Copper and zinc are two of the most ubiquitous contaminants found in many industrial and nonpoint source effluents that enter the marine environment. The sources of these toxic metals include discharges from facilities, ships, and small craft, as well as inputs from sediment fluxes and sediments disrupted during dredging operations and ship movements. Potential Department of Defense (DoD) sources of copper include storm waters, point sources, hull coatings, and discharges from DoD ships and facilities. Previous studies have shown that copper and zinc are highly toxic to some marine organisms. Copper and zinc discharges often exceed existing water quality criteria (WQC) or standards in the effluent and copper often exceeds WQC in the receiving systems. Compliance and cleanup actions associated with copper contamination are common at DoD and Navy facilities around the country. Regulatory compliance is challenging because of the many sources, both natural and anthropogenic, and the adoption of conservative water quality standards (WQS). Present WQC for these metals are based on concentrations of total or dissolved copper. In contrast, a large body of scientific data indicates that it is the concentration of the "free" or aqueous species (i.e., Cu(II)_aq) that correlates most closely with the toxicity of marine organisms.

The objective of this project was to develop a methodology for predicting the geochemical fate and ecological effects of copper and zinc in coastal embayments. Technical objectives were to (1) establish the overall copper budget in the San Diego Bay for use in developing a model that accounts for the non-conservative characteristics of copper, (2) evaluate the relationship between various copper species in a prototype system, (3) relate the observed speciation and lability to a range of biological and ecological indicators of bay health, (4) examine the seasonal variability of these processes, and (5) perform initial examinations of the distribution and lability of zinc.

San Diego Bay was studied as a prototype system, as it provided a unique range of hydrological conditions with a relatively constant distribution of total copper concentrations and well-defined chronic sources of copper. The bay was divided into 25 boxes or cells of about 1 km scale that matched the boxes used for the modeling effort. Also, there was a box for each Shelter Island and Commercial Basin, which are semi-enclosed marinas within the bay. Six sampling campaigns were performed in order to study spatial and temporal distributions of parameters indicating the health of the bay, as well as toxicity, complexation capacity, and physical, biological, ecological, and chemical conditions. The field investigations employed a combination of real-time and laboratory analytical tools to determine the bay-wide distribution of total copper and important fractions of the copper pool. These spatial distributions of copper in the bay reflected the balance of sources, flushing, and losses to the sediment. Modeling effort on these distributions allowed the development of an algorithm able to predict copper distributions and toxicity. This algorithm can also be used to estimate the effect on copper toxicity as a result of changes in the sources of copper to the bay.

The field program provided a unique and comprehensive view of water quality, with respect to copper and zinc, in relation to the hydrodynamics and residence time in the San Diego Bay over a period of approximately 2 years. The field effort was successful in establishing baseline water quality conditions and copper concentrations throughout the bay and in identifying locations and extent of contaminants. As part of this effort, a methodology was developed for measuring complexation capacity by ion selective electrode (ISE) in waters from marine harbors. This methodology was complemented with differential pulse anodic stripping voltammetry (DPASV) to measure the spatial and temporal variation of copper complexation capacity in the bay, which were examined in relation to other characteristics of the bay and toxicity. This analysis corroborated the use of the free ion model (FIM), as the concentration of the free ion (Cu(II)_aq) is the parameter most indicative of toxic conditions in San Diego Bay. Spatial and temporal variations of copper and zinc-spiked bay water toxicity to larvae of Mytilus galloprovincialis, Dendraster excentricus, and Strongylocentrotus purpuratu were characterized, and the results were cast in terms of the U.S. Environmental Protection Agency (EPA) water effects ratio (WER).

Two generalized models were developed to serve as predictive tools for the fate and effects of copper. The one-dimensional, steady-state box model SD-1D provided an initial assessment of the copper balance in San Diego Bay and estimates of partitioning coefficients and of copper loss rates to the sediment. This model gave a one-dimensional, steady-state solution to the balance of conservative and non-conservative constituents. It had the advantage of rapid formulation and run-times, but lacked the ability to simulate time-varying concentrations and had relatively coarse spatial resolution. The second numerical hydrodynamic model implemented for San Diego Bay was a depth-averaged tidal and residual circulation model known as TRIM-2D. The model, predicting water surface elevations and currents produced by astronomical tides, wind, and freshwater inflows, had been calibrated using measured data from 1995-2002. TRIM-2D had the advantages of providing high spatial resolution and accounting for time-varying flows and concentrations. TRIM-2D was modified to simulate contaminant fate and transport by adding the transport equation and associated kinetic subroutines.

The parameters assessed with SD-1D were used in TRIM-2D for predicting toxicity conditions in San Diego Bay. Data from the first four surveys was used for the assessment of partitioning coefficients and rate loss to the sediments with SD-1D. These coefficients were used in TRIM-2D for the replication of the distributions of total, dissolved, and particulate copper in the bay for the first four surveys. They were also used in conjunction with data for total suspended solids (TSS) and dissolved organic carbon (DOC) for the replication of the distributions of Cu(II)_aq in the bay for those surveys. TRIM-2D was validated by predicting the distributions of the different species of copper that were measured in the final two surveys, using the parameters developed for the first four surveys. This validation showed the capability of the model for these predictions, as the range values predicted included those measured.

The use of TRIM-2D as a management tool for sources of copper was also proven. The model was used to predict copper distributions in the case of theoretical changes in the sources of copper to the bay. While these theoretical changes were radical in nature and practically impossible to reach, the results were plausible in nature and indicated the most probable changes expected from these changes.

Results of this work have transitioned to support the development of an integrated model that can be used by the regulatory community. There is an effort at EPA on the development of the Biotic Ligand Model (BLM) for seawater. The BLM is already proposed for freshwater as an alternative for the WER approach; however, this BLM needs further development for its use in seawater. The advantage of BLM over WER is economical, as it requires a substantially lower effort to produce WQS specific for each body of receiving waters. Under ESTCP project ER-200523, an integrated model for the fate and transport of toxicity by copper in DoD harbors was developed and demonstrated, building on the efforts of SERDP project ER-1156.