It is difficult to even estimate the cost of progress. According to Barry Burgan, coordinator of the Environmental Protection Agency's National Water Quality Inventory, Washington, D.C., only a fraction of the rivers, lakes, wetlands, and ground water are sampled by the States. The EPA suggests that non-point sources of pollution, which by definition are diffuse or otherwise difficult to trace, are responsible for 50% of America's current water-pollution problems.

The most recent National Water Quality Report, issued in 1994, points out: "We know even less about the condition of our coastal waters than we do about our estuarine or inland waters...Only 6% of the Nation's estimated 56,121 miles of ocean coastline [have been assessed]...Data on pollutants and sources of pollution are too sparse to be included in this report."

In short, we have no idea how polluted our water really is.

However, the well-publicized clean-ups of Lake Erie, the Hudson River, and Massachusetts Bay give evidence to the assertion that what was done can be undone. According to Richard Signell, an oceanographer at the Woods Hole Field Center, Woods Hole, MA, run by the U.S. Geological Survey (USGS): "Modeling is the key to understanding. Before we can attack the threats posed by water pollution, we have to understand the mechanisms by which pollution is transported and diluted in water."

Signell has participated in a number of programs funded by the USGS designed to develop regional water models for use by government agencies, city planners, businesses, and as the basis for future research. The primary software tools used to construct these simulations are the Estuarine, Coastal, and Ocean Model, semi-implicit (ECOM-si) and the Hindcasting Shallow Water Waves (HISWA) prediction model.

Performing multiple runs is the key to accurate modeling with ECOM-si. Models developed by the USGS seem crude by mechanical engineering standards: The Massachusetts Bay models typically have 70 x 70 grid points and 15 levels while the most ambitious Gulf of Maine models have grids of 103 x 151 points and 19 levels. However, the time domains used quickly suck down processor power. "We often calculate 10 minute time steps for periods of years," Signell says. "And we do it repeatedly."

There are so many variables in a coastal fluid system, it is not obvious which ones are having a significant impact on pollution. For example, in order to accurately model Massachusetts Bay, the analysts must take into account the topology of the ocean floor and sedimentation, prevailing currents on a monthly basis, seasonal wind effects, outflow from rivers, stratification of the water column, and temperature. With ESOM-si, investigators can "turn off" rivers, winds, or particular currents on a given run.

Verification is a vital activity for accurate modeling. Engineers crash cars to verify predictions made with analysis codes, and oceanographers go out in boats to ping away with sonar and throw sensors in the water. Between 1990 and 91, the National Estuary Program of the Environmental Protection Agency sponsored a major field study of Massachusetts Bay.

A series of 25 current-measuring instruments were moored at nine locations throughout the bay. The instruments, developed by EG&G Inc., Wellesley, MA, posed vanes at right angles for measuring current velocities. EG&G offers a wide variety of environmental monitoring and remediation products and services.

In addition, the EPA survey acoustic doppler current profilers developed by what is now Sontek, San Diego, CA, were placed on the bottom. There, the devices shot sonar pulses upward and measured densities of sediments and organisms in the water from returns.

The Mass mess. According to Signell, the data gained from the EPA survey has proven invaluable in verifying regional pollution transport models. A case in point is the ongoing Boston Harbor Project, designed to reduce concentrations of sewage near the city.

Massachusetts plans to move the point where treated sewage (sanitized as "effluent" in official nomenclature) is discharged from the mouth of the Boston Harbor to a battery of diffusers nine miles offshore. There, it is argued, the forces of nature will dilute the effluent more effectively than in the estuarine environment of the harbor. Also, the expanse of Massachusetts Bay, cradled by Cape Cod, is expected to provide an effective depositional trap for fine-grained sediments.

However, the open flank of Massachusetts Bay is dominated by the Stellwagen Bank National Marine Sanctuary. This glacially formed underwater plateau supports zooplankton in profusion and thus serves as a feeding ground for many species, including humpback whales, basking sharks, endangered sea turtles, and shore birds of all descriptions. In addition, the Sanctuary provides an important resource for commercial fishing. It would not do for Boston to befoul an environmentally sensitive region while cleaning up its act.

According to Signell, simulations of both the current and proposed discharge sites showed that the latter resulted in a much more rapid dilution of effluent. Furthermore, the Gulf of Maine current and northeast wind patterns tend to move effluent south and west, away from the Stellwagen Bank. Signell reports that these contaminant transport studies were influential in defeating a court challenge to the project.

Similar techniques to those applied to modeling coastal waters are used for inland bodies as well. The USGS is using ECOM-si to develop models of Lake Ponchartrain, near New Orleans. HydroQual Inc., Mahwah, NJ, uses software and on-site analysis to develop water-quality models and pollution amelioration strategies in communities all over the U.S.

According to Kevin Keane, HydroQual's marketing director, the company's scientists and engineers have developed many models for eutrophication, sediment transport, contaminant fate and transport, and food-chain/bioaccumulation. "These models allow for a quantitative evaluation of potential impacts and for an assessment of the use of various management and/or remedial alternatives," Keane says. For standing bodies of water, the best (and often the only) pollution management option is prevention.

HydroQual Applications of the ECOM-si models

The code developed by HydroQual consultant Alan Blumberg has been instrumental in many pollution control activities around the country. Here are a few highlights:

The EPA, in conjunction with the States of New York and Connecticut, requested a study into the causes of low-dissolved oxygen in Long Island Sound. Model results have been used to propose nitrogen-reduction targets for waste treatment plants discharging into the Sound and immediate vicinity.