Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Use of Oil Dispersants: History and Issues Decisions about using dispersants are made in the context of the reasons for treating spilled oil. The image of a black, sticky mass covering a once clean shore or yacht hull is dramatic enough, but there are other less visible incentives to respond to a spill, such as threats to valuable habitat and nursery areas or to vulnerable early stages of marine life. Other concerns are: How great a threat do spills pose in quantity and frequency? What are the usual sources of of} in the sea and what can be done to counter a spill? These concerns establish the setting for using chemical dispersants and are discussed here. This chapter also reviews how dispersants have been used and developed and what their use is expected to accomplish. The potential use of chemical dispersants raises some primary questions, principally: Are they effective? and Do they reduce the potential damage caused by spired oil? These issues are discussed generally in this chapter and in more detail in later chapters. REASONS FOR TREATING OIL SPILLS Aesthetic and Ecological Damage Introduction of petroleum into the marine environment is a di- rect consequence of the production and transportation of crude of! and refined products. Even though natural seeps of crude oil occur 6
HISTORY AND ISSUES 7 in areas of the ocean floor and stable biotic communities are asso- ciated with them, the sudden introduction of high concentrations of hydrocarbons con kill or cause sublethal effects in some marine organisms. In the 1960s and 1970s, more and more of} was transported by sea and with the introduction of supertankers in larger vessels. The increased risk of a major of} spill was displayed dramatically when the first major tanker catastrophe the Torrey Canyon spin occurred. Public outcry from the Torrey Canyon spin and similar inci- dents stimulated development of a variety of response techniques to contain or remove spired oil before it could harm property or the environment. Unfortunately, some early attempts to clean up the of! for aesthetic reasons caused more ecological harm than good (Na- tional Research Council [NRC], 1985; Smith, 1968; Southward and Southward, 1978; Spooner, 1969~. Perhaps the most dramatic symbol of the consequences of an of] spill is an oiled seabird. Attempts by volunteers to rescue birds have been a response to many of} spills, yet despite enthusiastic care, the survival percentage is low. Diving birds, such as auks, are especially vulnerable. Some marine mammals may also be affected under certain circumstances. Nearshore marine waters arid shadow fishing banks are also rich in a variety of less visible organisms. Some areas, such as salt marshes, are among the most productive ecosystems known. The impact of spilled of} on these areas is obvious and often severe, and public awareness of these ecosystems has expanded concern about the danger from of! spites. The concern that of! spills pose a threat to commercial fisheries has been extended to open-ocean life in general (Boesch and Rabalais, 1987~. Economic Damage In addition to aesthetic and ecological concerns, coastal regions can suffer economically from damage done by of} spills to recreation areas, harbors and vessels, commercial shellfish grounds, and intake sources for desalination and power plants. During summer months, beaches along the coasts of most maritime countries are crowded with people on weekend outings and vacations. Thus, there is considerable economic incentive in coastal recreation areas to protect beaches from spills or to clean them up quickly. Cleanup of contaminated boats, seawalIs, and harbor equipment can be expensive. In a marina
8 USING OIL SPILL DISPERSANTS ON THE SEA where boat owners pride themselves on the appearance of their craft, pressure for cleanup is likely to be even more intense than in an industrial harbor. Because hydrocarbons from spilled of! can become entrained in sediments inhabited by many commercial shellfish species, shellfish can become contaminated with oil. They may then be unfit for human consumption due to high levels of hydrocarbons. Contami- nation can close a commercial shellfish bed for years, resulting in a considerable financial loss to shellfish producers. In some regions of the world- such as Bermuda, several Carib- bean islands, and the Middle East desalination plants provide an important source of drinking water for the surrounding sea. Oil, even when naturally dispersed, can enter the intake of such a plant and threaten to contaminate an entire drinking water supply with hydrocarbons (Kruth et al., 1987~. Power-plant cooling water intakes are similarly vulnerable. There have been several attempts to identify and analyze the costs for mechanical cleanup and disposal as well as for treatment with chemical dispersants. These analyses have produced a wide spectrum of results, reflecting many variables that are difficult to measure. For example, when of] is recovered by mechanical means what is usually measured includes large amounts of water and de- bris. Another variable is the cost of transporting equipment to the cleanup or staging site. Usually the cost of moving boats and other equipment for mechanical cleanup is omitted, but aircraft and equip- ment transport expenses are commonly included in evaluations of dispersant operations. Mechanical costs of oil cleanup range from $65 per bbl to $5,000 per bbl. Dispersant costs range from $15 per bbl to a maximum of $65 per bb] (Lasday, 1985~. These ranges are related to spin sizes of 10,000 to 100,000 bb} of oil. These statistics and other analyses, such as provided by White and Nichols (19831. show that the costs of of} spill cleanup are high ~ ,, . . ~ and that they are an order of magnitude more for mechanical cleanup than for treatment using chemical dispers~nts and still another order of magnitude more for cleanup of the immediate and obvious damage done once the oil has come ashore. Long-term residual effects have been identified in several reports (Teal and Howarth, 1984; White and Nichols, 1983), but attempts to place a monetary value on the loss of or damage to natural resources and to establish replacement or restoration costs have been based
HISTORY AND ISSUES 9 on sweeping assumptions about natural processes and usually lack long-term environmental data. Safety Hazards Crude of} contains a substantial fraction of volatile hydrocarbons, as do refined petroleum products such as gasoline and diesel oil. Spilled of} evaporates, producing high concentrations of flammable hydrocarbon-air mixtures. If such a mixture ignites, fire can spread to the bulk of! container and consume it. Although a fire may not pose a major hazard to aquatic systems, air pollution from partially burned crude oil can pose severe public health and ecological hazards (GundIach and Hayes, 1977; Thebeau and Kana, 1981~. Therefore, prevention of fires better yet, prevention of spills is a desirable goal. Some volatile components of crude and refined oils, such as benzene, are very toxic to humans. Indeed, a human health concern is that cleanup crews can receive doses of toxic fumes high enough to cause nausea and possibly other health effects. POTENTIAL SOURCES OF SPILLED OIL The worldwide input of petroleum to the marine environment was estimated by the National Research Council (1985), which noted the sources, probable ranges, and selected the best single number for the sources as shown in Table 1-1. The estimates, based on data acquired from 1971 through 1980, are provided only as indicators of major sources and their importance. Input of petroleum fluids from municipal and industrial wastes (urban and river runoff, sewage, re- fineries, and industrial sources), atmospheric-borne particulates, and natural sources (marine seeps, sediment erosion) is mostly continuous and, while the environmental effects are of serious concern, control or treatment technology usually differs greatly from that related to marine of} spill response. However, some accidental spills on inland rivers and waterways may be amenable to common marine oil spill response. Marine transportation tankers, barges, and lighters and off- shore operations for the exploration and production of hydrocarbons are the key sources of interest in regard to marine of] spills. On a global basis, marine transportation accounts for the largest single portion of petroleum inputs, more than 40 percent on the basis of
10 USING OIL SPILL DISPERSANTS ON THE SEA TABLE 1-1 Input of Petroleum Hydrocarbons Into the Marine Environment (Million Barrels Per Yearly ~ . Source Probable Range Best Estimate Transportation Operational 4.4-15.0 7.2 Accidents 2.2- 3.0 2.9 Subtotal 6.6-18.0 10.1 Municipal wastes 4.0-21.5 8.1 and runoff Atmosphere 0.3- 3.4 2.1 Natural sources 0.2-17.3 1.7 Offshore production 0.3- 0.4 0.3 Total 11.4-60.6 22.3 aOne barrel (bbl) equals 42 U.S. gallons or about 35 Imperial gallons. This report will refer to barrels an the standard volume measurement in reference to petroleum transport and spills. SOURCE: Converted from NRC (1985) estimates based on metric tons per year. "best estimates" (NRC, 1985) and 30 to 50 percent depending on the ranges of other sources. However, the bulk of this input is due to routine operational discharges. About 12 percent is due to acci- dents, but this can vary from year to year by factors of 2 or 3. In contrast, offshore of} and gas operations are estimated to contribute only a little more than 1 percent of the of] accidentally put into the sea. A single incident can dramatically alter these data, however; for example, the largest spin to date, the bloc ~ wed blowout, put 3.1 minion bb] in the Gulf of Mexico in 1979 and 1980. Apart from oil discharges caused by military action in the Per- sia~n Gulf, the incidence of major tanker spills elsewhere has been dramatically reduced from 1974 to 1986 (Table 1-2~. The majority of tanker spills over 5,000 bb} were from collisions and groun clings. All discharges reported in waters under the U.S. Coast Guard's jurisdiction or by the U.S. Environmental Protection Agency (U.S. EPA) from 1974 through 1983 range from 200,000 bb} to more than 500,000 bb! per year (Figure 1-1~. Of the reported discharges, only about 5 percent (about ll,OOO bb} in 1983) occurred in offshore waters, that is, in the territorial sea (shore to 3 ml), the contiguous zone (3 to 12 ml), and the high seas (beyond 12 ml). River channels,
HISTORY AND ISSUES 11 ports and harbors, and open sheltered waters receive most spillage, in contrast to offshore areas. In U.S. waters, waterborne transport accounted for about 28 percent of the tote] volume reported spilled in 1983. Most of this volume was attributed to tank barges, which accounted for 22 percent (43,000 bbl) of Al of} spiked. Much of this discharge occurred in ports and harbors. As more and more of} is refined outside of the United States, ocean transport of refined products to and along the U.S. East Coast by barge is increasing. About 35 to 50 percent of spillage in U.S. waters is crude oil. Spill volumes and number of incidents in U.S. onshore waters are shown in Figures 1-2 and 1-3. These statistics cover both trans- portation and offshore of! and gas operations in the three major U.S. coastal areas. Canadian offshore of} spins during the period from 1974 to 1984 averaged about 10,000 bb} per year, but the definition of offshore includes inlets and coastal areas, while the U.S. offshore definition excludes harbors and sheltered waters. Most Canadian discharges were fuel of] and distillates, in contrast to U.S. experi- ence, but statistical comparisons of incidents and volumes between the United States and Canada would require access to the raw data. Most spills are small and most spilled of! volume comes from only TABLE 1-2 Number of Oil Spills From Tankers Worldwide, 1974-1986 Barrels Year 50-5,000 >5,000 1974 92 26 1975 98 23 1976 67 25 1977 66 20 1978 57 24 1979 56 36 1980 50 13 1981 50 5 1982 45 1983 53 11 1984 25 7 1985 29 8 1986 23 7 SOURCE: International Tanker Owners Pollution Federation Ltd.
12 USING OIL SPILL DISPERSAN-TS ON THE SEA 600,000 550,000 500,000 450,000 400,000 an LLJ 350,000 m 300,000 250,000 200,000 1 50,000 1 00,000 A / A I I 1 1 1 ~ 1 1 1 1 1 - - 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 YEAR FIGURE 1-1 Volume of oil and the substances discharged, 1974-1983. All volumes have been converted from U.S. gallons to barrels. The 1983 figure includes 55,000 bbl attributed to "other," a category not previously used in Coast Guard reporting. Source: U.S. Coast Guard, 1987. . a few large incidents. For example, in 1983* the U.S. Coast Guard reported the following spill statistics in and around U.S. waters: · Small spills (less than 12 bbl) 5,923 spins (63 percent) ac- counted for 6,200 bb! (3 percent of total volume). These spills are usually not treated with dispersants nor cleaned up by mechanical means. . Moderate spills (12to 1,200 bbl) 617 spills(6.5 percent) accounted for 60,000 bbl (30 percent of total volume). Efforts to clean up these spills usually use mechanical means, but in some cases such spills could be treated with dispersants. *Although Coast Guard statistics for 1984 are available, they are preliminary. Variations between preliminary and final statistics can be significant, thus the figures for 1983 are the latest final data reported here (U.S. Coast Guard, 1987~.
HISTORY AND ISSUES 2,250 oh llJ 2,000 LL o 1,750 I oh iL 1,500 o 1,250 an <,0 1,000 oh ~ 750 o IL O 500 In m ~ 250 z o Total offshore - ,' Atlantic Ocean i' - Pacific Ocean __ ~ Gulf of Mexico I I _ 1980 1 981 13 1982 1 983 YEAR FIGURE 1-2 Number of oil spills in U.S. offshore waters (territorial, contiguous, and high seas). Source: U.S. Coast Guard, 1987. O Large spills (more than 1~200 bbl) 19 accidents (0.2 percent) released 133,000 bib! (67 percent of the total volume). These spills may require major response effort, including aerial application of dispersants. TREATMENT METHODS There are four major options for responding to oil spins: mechan- ical containment and collection; use of chemical dispersants; shoreline cleanup; and natural removal (no cleanup action). Countermeasures that are less widely used or have major limitations are burning, sink- ing, gelling, and enhanced biodegradation. In determining the best possible countermeasure for a given situation, availability and appli- cability must be carefully weighed against potential environmental (1am age (Table 1-3~.* *Comprehensive presentations of oil cleanup methods are provided in U.S. EPA (1981) and Wardley-Smith (1976).
14 USING OIL SPILL DISPERSANTS ON THE SEA 1 4,000 co a: ~ 1 2,000 tar I 1 0,000 co LO o oh JO 8,000 do C) LL rL 6,000 co o _ 4,000 O 2,000 \ l _ \, l _ ; ; -\ in, I>< Total offshore / \ \ /\ ,, i, ! , .\ ' Gulf of Mexico ,- \ Pacific Ocean O 1980 1981 1982 1983 YEAR Atlantic Ocean FIGURE 1-3 Volume of oil (bbl) spilled in U.S. offshore waters "territorial, contiguous, Ed high seam. Source: U.S. Coast Guard, 1987. Mechanical Containment, Recovery, or Removal Mechanical means for oil spill mitigation include barriers (booms) to contain or divert oil, and skimmers or sorbents to recover or remove it from the water surface. In addition to mechanical barriers (which include sorbent booms), surface-collecting agents ("oil herder" chem- icals), water jets, air jets, and air bubble barriers have been used to contain or divert spilled oil. Performance of any of these methods, however, can be severely limited by oceanic conditions and weather, including currents, waves, and wind, and by the nature of the oil slick.
HISTORY AND ISSUES TABLE 1-3 Cleanup Control Options and Their General Application Cleanup Operations and Options Application Water Under- Along- Shore- Surface water shore line Containment and diversion Major options Booms Sorbent booms Surface-collecting agents Water jets Air jets Other possibilities Air barriers Gelling Viscoelastic additives Underwater containment Removal x x x x x x x x x x x Major options Skimmers x Sorption x x Sorbent booms x Biodegradation x x Other possibilities Burning x Sinking x Dispersants x x Cleanup of stranded oil Major options Flushing x Beach cleaning x Manual x x x Vegetation cropping x x Organism x rehabilitation Other possibilities Substrate removal x x Burial x Sandblasting x Steam cleaning x Natural Cleansing x x x 15 Full containment using a boom cannot be assured unless max~- mum water speed at right angles to the boom is less than 1 kn. In faster currents, the barrier must be placed at an angle to the current to guide the floating oil toward slower current areas. Air and wa- ter jets and bubble barriers are sometimes used in restricted areas; in general, the current has to be fairly slow and the waters quies-
16 USING OIL SPILL DISPERSANTS ON THE SEA cent. Although surface-collecting agents do not move oil relative to underlying water, they can limit its spread regardless of water speed. Maximum wave height for complete containment should be less than equal to the freeboard of the barrier, or waves will splash water and oil over it. Mechanical barriers are thus severely limited on open seas. In practice, most barriers lose effectiveness in wave heights greater than 1.3 m (4 ft). In experiments, surface-collecting agents have kept of} from spreading in 1 to 2 m (4 to 6 It) seas, but duration of effectiveness decreases with increasing sea state. Limitations to operation of most containment options are related to current and wind-induced wave heights. Air and water jets are seriously affected by winds, especially when they blow toward the jet. Surface-collecting agents cannot move floating oil against the wind, but the ability of a surface-collecting agent to keep oil from spreading is not affected. Wind stress on boom barriers can be appreciable, and the boom must be strong enough to withstand such forces. The rate of oil recovery by any mechanical device decreases with decreasing of! thickness. Because the recovery rate of most skimmers is negligible at thicknesses of less than about 1 mm, booms are often used in conjunction with skimmers. This approach is limited, however, by problems of maneuvering, anchoring, and coordinating multiple vessels needed to handle such arrays. Mechanical collection devices have environmental limitations similar to those of containment devices because skimmers are affected by wind, waves, and currents in much the same way as barriers. Oil may pass under or by a skimmer and not be collected if the skimmer is moving through the water faster than 1 kn. Waves higher than 1.3 m may cause oil to splash over skimmers or otherwise cause them to lose effectiveness. More robust skimmers and barriers have been designed for the U.S. Coast Guard to use in the open ocean, and others for use in the North Sea. Some of these devices have suc- cessfully recovered oil in seas higher than 4 ft. In large spills, the effectiveness the percent of surface oil removed by treatment has been low for mechanical cleanup systems; for example, possibly 10 percent during the Bloc Iblowout (Teal and Howarth, 1984~. Shoreline Cleanup If oil strands on a shoreline, attempts are usually made to remove it using mechanical means, by flushing, by manual pickup, or by phys- ically removing the substrate. In some cases, oil-soaked vegetation
HISTORY AND ISSUES 17 has been cropped, but this method is usually regarded as environ- mentally undesirable. Hard surfaces, such as rocks or bulkheads, have been sandblasted or steam cleaned. In most cases, shoreline cleanup is expensive and may be environmentally damaging. However, the only method for cleaning and restoring public beaches and easily accessible shorelines near fisheries and industrial areas is removal of oil. Natural Removal Oil left alone is eventually removed from water surfaces and shorelines by a variety of natural means, including evaporation, pho- tooxidation, solution, physical dispersion, sedimentation on partic- ulate matter, and biological degradation. Although these natural processes may be slow, possibly as long as several years, they are generally conceded to be environmentally acceptable, and in some cases may be preferable to using active countermeasures. Other Countermeasures In addition to the mitigation measures discussed above, a num- ber of other countermeasures are available or have been proposed, including burning, sinking, and gelling. All of them have some limi- tations, however. A major limitation to burning oil at sea is that oil tends to spread on water and the sea is a very effective heat sink; it is difficult to raise the temperature of a thin layer of floating of! high enough to permit ignition. However, where spilled of} cannot flow well, as in the Arctic and on ice, there has been effective use of burn- ing as a cleanup technique. Even so, combustion is never complete and air pollution is a real concern. The addition of chalk or treated sand has been used or proposed as a means of sinking oil. However, sinking is seldom completely effective initially, n.nCl some of] tends to resurface. Moreover, of] that sinks to the bottom contaminates benthic life and degrades more slowly than when floating, dispersed or dissolved in water. Several gelling formulations have been proposed but have yet to be demonstrated in practice. Most formulations would require sub- stantial mixing energy to make them effective, which is not practical under actual spill conditions. A further limitation is that gelling formulations are expensive; the cost of cleanup using gelling agents is likely to be far higher than conventional means. A possible exception 1,
18 USING OIL SPILL DISPERSANTS ON THE SEA is a viscoelastic additive, however, it has yet to prove itself in a real spin; thus there is no basis as yet for judging its utility. Because constituents of of! degrade naturally when attacked by bacteria, algae, protozoa, and marine fungi, enhancement of biologi- cal degradation has been proposed using specially chosen or bioengi- neered microbes. However, microbes that degrade hydrocarbons are readily available everywhere in nature, except in polar waters where the rates of breakdown are very slow and variable (Atias, 1985~. It does not appear necessary in most cases to enhance their action. ROLE OF DISPERSANTS Much of the biological and ecotox~cological research on disper- sants, oils, and dispersed oils has been conducted in support of chemical dispersant response to oil spins in coastal waters. Real or perceived impacts on biological populations, habitats, water sources, and recreational areas have focused the attention of resource man- agers, cleanup specialists, policymakers, marine scientists, and the public on the need for better understanding of of! spill effects and development of techniques for dealing with them. Rationale for Dispersant Use An initial reason for using dispersants is to respond to public and governmental concerns by preventing potential damage to birds, fish, marine mammals, and other natural resources; fouling of shorelines and boats; and contamination of drinking water sources. Dispersing an of! spin will make it less visible, and may reduce its economic and ecological impact provided the water volume, which it disperses into, is great enough. If the of] is dispersed into a small volume of water with poor circulation, the ecological impact may in fact be increased. Some specific cases have been studied in adequate detail (see Chapter 4, "Biologically Oriented Mesocosm and Field Studies" ). Rapidly dispersing oil into the water column will, in most cases, be less costly environmentally than manual shoreline cleanup. Sea and Weather Conditions Dispersants may be especially valuable when other countermea- sures fail, for example when an open-ocean spill is moving oil onshore,
HISTORY AND ISSUES 19 but waves are too high to permit the use of booms and skimmers. Another example is if tidal currents are so strong that of} would be carried under a boom, resurface, and threaten a sensitive area. Be- cause severe weather enhances natural dispersion of oil, the results of chemical dispersant use may not be apparent as in calm seas. Nevertheless, reduction of interfacial tension by dispersants is likely to hasten dispersion. Logistics Application of dispersants can be accomplished more quickly than recovery of spired of} by mechanical means. Aerial spraying of chemical dispersants is usually the preferred method of application to use at sea, because it is more efficient and provides a wider range of coverage than application from boats. Like vessel spraying methods, aerial application is limited in high winds, fog, or darkness. The speed of aerial response itself is not the key goal, but rather the imperative to ensure protection of sensitive environments and public amenities. Protection of Ecologically Sensitive Areas and Organisms Marine ecosystems, such as salt marshes, mangroves, and coral reefs, and bird nesting areas are extremely sensitive to damage by oil, but the use of dispersants raises questions about the relative environmental effects of of! and dispersed of! (see Chapters 3 and 4~. Dispersants may be applied when it is judged that the impact of dispersed of! on organisms, habitats, and ecological processes wiD be less than that of of} alone. Determining relative environmental effects requires objectively assessing alternative treatments in terms of effectiveness, acute toxicity, ecological effects, and cost, including potential loss of natural resources. Accurate exposure assessment for surface and subsurface of! is critical to estimating the hazards of dispersed oils to seabirds, whose protection is a major reason for dispersant use over open waters (Peakall et al., 1987~. Protection of Fisheries Resources Fisheries resources are vulnerable to damage by oil, but the na- ture of this biological damage is thought to be largely indirect and
20 USING OIL SPILL DISPERSANTS ON THE SEA seasonal, via plankton~c eggs and larvae and through the contamina- tion of fish habitats. If a major spill occurs where a limited fish pop- ulation is spawning, an entire year-class could be exposed to damage. Although long-term damage could be ameliorated by strong previous or subsequent year-classes, losses to commercial fisheries might be substantial for a species under pressure from fishing, predation, or stresses from chronic pollution (Boesch and Rabalais, 1987; Howarth, 1987; Longhurst, 1982; NRC, 1985~. Using dispersants may mitigate this situation by moving of] into the water column and lowering its concentration and the time of exposure of gametes, developing em- bryos, and larvae that dweD on or near the surface. Dispersant use may also limit the damage to sensitive nearshore nursery waters, especially if of} is effectively dispersed in areas remote (i.e., where natural processes can treat oil) from the sensitive waters. Protection of Shoreline Amenities Beaches, harbors, marinas, and other shoreline amenities can suffer costly damage from of] spills. The same strategies suggested for protection of sensitive ecological areas can be applied to commercial areas by applying dispersants before a slick approaches too close to shore. In addition, booms may be set to protect lagoons and harbors. Dispersants as an Aid to Natural Cleanup Microbial degradation of of] appears to be enhanced by dispersal because of the larger surface area available. However, to some degree, the lower concentration of nutrients available in open water may limit the potential for growth of the hydrocarbon-utilizing microbial species. Laboratory and mesocosm studies show increased of} biodegra- dation rates when dispersants are used. Temporary inhibition of biodegradation with dispersed of! also has been observed in the lab- oratory, but appears to occur only at dispersed of] concentrations higher than would be observed in the field. Mesocosm studies in ponds suggest that when effectively used, dispersants should en- hance the rate of biodegradation of oil. Unfortunately, supportive field observations or case histories are not available. Finally, available information suggests that some compounds, such as tar bans, would remain undegraded regardless of the addition of dispersants (Lee and Levy, 19S6~. Prevention of the formation of tar balls and large oil- in-water emulsions accumulations caned mousse is a potentially
HISTORY AND ISSUES 21 important advantage of chemically dispersing oil. These forms of oil, with low surface area, tend to be resistant to biodegration, largely because there are few microbes and nutrients present in the mass. HISTORY OF DISPERSANT USE The Torrey Canyon SpiB Before about 1970, dispersant formulations were basically de- greasing agents that were developed to clean tanker compartments, bilges, and engine rooms. A number of "detergents," as they were called, were used to attempt to disperse the nearly 1 minion bb! of crude of} that spired from the tanker Torrey Canyon off the English coast in 1967. The extreme toxicity of these agents to marine life was attributed primarily to alky~pheno! surfactants and the aromatic hydrocarbons in the solvent(Portmann, 1970~. Over 14 days following the Torrey Canyon spill, about 10,000 bb! of various dispersants were sprayed on the water and along the shore. Investigations showed that the solvents evaporated (90 percent in 100 fur), but the denser surfactants did not evaporate, mix with, or dissolve in seawater. Instead, they formed stable "detergent-oil" emulsions, the most stable of which were produced by the most toxic dispersants Houghton, BPl002, and Gamien. Smith (1968:22) noted: As to spraying at sea, we have no information about its eventual effective- ness. It was generally agreed by those taking part in the sea operations that dispersal was often achieved in the immediate neighborhood of spraying. However, despite the large quantities of detergents used, large areas of undispersed oil persisted for weeks as extensive and discrete patches. The biological impacts along rocky shores were highly visible and devastating. Evidence of mortality was repeated all along the affected shoreline: empty limpet seats were conspicuous in pools, dead barnacle shells persisted for some time, and mussed sheds gaped. The rotting flesh did not take long to disappear, but even when the sheds had broken away, clumps of short straw-colored byssus threads persisted for a few weeks. Although the official position was that "the effects have not been catastrophic" (Smith, 1968:178), adverse publicity during and after the Torrey Canyon incident gave dispersants a bad reputation. Indeed, the experience led to a very cautious attitude toward dispersant use among several industrialized nations.
22 USING OIL SPILL DISPERSANTS ON THE SEA Development of New Formulations The main concerns supporting the cautious attitude toward dis- persant use after the Torrey Canyon spill were . toxicity of the products themselves, and · concern that effective dispersants would make of] constituents more available to biota and thus enhance toxicity of oil components. There was further concern about greater quantities of hydrocar- bons entering and persisting in shadow, low-energy marine environ- ments when dispersed with chemicals. Biological concerns focused on species important to fisheries, young life stages of marine animals, and littoral or shallow sublittoral habitats. Despite these concerns, the United Kingdom end several other European countries continued to test dispersants in a series of government-sponsored programs, and Canada and several other countries conducted research and prepared guidelines for acceptability and use. Second-generation dispersants were produced; generally they were much less acutely toxic than the earlier formulations but of variable effectiveness. In addition, a third generation of dispersants, with concentrates consisting mostly of surfactant with little solvent, was introduced to reduce volume for storage and transportation. The concentrates were intended to be diluted with seawater during spraying from boats, which was supposed to make spraying more uniform and increase operating time at sea (Morris and Martinelli, 1983~. Laboratory tests suggested, however, that dilution reduced effectiveness by as much as a factor of 5 (Crowley, 1984a,b; Doe and Wells, 1978; Cormack, private communication). Experience at sea did not agree with the laboratory prediction, however, and the large effect experienced may have been specific to the formulation used in the tests (Lindblom, private communication). Undiluted dis- persant is applied from aircraft, but some equipment on boats use water-diluted dispersant. Concerns about the environmental impact of dispersants stimu- lated considerable research in Canada, Europe, and the United States in the 1970s and 1980s. The Conference of the American Society for Testing and Materials (ASTM) in 1977 stimulated U.S. work lead- ing to major worldwide studies funded by the American Petroleum Institute (API) and Exxon (Koons and Gould, 1984; McCarthy et al., 1978), and the 1980 Toronto dispersant conference (Mackay et al., 1981) brought most of the major researchers together to consider new work. In part sparked by the Toronto discussions, a Canadian
HISTORY AND ISSUES 23 research and development program was continued, which built on earlier work, to gain better understanding of the fate and effects of chemically dispersed oils in cold, temperate, and arctic waters. This program continued into the 1980s, in particular with the Arctic Ma- rine Oilspill Program (AMOP), which supported the annual AMOP Technical Seminars and the Barn Island Oil Spid Project (BIOS) study. Considerable research was initiated in Norway and Sweden and continued in France, the Netherlands, and the United Kingdom, among others. Despite these efforts, many key questions asked 15 to 20 years ago are still the subject of controversy and research today. Development of Equipment Boat spraying systems were developed by the United Kingdom's Warren Spring laboratory (WSI`) shortly after the Torrey Canyon incident. They were designed to make (lispersant chemical spraying more controllable, and were especially useful for applying undiluted hydrocarbon-based dispersants at very high doses, 327 to SS9 T/ha (35 to 95 gal/acre) (Cormack, 1983c). Early dispersant formulations required agitation to promote dispersion. This was done using trail- ing woo(len "breaker boards," which required that the spray boom be mounted toward the rear of the vessel, often aft of where the bow wave breaks from the huh. This configuration caused some disper- sant spray to miss the of! that was pushed away. Development of dis- persant concentrates eliminated this problem because they reduced the need for externally applied agitation and made breaker boards obsolete. The dispersant concentrates also reduced the amount of dispersant needed to treat a given quantity of oil, and therefore re- duced the application rate and greatly extended the spraying time for a vessel or aircraft. A fixed-wing aircraft system, which uses undiluted concentrate, was developed by the WSL from crop-spraying equipment in the 1970s. Additional extensive experimental and developmental work on aerial spraying was done in Canada and the United States in the 1970s and 1980s (Cormack and Parker, 1979; Lindblom and Barker, 1978; Nichols and Parker, 1985; Ross et al., 1978~. Helicopter spraying systems were also developed in the late 1970s and were used to spray dispersants on the stick from the Rasbah 6 blowout along the coast of Saudi Arabia in 1980 (Martinelli, 1980; van Oudenhoven, 1983).
24 USING OIL SPILL DISPERSANTS ON THE SEA Case Studies Dispersants have been used at many spills (more than 50 have been recorded), including the very large operations such as the bloc ~ blowout in the Gulf of Mexico (Nelson-Smith, 1980, 1985~. A few of these spills are described briefly in Appendix B. However, lack of controls, ad hoc observations, poor documentation, and lack of objective criteria for effective dispersal have made these situations less informative than might be expected. Planned dispersant field trials are reviewed, assessed, and summarized in Chapter 4. ISSUES AND QUESTIONS This report focuses on two main themes that are amplified in the form of more detailed concerns or questions in this section, along with a brief discussion of related problems: Dispersant effectiveness: Do dispersants do any good? Are current formulations and techniques effective? Can more effective techniques be developed? Harmful effects of dispersants: Do dispersants do any harm? Is the toxicity of dispersant formulations significant to marine species and, if so, under what environmental conditions? Is the biological impact of dispersed oil greater or less than that of untreated oil? Using Chemical Dispersants to Remove Oil From the Surface of the Water Developing optimum techniques for applying dispersants under various conditions requires an understanding of the numerous factors affecting oil dispersion. To achieve this, a definition of effectiveness is required, as well as a knowledge of the physical and chemical mecha- nisms of dispersion, techniques and logistics of dispersant application, and possible methods for modeling effectiveness quantitatively. Defining Effectiveness A commonly accepted definition of effectiveness of spin treat- ment, and the one used in this discussion, is the fraction of of] removed from the surface of the water. Alternative definitions have been devised, based on laboratory measurements, in an attempt to provide an easily measured surrogate (Chapter 2, "Laboratory Stud- ies of Effectiveness"~. Field studies have measured the concentration
HISTORY AND ISSUES 25 of dispersed oil in the water column under a slick, to find how much oil might be encountered by organisms (Chapter 4, "Physical and Chemical Studies"~. Electiveness depends on many factors. The following are dis- cussed in detail later in this report: the physical and chemical inter- action of oil and dispersants to reduce interfacial tension (Chapter 2, "Composition of Dispersants"), slick structure, sea state, turbulent mixing of of} droplets in the subsurface water column (Chapter 2, "Fate of Oil Spilled on Open Water" and "Behavior of Oil-Dispersant Mixtures"), and efficiency of application techniques (Chapter 5, "De- sign of Dispersant Application Systems"~. Identifying Physical and Chemical Factors That Influence Effectiveness Many factors influence the dispersibility of oil and the effective- ness of treatment. Specific questions include: How is effectiveness influenced by · of! composition (amounts of aromatic and aliphatic hydrocar- bons, asphaltenes, and metalloporphyrins)? · dispersant composition (hydrophilic-lipophilic balance tHLBi) of surfactant~s), kind of surfactant~s), and type of solvents? · dispersant-to-oi} ratio? · energy input (breaking waves, subsurface turbulence, and mechanical mixing)? water salinity and temperature? · oil viscosity? Oii shck thickness and distribution on the water surface? weathering of the oil (Ioss of volatile hydrocarbons, photoox- idation, and water-in-oi} emulsion formation [mousses? Viscosity, slick distribution, and oil weathering are time-dependent variables. These environmental and time-dependent factors are dis- cussed in Chapter 2. Factors Affecting Dispersant Application Techniques New dispersant formulations have been developed through the years. in addition, the techniques of dispersant testing and appli- cation are continuously revised with experience. New developments range from testing new formulations for effectiveness and toxicity in the laboratory, to the design of spray equipment for boats and
26 USING OIL SPILL DISPERSANTS ON THE SEA aircraft, to the training of skilled response personnel in operating such equipment effectively. New formulations are usually proposed when a manufacturer is convinced of their effectiveness in treating oil spills. Government agencies normally test, or oversee the testing, of such new products for effectiveness and toxicity (see Chapter 3~. Questions raised are: Can one dispersant be used with a wide variety of oils with expectation of reasonably equal effectiveness? ~ Are there significant differences between the expected per- formance characteristics of different dispersants used with the same oil? ~ How important is it to apply dispersants before the oil weath- ers, and what factors control this? ~ At the time of a spill, is it important to test the proposed dispersant on a sample of the oil before proceeding to apply the dispersant? What are the alternatives? Is it important to spray dispersant uniformly? · How should dispersant effectiveness be measured? · What are the concentrations and persistence of dispersed oil under chemically dispersed and untreated oil slicks? What are the concentrations and persistence of dissolved hydrocarbons under slicks, and how do they differ from the Cal+ fraction? · Are current spray systems (circa 1987) adequate to control oil spills? Factors Affecting Toxicity of Chemical Dispersants and Dispersed Oil As discussed in later chapters, there is a wide diversity of use of chemical dispersants for the control of oil spills in various countries. A range of opinion exists concerning the biological effects resulting from the use or proposed use of chemical dispersants. The following questions (discussed in Chapters 3 and 4) have been raised, based on some real observations or on perceived concerns: . How toxic are dispersants alone, and chemically dispersed oil, to marine plants and animals in laboratory studies? ~ How do laboratory measured toxicity exposures compare with exposures measured in field studies? · Does chemically dispersed oil pose more or less hazard than untreated oil to organisms at the sea surface and in the immediate subsurface waters?
HISTORY AND ISSUES 27 · Does chemically dispersed of} pose more or less hazard than untreated of} to benthic organisms? · How do offshore biological effects compare with effects on shorelines habitats, including the intertidal and immediate subtidal zones? · Are there chronic effects from dispersants used to control oil spins? · r> slon ~ Is the of} biodegradation rate increased after chemical disper- Is photoox~dation greater or less after chemical dispersion? · Is there need to further refine and standardize toxicity testing in laboratory studies? · Does chemical dispersion increase or decrease the amount of oil that attaches to solids in the water column and the amount of of! that enters the sediments? Will chemical dispersion of the of] slick protect seabirds and mammals? . Should a more effective dispersant (less applied) be used even if it is more toxic? Answers to the above questions are addressed principally in Chapters 3 and 4.