Evaluating Storm Water Impacts-Monitoring the Receiving Environment Using a Floating Bioassay Laboratory System

The U.S. Navy is conducting an evaluation of impacts from facility storm water discharges to San Diego Bay. The investigation was prompted by the implementation of local regulations that require a 90% survival rate of fish or mysid shrimp in acute toxicity tests using undiluted storm water. An underlying conceptual approach was to monitor toxicity directly in receiving waters as well as in the undiluted discharge to evaluate impacts. Data collected to date have shown a full range in toxic response in outfall discharge samples. No toxic effects, however, have been observed in bay waters collected immediately outside the outfalls. These results, along with plume mapping, have suggested that the relatively small magnitude and ephemeral nature of these discharges were sufficient to explain the removal of toxicity of the storm discharge once it reaches the bay. One of the outstanding issues presented by standard toxicity testing is the relevance of 48- or 96-h exposure times to test organisms when actual storm exposures likely occur over much shorter times. To investigate this issue, we conducted toxicity tests with a boat-mounted flow-through bioassay system, which was positioned immediately outside an outfall before, during, and after a storm event. The bioassays included survival of the mysid (Americamysis bahia) and topsmelt (Atherinops affinis) as well as embryo-larval development of the mussel (Mytilus galloprovincialis). Surface bay water was continuously pumped to the test organism containers for the full 48- (mussel) or 96-h (mysid and topsmelt) exposure requirement. Bay water was analyzed continuously for salinity, temperature, dissolved oxygen, pH, light transmission, oil fluorescence, copper, and zinc. Additionally, dilution series toxicity and chemistry were conducted on first-flush and composite samples taken from the outfall prior to discharge. The floating-bioassay system results were consistent with previous monitoring that indicated toxicity of first-fl- - ush discharges but no toxicity in the receiving environment. Continuous monitoring showed that storm water was completely mixed out within minutes of discharging to the bay even though the observation point was only 15 feet away from the outfall. The reduction of toxicity in the receiving environment was a result of the very limited time exposure that occurs with this type of discharge