C Transport and Fate of Pollutants in the Coastal Marine Environment
INTRODUCTION
This appendix presents an assessment of current knowledge of the various physical, chemical, and biological processes that determine the transport and fate of pollutants associated with wastewater and stormwater inputs to coastal waters, and how well the behavior of these inputs can be modeled and predicted for engineering purposes. Specifically, how do the quantity, quality, and method of discharge of the wastewater to the coastal ocean affect the ambient water-quality and the quality of the sediments? With increasing knowledge of environmental engineering and marine sciences, it is now possible to design a waste management system by the water-quality and sediment-quality driven approach, namely finding the most cost-effective combination of source control, wastewater treatment, and outfall configuration. This process is explained in a later section on Overall Design following the next two sections which address Mechanisms of Input and Transport and Fate.
Coastal areas include a continuum from poorly-flushed small estuaries to the well-flushed open coastlines. This study focuses on larger estuarine and coastal systems subject to major urban impacts and that have significant exchanges of marine water with also the possibility of internal recirculation and entrapment of pollutants in the sediments within these bodies.
Federal law classifies inputs into point and nonpoint sources, according to whether discharge permits are required or not. As the regulations have changed (e.g., storm drains for cities over 100,000 people now require per-
mits), distinctions on the basis of physical characteristics have become blurred. Traditional point sources at the time of passage of the Clean Water Act of 1972 included only outfall discharges from defined municipal and industrial installations; these sources, the focus of most control efforts heretofore, are generally well characterized now by types and fluxes of pollutants, although that was not the case before the Clean Water Act was passed. Outfalls (with very few exceptions) are submarine pipelines or tunnels discharging from a few hundred meters up to 15 kilometers (10 miles) from shore depending on the volume and character of discharge and the nature of the receiving water body.
The term nonpoint sources is a poor descriptor because this term includes all inputs that are not point sources. Also, the definition of point sources changes with new laws and regulations. Here, the broad classification of diffuse sources is used to include all sources except the traditional point sources. This category includes (but is not limited to) streams, storm drains and flood control channels, combined sewer overflows (CSOs), discharges from boats, ground water seepage, and atmospheric deposition. These sources have three common features: 1) the original pollutant sources are widely distributed, 2) the rates of delivery to coastal waters are highly irregular depending primarily on the occurrence of rain, and 3) control measures other than at the original sources are limited. In some locations, the release of pollutants from existing contaminated sediments can be a significant diffuse source.
Inputs of storm runoff, CSOs, and streams occur in a very unsteady manner at, or close to, the shoreline. Storm drains and flood channels (separate from sewers) discharge significantly when it rains, bringing as pollutant loads whatever wastes have accumulated in the drainage basin since the last storm; but also, smaller dry-weather flows may be highly polluted by illegal or unregulated waste disposal practices. Combined sewer overflows occur when runoff combined with sewage flows exceeds the capacity of a system, which then discharges at numerous predesignated places into various bodies of water in an urban area, including into streams and estuaries as well as the open coast. Natural streams and rivers may bring other pollutants from upstream areas, such as agricultural chemicals, atmospheric deposits, and nutrients washed off the land.
Mathematical and conceptual models are used extensively to explain processes that disperse and modify pollutants in the ocean and to predict their effects on ecosystems. Various submodels may be combined to produce an overall model to relate pollutant inputs to water and sediment quality for single and multiple sources. These models are fundamental to management by the environmental-quality driven approach because the limits on emissions for any outfall discharge or diffuse source may be back calcu-
lated through the models. These models are analogous to the emissions-to-air-quality models used in developing air pollution control programs.
The main purpose of this appendix is to assess the knowledge of all the relevant processes and evaluate the modeling capability for management of the quality of coastal waters and sediments by environmental-quality driven approaches. To be successful there must be good predictive capability for the dominant factors that determine the engineering choices for satisfying the standards. These factors can be determined based on sensitivity analyses and the experience of the modeler. Thus, for engineering purposes it is not necessary to understand every process if more knowledge would have no effect on the choice of control strategy. For example, it is not necessary to understand the behavior of a certain pollutant at a location where the exposure is far below any possible threshold value of concern.
Since modeling for design of a management plan for pollution control always has some uncertainty covered by safety factors, it is cost-effective to implement a system (such as a waste treatment plant and an outfall) in a stepwise flexible way to allow for continuous feedback of the operating experience and the observed impacts on the coastal waters. In fact, there are very few situations where there is not already an existing discharge that serves as a prototype to study before and during upgrading the system. For example, the full effect of upgrading primary treatment on coastal water quality might well be observed before proceeding to secondary treatment levels if there is significant uncertainty about the need for secondary treatment. Or source control efforts for specific chemicals can be focused on those observed to be too high. This approach is always self-correcting as the discharger commits itself to take as many steps as necessary to solve any known problems. This incremental approach is one of the important features of integrated coastal management as proposed in the main body of this report.
MECHANISMS OF INPUT
Outfalls
An outfall is a pipeline that discharges liquid effluent into a body of water. In the last four decades, there have been great advances in technology for ocean outfalls to achieve high initial dilutions and submerged plumes that are trapped beneath the pycnocline (or by the density stratification of the ambient water). Outfalls have advanced from simple open-ended pipes not far from shore to long outfalls with large multiple-port diffusers discharging in deep water. Figure C.1 provides an example of a deep water ocean outfall with a long multiport diffuser. The characteristics of major
![](/openbook/0309048265/xhtml/images/img00019.jpg)
FIGURE C.1 Schematic plan and profile of the 120-inch outfall, County Sanitation Districts of Orange County, California. (In metric units, the overall length is 8.35 kilometers, the diffuser length is 1.83 kilometers, the diffuser depth is 53-60 meters, and the pipe diameter is 3.05 meters). (Source: Koh and Brooks 1975. Reproduced, with permission, from the Annual Review of Fluid Mechanics, Vol. 7, © 1975 by Annual Reviews Inc.)
outfalls on the Pacific coast of the United States constructed prior to 1978 are summarized in Fischer et al. 1979.
The construction of large outfalls in the marine environment has most commonly been accomplished with reinforced concrete pipe (RCP) with flexible joints. Recently, steel pipes have become more common because of improved manufacturing processes, better corrosion protection technology, proven constructibility (from technology transfer from the offshore oil industry), and construction costs for steel pipes, which can be significantly less than for RCP. One reason is that steel pipes are made into much longer lengths, requiring fewer junctions to be made in the marine environment. Two steel pipes of 64-inch diameter were used for the two outfalls of the recently built Renton outfall system in Puget Sound (Metro Seattle). They discharge through 500-foot long diffusers at a depth of about 185 meters (600 feet), which is probably beyond the capability of RCP construction. Tunnels have also become more competitive because of great advances in tunnel boring machines in the last 15 years. For example, the Boston outfall now under construction will be a 15 kilometer (9.4 mile) long tunnel, 7.39 meters (24.2 foot) in diameter, including a 2,000 meter (6,600 foot) long diffuser section with 55 vertical risers, each with 8 discharges ports. Three new successful outfalls in Sydney, Australia, are also tunneled.
The combination of source control, treatment plant, and outfall is an engineering system that has achieved often dramatic improvements in coastal water quality. Even today, however, while many major discharges have state-of-the-art systems, there are still others that discharge through short outfalls with poor initial dilution.
Figure C.2 shows schematically a typical multiport diffuser at the end of an ocean outfall discharging buoyant effluent into a density-stratified receiving water. Sewage effluent, being effectively fresh water, rises in the ocean, mixing intensely with the receiving water. The ocean is also usually density-stratified due to temperature and/or salinity gradients. Thus, the effluent mixing with the near-bottom denser ocean water can give rise to a mixture that is neutrally buoyant before the rising plume reaches the surface, leading to the formation of a submerged waste field, which is in turn advected by the prevailing currents (Figure C.2). This region of initial mixing is often called the near-field.
The mixing that occurs in the rising plume is affected by the buoyancy and momentum of the discharge and is referred to as initial dilution. It is typically completed within a matter of minutes. Dilution as used in the engineering community is defined to be the ratio of the volume of the mixture to that of the effluent (i.e., the reciprocal of the fraction of effluent in the mixture). This initial dilution phase of the mixing process is under some control by the design engineer since it depends on the diffuser details such as length of diffuser, jet diameter, jet spacings, and discharge depth.
The initial dilution is also controlled partially by nature since it depends on the density stratification and currents in the receiving water.
A typical large discharge diffuser (for a flow of 5 m3/s) might be a kilometer in length and located in 60-meter water depth at a distance of 10 kilometers offshore. There might be several hundred discharge jets (typical diameter 10 centimeters) spaced along the 1-kilometer-long diffuser. The initial dilution obtainable for such a diffuser would be expected to be in the hundreds to a thousand depending on details (mainly flow rate and density stratification).
The initial dilution and waste field submergence can now be estimated with a fair degree of confidence as a result of three decades of engineering research on the mixing processes in buoyant jets and plumes (Koh and Brooks 1975, Fischer et al. 1979). A number of computer models can provide such estimates of sufficient reliability as to make decisions on design choices. The most commonly available ones in the United States are the ones published by the U.S. Environmental Protection Agency (EPA) (Muellenhoff et al. 1985; Baumgartner et al. 1992, based in part on laboratory tank experiments at the EPA by Roberts et al. 1989a, b, and c).
All the models for dilution and submergence calculations are based on analyses of buoyant jets discharged into a large receiving body of water. Usually the equations of conservation of mass, momentum, and buoyancy fluxes are integrated across the plume cross-section, having first assumed similarity of cross-plume profiles for the velocity and density deficiency (usually Gaussian). The resulting equations in this integral method are a system of nonlinear ordinary differential equations with the auxiliary conditions being in the form of initial conditions. Such systems are readily solved numerically. This is indeed the backbone of the available models.
For approximate start-up calculations, it often suffices to use the formula for a simple line plume in a linearly stratified environment based on assuming that the multiple-port diffuser is well approximated by a line plume-a source of buoyancy flux only. For this case, Brooks and Koh (1965) derived simple formulas as follows:
![](/openbook/0309048265/xhtml/images/img00021.jpg)
where:
q = discharge rate per unit length of diffuser
g = gravitational acceleration
ρ = density of discharge
Δρ/ρ = relative density difference between effluent and receiving water
ymax = maximum height of rise of the plume
S = initial dilution (centerline or minimum time-averaged dilution)
∂ρa/∂z = average ambient density gradient
These formulas can be applied to the typical case described in the previous paragraphs. If we assume that the ambient density gradient is
![](/openbook/0309048265/xhtml/images/img00022.jpg)
(which can be due to a temperature difference of about 2°C over a depth of 50 meters), then ymax = 31 m (halfway from the discharge depth of 60 meters to the surface) and the initial dilution would be 200. Note that as the stratification increases, the plume rise and dilution are both reduced.
For actual design calculations with mathematical models, one also needs to examine many different ambient conditions such as density profiles and current speeds and directions. The effluent flow also varies. Thus the initial dilution for an outfall is not a constant value but fluctuates considerably depending on ocean conditions and the effluent flow rate.
It is important to point out here that dilution, being the ratio of the volume of the mixture to that of the effluent, can be converted to concentration c of a particular pollutant provided we know the concentration of that pollutant in both the effluent ce and the receiving water cb. Thus,
![](/openbook/0309048265/xhtml/images/img00023.jpg)
If cb, the concentration of the pollutant in the receiving water were zero, then
![](/openbook/0309048265/xhtml/images/img00024.jpg)
The value of cb includes the increase of the regional background concentration (background buildup) in the receiving water due not only to the continuous discharge from the outfall itself but also to all other sources.
Discharges from Barges and Ships
Ocean dumping from vessels has been practiced in the past by many coastal communities in various countries. It is still being practiced by some
and is being planned by others. Nations are not in total agreement regarding ocean dumping, although the practice has seen a dramatic decline, particularly in the developed countries.
By far the largest amount of material involved in ocean dumping is dredged material (formerly known as dredge spoil). In the past, other materials dumped have included digested sewage sludge, various industrial wastes (including acids), oil well drilling mud and cuttings, coal ash, and mine tailings. Refuse has also been dumped in the past, but the practice has ceased (except for occasional illegal acts). Bilge water and ballast water are also discharged by ships in coastal waters.
Ocean dumping has been mandated by law to cease in the United States, with the exception of dredged material. Other developed countries have also largely agreed to stop dumping of sewage sludge. In less developed countries, the status of ocean dumping is unclear. Rules and regulations may not exist. It is unrealistic to expect ocean dumping of nonhazardous polluted materials to be eliminated worldwide any time soon or even in a few decades. Clandestine dumping and dumping where not allowed are difficult to police in most areas of the world's ocean, usually for the simple reason that the necessary infrastructure is inadequate or nonexistent.
Procedurally, the barge (or ship) is loaded with the waste material by placement into the ship's compartments. The vessel is moved to the designated dump site, which is generally a rectangular area with typical linear dimension of several kilometers. As long as the vessel is in the dump site, the material is allowed to be discharged into the ocean. Frequently a bottom-opening hopper barge is used. Here the barge bottom is equipped with doors, which can be opened to permit the material to fall out by gravity. Sometimes, the material is pump-discharged into the wake of the moving vessel to take advantage of the high turbulent energy that increases the initial dilution.
Modeling of the mixing, transport, and fates of materials after disposal from barges and ships is less well developed and much less well verified than the corresponding models for outfalls. While the physical processes involved in the two cases are similar, the situation for ocean dumping is less amenable to analyses because the discharge conditions (discharge rate, bulk density of the material, and characteristics of its contents) may be ill-defined. This has the most effect on near-field predictability but extends also to the intermediate and far-field because the near-field equilibrium vertical location of the discharged material depends on the discharge condition.
A detailed discussion of ocean disposal of digested sewage sludge has been presented with policy recommendations in a previous NRC report (NRC 1984).
Diffuse Sources
The term diffuse sources describes the inputs other than municipal and industrial wastewater to coastal water bodies. These pollutant sources include urban storm drains, combined sewer overflows, natural streams and rivers, ground water outflow (under the sea), discharges from recreational boats and commercial shipping, and atmospheric deposition. The input of pollutants into these delivery pathways is widely distributed, and more challenging to control at the points of origin. Furthermore, there is little opportunity to manage the hydraulics of the inputs to achieve high dilution far from shore as for wastewater from publicly owned treatment works (POTWs). Nonetheless, the same principles of transport and fate apply. For modeling the water and sediment quality, it is, of course, important to include all these diffuse sources along with the outfall discharges from publicly owned treatment works.
TRANSPORT AND FATE
Following plume rise and the attainment of initial dilution, the diluted effluent cloud (often submerged below the thermocline) is advected with the currents and undergoes a variety of physical, chemical, and biological processes, referred to as transport and fate of pollutants. These processes occur in the natural environment and are beyond the direct control of the engineers, other than the initial conditions determined by the characteristics of the outfall and the effluent. For example, if a plume is kept submerged below the surface mixed layer, the subsequent transport, fate, and effects may be greatly different from a surface plume in the near-term (on the order of days to weeks). This region, which is dominated by natural processes beyond the near-field, is called far-field. This section describes the major processes affecting the behavior of pollutants in the coastal ocean, that is, transport and fate.
Far-Field Transport and Dispersion of Contaminants
Scientific knowledge of far-field transport and dispersion of contaminants has advanced significantly in the last several decades. When this knowledge is coupled with modeling and site-specific programs to measure currents, density stratification, and dispersion, engineering designs for outfall diffusers can be made by the water-quality driven approach. Far-field transport and dispersion can be modeled for design purposes with reasonable factors of safety to cover uncertainties. This section addresses the current knowledge and gaps in science and modeling. While predictions now are adequate for project design, increased knowledge will lead to improved management techniques and predictions.
Beyond the near-field region where the delivery of the effluent has a dominant influence on its dispersal, the subsequent transport and dilution depends primarily on the currents in coastal waters. Persistent currents cause advection away from the outfall site, while currents that fluctuate over short time and space scales result in dispersion of the effluent. Dispersion results in dilution of the effluent, while advection carries it away from its point of entry. The exposure of the receiving waters to the environmental hazards introduced in the effluent depends sensitively on the advection and dispersion rates. Higher concentrations, and hence greater exposures, occur in regions of sluggish transport, and lower concentrations and exposure occurs in rapidly flushed environments.
There has been considerable effort mounted over the last 20 to 30 years to measure, model, and better understand transport and dispersion processes in coastal waters with application to the siting of outfalls and assessing the risks of oil spills and other toxic contamination, as well as developing an understanding of the interaction between the physics and the ecology of coastal waters. Because of the diversity of coastal water bodies and the complexity of the interactions between topography, density stratification, freshwater inflows, tidal motions, and the wind, it is not possible to predict a priori the magnitude of advective and dispersive transports at a given location. However, as will be discussed in more detail later in this section, it is possible to combine our general understanding of coastal processes with site-specific measurements to yield quantitative estimates of these processes that are accurate at least to an order of magnitude and often within a factor of two. This level of confidence is usually adequate to support water- and sediment-quality based analyses, with suitable safety factors to cover any errors of prediction.
A general discussion of the transport and dispersion in coastal waters must first acknowledge the great diversity in the physical characteristics of coastal environments, from lagoons to estuaries and bays of various sizes to continental shelves with widths that vary from several kilometers along the southern California coast to more than 100 kilometers on the east coast of the United States. The driving forces vary tremendously from place to place as well. For example, the currents on the west coast of the United States are driven primarily by the along-shelf winds, while in other areas, such as the Gulf of Alaska and the South Atlantic Bight (Georgia, South Carolina, and part of North Carolina), the currents are strongly influenced by the input of freshwater from rivers. Tidal motions, which are more important with respect to dispersion than net transport, are also highly variable in strength and relative importance, being tremendously important, for example, in Puget Sound and the Gulf of Maine. Finally, the currents in the ocean margins adjacent to the continental shelf often influence the transport on the shelf-the most famous example being the Gulf Stream, which inter-
mittently spews warm-core rings (100-kilometer diameter rings of warm water originating from the subtropical Atlantic) against the continental shelf.
Transport also varies considerably depending on the location and timing of the delivery of effluent. Effluent being discharged through an outfall at the edge of the continental shelf will be directly exposed to oceanic currents, while stormwater, discharged at the shore, may be trapped in a nearshore region for a considerable period before it is exposed to the more energetic and dispersive motions further offshore. Discharge rates of effluent from POTWs are relatively constant, but nonpoint source discharges are highly intermittent, occurring during periods of high freshwater input. Thus, the fate of nonpoint source inputs is sensitive to the buoyancy-driven motions occurring during runoff events. Transport in coastal waters varies as a function of depth, so the fate of the effluent may depend sensitively on the vertical location of the discharge. Surface waters tend to be more energetic than deep waters, hence the advection and dispersion tend to be more rapid there. In regions with significant freshwater input and upwelling zones, there is a net offshore transport in the near-surface layer, favoring the dispersion of effluent. Because the introduction of nutrients in the euphotic zone may have the undesirable consequence of stimulating algal production, the recent tendency is to design submerged outfall plumes.
The fate of different waste substances varies considerably depending on whether they are dissolved or part of the particulate fraction of the effluent. Virtually all of the toxic metals and organic compounds in effluent are strongly particle-reactive. Hence their fates depend on sedimentary processes as well as fluid motion. The rapid coagulation or flocculation that occurs soon after the effluent is exposed to seawater causes settling of much of the solid fraction on time scales of I to 3 days. Subsequent transport of that material depends on resuspension from the seabed. Dissolved material, such as nutrients, is carried with the ambient water, but its distribution may change rapidly as a result of biogeochemical processes within the water column. Both in the case of the dissolved and the particulate constituents, residence times are generally short enough that the transport processes that occur within the first several days of their entry into the coastal environment are most important. Given typical transport rates, this represents a region extending 10 to 20 kilometers from the outfall. Transports of larger spatial and temporal scales are important with respect to the ultimate fate of substances and for basin-scale or regional ecological impacts.
Dispersion
Dispersion refers to the tendency of a parcel of water to increase in spatial dimensions, and hence be diluted, with time. In small-scale fluid dynamics, this tendency is referred to as diffusion (either molecular or tur-
bulent). Dispersion is distinguished from diffusion in that it includes motions at various scales that may not be formally defined as turbulence but that have the effect of spreading out the fluid in a manner analogous to diffusion. Examples of processes leading to dispersion include tidal motions, eddies shed from coastal currents, and vertical or horizontal shears in the mean or low-frequency flow. In the context of far-field dispersion, vertical spreading is generally much less significant than horizontal spreading due to the smallness of vertical scales (10 to 100 meters) as compared with the horizontal scales (0.1 to 100 kilometers).
Dispersion has been notoriously difficult to predict, whether in estuaries, coastal waters, or the deep ocean, due to the complexity and wide range of scales of motion that may contribute to the mixing. The dispersion rate can be estimated by the time rate of increase in the size of a patch of fluid, with the dispersion coefficient defined as
![](/openbook/0309048265/xhtml/images/img00025.jpg)
where σ2 is the spatial variance of the patch. Stronger flows as well as flows of large spatial scales tend to disperse more rapidly. Dispersion coefficients have been estimated directly in a variety of coastal environments by measuring the spreading of dye patches. Okubo (1971) combines the results of a number of studies, obtaining a consistent relationship between the diffusion coefficient (K) and the scale a of the patch, from scales of 100 meters up to 100 kilometers, over which the dispersion coefficient increases from 10-1 m2/sec to 1000 m2/sec.
Alternative formulations of dispersion in coastal waters express the rate of patch growth as ∂σ/∂t = P where P is a dispersion velocity that has been found to be in the range 10-3 to 10-2 m/sec (Kullenberg 1982). Empirical correlations have also been developed (usually in terms of a dispersion coefficient) for shear-induced dispersion in rivers and longitudinal dispersion in estuaries (Fischer et al. 1979). These studies have succeeded in demonstrating consistency among different experiments in different environments, but the scatter in the data generally reflects a five to ten fold uncertainty in the magnitude of dispersive transport. Consequently, many studies of potential environmental impacts have been based on site-specific dispersion measurements using suitable tracers or drifters.
How critical is this uncertainty for predicting the reduction in concentration and increase in horizontal extent of a pollutant distribution over a time scale of several days? In relatively open coastal waters, the tracer data indicate that a pollutant patch grows at a rate of about 1 kilometer per day. So a continuous plume that is initially 100 meters in size, typical of a relatively small discharge, will experience a ten fold increase in width and a ten fold decrease in average and peak concentration that may be a signifi-
cant addition to the initial dilution. On the other hand, a larger discharge, say from a submerged diffuser on the order of a kilometer in length, will only about double in width over a day or so. The consequent dilution factor of two increases dilutions from initial values of the order of 50-200 to 100400. In this case, the uncertainty in the estimated rate of dispersion is less critical to the analysis.
Where net advection is absent or weak, the more rapid dispersion that occurs over longer times and larger spatial scales, due to fluctuating tidal and wind-driven currents, may determine the residual background concentration of effluent (Csanady 1983, Koh 1988). In open water, the background concentration may be a negligibly small quantity. However, in relatively enclosed coastal regions, including estuaries, significant accumulation of pollutant mass may be controlled by large-scale dispersive processes that should be quantified by tracer studies. As discussed in the previous section, an increase in background concentration reduces the effective initial dilution.
Net Advective Processes
Currents occur at a broad range of time scales, from seconds (e.g., surface waves and turbulence) to hours (e.g., tidal motions) to days (e.g., wind-driven motions) to seasonally varying flows and finally to steady flows. Generally the spatial scale of the motion increases as the temporal scale increases simply because water of a given velocity will be carried farther in a longer time period. Motions of short time scales do not carry water large distances, hence they do not contribute to net advection except at small scales (although as mentioned above, they may be important with respect to dispersion). In moderate sized embayments and the continental shelf, where spatial scales are at least tens of kilometers, the motions responsible for net transport tend to have time scales longer than 24 hours; these include wind-driven motions, buoyancy-driven flows, and flows forced by oceanic motions.
Buoyancy-Driven Flows. Buoyancy-driven flows predominate in estuarine environments, where the input of fresh water whose density contrasts with seawater produces a pressure gradient, which drives the less dense water seaward in the upper water column and pulls more dense seawater landward underneath. This so-called estuarine circulation is generally the most important flushing mechanism in estuarine systems. The well-documented variations in water quality in San Francisco Bay as a function of freshwater input clearly indicate the important role of estuarine circulation to the flushing and hence the water quality of estuarine systems.
The transport of suspended solids is highly coupled to density-induced
vertical variations in advective transport. In many estuaries, the locations where fine sediments accumulate are known to be determined by the estuarine circulation pattern.
In larger estuarine systems and in coastal environments with large freshwater inputs, such as the Gulf of Maine and the Gulf of Alaska, buoyancy effects are still important for net advection, but rather than developing a two-layer, estuarine circulation, the buoyancy-driven flow is manifested as a coastal current, which flows along the coast to the right (in the northern hemisphere) of the offshore direction. Thus the coastal flow in the Gulf of Maine is southwestward, while in the Gulf of Alaska, it is northwestward. While these flows provide a substantial along-shelf component of flow, they do not by themselves contribute significantly to cross-shelf exchange. Effluent that is introduced in one portion of the coast may be carried away from its source region only to impact coastal areas downstream (after further dilution). A natural example of this type of downstream influence is the transport of toxic red-tide organisms from the estuary of the Kennebec River to the coasts of New Hampshire and Massachusetts by the buoyancy-driven coastal current (Franks and Anderson 1992).
Instabilities in buoyancy-driven coastal currents (e.g., Chao 1990) provide a mechanism of cross-shelf transport. The instabilities start out as undulations in the front, growing into eddies and extrusions into the adjacent oceanic waters. In addition, buoyant plumes are very sensitive to wind forcing due to their shallow vertical expression. Significant cross-shelf transport can occur when winds act on buoyant flows.
Wind-Driven Motions. The wind is often the most important driving force of net transport on the continental shelf, and it is often a major contributor to exchange between embayments and coastal waters. In the upper few centimeters of the water surface, the flow tends to proceed in the same direction as the wind at approximately 3 percent of the wind speed (Wu 1983). This rule of thumb is useful for the prediction of the trajectories of oil spills and floatable wastes. Below this very thin surface layer, the influence of the earth's rotation tends to turn the wind-driven transport to the right. In the absence of other forces, a steady wind will result in transport exactly 90 degrees to the right of the wind direction (or left in the southern hemisphere). This is known as Ekman transport, and the portion of the upper ocean in which it occurs, typically the upper 30 meters, is called the surface Ekman layer. The strongest wind-driven currents on the continental shelf occur as a result of along-shelf winds, which result in cross-shelf Ekman transport, but the pressure gradient induced by the coast causes the dominant flow to be in the along-shelf direction, in the same direction as the wind. Along-shelf current speeds tend to be 1 to 2 percent
of the wind speed, their magnitudes varying with the shelf geometry, stratification, and other factors.
The cross-shelf flow based on Ekman theory should be a few centimeters per second (kilometers/day) for a moderate along-shelf wind stress. This is weak enough relative to the along-shelf flow that it has not been well resolved in measurements. It is also weak enough that it takes on the order of a week of persistent winds to transport material across a 20-kilometer-wide shelf and a full month to cross a 100-kilometer-wide shelf. A week of persistent winds is not uncommon, particularly along the west coast of the United States, hence the cross-shore Ekman transport often provides a relatively rapid offshore transport on the west coast. The wider shelf on much of the east coast and the absence of persistent winds renders cross-shelf Ekman transport less effective there.
A potentially important cross-shelf transport process is the convergence caused by a change in the along-shelf wind forcing, such as a change in wind direction. Relaxation from upwelling on the west coast causes a sudden shift from southward to northward currents. Where the oppositely-directed currents collide, a strong cross-shelf flow results. Such cross-shelf flows tend to be short lived, but they are strong enough to transport material all the way across the shelf, and they may be volumetrically as important as the Ekman transport itself in contributing to cross-shelf exchange. Similar convergences can also be caused by changes in coastline geometry such as headlands.
The vertical shear associated with wind-driven currents has consequences for effluent plumes that may be located near the surface, below the pycnocline, or near the bottom. In addition, the near-bottom currents may regulate the long-term transport of sediments (see later section on Sediment Processes).
Oceanic Currents Impinging on the Coast. Ocean margins are often the sites of major current systems, the most notable one bordering North America being the Gulf Stream. A strong steering influence of bathymetry generally keeps these currents from riding up onto the continental shelf, but there are often instabilities in these current systems that cause eddies to be shed, which impinge on the coast and may influence the cross-shelf transport. On the east coast, instabilities of the Gulf Stream generate warm core rings and smaller eddies called shingles, which impinge on the shelf along the east coast. They have been found to have a strong influence on the flow on the outer shelf and are likely important agents in fluid exchange between the outer shelf and the adjoining ocean. On the west coast, instabilities in the southward flowing California Current result in a complex field of eddies adjacent to the continental shelf. These eddies result in strong offshore flows between the outer shelf and the ocean, which carry cold, upwelled water from the shelf into the ocean interior.
Modeling and Measurements
Numerical models are well suited to investigating the nature of physical processes in complex coastal environments in which the mathematical problem does not lend itself to an analytical solution. Numerical models are particularly useful for describing flows in regions of complex geometry (Signell and Butman 1992) and for performing simulations of effluent motions (Baptista et al. 1984). There are many examples of accurate numerical model predictions of sea level variations, including tidal and wind-forced regimes and recently the problem of coastal trapped waves. However, sea level is of no relevance to effluent transport. Tidal motions have been modeled reasonably well (Tee 1987, Sheng 1990), and wind-driven, along-shelf motions can be predicted accurately under certain circumstances (Allen 1980). Nontidal currents in general are more difficult to model than sea level, and there are few examples in which models have accurately predicted the variations in currents in a particular environment, based on prescribed forcing variables such as winds and freshwater input. The quantities most relevant to the fate of effluent, such as dispersion and cross-shelf motions, are considerably more difficult to model since they depend on spatial gradients of the dominant currents.
It is clear from the above discussion that the scientific understanding of transport processes has not advanced enough to be able to model, a priori, the rates of the various processes at a given site. However, with an appropriate mix of measurements, theoretical calculations, and numerical modeling, the advection and dispersion rates can be determined with acceptable levels of uncertainty for use in environmental risk assessment.
In order to proceed effectively with a site assessment, it is important to start with a solid understanding of the regional transport processes. Once a region is well enough understood to focus an investigation on candidate outfall sites, a combination of field measurements, theoretical calculations, and modeling allows the transport to be quantified and the distribution of effluent components to be predicted. The measurements typically include moored measurements of currents and water properties (temperature, salinity, and in recent studies suspended sediment and dissolved oxygen), drifter releases, and shipboard survey measurements of relevant water column and sediment parameters. The measurements should (but generally do not) extend over a complete annual cycle. The outcome of such studies is usually a very good characterization of the dominant current regime, most often the tidal currents, and a somewhat less certain picture of the other transport processes. Even without the aid of a numerical model, such a measurement program can provide the basis for predictions of effluent transport (e.g., Koh 1988). By combining current measurement results with analyses of water column and sediment distributions of various measured constituents,
the more subtle problems such as sediment transport and horizontal dispersion may be quantified.
This combination of tools for assessing physical transport processes will yield predictions with large uncertainty factors, but the errors in predicting advection and dilution will typically be well inside an order of magnitude. Considered in context with the overall risk management matrix, the assessment of water-column transport processes has a fairly narrow range of uncertainty. The fate of sediment is less easily predicted, but it is still amenable to prediction in a probabilistic sense, albeit with order-of-magnitude uncertainty in transport rates. Further discussion of sediments is provided in a later section, Sediment Processes.
Conclusions
-
Scientific knowledge of far-field transport and dispersion of contaminants has advanced significantly in the last several decades. When this knowledge is coupled with modeling and site-specific programs to measure currents, density stratification, and dispersion, the water- and sediment-quality driven approach incorporating transport and dispersion is feasible with reasonable safety factors. This approach is the only one that can be improved and adjusted in response to new research information.
-
Ongoing programs of research will provide continual refinements in the predictive capability of the transport and fate of effluent constituents, which will be useful for making future engineering decisions.
Recommendations
-
Coastal physical oceanographers should be encouraged to address questions of particular relevance to water quality, particularly dispersion and cohesive sediment transport.
-
The investigation of the fate and transport should not stop upon selection of an outfall site, but it should continue after a new or modified outfall is put into service, particularly to learn how good the preconstruction modeling and predictions were.
Behavior of Particles from Wastewater and Sludge: Flocculation and Sedimentation
Wastewater effluents and sludges contain particulate matter and particle-reactive pollutants. These particles vary in size over a broad range, from submicron dimensions to several tens of microns. These particles and the particle-reactive pollutants that accompany them can be deposited near the point of discharge or transported long distances in coastal zones. After
deposition, substantial recycling and release of contaminants from bottom sediments to overlying waters can occur. Cycling of contaminants between suspended solid phases and seawater may also occur in the water column. While a substantial amount is known about these processes, much remains to be understood.
Knowledge of the transport and fate of particles and particle-reactive pollutants in marine environments is limited. This limitation constrains, but does not negate, the ability to design waste disposal systems that protect the environment. Conservative approaches can be used to meet appropriate water-quality and sediment-quality criteria. For example, an approach to addressing sediment quality criteria could assume that all of the particles and particle-reactive pollutants in a waste discharge are deposited in the region around a discharge site. This assumption would provide conservative estimates of the concentrations of contaminants to be expected in sediments and pore waters. If the resultant estimates did not meet sediment quality criteria, the contaminant discharge could be reduced to provide the needed environmental protection. Additional and more detailed modeling could be conducted to improve the estimates as necessary. A similar approach, in which all of the particles and contaminants are assumed to remain in the water column, could be used in addressing water quality criteria.
Particles in Marine Environments
Particulate matter in the ocean is comprised largely of aggregates of algae, bacteria, organic detritus, and inorganic particles (including natural sediments). These aggregates vary in size from submicron dimensions to many centimeters. Settling rates can be high. Hill (1992) has summarized published observations indicating that settling velocities of aggregates following phytoplankton blooms are in excess of 100 meters per day. These aggregates are formed by two major pathways: 1) biological aggregation by animal grazing and 2) physicochemical aggregation involving interparticle collisions and attachment. Biological activity can affect interparticle collisions and attachment as, for example, in the formation of the large aggregates known as marine snow. The relative importance of physicochemical and biological processes in aggregate formation and destruction has been debated for some time and remains unresolved at present. There are models for these processes, but their validity and accuracy require testing.
The production of these aggregates is important in the transport and cycling of carbon, energy, pollutants, and nutrients in marine environments. Large aggregates such as marine snow (Silver et al. 1978, Smetacek 1985, Alldredge and Gotschalk 1989) are important in particle transport and in marine chemistry (Fowler and Knauer 1986). Recently Hill and Nowell (1990) have assessed the role of rapidly settling particles in clearing nepheloid
layers by physicochemical coagulation. To these can be added the scavenging of metals, including U and Th by rapidly settling particles in the ocean (Honeyman et al. 1988) and the importance of physicochemical coagulation in this process (Honeyman and Santschi 1991). Jackson (1990) has combined theory for the kinetics of coagulation with expressions for gravity sedimentation and the kinetics of algal growth to describe algal production, aggregation, and sedimentation during a bloom. The results indicate that physicochemical coagulation can place an upper limit on the accumulation of biomass in such blooms, leading to the formation of algal flocs commonly observed by divers.
Particles and Particle-Reactive Pollutants in Wastewaters and Sludges
The particles in wastewater effluents from primary or secondary treatment plants can be expected to contain particles that have settling velocities in fresh water that are less than about 40 m/day, corresponding to the hydraulic loadings used in the design of the settling basins in these treatment systems. (Hydraulic loadings for settling basins range from 300 to 2,400 gal ft-2 d-2 as discussed in Appendix D. A representative value used for comparison purposes here is 1,000 gal ft-2 d-1.) This agrees fairly well with laboratory measurements of the settling velocities of these particles in seawater. For example, Faisst (1980) reviewed reported experimental results for sludge and wastewater effluent particles in seawater and found settling velocities to range from about 10-2 m/day to 25 m/day. Wang (1988) reported that 90 percent of the particulate mass in a secondary wastewater effluent had settling velocities of ≤ 1 m/day; for a digested sludge the corresponding figure was about 3 m/day. As with particles in the open ocean, particles in wastewater effluents and sludges range in size over several orders of magnitude, from the submicron range to several tens of microns.
Many pollutants in wastewater effluents and sludges are associated with particles in these suspensions. Karickhoff (1984) has reviewed the thermodynamics and kinetics of the sorption of organic pollutants in aquatic systems. The sorption of uncharged organic chemicals to particles is dominated by hydrophobic interactions and depends primarily on a chemical's affinity for water, typically described by an octanol-water partition coefficient, and on the organic carbon content of the solid sorbent phase. An estimate of the sorption of an uncharged organic pollutant in a wastewater or sludge can be made on the basis of the chemical's octanol-water partition coefficient, the organic carbon content of the solid phase, and the concentration of solids in the water. For hydrophilic organic pollutants, nonhydrophobic contributions to sorption can be important and inorganic surfaces, such as clays and metal oxides, can be significant adsorbents. The sorption
of organic pollutants to suspended particles can generally be viewed as rapid, although true equilibrium may take weeks or more to achieve, and the process is often not completely reversible.
Several investigators have studied the transport and fate of nonpolar organic compounds in natural waters. Representative studies are given by Schwarzenbach and coworkers (Schwarzenbach and Westall 1981, Imboden and Schwarzenbach 1985). These studies and others show that the sorption of many organic compounds is proportional to the organic carbon content of the solids when this is greater than 0.1 percent and indicate that sorption can be predicted from the octanol-water partition coefficient of the nonpolar organic solute and the organic carbon content of the natural solid sorbent.
The adsorption of inorganic pollutants in aquatic systems has been reviewed by Dzombak and Morel (1987). In contrast to the nonspecific adsorption observed for hydrophobic organic pollutants, the adsorption of inorganic solutes is viewed as a site-specific process in which ions bind chemically at functional groups on solid surfaces. Surface complex formation models of varying complexity are available to model pollutant adsorption. The reaction is dependent upon pollutant type and concentration, pH, ionic strength, and solid concentration.
Particle-reactive pollutants from anthropogenic sources accumulate in marine sediments. An example of the effects of the organic particulate matter in a wastewater effluent on the transport and fate of synthetic organic compounds and of the deposition of metals in the discharge is given by Olmez et al. (1991). The site considered is the San Pedro Shelf off Southern California in the area of the outfalls of the Joint Water Pollution Control plant of Los Angeles County Sanitation Districts. Sediment distributions of organic carbon, DDT, and hydrocarbons as functions of depth or time correlate well with inputs of effluent particulate matter. Upper sediments are also enriched in light rare earth elements (La, Ce, Nd, Sm) and reflect use of fluid-cracking catalysts in the petroleum industry, and some release into the sewer system.
Transport and Fate of Wastewater Particles
When a wastewater effluent or sludge is discharged to the ocean, light, fresh water is introduced into dense, salty water. Three spatial regions are summarized here: 1) a zone of initial mixing or entrainment, often accomplished with a diffuser and characterized by a time scale of minutes; 2) subsequent transport and further dilution by tidal and wind-driven currents with time scales in the order of several hours to days; and 3) far-field transport driven by large scale circulation with time scales of several days to a few months. Brief descriptions of each follow.
The Outfall: Initial Mixing. In a point source discharge of wastewater to the ocean, fresh water containing particles at high concentration and that are moderately stable with respect to aggregation are mixed with seawater to produce a dilute suspension of particles that are probably chemically unstable with respect to their ability to attach to each other and form aggregates. When a diffuser is used to provide rapid effective initial dilution of the wastewater with seawater, the energy dissipated in the mixing process can provide contact opportunities between particles and contribute to particle aggregation. Extensive aggregation, if it occurs, will enhance the deposition of particles and particle-reactive pollutants in the sediments at the discharge site.
Extensive theoretical and laboratory studies of particle aggregation in wastewater plumes have been made by Wang (1988) and by Holman and Hunt (1986). These investigators disagree about the occurrence of coagulation during initial mixing in an outfall plume. Holman and Hunt concluded that there is a subregion of the zone of initial mixing within which coagulation occurs. It occurs because dilution is significant enough to provide salt for particle destabilization, while at the same time mixing and energy dissipation are fast enough and particle concentrations are high enough to provide particle contacts for aggregation. Further dilution with sea water slows coagulation appreciably because particle destabilization is not increased while particle concentrations continue to be lowered and energy dissipation continues to be reduced. Wang (1988) considered the changes in particle stability in the plume (from stable particles in the wastewater to unstable particles in the mixed discharge) and also the reductions in particle concentration and in energy dissipation that occur as dilution of the effluent proceeds; she concluded that coagulation is insignificant in the area of initial mixing or entrainment of a typical wastewater or sludge discharge because favorable conditions last for such a short time. Both of these studies are based on conceptual models and laboratory experiments; the occurrence and significance of coagulation in the zone of initial mixing of a wastewater plume have not been determined with accuracy.
The Outfall Region. The sedimentation of particles from sewage sludge discharged from ocean outfalls to coastal waters off southern California was modeled by Koh (1982), who concluded that sludge particles would be widely dispersed. Aggregation to enhance sedimentation was not considered. Coagulation of particles in wastewater and sludges has been studied conceptually and in the laboratory; results indicate that the process may affect particle transport and enhance particle deposition after discharge into coastal waters (Hunt 1980, 1982; Hunt and Pandya 1984). Farley (1990)
has considered particle decomposition, aggregation, and settling together with advective and dispersive transport in modeling the deposition of particles and the accumulation of organic matter in the regions of the Orange County and Los Angeles County outfalls in southern California. Model predictions of both particle deposition rates and organic accumulations compared well with observations at these two locations. It was concluded that coagulation, sedimentation, and tidal motion affect particle deposition and sediment accumulation around outfalls. The model contains empirical coefficients that must be determined from laboratory or field data.
Far-Field Transport. Modeling of the transport of particles in waste-waters at the scale of tens or hundreds of kilometers has been sparse. O'Connor et al. (1983a) considered the fate of wastewater sludge dumped at a deep water site (Dumpsite 106) in the New York Bight and predicted little effect on the bottom sediments. These authors used information about the settling properties of Los Angeles sludge in arriving at this prediction. Some observations indicate, however, that sediments below the dump site have been affected by the sludge inputs (Van Dover et al. 1992).
The Sediment-Water Interface
Wastewater and sludge particles that reach the sediment-water interface are then subject to a variety of physical, chemical, and biological processes that can enhance deposition, lead to resuspension, produce chemical dissolution or biological mineralization, and lead to the release or burial of particle-reactive pollutants and nutrients. For example, Stolzenbach et al. (1992) report that particles, including fine submicron particles, are removed from suspension by aggregation in a porous and mobile layer at the sediment-water interface driven by near bottom flow and also by the activities of benthic organisms. The authors suggest that this process may be particularly important in shallow waters.
A study of benthic recycling in Lake Superior by Baker et al. (1991) provides a paradigm for pollutant cycling at the sediment-water interface in coastal waters. As in the ocean, degradation of organic particles in Lake Superior is efficient so that only a small fraction of the primary production in the epilimnion is incorporated permanently into the sediments. Inputs of hydrophobic organic contaminants to the lake are dominated by atmospheric deposition, while burial in the sediments and volatilization are the major removal mechanisms. Contaminants include polyaromatic hydrocarbons (PAHs) and polychlorobiphenyls (PCBs). Most of the hydrophobic organic compounds introduced into the lake are incorporated into rapidly settling particles produced in the epilimnion. On reaching the sediments, many of these contaminants are released and mixed back into the water column by
biological processes in the benthic food web. Higher molecular weight PAHs, the most particle-reactive of the contaminants, are retained in the sediments and efficiently buried.
Discussion and Conclusions
A hypothetical description of the behavior of particles and particle-reactive pollutants in wastewater effluents and sludges prior to and after discharge to coastal waters is as follows.
-
Particles in both wastewater effluents and sludges can be enriched in hydrophobic organic contaminants such as PAHs and PCBs. In aerobic discharges, many metals will partition so that substantial fractions occur in both the solution and the particulate phases; in anoxic discharges, such as sludges and some primary effluents, most metals of concern will reside in solid phases such as particulate organic matter and metal sulfides.
-
Upon discharge, the particles will be chemically destabilized, thus favoring aggregation, but the extent of aggregation and enhanced settling that may occur during the initial dilution in the plume is not known. Chemical mineralization, oxidation, and sorption reactions are probably not important in this step, nor are biodegradation processes.
-
Some particles, perhaps most of them, will be deposited in a region up to several tens of square kilometers in size in the vicinity of the discharge. Aggregation in the water column among the wastewater or sludge particles and with marine particles may and probably does occur, but the process cannot be predicted accurately. Horizontal transport of particles by currents is dominated by tides, wind, and density gradients; vertical transport is driven by gravity, perhaps enhanced by aggregation and by processes at the sediment-water interface. The time scale in the water column is from several hours to a few days. Chemical dissolution, oxidation, and desorption reactions may occur, releasing some contaminants to the water column, but the results of these reactions cannot be forecast with accuracy. Biological degradation of particulate material in the water column is probably not great on this time scale.
-
The sediments in the region of the discharge will receive inputs of particles and particle-reactive pollutants. Rates of deposition and sediment accumulation cannot be predicted accurately. Biological degradation of organic matter and other processes can lead to a release of particle-reactive pollutants (metals, nutrients, and hydrophobic organic contaminants) to the overlying waters. Some fraction of many contaminants will be buried permanently in these sediments. The rates and effects of these reactions cannot be predicted quantitatively.
-
Some particles and particle-reactive pollutants in the wastewater or
-
sludge discharge will be transported horizontally for long distances. Predictions are few, and their accuracy has not been demonstrated. To these materials in a discharge will be added those pollutants previously deposited in the discharge region and then released from bottom sediments by biological and chemical processes.
This summary of the behavior in marine environments of particles from wastewater and sludges indicates that, while a substantial amount is known, much remains to be understood. The science is challenging. The summary also indicates that our ability to make accurate and precise predictions of the fate of these particles and of particle-reactive contaminants in the coastal zone has not been tested sufficiently. The design of wastewater or sludge discharges to coastal waters is constrained by our limited knowledge of these environments. This does not mean that discharges cannot be designed to protect the marine environment; it does mean that these designs must be conservative, incorporating factors of safety into the design process.
Two different approaches can be taken in evaluating the effects of particles and particle-reactive pollutants in the coastal zone. One is directed at sediment quality criteria and the other addresses water quality criteria. Considering sediment quality, all of the particles in the wastewater or sludge discharge can be assumed to settle to the bottom in a zone around the outfall, with the area of this zone determined by depth and tidal motion. The rate of deposition will depend on the mass rate of input of particulate matter by the discharge. Estimates of the pollutant concentrations in the sediments and the pore waters can be made and compared with appropriate sediment-quality criteria. Where a sediment quality problem is anticipated, pollutant emission can be reduced to meet the standard. Alternatively, additional and more detailed modeling and testing can be performed to improve the estimates. This approach to the problem is conservative because dilution of contaminated particles with ambient particles is neglected, transport of particles away from the discharge area is ignored, and release of contaminants from the bottom sediments is not considered. An example of this type of calculation is provided in the subsequent section on Sediment Quality Modeling.
With respect to water quality criteria, all of the particles and particle-reactive contaminants can be assumed to remain in the water column and be transported by wind, tides, and large-scale circulation. Initial concentrations in the water column would be determined by the initial dilution of the plume, corrected for background buildup. Transport and fate would be modeled assuming that the contaminants are conservative with respect to chemical or biological degradation and are not removed from the water by sedimentation. Where a water quality problem is anticipated, site-specific
modeling and testing can be performed to achieve more reliable predictions, and pollutant emission can be reduced to meet the standard.
Chemical and Biological Conversions of Toxics in situ; Biological Availability/Bioaccumulation
Chemical speciation serves as the common conceptual foundation on which geochemical, biochemical, and modeling studies have built our present understanding of the environmental chemistry of both organic contaminants and trace elements. Advances in the design of wastewater management systems will necessarily utilize this foundation. The term speciation includes important physical and chemical distinctions in the form in which an organic compound or metal ion is found. The most commonly measured type of chemical speciation of a toxicant involves separation of particulate and dissolved forms of a substance. Characterization of the chemical nature of particle-associated substances is currently the subject of intensive research. A second type of speciation involves identifying the chemical nature of solutes by distinguishing species engaged in reversible equilibrium reactions, i.e., acidic or basic forms of organic acids and inorganic or organic complexes of metal ions. Although these reactions are readily described by classical solution thermodynamics, the speciation of metals in the environment has only begun to be elucidated in recent years. Finally, speciation can involve distinguishing forms of trace elements that are not readily interconvertible by equilibrium reactions, e.g., alkylated metals and different redox states of metals. All of these kinds of speciation come into play when modeling the environmental fate of organic compounds and trace elements.
Speciation-based models of the environmental fates of potentially toxic trace metals and organic compounds couple equilibrium representations of the reversible reactions with rate laws for the transport and transformation processes. The importance of speciation lies in the fact that different species behave differently, e.g., particulate species settle and those associated with dissolved or colloidal organic matter are generally unavailable to biota. Consequently, the rates of the transport and transformation processes making up the environmental cycle of a toxicant are determined by the availability of its various species to the governing biogeochemical processes and the magnitude of those processes. Speciation-based models have been developed for organic solutes (Imboden and Schwarzenbach 1985) and trace metals such as manganese (Johnson et al. 1991) and mercury (Hudson et al. 1992) in lakes. The major limitation that these models currently face, however, is that in many cases our understanding of the principles governing metal cycling is ahead of our knowledge of the mechanisms and rate dependencies governing the processes, particularly transformation processes.
Transformation Processes
Waste materials entering the marine environment can undergo a wide variety of transformations such as photolysis, biodegradation, and hydrolysis. Transformation processes can be biotic or abiotic and are influenced by a variety of physical and chemical conditions. Biodegradation processes, for example, depend upon the population of microorganisms present as well as the structure of the particular organic compound being degraded.
The transformation of wastewater constituents in the marine environment can influence the bioavailability, transport, and fate of the materials. Degradation products can be more or less hazardous than the original compound. With metals, for example, methylated mercury is far more hazardous than inorganic mercury species. The debutylization of tin, however, results in a less toxic form of the metal.
There has been significant progress in understanding and unravelling transformations in the marine environment over the past decade. For example, understanding has evolved from the concept of partitioning between the particulate and dissolved phases to a recognition that there is a continuum of phases from truly dissolved through the colloidal state to true particulate. In this case, improved recognition of the importance of colloids in natural waters reveals that concepts commonly accepted ten years ago were simplistic and somewhat arbitrary, having been based on separation techniques using filters with pore sizes ranging from 0.2 to 0.5 pm.
Organics in Sediments
A considerable amount of research has recently been directed toward understanding the distribution of organic contaminants in sediments and the biological effects of the contaminants in the various sediment phases. The observation that bulk chemical concentrations of hazardous chemicals in sediments do not necessarily reflect the biological availability of those substances was a major impetus for the effort. The result is the equilibrium partitioning theory to determine concentrations of a substance in question among the various phases that exist in sediments. These levels are then correlated with concentrations known or thought to be toxic (EPA 1991). The equilibrium partitioning (EqP) concept assumes that, for non-ionic organic substances, most of the chemical in the sediment solid phase will be sorbed to organic carbon. It also assumes that pore water concentrations of the chemical correlate best to biological effects. The partitioning coefficient between the organic carbon and pore water (Koc) has been shown to be approximated by the partitioning coefficient between n-octanol and water (Kow) (EPA 1991). The relationship is given by:
![](/openbook/0309048265/xhtml/images/img00026.jpg)
Therefore, if the bulk sediment concentration, the organic carbon content of the sediment, and the Kow for a particular non-ionic chemical are known, one can calculate the pore water concentration of that substance at equilibrium. In a retrospective mode, one can compare this number to some acceptable concentration to determine if the sediment is likely to be harming the environment. In a prospective mode, knowing the acceptable pore water concentration and the Kow for a non-ionic organic, one can calculate a sediment quality criterion on a sediment organic carbon basis. An example of this type of calculation is provided in the subsequent section on Sediment Quality.
Trace Elements and the Importance of Speciation
Although knowledge of the concentrations and distributions of trace elements in coastal waters has advanced dramatically, it has become increasingly clear that information on just total concentrations is insufficient for providing an adequate understanding of a trace element's biological and geochemical interactions. Much of the uncertainty about the relationship between total metal concentrations and their toxicity to aquatic organisms results from a lack of definitive knowledge of the chemical forms of these metals in natural waters. Trace elements dissolved in seawater can exist in different oxidation states and chemical forms (species) including free solvated ions, organometallic compounds, complexes with inorganic ligands (e.g., with etc.), and complexes with organic ligands (e.g., with phytoplankton metabolites or humic substances). Particulate forms include metals adsorbed onto or incorporated into a range of particles from small colloids on the order of 10 nm to large particles resuspended from the bottom sediments by episodic events such as storm activity or tidal flushing. Advances made over the past 15 years in the understanding of metal speciation and its relevance to toxicity in marine ecosystems have important implications for the development of wastewater management strategies for coastal urban areas.
Organometallic Compounds. Organometallic forms of trace elements are those in which the trace element is covalently bound to carbon (e.g., methyl forms of As, Ge, Hg, Sb, Se, Sn, and Te; ethyl-Pb forms; butyl-Sn forms). The existence of naturally-occurring methylated forms of As (Andreae 1977, 1979), Ge (Hambrick et al. 1984, Lewis et al. 1985), Hg (Mason and Fitzgerald 1990), Sb (Andreae et al. 1981), and Sn (Byrd and Andreae 1982, Andreae and Byrd 1984) has been demonstrated in seawater. For example, a most astonishing discovery is that 90 percent of the oceanic Ge exists as methylated forms (CH3 Ge(OH) and (CH3)2Ge(OH)) that are so stable to degradation that they have been called the ''Teflon of the sea" (Bruland 1988).
Examples of important highly toxic organo-metal species include methyl-mercury and butyl-tin compounds. Recent studies have indicated that methyl-mercury species, rather than inorganic mercury species, accumulate in fish and pose a potential problem in human diets (Clarkson 1990, Grieb et al. 1990, Weiner et al. 1990). Methyl-mercury, which can be formed in aquatic environments by microbial activity, is far more hazardous than inorganic mercury species. Butyl-tin and phenyl-tin compounds are used as algicides and fungicides in paints and in agriculture, respectively. These highly toxic organo-tin forms degrade into inorganic tin species, which are less toxic.
Examples of important organ-metalloid species include organo-selenium compounds and methylated arsenic acids. Organo-selenium species can be bioaccumulated much more effectively than inorganic forms of selenium such as selenate or selenite (Fisher and Reinfelder 1991, Luoma et al. 1992). Water quality criteria need to take into account the chemical form or species of the various elements in both source waters and receiving waters, together with a knowledge of the potential transformations that can occur in natural water systems. Better understanding must be gained about the kinetics and mechanisms of production, degradation, and transformation of these organometallic compounds to improve modeling and standard setting.
Trace MetallOrganic Ligand Complexation. Potentially toxic metal cations include Cu, Zn, Pb, Ag, Hg, Cd, Ni, etc. These metals can occur in the marine environments as free cations, as relatively labile inorganic complexes, or as relatively inert coordination complexes (or chelates) with various organic ligands. Recent research has demonstrated that the biological response of planktonic organisms to these metals is related to the free metal cation concentration (Sunda 1988-1989). The toxicity of Cu and Cd to phytoplankton (Sunda and Guillard 1976, Brand et al. 1986); cadmium to grass shrimp (Sunda et al. 1978); and even the availability of the nutrient metals, Fe, Nm, and Zn, to phytoplankton (Brand et al. 1983) are believed to be related to the concentrations of the respective free metal ions, rather than by their total concentrations or the concentrations of specific organic complexes.
It is the lowest trophic level organisms that are most sensitive to these hazardous trace metals. For example, ciliate protozoans isolated from estuarine waters are among the most sensitive organisms to free copper (Stoecker et al. 1986); certain phytoplankton species (Brand et al. 1983) and zooplankton such as copepods (Sunda and Hanson 1987) have exhibited toxicity at very low levels of free copper.
Recently, marine chemists have advanced their ability to characterize the forms (i.e., chemical species) in which some of these trace metals exist in seawater (Bruland et al. 1991). For example, recent studies demonstrate
that greater than 90 percent (generally closer to 99 percent) of dissolved copper is organically complexed in coastal and estuarine waters (Sunda and Ferguson 1983, van den Berg 1984, Moffett and Zika 1987, Sunda and Hanson 1987, Coale and Bruland 1988, 1990). Complexation with these organic ligands appears to render copper essentially unavailable biologically and therefore, nontoxic. Consequently, the degree to which the metals in source waters from sewage effluents and urban runoff are complexed with organic ligands can determine their biological effect in the receiving waters. Measures of total copper concentrations, while easier to obtain, may have little relevance in predicting biological effects.
Bioavailability of Trace Metals. The factors controlling the biological availability of trace metals influence both their uptake by the food chain and their involvement in the biologically mediated reactions of their biogeochemical cycles. Because most biologically mediated processes require uptake of the element, either into cells or by surface sites of cells, biological availability can often be reduced to the relative rates of uptake of different metal species. A variety of mechanisms, each exhibiting different dependencies on metal speciation, can control the rates of biological uptake. Consequently, a complete understanding of the environmental fate of trace metals awaits the determination of the mechanisms by which metals are assimilated.
For example, the central mechanistic issue for mercury assimilation is whether it enters the cells in question by passive diffusion or by facilitated transport. Transport of neutral forms of some metals, e.g. elemental mercury and neutral complexes of methyl and divalent mercury, by passive means is rapid in lipid bilayer membranes (Gutknecht 1981, Boudou et al. 1983). Metal complexes with naturally-occurring organic chelators are generally not membrane permeable. Passive diffusion provides a baseline uptake rate that facilitated transport may supplement. In aquatic organisms, facilitated transport of toxic metals likely involves the uptake systems for nutrient metals, i.e., copper may enter cells through the uptake systems whose normal physiological function is to acquire zinc, manganese, or other essential metals. Such interactions have been previously observed in marine phytoplankton (Sunda et al. 1981, Harrison and Morel 1983). Facilitated transport generally involves complexation of the metals by specific ligands or binding sites. When this mechanism is significant for a toxic metal, the rate of its uptake would be influenced by both competition with the nutrient metal for transport and feedback between transport system kinetics and the nutritional status of the organisms. Both of these factors would cause the rates of toxicant uptake to be enhanced under conditions of low essential metal availability.
The best known relationship of facilitated transport rates to chemical
speciation involves control by free metal ion concentration or activity. This observation reflects the control of a variety of biological effects by the equilibrium binding to specific cellular sites. In some cases, equilibrium between the site and solution are not attained, and the rate of the process reflects the complexation rates of labile species, not just that of the free metal ion (Hudson and Morel 1990). In both equilibrium- and rate-controlled cases, strong chelators reduce the metal's availability by reducing the concentration of free metal ions and labile species.
Finally, whenever the rate of cellular uptake is high enough, slow diffusion through the extracellular medium can cause the transported species to become depleted at the cell surface. When species interconvert rapidly within the boundary layer, the uptake rate under diffusion-limitation will depend on the total concentration of labile species rather than the species that controls the rate of the transport step (Jackson and Morgan 1978). Generally, inorganic and weak organic complexes of a metal interconvert rapidly enough that they remain in equilibrium throughout the diffusive boundary layer.
Nutrient Cycling and Biostimulation
Eutrophication is perhaps the greatest and most obvious impact of wastewater disposal in estuaries and some shallow coastal areas (see Appendix A). Eutrophication results from a damaging excess production of phytoplankton or aquatic plants due to abnormally high inputs of nutrients. Nutrients are elements required for plant growth and include nitrogen, phosphorus, silicon, and sulfur as well as trace metals such as iron and molybdenum. Damage caused by excess growth may simply include reduced transparency or, at the extreme, hypoxia or anoxia caused by respiration and decay of dying plants and animals. Intermediate effects can include damage to coral reefs and blooms of noxious algae.
Phytoplankton production in most coastal areas is nutrient limited; i.e., increased inputs lead to increased net production. Phosphorus is the limiting nutrient in some temperate estuaries and in most tropical lagoons and estuaries. However, nitrogen is the limiting nutrient in most of the United States coastal zone, including most estuaries.
Because nitrogen is quickly assimilated by phytoplankton and plants in the coastal zone, it is nearly impossible to define a safe concentration of dissolved nitrogen. Instead, researchers have had to develop criteria and guidelines based on relationships between rates of nitrogen input and phytoplankton production or standing crop. Assessment of actual conditions in various estuaries of the United States, coupled with controlled experiments in large tanks (mesocosms), have lead to what this Committee believes to be a convergence of opinion on rates of nitrogen inputs that may be of
concern. This body of information is summarized in Appendix A, Figure A.6a, which suggests several ranges in inputs of interest. First, estuaries may maintain low phytoplankton standing crops (i.e., below 5 mg/m3 chlorophyll a, the upper end of the oligiotrophic range) noneutrophic when the rate of dissolved inorganic nitrogen (DIN) inputs is below about 100 mmol/ m3/y. Estuaries and coastal areas with DIN loadings between about 100 and 500 mmol/m3/y are mesotrophic, with phytoplankton standing crops on the order of 5 to 15 mg/m3 of chlorophyll a. Above about 500 mmol/m3/y of DIN (to 8,000) phytoplankton standing crop may range from about 15 to over 60 mg/m3 (or 3 to 120 times the production in oligiotrophic estuaries). Interestingly, Jaworski (1981) proposed that inputs to estuaries be kept below 380 mmol N/m2/y (a surface area criterion); this corresponds to a loading of 40 to 95 mmol/m3/y for shallow estuaries, 4 to 9 meters deep, a range of noneutrophication rates not inconsistent with the first category (i.e., less than 100 mmol/m3/y).
How do these rates relate to other coastal areas not included in the estuary studies, and how do they translate into actual emissions from real waste streams? An example is Santa Monica Bay. At present, no sewage DIN is discharged directly to the upper mixed layer, which on average is 20 meters deep (twice Jaworski's shallow estuary depth). In 1989, the Hyperion Treatment Plant (city of Los Angeles) discharged 9900 Mt DIN/yr, mostly below the mixed layer (data from SCCWRP 1991). If all the N went into the mixed layer (only a fraction may, due to accumulation by motile phytoplankton, Hendricks and Harding 1974), and if exchange and transport were in the same range as for shallow estuaries (i.e., Nixon 1988, Appendix A), the bay would still be in an oligotrophic range (DIN input = 70 < 100 mml/m3/yr) insofar as predicted chlorophyll a is concerned (assuming no other nutrient sources). This hypothesis is not inconsistent with observation (Eppley 1986) that suggests that the bay is, at most, mesotrophic. However, transport in this bay is in fact greater than in the estuaries reviewed in Appendix A. Only a fraction of the nitrogen input from the Hyperion Treatment plant is expected to enter the mixed layer in Santa Monica Bay. Again, there has been no evidence of damaging excess phytoplankton blooms in Santa Monica Bay over the past two decades (no dissolved oxygen depression, no green epicenter, no depression of secchi disc depth readings (light penetrations)), consistent with the expected lack of nutrient biostimulation.
The alternatives for controlling nutrient inputs from wastewaters include 1) no action, 2) relocation of the discharge to increase dilution and plume submergence and far-field dispersion, 3) source control, 4) treatment to remove limiting nutrients, and/or 5) changing the ratio of nutrients (provided that the limiting nutrient is at least partly removed). As noted above for Santa Monica Bay, the no action alternative for sewage wastewater is a reasonable one; at current sewage DIN input rates and location, there is no
damaging excess production and, therefore, there is no action that would make an improvement. However, there may be a need to evaluate shallow nearshore conditions to make sure that this is the correct conclusion when all possible sources to the mixed layer are taken into account. These would include non-point source urban runoff and other diffuse sources (including upwelling), as well as other point sources.
The second alternative is discharge relocation. Officer and Ryther (1977) argued that eutrophic conditions in some east coast estuaries would not be relieved by secondary treatment (which removes little DIN) but could be greatly relieved by diverting primary-treated effluents to the ocean. During the 1950s, there was biostimulation of plankton (whole plankton volume) nearshore in Santa Monica Bay associated with a shallow-water (20 meters) nearshore (1.6 kilometers) discharge; the area of excess phytoplankton disappeared when effluent was diverted to a deepwater (60 meters) diffuser (2,440 meters [8,000 feet] long) located 8 kilometers (5 miles) offshore (SCCWRP 1973), with greatly increased initial dilution and plume submergence below the thermocline (except in winter).
Such opportunities for diversion exist along much of the United States west coast, Alaska, the Pacific islands, and Florida but not within east coast embayments such as the Long Island Sound or the Chesapeake Bay. In these areas, source control (of agricultural fertilizer applications, farm practices) and treatment (beyond secondary) are required. These actions are now recognized in many area plans that include specific goals for reducing nutrient input rates.
To control biostimulation due to excessive nutrients in areas where it is or may be a problem, it is necessary to use an integrated management approach using the water- and sediment-quality driven approach to devise appropriate control strategies.
Sediment Processes
Many of the chemicals that have caused environmental impacts in estuarine or marine systems readily partition to particles, either suspended in the water column or on the bottom in sedimentary deposits. As such, there are numerous cases of sediment contamination that exist even though the original source of the substance has been eliminated. These deposits have become persistent sources of the contaminants to the overlying waters and the biota that reside in them. Therefore, not only must this source term be considered in modeling efforts for some areas but also the factors that govern the sorption and desorption of contaminants from particles must be understood to allow accurate fate and transport models to be constructed.
In general, fine-grained sediments, such as silts and clays, exert a greater influence on the quality of overlying waters than coarse-grained sediments
such as sand and gravel. This is due to a number of factors. The surface area per unit mass of the sediments is greater for finer grained sediments, which facilitates adsorption of some cationic species. In addition, the organic carbon content of sediments generally increases with decreasing sediment particle size, which enhances sorption of hydrophobic organic compounds. Therefore factors that enhance either transport, deposition, or resuspension must also be carefully considered in the formulation of fate and transport models.
Sediment Deposition, Resuspension, and Transport
Sediment deposition by gravitational settling is regulated by the size and density distribution of suspended particles. Particle coagulation by physical or biogenic processes may alter the size distribution and thus affect the rate of deposition by increasing the effective settling of fine particles that would otherwise remain suspended (Weilenmann et al. 1989). Because of the difficulty of making quantitative measurements of particle size, density, and coagulation efficiency, estimates of deposition are often based on observations of mass accumulation in sediment traps or the decrease of suspended mass in laboratory settling columns. Although neither of these methods is free from methodological problems (Wang 1988, U.S. GOFS 1989), they can provide valuable information on spatial and temporal variations in potential particle deposition.
Resuspension of deposited sediments occurs whenever the shear stress exerted on the bottom exceeds a critical value. Although the critical shear stress for larger, non-cohesive sediment particles (i.e., sand) can be fairly well predicted from an empirical relationship known as the Shields diagram, this method has been of little use for cohesive fine sediments, particularly when they are colonized by benthic organisms (Cacchione and Drake 1990). The rate of erosion once the critical stress has been exceeded has also been observed to vary from site to site for sediments with similar physical characteristics (Lavelle et al. 1984). Site-specific measurements of the critical stress have been obtained using inverted flumes in situ to produce controlled flow velocities over the bed surface for scour observations (Gust and Morris 1989).
Reasonably well-accepted methods for estimating the bottom stress from oceanographic measurements are now available and have been used in attempts to predict sediment resuspension in relatively open coastal waters (Grant and Madsen 1986). To date, these efforts seem to have been limited in predicting the long-term sediment accumulation by the variation in the critical shear stress as a function of grain size and depth in the sediment as well as by interactions between different grain sizes (armoring of fine sediment by larger grains) (e.g., Lyne et al. 1990). However, these methods are
useful in predicting the frequency of episodic erosion by large storms (Drake et al. 1985).
Once eroded, particles may be transported as bed load (rolling, hopping, or sliding along the bottom) or as suspended load (moving with the water) with the latter being the most likely mode for contaminant-laden fine sediments. Predictions of suspended load transport are often based on the assumption of local equilibrium, i.e., that the upward flux of particles is balanced by the downward settling flux. However, this model has been difficult to apply in practice because of the uncertainties in particle settling speeds and in the suspended concentration at the bed-water interface (Hill et al. 1988). Direct measurement of the suspended flux is feasible using a combination of current meters and transmissometer deployments, but the transmissometer must be carefully calibrated to account for particle size effects (Lyne et al. 1990).
Finally, episodic events such as spills or hurricanes can have a major influence on the ultimate fate of pollutants in sediments. Materials that have been stored in unconsolidated sedimentary deposits for decades can be dispersed in a matter of hours by abnormal waves or currents. Therefore, these phenomena must be considered in treatment and disposal system designs. The effects of episodes depend not only on the intensity of the episode but also on the water depth, basin geometry, and other factors.
As a result of the difficulties noted above in quantifying the deposition, erosion, and transport of fine sediments, it is not surprising that virtually all estimates of net sediment accumulation or loss are made on the basis of observations of the deposited sediment. Regions of relative sediment deposition or erosion may be inferred from an examination of sediment characteristics such as grain size and organic content (Nichols 1990). Net accumulation is commonly determined from the measured profiles of radioactive elements (Thorbjarnarson et al. 1985) or back-figured from models of diagenetic processes in the sediments (Berner 1980).
Sediment Mixing, Contaminant Reactions, and Release to the Water Column
The subsurface distribution of contaminants associated with deposited sediments may be altered by biogenic or physical mixing. In the marine environment, sediment mixing by organisms, where present, usually exceeds that resulting from waves and currents. It may transport sediment particles and porewater from the bed-water interface to a depth of 10 cm or more on a time scale of weeks to years depending upon the intensity of organism activity (Berner 1980). This mixing has been modeled as both a diffusive (Officer and Lynch 1989) and an advective (Fisher et al. 1980) process in the sediments, and, although experience has established some
bounds on the likely magnitude of mixing rates, the most reliable estimates are based on site-specific observations. Biogenic mixing may be particularly important in transporting contaminants that would otherwise be buried by net sediment accumulation to the bed-water interface where exchange with the water column and exposure of pelagic and benthic organisms may occur.
The subsurface distribution of a contaminant may also be affected by chemical and biochemical transformations, including, but not limited to, sorption and desorption of organic and inorganic compounds and elements, oxidation-reduction reactions, and microbial degradation of organic matter (Berner 1980). These diagenetic processes may result in remobilization of heavy metals and the detoxification of organic contaminants such as hydrocarbons and pesticides. Changes in the contaminant profile in the sediments and release of contaminants to the water column may also involve the processes of molecular diffusion and colloidal advection (Gschwend and Wu 1985). A variety of models of varying complexity have been developed to describe such reactions in conjunction with bioturbation and sediment accumulation (see Berner [1980] for a summary). Because of uncertainty in key parameters, these models have been used mostly to infer rates of sediment processes from measurements rather than to make quantitative predictions. Predicted rates of contaminant release from the sediments have been compared with observations obtained from flux chambers deployed on the bottom (Sayles and Dickinson 1991).
The United States Experience
Experience with contaminated sediments in the United States has been gained as a result of the Dredged Material Research Program of the Corps of Engineers (Palermo et al. 1989) and from studies at sites where PCB (Ikalainen and Allen 1989, Sanders 1989), DDT (Logan et al. 1989), or Kepone (Huggett 1989) are found at high concentrations in subsurface sediments. Future conditions at these sites have been predicted using models that parameterize advective and dispersive transport, sediment deposition and erosion, sediment mixing and reaction, and even contaminant uptake in the food chain (O'Connor et al. 1983b, Connolly and Tonelli 1985, Connolly 1991, Thomann et al. 1991). In most cases, both models and measurements indicate that, after source reduction has occurred, the contaminant will continue to be buried by subsequent accumulation of cleaner material and that the maximum concentration will be found some distance below the surface and may be reduced by dilution with the cleaner material (Logan et al. 1989, O'Connor 1990) (Figure C.3). However, even where the contaminant peak is buried, the flux of contaminant mass from the sediments to the water column may still be sufficient to maintain undesirable levels in aquatic organisms (Stull et al. 1986, Ikalainen and Allen 1989). Also, the effect of future source
controls can depend on whether the contaminant source contributes significantly to the balance of sediment accumulation (Stull et al. 1986).
Conclusions and Recommendations
-
Because of the difficulty of measuring the separate components of the net sediment balance, accumulation (or erosion) rates should be estimated from site-specific observations of sediment properties and profiles of appropriate tracers.
-
Improved methods of predicting the flux of contaminants through the sediment-water interface will depend upon continued collaboration between specialists in sediment transport, contaminant transformation, and benthic biology.
-
Sites of historical contamination may be self-mediating in that natural accumulations may result in continuing burial of the highest levels of contamination. However, such sites should be carefully monitored to assess the potential for changes in sediment dynamics.
-
Quantitative modeling of future conditions is feasible but requires an extensive data set documenting historical changes for the purpose of calibrating the models.
Sediment Quality
Definition and Criteria
Until recently, environmental agencies in this country relied on either Best Available Technology (BAT) or aqueous chemical concentrations in receiving water to regulate hazardous chemicals in marine environments. In the latter case, concentrations that result in acute or chronic toxicities to aquatic organisms (called Water Quality Criteria) are determined through laboratory exposures, and these are used to establish acceptable levels, called Water Quality Standards. Permits for discharges into coastal waters are often based on these regulatory levels or treatment technologies.
Many of the chemicals that have the potential to adversely affect aquatic organisms are hydrophobic and sorb to sediments. Contaminated sediments can impact not only the organisms that live in direct contact with the solids but also those that reside in the overlying water since the sediments themselves can act as a source of the toxic substances. Water quality or BAT strategies do not directly address hazardous substances in sediments. This and the fact that there are numerous coastal areas that already have contaminated sediments (NRC 1989) have led to the development of methodologies to establish protection criteria for toxics in bottom materials. These levels, sometimes called Sediment Quality Criteria or Sediment Quality Values, can be used in engineering calculations to determine whether a
given mass loading of chemicals in an effluent will likely result in toxic sediments or what measures are needed to prevent toxic conditions. They can also be used to gauge the existing or potential adverse biological impacts of in-place contaminated sediments.
The EPA has compiled ten methodologies that have the potential to assess sediment quality relative to chemical contaminants (EPA 1989). Some of the methods involve chemical analyses that allow for the establishment of chemical specific criteria, e.g., an acceptable level for phenanthrene in sediments. Others involve only biological observations that limit the results to assessing whether or not a sediment is toxic. Others combine chemical and biological measurements. A brief description of the ten methods is given in Table 4.4 (EPA 1989).
The Equilibrium Partitioning (EqP) approach is the only method that relies, in part, on a fundamental thermodynamic parameter, i.e. fugacity, to derive a numerical estimate of sediment quality. Fugacity is the tendency of a chemical to flow from one phase of a system to another until the free energy of that particular system is at its lowest and equilibrium is attained. Under such conditions, one can calculate the equilibrium concentration of a substance in any one phase of the system. The concentrations in the remaining phases can be calculated if the distribution of partition coefficients among the phases is known.
The EqP method assumes that only chemicals dissolved in the pore water phase are biologically available and therefore potentially toxic. It further assumes that for hydrophobic organic chemicals, the organic carbon fraction of bottom sediments will contain most of the sorbed material. With these assumptions, one can calculate acceptable sediment levels if one knows: 1) the pore water concentration above which toxicity is exhibited, 2) the organic carbon content of the sediments, and 3) the partition coefficient between organic carbon and pore water. Experimental evidence indicates that for many hydrophobic organic compounds, the partition coefficient of the substance between n-octanol and water approximates the partitioning between sediment organic carbon and water. Therefore the sediment quality value (SQV) on an organic carbon basis is: SQV = KocATV, where Koc is the partitioning coefficient of the chemical between sediment organic carbon and water (= partitioning coefficient between n-octanol and water) and ATV is a chosen Acceptable Toxicity Value. The EPA is in the process of establishing sediment quality criteria for a number of organic compounds based on the EqP methodology (for example, for acenaphthene see EPA 1991).
Sediment Quality Modeling
A basic question is whether it is possible to predict sediment concentrations of hazardous chemicals due to a particular discharge in order to pro-
tect the coastal environment as well as is presently done by BAT. The answer is that it is possible for a given discharge to calculate quite simply an upper bound for sediment quality on an organic carbon basis or, conversely, for a specified SQV standard to back-calculate a safe limit for the effluent quality (the sediment-quality driven approach).
The calculation of an effluent limit is illustrated for a POTW discharge by considering acenaphthene, a polynuclear aromatic hydrocarbon (PAH), as an example. In EPA 1991, the proposed criterion for ''acceptably" protecting benthic organisms in saltwater sediments is a concentration less than SQV = 240 glg acenaphthene/g organic carbon (goc), with confidence limits 110-520 µg/goc. To be safe, SQV = 110 µg/goc is used as the sediment quality standard to be met. Calculating backward to the source (for this example only one source is considered), the next question is: what fraction of the organic carbon in the sediment is derived from a particular POTW discharge; again to be conservative, assume 100 percent. If it is now conservatively assumed that there are no losses of the organic pollutant adsorbed on the suspended particles from point of discharge to sediment deposition, then the concentration of acenaphthene on the suspended solids in the effluent should be limited to 110 µg/goc. The concentration of particulate organic carbon co, is estimated to be ~ 0.4 css, where css is the concentration of suspended solids. Therefore, on the basis of suspended solids, the limit should be reduced to 44 µg/gss For the next step, it is necessary to know the suspended solids concentration of the POTW effluent (well known at every plant). For this example, assume that css = 75 mg/l, which is the effluent standard (average value) of Table A in the California Ocean Plan (CWRCB 1990). The safe effluent concentration limit for acenaphthene (cas)adsorbed to solids can now be calculated as follows:
![](/openbook/0309048265/xhtml/images/img00030.jpg)
To complete the calculation, one must estimate the dissolved concentration in the effluent, which would be in equilibrium with an adsorbed concentration of 110 μg/goc. Based on an organic carbon partition coefficient for acenaphthene of 6030 L/kg (log = 3.78) given by the EPA (1991), the calculated equilibrium dissolved concentration (caw) is
![](/openbook/0309048265/xhtml/images/img00031.jpg)
The total effluent safe limit is then
![](/openbook/0309048265/xhtml/images/img00032.jpg)
It is of interest to note that only 15 percent of the pollutant is attached to effluent particles in spite of its high Koc value because of the low particle concentration in the effluent.
Note that most changes in the assumptions used could only increase the calculated effluent limit. For example, if the deposited organic carbon from the POTW is diluted with equal amounts of clean natural organic carbon, then the calculated answer would be cat = 43 µg/L. Furthermore, if it was found that 10 percent of the particle-bound pollutant in the effluents is released to the ocean water before depositing to the sediments, the result would be multiplied by 1/0.9. The point is that with simple assumptions a safe effluent limit can be calculated, but if more were known, the required effluent limit would be larger-i.e., less stringent. Only if there were other sources-point of diffuse-would a lower effluent limit for the POTW be calculated to allow for the other sources or pre-existing contamination of sediments.
It is now interesting to compare the above result with the effluent limit derived from the California Ocean Plan (CWRCB 1990) for all PAHs (including acenaphthene). Table B gives a value of 8.8 ng/L as the standard for receiving water (specified for protection of human health [carcinogen]). For a typical initial outfall dilution of 100:1, this corresponds to an effluent limit of 0.88 µg/L. This value for all PAHs is about 25 times smaller than the limit for acenaphthene alone calculated above to meet the conservative sediment-quality standard. Thus, it would not have made any difference in the setting of the effluent limit to improve the conservative assumptions above because the health standard is much more stringent in this case. It is expected, however, that there will be other cases where the opposite is true, i.e., where the effluent standard needed to protect benthic organisms is smaller than that needed to protect human health. The sediment-quality driven approach provides a rational way to work this out in conjunction with water-quality driven analyses.
The reader is reminded that the above example is based on the Equilibrium Partitioning (EqP) approach, which is only one possibility. Long and Morgan (1991), in a National Oceanic and Atmospheric Administration study, give an extensive analysis and comparison of different methods of assessing toxicity using total sediment concentrations (mass of pollutant per unit mass of sediment) for a variety of trace metals, petroleum hydrocarbons, and synthetic organic compounds. The state of Washington has recently adopted sediment standards based on total sediment concentrations (see Becker et al. 1989 for discussion of criteria and approaches). The calculations to derive a corresponding effluent limit from a total sediment concentration standard would be different from the example above because the rate of deposition of organic carbon of sewage origin must be predicted, along with the losses from the sediments back into the water column. The EqP approach is
simpler to model because it deals only with concentrations of contaminant per unit mass of organic carbon, thereby allowing an easy link to wastewater quality.
Ultimate Sinks for Pollutants: Distribution of Pollutants in Water and Sediments in the Ultra-Far-Field
The question of what constitutes the ultimate fate of material discharged into the marine environment has no simple answer, in part because most of the constituents also occur naturally, if not necessarily at the high concentrations present in municipal effluent. This is particularly true of the elements associated with the enrichment of biological productivity in receiving waters, such as carbon, nitrogen, and phosphorus, but is also true of such potential toxicants as trace metals. Furthermore, some substances of concern, such as chlorinated hydrocarbons, can be broken down chemically into compounds of no concern and thereby have an ultimate fate, while the carbon atoms of which they are composed continue to be recycled. Even material that is buried in the sediments, usually considered to be the ultimate fate of any material, can return to the ecosystem under the right circumstances.
Part of the problem in answering the question of "What is determining the ultimate fate of material?" is that it represents the poorly formulated question about what are the long term, subtle effects associated with a discharge—the ultimate impacts—for which we do not, and cannot, have enough information to answer for any anthropogenic interaction with the environment. Despite the impossibility of ever completely understanding all environmental effects of a given discharge, the major effects can usually be quantified and studied in order to learn about potentially important subtle effects. The fact that it is not possible to effectively understand all the cascading environmental effects associated with a discharge, be it on land, into the atmosphere, or into the ocean, should keep us from getting complacent about such a discharge.
Most studies of environmental effects associated with marine discharge have focused on changes in the sediment composition and associated ecosystem effects. The benthos is a particularly good place to study because particle sedimentation concentrates pollutants, providing larger signals from a discharge. Furthermore, the relative immobility of benthic organisms makes it easier to relate ecosystem responses to a documented environmental change. Such is not the case for planktonic systems, where it is difficult to separate those organisms that have been exposed to a discharge from those that have not and for which the exposure need not continue for life-
times of organisms. This does not mean that planktonic effects cannot be important, as shown by the food chain transmission of DDT to pelicans off the coast of California until the 1970s, presumably mediated by the plankton.
From the narrow perspective, what constitutes the ultimate fate of a substance? For an undesirable substance with no natural sources, it would be its chemical degradation to an innocuous form or its permanent immobilization in the sediments at a location unlikely to be disturbed. The ultimate fate of a naturally occurring substance would be its dilution and dispersion into the environment to the point where its concentration is indistinguishable from the naturally occurring concentration. Here, again, there is ambiguity in the question of what is indistinguishable. On the one hand, it may be possible to measure even small differences between natural and altered systems. On the other hand, the natural spatial and temporal variability in natural systems can be so large that some anthropogenically induced differences may be meaningless when compared to the larger natural variability in total concentrations and ecosystem processes.
Dumpsite 106
Dumpsite 106 is a region offshore New York that has been designated for the surface disposal of sewage sludge. The surface is an average of 2,200 meters off the bottom (O'Connor et al. 1983a). The amount of material that reaches the bottom depends largely on the particle's sedimentation rates as well as the current patterns. O'Connor et al. (1983a) used simple settling and horizontal dispersion models to argue that only 20 percent of the sludge dumped there would reach the bottom in 200 days. The other 80 percent would continue to drift with the water. They argued that in the region underneath the actual discharge site, the anticipated impact of municipal dumping would be an increase in the benthic sedimentation rate of about 10 percent and in organic carbon sedimentation rate of about 20 percent.
Fry and Butman (1991) used more extensive information about sludge settling rates and about regional currents to predict that about 23 percent of the sludge would settle within 350 kilometers of the discharge site. This represented maximum increases in sedimentation rates of about 40 to 60 percent. Because most of this sedimentation is by faster settling particles, it would contain disproportionate amounts of denser grit and other inorganic particles.
Recent measurements of deposition rates and sedimentary composition in the region suggest that there are measurable increases (Van Dover et al. 1992). Preliminary results suggest that sedimentation rates of organic material are considerably larger than estimated above and represent a substantial
fraction of the natural sedimentation rates. There is no evidence of ecological impacts of this enhanced flux.
It appears that models for the fate of sludge discharged at Dumpsite 106 do not yet adequately predict the fate of material discharged there. Measured values for sedimentation appear to be higher than predicted. Efforts are under way to increase the accuracy of these model predictions and to compare them against field data. This is a joint effort of the EPA, the National Oceanic and Atmospheric Administration, and the United States Coast Guard (EPA 1990).
Understanding from Seattle Puget Sound
The southern end of Puget Sound, Washington is dominated by the anthropogenic inputs from the Seattle and Tacoma metropolitan regions. Paulson et al. (1988, 1989) used the narrow, fjord nature of the sound to make budgets of the fate of trace metals discharged into Puget Sound. About 70 percent of the lead, and 40 percent of the copper and zinc discharged in the region is deposited in the sediments of the 70-kilometer-long central basin of Puget Sound. The rest moves out toward the Strait of Juan de Fuca and Pacific Ocean.
Understanding from Southern California
The chemical oceanography of the Southern California Bight has been the subject of a recent review by Eganhouse and Venkatesan (1991).
Trace metals are mostly advected from sites of municipal waste discharge. There have been several estimates that more than 85 percent of the trace metals discharged at the Whites Point outfall are not deposited in the sediments of the surrounding Palos Verdes Shelf. Evidence of enhanced trace metal deposition has been found in several basins offshore. Bruland et al. (1974) estimated the fraction of different trace metals associated with anthropogenic sources that were settling within three inner basins of the Southern California Bight: Santa Barbara, Santa Monica, and San Pedro. The fraction of released metals accounted for in the sediments varied from 15 percent for cadmium to 69 percent for silver.
Fallout of particulate organic carbon is locally dominant but represents a smaller fraction of deposition in local basins. This is due, in part, to the greater sedimentation of natural organic matter that could dilute out the anthropogenic component. There is evidence of the deposition patterns from different outfalls overlapping each other in the changes in benthic organism composition found along the 60 m bottom contour (Mearns and Young 1983). This suggests that there can be small but detectable effects of discharge over large areas.
Conclusion
Researchers have been able to establish the ultimate fate of only a fraction of material discharged into the ocean.
Afterword
The ability and effort put into the prediction of environmental impacts associated with marine discharges before a discharge is permitted are quite sophisticated in comparison to the limited data collection and analysis used to test predictive models after discharge commences. From a regulatory point of view, that of designing discharge systems that meet regulatory requirements, current models can predict environmental impacts. However, at the scientific level, the ability does not exist to predict detailed distributions of pollutants and their movement through the ecosystem. This disparity stems from the fact that there is no systematic program to test predictive models against actual discharges. Note that testing a model is distinct from running a monitoring program because the intensive data collection and analysis needed to test a model is not required to determine whether a system is complying with environmental regulations.
The main consequence of poor model predictions is likely to be overdesign of facilities. When models have uncertainties, safety factors are applied in the design of a facility to ensure that environmental objectives will be met. The application of excessive safety factors results in overbuilt facilities and excessive expenditures. Model comparison with real prototypes provides valuable information for the improvement of models. In the event a project was underdesigned, model prototype comparison will indicate what needs to be done to correct the problem.
Recommendation
Waste dischargers should institute programs to verify the models used to predict environmental impacts of discharges. These efforts should be distinct from monitoring requirements and of limited duration. Their costs might be considered as part of design and construction costs rather than as monitoring costs.
OVERALL DESIGN OF DISPOSAL SYSTEMS, CONTROL OF DIFFUSE SOURCES, AND USE OF MODELS
For a coastal community planning an ocean discharge, the task of choosing an outfall location, depth, design, and allowable mass loadings is basically an iterative one subject to definite but somewhat elastic boundary condi-
tions. Some of these boundary conditions pertain to water and sediment quality, others to factors associated with such issues as foundation stability; earthquake loadings; forces due to waves, currents, and possible tsunamis; construction technology; and cost of construction. Additional factors that enter the overall design task of the outfall are related to system components associated with the outfall within the overall wastewater infrastructure system including interceptors, treatment plants, and pumping stations. This section addresses only issues related to water quality and not those associated with the other factors.
Steps in the Design of a Disposal System (new systems and upgrading existing systems)
The design of a system for wastewater disposal into the coastal wastes depends on three basic inputs:
-
the water and sediment quality objectives,
-
the characteristics of the receiving water, and
-
the volume and composition of the wastewater to be discharged from an outfall.
Each of these items is a large array of information with temporal variations, or frequency distributions. Designs should not be for a single condition; they should be robust and work for all conceivable conditions with perhaps some intentional exceptions.
Furthermore, these inputs are not fixed and often change during the design in a feedback process. Consider the following examples. First, the regulatory agency or the discharger may set very high goals, but when faced with the engineers' estimate of the costs, decide to back off. This happened in San Francisco to the program to control combined sewer overflows to an average of only one spill per year; then, because of cost, it was relaxed in several steps to an expected average of 8 overflows per year! But if the cost was considered reasonable for the target of one per year, they would have stuck to it.
Second, consider the environmental characteristics of the receiving waters. These are not fixed either before the design starts because the outfall designer has some choice of location of the discharge point, i.e., depth and location, to optimize the system.
Third, the quality of the wastewater depends on the treatment processes chosen and the stringency of the measures for source control of toxics. From the overall systems point of view, the more treatment and source control implemented, the less that is required of the outfall to achieve the same water quality; conversely, a really good outfall reduces the need for
treatment and source control measures. Thus, for a water-quality/sediment-quality driven approach, neither the optimum design of the outfall nor the appropriate selection of treatment levels and source controls can be made without analyzing the whole system for various combinations. Even the flow capacity of the system may be uncertain due to plans for storm water detention for later treatment or future wastewater reclamation for reuse.
The Water-Quality Driven Approach
The next step in risk management is to design engineering control systems that may be used to achieve compliance with various water-quality objectives established to manage risks to human health and ecosystems. In the use of water quality objectives, sediment quality objectives are also included. Sediments can also be significant routes of exposure to ecosystems and humans (through shellfish and benthic fish).
The design of engineering systems based on water quality objectives is called the water-quality driven approach, shown schematically in Figure C.4. The figure and the following discussion refer to a municipal wastewater disposal system in order to illustrate the concepts; however, the same concepts and procedures can be extended to water bodies with multiple inputs from point and diffuse sources.
The first step (top of Figure C.4) is the expression of the environmental objectives, in terms of water quality objectives or standards from which the engineering design may proceed. The chart then displays what appears as a backwards calculation: from these standards is derived the combination of engineering measures that is best suited to reach the prescribed water quality. There are three basic types of engineering components for a municipal wastewater system:
-
the outfall(s), including location and characteristics of multiport diffusers used for high dilution.
-
the treatment works (POTWs), with various possible components and levels of treatment (see Appendix D).
-
source control (or source reduction) to limit the amount of toxic substances or other pollutants entering the sewer system and the treatment plant (see Appendix D).
These three parts constitute a system in which changes in one part will change the need for the others. For instance, a long outfall with high initial dilution generally reduces the need for secondary treatment; or better source control reduces the need for toxics removal at the POTW and simplifies the sludge disposal problem. Only by considering all three components and their associated environmental and financial costs and benefits at once can
the optimum combination be found. The conceptual design then proceeds with a trial choice of outfall and related treatment levels and source control programs deemed necessary to meet the ambient water-quality standards. The process involves complex modeling of transport and fates of contaminants in the ocean after initial dilution. Given a certain outfall configuration, it can be determined in this way the effluent limits needed at the POTW. To meet these, appropriate treatment components and upstream source control measures are then selected.
As a practical matter, such modeling must be done in the forward sense (sources→treatment→outfall→transport and fate→effects), but with iterations it is conceptually the same as Figure C.4 with the reversed order of conceptual steps. However, with experience it is not difficult to work backwards from water quality and sediment quality standards to get approximate solutions for the three system components, which are then used as the initial iteration for the detailed forward water-quality modeling (Figure C.5).
Transport and Fates Modeling: Predicting Ambient Water and Sediment Quality
Mathematical and conceptual models are used extensively to explain observed processes that disperse and modify pollutants in the ocean. Various submodels may be combined to produce an overall model to relate pollutant inputs to water and sediment quality for single and multiple sources. These models are fundamental to management by the water-quality driven approach because the limits on emission for any discharge or nonpoint source may be back-calculated through the models. Because of the various length and time scales associated with different pollution problems, a variety of models is needed—one mathematical model cannot provide answers to all questions.
The following discussion relates to the wastewater disposal system for a municipality including source control, a treatment plant (POTW), and outfall. However, the three kinds of required information apply equally well to all other types of pollutant sources and the approach to devising an engineering system. For example, in the case of combined sewer overflows, there is a set of water quality objectives; some knowledge of the environment; and information on the amount, quality, frequency, and distribution of existing combined sewer overflows. If there are multiple contributing sources to the water quality, then the environmental modeling must integrate the effect of all sources and develop scenarios for different degrees of control and handling of different sources. Sources of the same kinds can be combined into classes to simplify the modeling, such as one for a large number of small POTWs all affecting the water quality of a large body of water such as Long Island Sound.
Predictive models have a number of uncertainties and need improvement, but nonetheless appropriate engineering systems for wastewater disposal and diffuse source control can be designed to meet prescribed waterquality and sediment-quality objectives.
Choice of System Components: Alternative Systems
The actual design process for an ocean disposal system for a POTW is illustrated in Figure C.5. It is an iterative process based on the waterquality driven approach. The following steps correspond to the numbered boxes in the figure.
Box 1. Water quality and sediment quality objectives are established. These objectives may either be regulatory standards for the receiving water and sediments or in some cases the effluent or they may be qualitative mandates related to health of ecosystems.
Box 2. The assessment of the current situation is the starting point for improvement, including identifying the problems that need to be addressed, such as those enumerated in Table 4.8 on page 122, with a full knowledge of the wastewater characteristics, including the whole range of toxics, the evaluation of existing facilities (whether they are to be used or replaced or upgraded), and a full accounting of all the sources and stressors for the location of the particular project. The latter is necessary in order to do the multiple source effects evaluation as required by the right-hand boxes adjacent to Boxes 8 and 13 in the diagram.
Box 3. It is important for the designer to make a full evaluation of the existing ocean data on temperature, salinity, and density profiles with seasonal variations; ocean currents; chemical constituents, including nutrients; dissolved oxygen profiles; present state of contamination of sediments; and a description of the biota and living resources to be protected and where they are. There is no use in acquiring new oceanographic data before evaluating what is already known and where the gaps are and are not.
Box 4. Based on experience, the designer can make a preliminary choice based on the logic of Figure C.4 of what treatment components and source control measures would most likely be required. For example, if a body of ocean water is poorly flushed and not very deep, then secondary treatment is very likely to be necessary; on the other hand, if the discharge is in deep water relatively well flushed (as Congress envisioned in establishing the 301(h) waiver provision), then enhanced primary treatment would be a likely choice rather than full secondary because the biochemical oxygen demand does not need to be removed for such a case.
Box 5. Different outfall choices are usually considered on a parametric basis, for example, maybe two outfall lengths (and different depths) and
three diffuser lengths. Engineering constraints include, for example, the bottom bathymetry and geology, which may preclude siting in certain places. The type of outfall, such as tunneled versus steel pipe versus concrete pipe, may be dependent on various engineering considerations and costs. The longer outfalls also cost more so that different candidates will have different costs as well as different benefits associated with them.
Box 6. The near-field modeling consists of initial dilution calculations by various buoyant plume formulas of which there are a number of acceptable models (e.g., the EPA's group of computer models: Muellenhoff et al. 1985, Baumgartner et al. 1992). Also calculated is the height of rise of the wastewater plume, which may be kept submerged below the pycnoline due to the density stratification of the ambient water. Submergence is highly advantageous in many situations to control or completely prevent impacts to the surface-mixed layer or the shoreline. Initial dilution and sewage field submergence can be predicted reasonably well (~±25 percent for no current, worse for currents). Further research can improve the predictions but probably will have little impact on design choices.
Box 7. At this stage, it is useful to do a quick scan to see which water quality objectives and sediment objectives are already met. For example, if there are requirements for 50 constituents, it is no use to carry forward the modeling for those that are obviously going to meet the water and/or sediment quality objectives. Rather, attention should be focused on those that look as though they are going to be the key drivers of the design. In the past, coliform concentrations were often the critical factors, whereas now it is likely to be nutrients or toxics in sediments.
Box 8. The next step is approximate far-field modeling considering further mixing, advection, biological and chemical conversions, sedimentation, and final fate of particles. This may be done initially in an approximate way in order to screen out outfall locations that are going to turn out to be unsatisfactory so that no further time and effort is wasted in data collection at such sites in the ocean or doing analyses that will lead to unacceptable results. This is an extremely important step in order to focus resources on engineering alternatives that are going to be successful. One cannot wait until after one gets the full-blown oceanographic study because one needs this preliminary judgment in order to plan and focus the oceanographic study (Box 12). The overlap with far-field effects of other point or diffuse sources is also considered here in a preliminary way.
Boxes 9 & 10. The requirements can be further screened and it is possible to see which remain to be satisfied and what to do about them. For example, if only coliforms (pathogens) are still excessive at certain target areas, one needs to decide what to do next. One choice would be to disinfect the effluent full time or intermittently—that represents the arrow that goes from Box 10 back to Box 4. Or alternatively, one may decide to make
the outfall longer and deeper, which is the arrow from Box 10 back to Box 5. Also, if the chosen candidate outfalls and treatment seem to be excessive, i.e., run up more cost than is necessary to achieve the desired objectives, then revisions in the opposite direction might be made. Finally, if it appears that the objectives are impossible to satisfy at any reasonable cost, then the arrow from Box 10 back to Box 1 indicates a possible change of objectives. For example, a small kelp bed or shellfish bed near the proposed discharge might be closed to beneficial use at a small environmental cost compared to the possibly large monetary cost to do otherwise.
Boxes 11 & 12. If the planned project appears viable and cost effective, then a major discharger would proceed typically with a year-long detailed environmental survey of the physical, chemical, and biological characteristics in the ocean in the area of the proposed outfall site and the target areas where water quality is to be protected. Included would be continuously recorded currents at different depths; profiles and transects of temperature, salinity, density, dissolved oxygen, and nutrients; measurement of toxicants in organisms and sediments; and biological assessments.
Box 13. With this information then, a full-blown mathematical modeling is undertaken with the near-field and far-field behavior of the projected waste discharge with full consideration of the combination with other sources. The predicted results will include spatial variations and frequency distributions of various water-quality parameters in the water column and rates of accumulation of pollutants in sediments.
However, the ability to do this is not perfect, but it is growing and it would be advantageous for the profession to have more post-construction evaluations of designs of systems that have been put into operation. The whole water-quality driven approach can easily assimilate new scientific information and oceanographic data as it becomes available for future adjustments or corrections of management plans.
Box 14. By this time, a completely satisfactory design has been developed with perhaps only minor refinements needed.
Box 15. These refinements are then carried in an iterative process by the arrows shown back to Boxes 1, 4, and 5 from Box 15.
Box 16. The next question is whether the system proposed is the most cost-effective system to meet the objectives, and, if not, adjustments can be made as needed.
Box 17. The proposed design and some viable alternatives can be presented to decisionmakers. Sometimes systems are developed that will go well beyond existing requirements in anticipation of future upgrading or tightening of requirements. Decisionmakers may often prefer such an alternative because the life of outfalls may be 50 to 100 years, far longer than the lifetime of many regulations.
Many coastal outfalls have been designed and built in the past two to three decades. Those that were designed generally using procedures outlined above have performed well and largely in accordance with predictions or better; some have probably been over designed because of large safety factors to cover uncertainties. High dilutions and plume submergence have been obtained when predicted.
Present day upgradings are being driven mostly by more stringent water-quality requirements and increased loadings, rather than incorrect predictions of performance at the time of initial design. A new factor is sediment quality, which was usually not directly included in design considerations before about 1980.
Quality driven approaches are not as well developed for nonpoint sources, but the principles are basically the same. Multiple point and diffuse sources can all be logically integrated into environmental-quality driven calculations as part of integrated coastal management.
Discussion
The Quality-Driven Approach
The preceding sections address the range of scientific knowledge and engineering techniques related to the processes by which wastewater treatment plant effluents can be discharged to coastal waters safely. Much of this knowledge has evolved within the past three decades. Since scientists and engineers now have a good basic understanding of these processes, the management of coastal water and sediment quality can be addressed through a logical scientific framework. Predictions can be made of the benefits and costs of various control actions on a case-by-case basis or by classes. Future research will contribute to this improved understanding of coastal waters, and allow for a shift from mandated technologies to the water-quality and sediment-quality driven approach. It is only through the latter approach that the most technically and cost effective control measures can be identified.
Because many uncertainties have been described, it may seem that the integration of all the necessary scientific information and engineering techniques for the purpose of making a decision is a hopeless task. But scientists, engineers, and other professionals working together with the public can sort out the key factors and focus on solving the most important problems in a cost effective manner through the integrated coastal management process discussed in Chapters 3, 4, and 5.
It is clearly possible to design pollution control systems to achieve water and sediment quality sufficient to meet specified standards with appropriate safety margins. The outlines of this approach are provided in this
appendix with additional information presented in Appendix A on nutrients, Appendix B on pathogens, and Appendix D on wastewater treatment and stormwater management. The water-quality and sediment-quality driven approach is the only one that can be applied logically to multiple point and diffuse sources and rapidly assimilate new research and monitoring results.
While water quality modeling is well developed, sediment quality modeling is new within the past decade and is not fully developed yet. While currently there is limited agreement on the way in which sediment quality standards should be specified, it can be anticipated that a consensus will develop over the next few years.
Toxicants
Toxicants have received great attention during the last two decades (but little before that), and strong control measures have been implemented in many areas. Source control and source reduction have proved to be effective measures for many POTWs. As effective source control programs are implemented the toxics problems will evolve toward one primarily associated with either sediment beds contaminated with past deposits or diffuse sources that are still unregulated or uncontrolled. Further work on the chemical speciation of metals in relation to toxicity will help to focus and refine requirements. In the meantime, toxicity limits for metals are established without regard to speciation. Many POTWs have found source control programs to be easier to implement than expected. Still illegal discharges continue to pour into many storm drains and are polluting the shorelines.
Particles
Residual particles (or suspended solids) in the effluent may be a concern for several reasons: they may be carriers of adsorbed pollutants; they may reduce light levels; they may contribute to nutrient enrichment; and they may, by settling, decrease the dissolved oxygen in the water column and sediments. But whether these problems actually exist depends on the circumstances. For example, if toxics are well controlled by source control in the sewer system, then toxics transport by particles may be below the levels where any standards would be violated. Similarly if the wastewater plume is confined below the thermocline where light levels are already low, then there is no effect on the euphotic zone above the thermocline. Also, when a region is well-flushed, then nutrient buildup will not likely be a problem. Finally, for outfalls producing high dilutions, all of the effects are reduced.
Thus, acceptable limits for suspended solids concentrations are site-
specific, but they can be worked out by the water quality/sediment quality approach explained in this appendix.
Nutrients
The need to limit nutrient inputs to coastal waters is also site or region specific. Coastal waters that are impacted by excessive nutrients (usually nitrogen) usually receive these inputs from a variety of sources, including some natural inputs (for example Long Island Sound). In such cases, it is absolutely essential to follow an integrated coastal management plan (as explained in the main report) in order to achieve any results. Tightening up on minor sources may be a real waste of effort if major sources are left uncontrolled. An overall water-quality modeling including all sources is necessary to first understand the system then to devise the most effective control measures.
In the long run, the nutrient enrichment problems in some areas may be the most difficult and expensive problems to solve; by comparison, toxics appear to be coming under control, with the residual in sediments being the remaining issue.
Better Integration of Field with Laboratory and Computer
An existing outfall discharge (or multiple discharges) is a full-scale prototype that can be studied and compared with mathematical and laboratory models. If models can reproduce what occurs now, then they reduce the uncertainties in planning the next level of improvement of environmental quality. Furthermore, post-construction field investigations are valuable to compare predictions with the actual performance. Such information provides a valuable feedback for planning future control measures.
SUMMARY
-
Integrated coastal management requires the use of the water-quality and sediment-quality driven approach to model and manage the effects of single or multiple discharges and diffuse pollution sources and to make effective regional control strategies.
-
Predictive models have a number of uncertainties and need improvement, but nonetheless appropriate engineering systems for wastewater disposal and diffuse source control can be designed to meet prescribed water-and sediment-quality objectives.
-
There is much to be learned from existing problem discharge situa-
-
tions that is useful to support modeling and engineering efforts for designing new or upgraded facilities.
-
Our ability and effort to develop and use mathematical and conceptual models is ahead of our field confirmation of the accuracy of models. More effort is needed to study prototype systems after construction to evaluate the pre-construction modeling and analysis.
-
A continuous, responsive approach is needed for future management of major discharge areas, including on-going ocean studies and flexibility of management to modify the discharge system as needed in response to new research findings, new problems, or new environment objectives.
-
Coastal water-quality management must be site (or region) specific because of widely varying conditions along the coastline of the United States.
-
Because of the wide range of length and time scales of various ocean processes, and the various time scales of various water-quality problems, different modeling approaches are required for different pollutants.
REFERENCES
Alldredge, A.L., and C.C. Gotschalk. 1989. Direct observations of the mass flocculation of diatom blooms: Characteristics, settling velocities, and the formation of diatom aggregates. Deep-Sea Research 36:159-171.
Allen, J.S. 1980. Models of wind-driven currents on the continental shelf. Annual Review of Fluid Mechanics 12:389-433.
Andreae, M.O. 1977. Determination of arsenic species in natural waters. Anal. Chem. 49:820.
Andreae, M.O. 1979. Arsenic speciation in seawater and interstitial waters: The influence of biological-chemical interactions on the chemistry of a trace element. Limnol. Oceanogr. 24:440.
Andreae, M.O., and J.T. Byrd. 1984. Determination of tin and methyltin species by hydride generation and detection with graphite-furnace atomic absorption or flame emission spectrometry . Anal. Chim. Acta 156:147.
Andreae, M.O., J.F. Asmod, P. Foster, and L. Van't dack. 1981. Determination of antimony (III), antimony (V), and methylantimony species in natural waters by graphite furnace atomic absorption spectrometry with hydride generation. Anal. Chem. 53:287.
Baker, J., S.J. Eisenreich, and B.J. Eadie. 1991. Sediment trap fluxes and benthic recycling of organic carbon, polycyclic aromatic hydrocarbons, and polychlorophenyl congeners in Lake Superior. Environmental Science and Technology 25:500-508.
Baptista, A.M., E.E. Adams, and K.D. Stolzenbach. 1984. The solution of the 2-d unsteady convection-diffusion equation by the combined use of the DE method and the method of characteristics. Proc. 5th International Conf. on Finite Elements in Water Resources. Burlington, Vermont: University of Vermont.
Baumgartner, D.J., W.E. Frich, P.J.W. Roberts, and C.A. Bodine. 1992. Dilution Models for Effluent Discharges. EPA Draft Report.
Becker, D.S., R.A. Pastorok, R.C. Barrick, P.N. Booth, and L.A. Jacobs. 1989. Contaminated Sediments Criteria Report. Olympia, Washington: PTI Environmental Services, for the Washington Department of Ecology.
Berner, R.A. 1980. Early Diagenesis. Princeton, New Jersey: Princeton University Press.
Boudou, A., D. Georgescauld, and J.P. Desmazes. 1983. Ecotoxicological role of the membrane barriers in transport and bioaccumulation of mercury compounds. Pp. 117-136 in Aquatic Toxicology, J. Nriagu, ed. New York: Wiley-Interscience.
Brand, L.E., W.G. Sunda, and R.R.L. Guillard. 1983. Limitation of marine phytoplankton reproductive rates by zinc, manganese, and iron. Limnol. Oceanogr. 28:1182-1195.
Brand, L.E., W.G. Sunda, and R.R.L. Guilland. 1986. Reduction of marine phytoplankton reproduction rates by copper and cadmium. J. Exp. Mar. Biol. Ecol. 96:225-250.
Brooks, N.H., and R.C.Y. Koh. 1965. Discharge of Sewage Effluent from a Line Source into a Stratified Ocean. Paper 2.19. Leningrad: XIth Congress of Intern. Assoc. for Hydr. Research.
Bruland, K.W. 1988. Trace element speciation: Organometallic compounds and metalorganic ligand complexes. Applied Geochemistry 3:75.
Bruland, K.W., J.R. Donat, and D.A. Hutchins. 1991. Interactive influences of bioactive trace metals on biological production in oceanic waters. Limnol. Oceanogr. 36:1555-1577.
Bruland, K.W., K. Bertine, M. Koide, and E.D. Goldberg. 1974. History of metal pollution in southern California coastal zone. Env. Sci. Technology 8:425-432.
Byrd, J.T., and M.O. Andreae. 1982. Tin and methyltin species in seawater: Concentrations and fluxes. Science 218:565.
Cacchione, D.A., and D.E. Drake. 1990. Shelf sediment transport: An overview with applications to the northern California continental shelf. In The Sea, Vol. 9, Part B, B. Le Mehaute and D.M. Hanes, eds. Wiley Interscience.
Chao, S. 1990. Instabilities of fronts over a continental margin. Journal of Geophysical Research 95:3199-3211.
Clarkson, T.W. 1990. Human health risks from methylmercury in fish. Environmental Toxicology and Chemistry 9:957-961.
Coale, K.H., and K.W. Bruland. 1988. Copper complexation in the northeast Pacific. Limnol. Oceanogr. 33:1084-1101.
Coale, K.H., and K.W. Bruland. 1990. Spatial and temporal variability in copper complexation in the north Pacific. Deep-Sea Res. A 37:317-336.
Connolly, J.P. 1991. Application of a food chain model to polychlorinated biphenyl contamination of the lobster and winter flounder food chains in New Bedford Harbor. Environmental Science and Technology 25:760-770.
Connolly, J.P., and R. Tonelli. 1985. Modelling kepone in the striped bass food chain of the James River Estuary. Estuarine, Coastal and Shelf Science 20:349-366.
Csanady, G.T. 1983. Dispersal by random varying currents. Journal of Fluid Mechanics 132:375-394.
CWRCB (California Water Resources Control Board). 1990. California Ocean Plan, Water Quality Control Plan, Ocean Waters of California. Sacramento, California: CWRCB.
Drake, D.E., D.A. Cacchione, and H.A. Karl. 1985. Bottom currents and sediment transport on San Pedro Shelf, California. Journal of Sedimentary Petrology 55(1):15-28.
Dzombak, D.A., and F.M.M. Morel. 1987. Adsorption of inorganic pollutants in aquatic systems. Journal of Hydraulic Engineering 113:430-475.
Eganhouse, R.P., and M.I. Venkatesan. 1991. Chemical oceanography and geochemistry. In Ecology of the Southern California Bight.
EPA (U.S. Environmental Protection Agency). 1989. Sediment Classification Methods Compendium. Draft Final Report.
EPA (U.S. Environmental Protection Agency). 1990. Monitoring, Research, and Surveillance Plan for the 106-Mile Deepwater Municipal Sludge Dump Site and Environs. Tech Rept. EPA-503/4-91/001. Washington, D.C.: Office of Water, U.S. Environmental Protection Agency.
EPA (U.S. Environmental Protection Agency). 1991. Proposed Sediment Quality Criteria for
the Protection of Benthic Organisms: Acenaphthene. Washington, D.C.: U.S. Environmental Protection Agency.
Eppley, R.W., ed. 1986. Plankton Dynamics of the Southern California Bight. Lecture Notes on Coastal and Estuarine Studies. New York: Springer-Verlag.
Faisst, W.K. 1980. Characterization of particles in digested sewage sludge. Pp. 259-282 in Particulates in Water: Characterization, Fate, Effects, and Removal, M. Kavanaugh and J. O. Leckie, eds. ACS Advances in Chemistry Series No. 189. Washington, D.C.: American Chemical Society.
Farley, K.J. 1990. Predicting organic accumulation in sediments near marine outfalls. Journal of Environmental Engineering 116(1): 144-165.
Fischer, H.B., E.J. List, R.C.Y. Koh, J. Imberger, and N.H. Brooks. 1979. Mixing in Inland and Coastal Waters. New York: Academic Press.
Fisher, N.S., and J.R. Reinfelder. 1991. Assimilation of selenium in the marine copepod Acartia Tonsa studied with a radiotracer ratio method. Marine Ecology Progress Series 70:157-164.
Fisher, J.B., W.J. Lick, P.L. McCall, and J.A. Robbins. 1980. Vertical mixing of lake sediments by tubificid oligochaetes. Journal of Geophysical Research 85(C7):3997-4006.
Fowler, S.W., and G.A. Knauer. 1986. Role of large particles in the transport of elements and organic compounds through the oceanic water column. Progress in Oceanography 16:147-194.
Franks, P.J.S., and D.M. Anderson. 1992. Alongshore transport of a toxic phytoplankton bloom in a buoyancy current, Alexandrium Tamarense in the Gulf of Maine. Marine Biology 112(1):153-164.
Fry, V.A., and B. Butman. 1991. Estimates of the seafloor area impacted by sewage sludge dumped at the 106-mile site in the Mid-Atlantic Bight. Mar. Env. Res. 31:145-160
Grant, W.D., and O.S. Madsen. 1986. The continental-shelf bottom boundary layer. Annual Review of Fluid Mechanics 18:265-30.
Grieb, T.M., C.T. Driscoll, S.P. Gloss, C.L. Schofield, G.L. Bowie, and D.B. Porcella. 1990. Factors affecting mercury accumulation in fish in the upper Michigan peninsula. Environmental Toxicology and Chemistry 9:919-930.
Gschwend, P.M., and S.C. Wu. 1985. On the constancy of sediment-water partition coefficients of hydrophobic organic pollutants. Environmental Science and Technology 19(1):90-96.
Gust, G., and M.J. Morris. 1989. Erosion thresholds and entrainment rates of undisturbed in situ sediments. Journal of Coastal Research 5:87-100.
Gutknecht, J. 1981. Inorganic mercury (Hg2+) transport through lipid bilayer membranes. J. Membrane Biology 61:61-66.
Hambrick, G.A. III, P.N. Forelich Jr., M.O. Andreae, and B.L. Lewis. 1984. Determination of methyl-germanium species in natural waters by graphite furnace atomic adsorption spectrometry with hydride generation. Anal. Chem. 56:421.
Harrison, G.I., and F.M.M. Morel. 1983. Antagonism between cadmium and iron in the marine diatom Thalassiosira weisflogii. J. Phycol. 19:495-507.
Hendricks, T.J., and J.M. Harding. 1974. The Dispersion and Possible Uptake of Ammonia in a Wastefield, Point Loma. Technical Memorandum 210. El Segundo, California: Southern California Coastal Water Research Project.
Hill, P.S. 1992. Reconciling aggregation theory with observed vertical fluxes following phytoplankton blooms. Journal of Geophysical Research (Oceans) 97(C2):2295-2308.
Hill, P.S., and A.R.M. Nowell. 1990. The potential role of large, fast-sinking particles in clearing nepheloid layers. Philosophical Transactions of the Royal Society London, Series A. 331:103-117.
Hill, P.S., A.R.M. Nowell, and P.A. Jumars. 1988. Flume evaluation of the relationship
between suspended sediment concentration and excess boundary shear stress. Journal of Geophysical Research 93(C 10):12,499-12,509.
Holman, H.Y.N., and J.R. Hunt. 1986. Particle Coagulation in a Turbulent Plume, Report No. 86-6. Berkeley, California: Sanitary Engineering and Environmental Health Laboratory, University of California, Berkeley.
Honeyman, B.D., and P.H. Santschi. 1991. Coupling adsorption and particle aggregation—laboratory studies of colloidal pumping using FE-59 labelled hematite. Environmental Science and Technology 25(10):1739-1747.
Honeyman, B.D., L.S. Balistrieri, and J.W. Murray. 1988. Oceanic trace metal scavenging: The importance of particle concentration. Deep-Sea Research 35:227-246.
Hudson, R.J.M., and F.M.M. Morel. 1990. Iron transport in marine phytoplankton: Kinetics of cellular and medium concentration reactions. Limnol. Oceanogr. 35:1002-1020.
Hudson, R.J.M., S.A. Gherini, C.J. Watras, and D.P. Porcella. 1992. Modelling the biogeochemical cycle of mercury in lakes: The mercury cycling model and its application to the MTL lakes. In Mercury as a Global Pollutant: Toward Integration and Synthesis, R.C.J. Watras, and J.W. Huckabee, eds. Chelsea, Michigan: Lewis (in press).
Huggett, R.J. 1989. Kepone and the James River Estuary. Pp. 417-424 in Contaminated Marine Sediments—Assessment and Remediation. Washington, D.C.: National Academy Press.
Hunt, J.R. 1980. Predictions of oceanic particle size distributions from coagulation and sedimentation mechanisms. Pp. 243-257 in Particulates in Water: Characterization, Fate, Effects, and Removal, M. Kavanaugh and J. O. Leckie, eds. ACS Advances in Chemistry Series No. 189. Washington, D.C.: American Chemical Society.
Hunt, J.R. 1982. Particle dynamics in seawater: Implications for predicting the fate of discharged particles. Environmental Science and Technology 16:303-309.
Hunt, J.R., and J.D. Pandya. 1984. Sewage sludge coagulation and settling in seawater. Environmental Science and Technology 18:119-121.
Ikalainen, A.J., and D.C. Allen. 1989. New Bedford Harbor Superfund Project. Pp. 312-350 in Contaminated Marine Sediments—Assessment and Remediation. Washington, D.C.: National Academy Press.
Imboden, D.M., and R.P. Schwarzenbach. 1985. Spatial and temporal distribution of chemical substances in lakes: Modeling Concepts. Pp. 1-30 in Chemical Processes in Lakes, W. Stumm, ed. New York: Wiley-Interscience.
Jackson, G.A. 1990. A model of the formation of marine algal flocs by physical coagulation processes . Deep-Sea Research 37:1197-1211.
Jackson, G.A., and J.J. Morgan. 1978. Trace metal-chelator interactions and phytoplankton growth in seawater media: Theoretical analysis and comparison with reported observations. Limnol. Oceanogr. 23:268-282.
Jaworski, N.B. 1981. Sources of nutrients and the scale of eutrophication problems in estuaries. In Estuaries and Nutrients, B.J. Neilson and L.E. Cronin, eds. New York: Humana.
Johnson, C.A., M. Ulrich, L. Sigg, and D.M. Imboden. 1991. A mathematical model of the manganese cycle in a seasonally anoxic lake. Limnol. & Oceanogr. 36:1415-1426.
Karickhoff, S.M. 1984. Organic pollutant sorption in aquatic systems. Journal of Hydraulic Engineering 10:707-735.
Koh, R.C.Y. 1982. Initial sedimentation of waste particulates discharged from ocean outfalls. Environmental Science and Technology 16:757-762.
Koh, R.C.Y. 1988. Shoreline impact from ocean waste discharges. Journal of Hydraulic Engineering, American Society of Civil Engineers 114(4):361-376.
Koh, R.C.Y., and N.H. Brooks. 1975. Fluid mechanics of waste-water disposal in the ocean. Ann. Rev. of Fluid Mechanics 7:187-211.
Kullenberg, G., ed. 1982. Pollutant Transfer and Transport in the Sea. Vol. 1. Boca Raton, Florida: CRC Press.
Lavelle, J.W., H.O. Mofield, and E.T. Baker. 1984. An in situ erosion rate for a fine grained marine sediment. Journal of Geophysical Research 89(C4):6543-6552.
Lewis, B.L., P.N. Froelich, and M.O. Andreae. 1985. Methylgermanium in natural waters. Nature 313:303.
Logan, B.E., R.G. Arnold, and A. Steele. 1989. Computer simulation of DDT distribution in Palos Verdes sediments. Pp. 178-198 in Contaminated Marine Sediments-Assessment and Remediation. Washington, D.C.: National Academy Press.
Long, E.R., and L.G. Morgan. 1991. The Potential for Biological Effects of Sediment-Sorbed Contaminants Tested in the National Status and Trends Program. NOAA Technical Memorandum NOS OMA 52. Seattle, Washington: National Oceanic and Atmospheric Administration.
Luoma, S.N., C. Johns, N.S. Fisher, N.A. Steinberg, R.S. Oremland, and J.R. Reinfelder. 1992. Determination of selenium bioavailavility to a benthic bivalve from particulate and solute pathways. Environmental Science & Technology 26(3):485-491.
Lyne, V.D., B. Butman, and W.D. Grant. 1990. Sediment movement along the U.S. east coast continental shelf-II. Modelling suspended concentration and transport rate during storms. Continental Shelf Research 10(5):429-460.
Mason, R.P., and W.F. Fitzgerald. 1990. Alkylmercury species in the equatorial Pacific. Nature 347:457.
Mearns, A.J., and D.R. Young. 1983. Characteristics and effects of municipal wastewater discharges to the Southern California Bight, a case study. Pp. 763-819 in Ocean Disposal of Municipal Wastewater: Impacts on the Coastal Environment, Vol. 2, E.P. Myers and E.T. Harding, eds. Cambridge, Massachusetts: MIT Sea Grant.
Moffett, J.W., and R.G. Zika. 1987. Solvent extraction of copper acetylacetonate in studies of copper (II) speciation in seawater. Mar. Chem. 21:301-313.
Muellenhoff, W.P., A.M. Soldate, Jr., D.J. Baumgartner, M.D. Schuldt, L.R. Davis, and W.E. Frick. 1985. Initial Mixing Characteristics of Municipal Ocean Discharges: Vol. I, Procedures and Applications, EPA/600/3-85/073a; Vol. II, Program Listings, Computer Programs EPA/600/3-85/073b. Narragansett, Rhode Island: U.S. EPA, Environmental Research Laboratory. Available from NTIS as PB86-137478 and PB86-137460.
Nichols, M.M. 1990. Sedimentologic fate and cycling of kepone in an estuarine system: Example from the James River Estuary. The Science of the Total Environment 97/ 98:407-440.
Nixon, S.W. 1988. Physical energy inputs and the comparative ecology of lake and marine ecosystems. Limnol. Oceanogr. 33:1005-1025.
NRC (National Research Council). 1984. Ocean Disposal Systems for Sewage Sludge and Effluent. Washington, D.C.: National Academy Press.
NRC (National Research Council). 1989. Contaminated Marine Sediments—Assessment and Remediation. Washington, D.C.: National Academy Press.
O'Connor, T.P. 1990. Coastal Environmental Quality in the United States, 1990. Chemical Contamination in Sediments and Tissues, A Special National Oceanic and Atmospheric Administration 20th Anniversary Report. Rockville, Maryland: National Oceanic and Atmospheric Administration.
O'Connor, D.J., J.A. Mueller, and K.J. Farley. 1983b. Distribution of kepone in the James River Estuary. Journal of Environmental Engineering 109(2):396-413.
O'Connor, T.P., A. Okubo, M.A. Champ, and P.K. Park. 1983a. Projected consequences of dumping sewage sludge at a deep ocean site near New York Bight. Can. J. Fish. Aquat. Sci. 40(Sup. 2):228-241.
Officer, C.B., and D.L. Lynch. 1989. Bioturbation, sedimentation and sediment-water exchanges. Estuarine, Coastal and Shelf Science 28:1-12.
Officer, C.B., and J.H. Ryther. 1977. Secondary sewage treatment versus ocean outfalls: An assessment. Science 197:1056-1060.
Okubo, A. 1971. Ocean diffusion diagrams. Deep-Sea Res. 18:789-802.
Olmez, I., E.R. Sholkovitz, D. Hermann, and R.P. Eganhouse. 1991. Rare earth elements in sediments off southern California: A new anthropogenic indicator. Environmental Science and Technology 25(2):310-316.
Palermo, M.R., C.R. Lee, and N.R. Francingues. 1989. Management strategies for disposal of contaminated sediments. Pp. 200-220 in Contaminated Marine Sediments—Assessment and Remediation. Washington, D.C.: National Academy Press.
Paulson, A.J., R.A. Feely, and H.C. Curl. 1989. Separate dissolved and particulate trace metal budgets for an estuarine system: An aid for management decisions. Env. Poll. 57:317-339.
Paulson, A.J., R.A. Feely, H.C. Curl, E.A. Crecelius, and T. Geiselman. 1988. The impact of scavenging on trace metal budgets in Puget Sound. Geochim. Cosmochim. Acta 52:1765-1779.
Roberts, P.J.W., W.H. Snyder, and D.J. Baumgartner. 1989a. Ocean outfalls. I: Submerged wastefield formation. Journal of Hydraulic Engineering, American Society of Civil Engineer 115:1-25.
Roberts, P.J.W., W.H. Snyder, and D.J. Baumgartner. 1989b. Ocean outfalls. II: Spatial evolution of submerged wastefield. Journal of Hydraulic Engineering, American Society of Civil Engineers 115:26-48.
Roberts, P.J.W., W.H. Snyder, and D.J. Baumgartner. 1989c. Ocean outfall. III: Effects of diffuser design of submerged wastefield. Journal of Hydraulic Engineers, American Society of Civil Engineers 115:49-70.
Sanders, J.E. 1989. PCB pollution in the upper Hudson River. Pp. 365-400 in Contaminated Marine Sediments—Assessment and Remediation. Washington, D.C.: National Academy Press.
Sayles, F.L., and W.H. Dickinson. 1991. The ROLAI2D lander: A benthic lander for the study of exchange across the sediment-water interface. Deep-Sea Research 38(5):505-529.
SCCWRP (Southern California Coastal Water Research Project). 1973. The Ecology of the Southern California Bight: Implications for Water Quality Management. Long Beach, California: Southern California Coastal Water Research Project.
SCCWRP (Southern California Coastal Water Research Project). 1991. Annual Report. Long Beach, California: Southern California Coastal Water Research Project.
Schwarzenbach, R.P., and J. Westall. 1981. Transport of nonpolar organic compounds from surface water to groundwater. Laboratory Sorption Studies. Environmental Science and Technology 15:1360-1367.
Sheng, Y.P. 1990. Evolution of a three dimensional curvilinear-grid hydrodynamic model for estuaries, lakes and coastal waters: CH3D. Pp. 40-50 in Estuarine and Coastal Modeling, M. Spaulding ed. New York: American Society of Civil Engineers.
Signell, R.P., and B. Butman. 1992. Modeling tidal exchange and dispersion in Boston Harbor. Journal of Geophysical Research 97(C10): 15591-15607.
Silver, M.W., A.L. Shanks, and J.D. Trent. 1978. Marine snow: Microplankton habitat and source of small-scale patchiness in pelagic populations. Science 201:371-373.
Smetacek, V.S. 1985. Role of sinking in diatom life-history cycles: Ecological, evolutionary, and geological significance. Marine Biology 84:239-251.
Stoecker, D.K., W.G. Sunda, and L.H. Davis. 1986. Effects of copper and zinc on two planktonic ciliates. Marine Biology 92:21-29.
Stolzenbach, K.D., K.A. Newman, and C. Wong. 1992. Aggregation of fine particles at the sediment-water interface. Journal of Geophysical Research 97(C 11): 17889-17898.
Stull, J.K., R.B. Baird, and T.C. Heesen. 1986. Marine sediment core profiles of trace constituents offshore of a deep wastewater outfall. J. Water Pollut. Control Fed. 58(10):985-991.
Sunda, W.G. 1988-1989. Trace metal interactions with marine phytoplankton. Biol. Oceanogr. 6(5-6):411-442.
Sunda, W.G., and R.L. Ferguson. 1983. Sensitivity of natural bacterial communities to additions of copper and curic ion activity: A bioassay of copper complexation in seawater. Pp. 871-891 in Trace Metals in Seawater, C.S. Wong, ed. New York: Plenum Press.
Sunda, W.G., and R.R.L. Guillard. 1976. The relationship between cupric ion activity and the toxicity of copper to phytoplankton. J. Mar. Res. 34:511-529.
Sunda, W.G., and A.K. Hanson. 1987. Measurement of free cupric ion concentration in seawater by a ligand competition technique involving copper sorption onto C-18 SEPPAD cartridges. Limnol. Oceanogr. 32:537-551.
Sunda, W.G., D.W. Engel, and R.M. Thuotte. 1978. Effect of chemical speciation on toxicity of cadmium to grass shrimps. Palaemonetes pugio: Importance of free cadmium ion. Environ. Sci. Technol. 13:213-219.
Sunda, W.G., R.T. Barber, and S.A. Huntsman. 1981. Phytoplankton growth in nutrient rich seawater; importance of copper-manganese cellular interactions. J. Mar. Res. 39:567-586.
Tee, K.T. 1987. Simple models to simulate three-dimensional tidal and residual currents, in three-dimensional coastal ocean models. Pp. 125-147 in Three-Dimensional Coastal Ocean Models, N.S. Heaps, ed. Washington, D.C.: American Geophysical Union.
Thomann, R.V., J.A. Mueller, R.P. Winfield, and C.R. Huang. 1991. Model of fate and accumulation of PCB homologues in Hudson Estuary. Journal of Environmental Engineering 117(2):161-178.
Thorbjarnarson, K.W., C.A. Nittrouer, D.J. DeMaster, and R.B. McKinney. 1985. Sediment accumulation in a back-barrier lagoon, Great Sound, New Jersey. Journal of Sedimentary Petrology 55(6):856-863.
U.S. GOFS (U.S. Global Ocean Flux Study). 1989. Sediment Trap Technology and Sampling, Planning Report No. 10. Woods Hole, Massachusetts: U.S. GOFS Planning and Coordination Office, Woods Hole Oceanographic Institution. 94p.
van den Berg, C.M.G. 1984. Determination of the complexing capacity and conditional stability constants of complexes of copper (II) with natural organic ligands in seawater by cathodic stripping voltammetry of copper-catechol complex ions. Mar. Chem. 15:1-18.
Van Dover, C.L., J.F. Grassle, B. Fry, R.H. Garritt, and V.R. Starczak. 1992. Stable isotope evidence for entry of sewage-derived organic material into a deep-sea food web. Nature 360:153-155.
Wang, R.F.T. 1988. Laboratory Analysis of Settling Analysis of Wastewater Particles in Sea Water Using Holography, Report No. 27. Pasadena, California: Environmental Quality Laboratory, California Institute of Technology.
Weilenmann, U., C.R. O'Melia, and W. Stumm. 1989. Particle transport in lakes: Models and measurements. Limnol. Oceanog. 34(1):1-18.
Weiner, J.G., W.F. Fitzgerald, C.J. Watras, and R.G. Rada. 1990. Partitioning and bioavailability of mercury in an experimentally acidified Wisconsin lake. Environmental Toxicology and Chemistry 9:909-918.
Wu, Jin. 1983. Sea-surface drift currents induced by wind and waves. J. Phys. Oceanogr. 13:1441-1451.