRRC results, although not specifically required by the state of Alabama.

Both the NRRC and ANAD systems met nearly all of the performance criteria. Both systems were shown to be effective, reliable, and ran operator-free during the demonstration. The bioassay tests on the NRRC system were met once modifications were made to the filter media design, and while the concentrations of copper and zinc were initially high, the concentrations were lowered dramatically after the system modification. Additionally, the 122 μg/L zinc concentration falls within 4% of the 117 μg/L performance criteria. 

The major factor that affected the performance of the NRRC unit was that it was hypothesized that storm water was bypassing the media bed along the perimeter walls of the vaults (edge effects), resulting in partial treatment of the influent, and non-optimal media bed performance. Additional flaps of geofabric were glued to the perimeter walls of the vaults, above an existing sheet of geofabric, to decrease edge effects and redirect the influent to flow through the center of the media bed. The reconfiguration did redirect flow through the center of the media bed, but significantly decreased the porosity of the top fabric layer. Therefore, the top layer of geofabric was replaced with a more porous mesh to reduce the flow restriction through the top fabric layer, and prevent premature clogging of the media bed. The high removal efficiencies for copper and zinc are a direct result of minimizing undesirable edge effects, and redirecting flow through the center of the media bed. The 45% survivability rate observed during toxicity testing could be attributed to low dissolved oxygen levels in the acute toxicity test chambers, but as Nautilus Environmental states, it is not clear whether low dissolved oxygen levels or a toxic substance in the sample caused mortality.