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.
6 Technical Basis of Decision Making This chapter considers the scientific and technical information reviewed in the previous chapters and uses that information to rec- ommend what to do when an of! spill occurs. FINDINGS FROM PREVIOUS CHAPTERS The preceding chapters have shown the following: . Recent chemical formulations can effectively disperse an oil that spreads on water if the of] viscosity is lower than approximately 2,000 cSt. Dispersion becomes progressively more difficult with in- creasing viscosity until, at viscosities higher than around 10,000 cSt, little of} is dispersed. For small, medium, and most large spills, dispersed oil con- centrations in open waters tend to decrease rapidly owing to tidal currents and other transport processes. ~ Very large spills, such as Bloc I, may introduce such a large, continuous flow of oft that normal, open-sea current cannot provide rapid dispersal. However, for most spills, unless water circulation is limited, organism exposure to dispersed of] is likely to be low compared with the exposures required to cause behavioral changes or mortalities. · The principal benefit of of} spin control by chemical disper- sion or mechanical recovery is the prevention of of! from stranding 239
240 USING OIL SPILL DISPERSANTS ON THE SEA on shore, entering sensitive shoreline habitats, or entering sensitive areas such as seabird colonies or sea otter locations. Serious ad- verse biological effects from untreated of} have been documented on seabirds (if present) at many spins, and by of} that concentrates on shores. · Dispersants are most effective when applied early. Oil be- comes progressively less dispersible with time as its viscosity in- creases by loss of volatile hydrocarbons and by formation of water- in-oi] emulsions (for a number of oils). Thus, the decision to use dispersants should be made as rapidly as possible after a spill occurs, preferably within the first few hours. · Spilled oils generally attain an average slick thickness of 0.1 mm or less in an hour or two, and this thickness appears to be relatively independent of spin size for those oils that spread on water. However, it should be noted that the distribution of of} on water is usually not uniform, and there may be some areas within the slick that are significantly thinner or thicker than 0.1 mm. · As water temperature decreases, of! viscosities increase. Thus, oils that spread in tropical or temperate climates are less able to spread at arctic water temperatures. Lower temperatures may also cause additional oils to be solid or semisolid because the temperature is below their pour point. Some oils have pour points in excess of the highest likely ambient temperatures; little spreading occurs when they spill. · The dispersant spray must hit the thicker part of the slick. Aerial or boat spraying usually requires direction by spotter aircraft. . Dissolved hydrocarbons in the water column after dispersion of an oil slick are largely limited to areas close to the spin source, because most of the volatile and soluble hydrocarbons in the oil evaporate rapidly from the slick before dispersion. Hydrocarbons dissolved in the water also evaporate into the atmosphere and are diluted rapidly in the water column. These dissolved hydrocarbons (many of which are aromatic) appear to produce the most immediate biological toxicity. TECHNICAL QUESTIONS A number of technical questions must be answered when consid- ering dispersant use as an of! spill countermeasure. These questions are discussed in this section.
TECHNICAL BASIS OF DECISION MAKING 241 Response Options Whether a countermeasure is needed or whether the spin wiB be dissipated by natural forces before it can impact a sensitive resource must be determined. Natural dissipation can be expected if the seas are rough, the of} is thinly spread on the water surface, the spin is not threatening a shore or sensitive area, or the volume of of! spilled is small. Alternative countermeasures, their availability, and determina- tion of their ability to remove more or less of} than dispersants are further considerations. It should be noted that mechanical conta~n- ment and recovery are generally ineffective if the oil layer is relatively thin (less than about 0.05 mm), or if the sea is moderately rough, typically sea state 4 or greater. Environmental Considerations The use of a chemical dispersant may not be appropriate on a] portions of a spill. While laboratory and mesoscale tests have shown that the acute biological effect of dispersed oil is no worse than of untreated of} per unit of oil, there are species and habitats, such as benthic organisms and mollusks, that may suffer greater damage than that caused by untreated oil. However, several nearshore studies (Chapter 4) have shown that dispersal of of] offshore reduces its impact on intertidal and benthic communities. The problem of anticipating environmental damage is tied to an assessment of natural populations and habitats that could be threat- ened by an of} spill. This environmental assessment should be done and the results incorporated into scenarios for areas of concern as a component of the prespill information base supporting the decision- making process. Since inaction in undertaking spin treatment may cause the greatest environmental harm, the environmental assess- ment data and information base should be sufficient, and operational scenarios that include this information should be understood and accepted as part of prespill planning. The desirable objective in the decision-making process is to be able to focus on operational details, such as the location of aircraft and boats relative to the spill, at the time of an accidental spill. Other Factors That Affect Decision Making Spill size is important because the area covered by the slick may be so great that it overwhelms mechanical response capabilities and
242 USING OIL SPILL DISPERSANTS ON THE SEA possibly even dispersant spray capabilities. Thus, malting logical decisions concerning oil spin control requires evaluation of the ca- pabilities of available methods. For purposes of this discussion, an average slick thickness of 0.1 mm is used. Method capabilities are limited also by operating conditions, which imply that operations should be carefully monitored during a spin. Monitoring, control, and evaluation usually can best be done from the air by spotter aircraft. Thus, operations, whether by skimmers, spray boats, or spray aircraft, are limited to daylight Bent adequate flying conditions. Night operations are seldom possible, except possibly for spray barges (and boats) and skimmers operating at the source of a continuous spill. Skimmers with 100 percent efficiency encountering a O.~-mm- thick slick at ~ kn, with sweep widths of 10 m (3.3 ft) and 100 m (33 ft), would collect, respectively, il6 bbl, and 1,160 bb! of oil in a 10-hr day. Thus, it would take aD day for one skimmer with a 10-m sweep width to collect about 100 bbl of oil unless it can operate in areas where the oil thickness is greater than 0.! mm. A large oceangoing skimmer system with a lOO-m encounter width (heavy seaboom, three ships, and collection barge) might handle a 1,000-bbl spill in a day under ideal conditions. If the oil has a high viscosity, and has not been spread by wind and waves, skimmers may have greater collection potential. Skimming systems are also limited by wind, currents, and sea state. It should be noted that the percentage of oil recovered at accidental spills has been low, particularly with large spills. Spray boats, moving through a slick at 6 kn with spray widths of 5 to 10 m (16 to 33 ft) and operating with 100 percent efficiency (although this is unlikely), might disperse, respectively, 350 to 700 bbT of oil over 10 hr. Although a spray boat can operate in sea states where skimming systems are ineffective, larger waves reduce its efficiency. The boat may have to decrease speed, and the outboard nozzles may dip into the water. Larger boats rod less and can carry large amounts of dispersant. Spray planes have the advantage of spraying dispersant rapidly, but may have the disadvantage of not carrying large amounts of dispersant. They also are capable of rapid response, and of response to more remote areas (perhaps the only response). Small planes and helicopters have limited range from a support base. A large plane flying at 140 kn with a spray swath width of 100 m could cover 28.5 km2 in 1 hr. Thus, the capacity of the spray tanks, not the slick area, is the controlling factor. Ideally,
TECHNICAL BASIS OF DECISION MAKING 243 a C-130 (Hercules) aircraft with ADDS, which has a 130-bb! tank that can spray 2,600 bb] of oil with a dispersant-oi} ratio of 1:20 on each flight, could make six to eight flights per day depending on the distance from base to slick. The above analysis for spray boats and aircraft has assumed 100 percent dispersion of the slick. Generally, that is not the case. Higher dispersant application rates might be required, and corre- spondingly larger spray capabilities required for oils that are not so readily dispersible. Because water-in-oi} emulsion formation hinders or prevents effective chemical dispersion, to be effective, of} slicks should be sprayed before the oil incorporates water. In practice if control of the entire slick is not possible, spraying should be directed to the slick closest to shore or a sensitive resource. Weather Conditions In general, oil is dispersed more readily when the sea is rough than when it is calm. Mackay (1986) suggests that chemical dis- persion may be less effective at wind speeds under about 7 m/see, although this is not a precise threshold nor is its value firmly estab- lished. This does not mean that dispersants should not be applied, but they are likely to be less effective. Conversely, if the seas are very rough (sea state 5 or higher), treatment may not be necessary because wind and wave action might be adequate to remove the spilled oil from the water surface quickly and application may not be practical under rough conditions. However, two other factors should be considered in rough seas: 1. The spill will move relatively quickly (rapid advection) at high wind speeds, so time available for response may be less. 2. Some of the naturally dispersed of! may resurface as the weather moderates and the seas subside. ADVANCE PLANNING Although some of the information needed for decision malting will only be available at the time of the spin, much can be obtained well in advance and incorporated into an advance plan for of! spill control. The following information would be desirable for dispersant use, but much of it applies to other control methods as well:
244 USING OIL SPILL DISPERSANTS ON THE SEA · potential sources of crude oils and products that may be spiked- type of oils produced in or transported through the area of interest, volumes involved, routes traveled (tankers and pipelines), and locations of of! production platforms; . environmentally sensitive resources that might be impacted by spilled of]relative sensitivities, local priorities for protection, and relative importance, that is, to the resource management agencies; · available dispersants and storage locations dispersant prop- erties and performance with oils of concern, and appropriate app~i- cation rates; · available equipment- type and location, with proper calibra- tion for dispersants to be used, and availability of adequately trained operators; and monitoringavailable means to monitor dispersant applica- tion and their effectiveness, other appropriate measurements or ob- servations, needed instruments, and trained operators. Additional site-specific data are also needed, such as spin loca- tion, volume and type of oil, and local meteorological and hydro- graph~c information. Finally, one more component is needed in order to prepare for dispersant use: a weD-conceived system for making the dispersant-use decision, and acceptance of this system by the regulatory agencies that are involved. DECISION SCHEMES The use of the technical information discussed above may be illustrated by decision-making diagrams, accompanied by extensive footnotes and text. Examples are shown in Figures 6-l to 6-4. They are similar in some ways, but each was developed for a different purpose and each emphasizes different aspects of spin response. (It should be noted that these decision diagrams are used for illustra- tive purposes and do not by themselves comprise complete decision- making tools.) The decision-making diagrams shown have been selected from those that are in use primarily in the United States. However, they are similar to diagrams that have been published elsewhere, e.g., by the International Maritime Organization (1982) and the Interna- tional Petroleum Industry Environmental Conservation Association (1980). These diagrams have been proposed for use by spill response coordinators at the time of a spin, but it appears likely that such use
TECHNICAL BASIS OF DECISION MAKING 245 will only be effective if the spin response coordinator has experience with their use, for example, through training sessions in advance of a spin. This is because dispersant-use decisions should be made promptly; any delays can result in serious loss of dispersant effective- ness. Thus, those who provide and assemble the background data should be trained in its use, and regulatory decision makers should also be trained so that they wiD understand the decisions made and the need for speedy action. Ideally, the decision to use dispersants should be made prior to a spill. U.S. EPA Oil SpiU Response Decision Tree The U.S. EPA procedure, programmed for use on personal com- puters, is one of the more detailed and complete decision-making procedures available (Flaherty et al., 1987~. At each node in the decision diagram the user may request an explanation of the factors involved in each option. Help menus include information on mechani- cal containment and recovery, observation techniques and needs, and conditions that would lead to a decision to let natural processes clean up the spin. Consideration is given to the effectiveness of different countermeasures, weather conditions, spill site, oil type, and other factors. Although it is not shown in Figure 6-1, the text of the program explains that simultaneous use of more than one countermeasure may be appropriate. Little or no guidance is Even on evaluating the environmental trade-offs that usually must be made between untreated versus dispersed oil. The most time-consum~ng component of a dispersant-use deci- sion is the question of environmental damage: Will dispersant use result in more or less damage than nonuse? This question should preferably be addressed prior to any spill, when decisions should be made about the locations and the conditions under which dispersant use should be considered or when their use would be inappropriate. AP! Decision Diagram The APT decision diagram is one of the less complex. It is based on the concept that spraying the of] slick will have little or no adverse biological effects based on a comparison of field hydrocarbon exposures with laboratory bioassays and behavioral studies. It also brings in spill size as it relates to the spill control capabilities of skimmers, spray boats, and spray aircraft.
246 1 Togo | OPENING REMARKS l 1100 AERIAL SURVEILLANCE 1200 IS OIL VISIBLE ON yes THE WATER SURFACE? no 4100 CONTINUE SURVEILLANCE 1300 IS OIL A SAFETY HAZARD TO PERSONNEL? 1400 ATTEMPT TO REMOVE HAZARDS , ,, ~ ., .__ 1500 ~ WILL OIL IMPACT - yes SENSITIVE AREAS? no 1 4200 ALLOW FOR NATURAL REMOVAL AND CONTINUE SURVEILLANCE 1700 IS THE OIL SLICK GREATER THAN 0.05 MM THICK? 4300 ALLOW FOR NATURAL REMOVAL AND CONTINUE SURVEILLANCE USING OIL SPILL DISPERSANTS ON THE SEA Ayes 1800 IS THE SEA STATE -yes GREATER THAN 3? no 1900 ' 0 _ IS THE OIL SLICK GREATER THAN 0.50 mm THICK? yes _ DEPLOY MECHANICAL EQUIPMENT no_ IS THE MECHANICAL EC;IUIPMENT EFFECTIVE? yes 2400 CONTINUE WITH MECHANICAL RECOVERY AND SURVEILLANCE r o- 2300 IS THE MECHANICAL EC;IUIPMENT EFFECTIVE? 2400 CONTINUE WITH MECHANICAL RECOVERY AND SURVEILLANCE l 1 2600 | IS THE SEA STATE l | GREATER THAN 5? l yes 2500 I ALLOW FOR NATURAL l REMOVAL AND CONTINUE | SURVEILLANCE I l 2700 1 IS THE OIL ~ not DISPERSIBLE? I l yes 2600 IS THE USE OF DISPERSANTS | ACCEPTABLE? l Knot yes l 1 3200 I EVALUATE DISPERSANT I AVAILABILITY I . . . _ r ' I ~ 1 3300 ~ no | ARE AVAILABLE l I DISPERSANTS I | EFFECTIVE? Ayes . 3000 PREPARE FOR IMPACT I ON SENSITIVE AREAS I AND CONTINUE | SURVEILLANCE 3400 ARE DISPERSANTS I SAFE? 3500 | OBTAIN APPROVAL FOR I DISPERSANT USE 3550 WERE APPROVAL AND I CONCURRENCE I OBTAINED I FOR DISPERSANT USE? I yes 3600 | IS THE SPILL AREA L o | GREATER THAN 60 . n | ACRES? . yes 1~ nc - no 1 3700 I APPLY DISPERSANT BY | AERIAL SPRAY FROM | FIXED-WING AIRCRAFT 3800 , APPLY DISPERSANT BY | BOAT OR HELICOPTER I SPRAY 3900 IS DISPERSANT EFFECTIVE? 1 4000 | CONTINUE WITH | DISPERSANT | APPLICATION | AND CONTINUE WITH | SURVEILLANCE 44oo | CONSIDER OTHER | ALTERNATIVES, | PREPARE FOR | IMPACT ON | SENSITIVE AREAS l 1~ no FIGURE ~1 U.S. Environmental Protection Agency Oil Spill Response Decision Tree. Source: Flaherty et al., 1987.
TECHNrICAL BASIS OF DECISION MAKING 247 Figure 6-2 is an oil spin control diagram that outlines the real- isticaDy available options. If the estimated spin volume is less than 1,000 bbI, a choice can be made between mechanical recovery and dispersant spraying. This choice depends on availability of mechani- cal equipment and suitability of winds, waves, currents, and response time; or availability of spray planes and dispersibility of the of! (Fig- ure 6-2, lower left). If neither option is available, the shoreline or sensitive habitats can be cleaned using appropriate methods, such as those suggested by AP] (1985), or the of} can be left to weather naturally. Spills much over 1,000 bb] per day have little possibility of being controlled by mechanical means unless conditions are ideal (waves less than 1.3 m and surface currents less than ~ kn) and a large amount of equipment is available. Dispersant application by large aircraft spraying systems would appear to be the only serious control possibility for large of} spills (Figure 6-2, lower right). Because it is unlikely that there will be sufficient mechanical equipment available to control larger of} spills, equipment that is available should be used to collect or divert spired of] as it approaches critical locations. Mechanical equipment can be used effectively on spills of oils that have pour points above the ambient temperature, are highly viscous, do not spread, or have formed a viscous mousse. If the oils have not spread, mechanical recovery devices have less area to cover. Health hazards must be considered. Mechanical cleanup and spray boat personnel must be protected from volatile hydrocarbons when operating in an oil slick downwind near, for example, a wed blowout. Special precautions must be taken if the of] and associated gas contain hydrogen sulfide (H2S). Operations also must be outside the zone in which gas and air forms an explosive mixture. S[R Dispersant Dec~sion-Making Workbook The objective of this decision-making method is solely to indi- cate whether or not dispersant use is environmentally appropriate. The S. L. Ross (SIR) workbook (Figure 6-3) gives methods for characterizing, on a numerical basis, the environmental impacts on populations that may be at risk from either dispersed or untreated oil (Truce! and Ross, 1987~. Using these computed values, methodical and objective decisions can be made regarding the advisability of dispersant use or nonuse from an environmental perspective. Other
248 ·Q O _ 0 ° ~ ~ -O c a o' .0 0 ~ ~ _ g mC C ~ E o .= ._ (D O ~ 0 ~ ~D al Q _ _ ._ ~ D .c 0 3 \ 4 ;- ,, ' ~, ___ - - a' ~o - ~ <~' ~o Q Q v' CO ~ o' C {o ·_ n ~ 0 ttSQ~ (D ~i52~- ~ ~ ~ I ~ ~ C~ C~.0 DO E ~ 3 o._ O '~c, ~ ~ c ~ 0 .> ~ 0' .~ u, c i~ c ~ ,_ ~D _ t 0 s ~ ,O0 >-~`,o s:3 l _ 1 mo .C) C C s <' o (D L) - = ~ . a) n O (15 o~ s c ~ a, ~ ~ a~ > ~ ~ a, ~ (~) ID C 73 E ~D u~ 4) .o > _= ~ ~ c c co ~ a' a) - ~ ~n c `_ ~n _ D a <t = 0 d.)-='S 0' ~ a, c nm°. n ,,, s" ~, O a, (D O ~ ~s Q ~ o> cr7 ~ ~ a, O >~_ ~ _ td Q <D Q Q 4) D O ~ ~ `._ C oo ~a ~ . o, ~ `,, o ~ 6 .~) {~, _ 0 Q ~ 43D cc, CO o O (D X . I ~ tD o ~ o ° C,) C ~ Q to O - Q~ _ _ ~D ~D C o ~n ~n .cO ~ C c C O ~ _ ~ _ ~ . C o ~ o ·t C $ mc =; 3 <t 3 _ .c o O aC' ~ c I C o 8 ~ ~ .,, `, o C _ C .<~ ~ C ~5 c mc 0 8 ~ 'c c <~s <D c ° ~ 0~ n a) ,c 1 c c ~ ~ c O ~ ~ 0 a' C CQ o C o ~ ° c ~ c ~6 ~ a~ ~~~ ~ ~ ~ t`] c c _ ~ ~ c ~ c O _ ~ _ CD 00 b' c~ · U) ·_ ~n ·C' P4 ·. 1_ P~ 53 C~ :D U
TECHNICAL BASIS OF DECISION MAKING 249 aspects of spill response (i.e., mechanical recovery and natural re- moval) are deliberately not considered, because they are considered to be separate parts of the of} spin countermeasures problem. No guidance is given on dispersant application rates, effects of weather conditions, spin size, or of! condition. State of Alaska Dispersant-Use Guidelines The State of Alaska's guidelines are illustrated in Figure 6-4. The user must assemble a significant amount of information prior to making a dispersant-use decision, including a comparison of the effects of dispersed of! and untreated of] on populations at risk (Re- gional Response Team Working Group, 1986~. However, this system gives no guidance as to how to make the comparison and appears to assume a fairly high level of expertise by the user. Accompanying the decision tree are maps and text showing zones in which dispersants ~ may be used with approval by the federal on-scene coordinator (OSC); · may be used only with concurrence of the EPA and the state plus consultation with the Regional Response Team; or · may not be used. Federal Region OX (California) has dispersant-use guidelines that are similar in many ways to those of the State of Alaska, except that maps have not been prepared in California showing areas where the OSC may approve dispersant use unilaterally. It may be noted that the Region OX guidelines have been used on two occasions to reach decisions favorable to dispersant use in 1984 at the Puerto Rican spin (Zawadzki et al., 1987) and at the 1987 M/V PacBaroness spin (Oi} Spill Intelligence Report, 1987b,c). However, it should also be noted that on both occasions it took more than 24 Or to come to this decision (Onstad, private communication). The objective of this method is solely to indicate, from a regula- tory perspective, whether dispersant use is or is not appropriate to consider. Note that the OSC must notify the U.S. EPA and the State of Alaska as soon as possible if he or she authorizes dispersant use. The zones are defined by bathymetry and currents, biological pa- rameters, nearshore human activities, and time required to respond. The zones were defined by a subcommittee of the Alaska Regional Response Team. The zones were not evaluated by procedures such as those in the SER workbook. In the event that dispersant use may
250 USING OIL SPILL DISPERSANTS ON THE SEA OIL SPILL TREATMENT OPTIONS CHEMICAL DISPERSION OIL FATE & MOVEMENTS IDENTIFY THREATENED RESOURCES RELATIVE IMPORTANCE OF RESOURCES( ) 1 QUANTITATIVE IMPACT O ON RESOURCES SUMMARIZE COMPARE DECIDE( ) DOCUMENT (a) Relative importance of sensitive resources is determined by the affected regulatory agencies in terms of "High," "Medium," and Low"; the determination is based on local priorities. (b) Quantitative impact on resources is calculated using environmentally based algorithms; these algorithms yield a quantitative estimate of the degree of impact on each resource in terms of "Major," "Moderate," "Slight," or "Negligible." (c) The dispersant use decision is based on a comparison of the impacts on affected resources by the spilled oil if chemically dispersed versus the impacts (usually on a different set of affected resources) by the untreated oil. NO DISPERSION OIL FATE & MOVEMENTS IDENTIFY THREATENED RESOURCES RELATIVE IMPORTANCE OF RESOURCES(a) QUANTITATIVE IMPACT ON RESOURCES(b) FIGURE ~3 SLR dec~sion-~r~king method. Source: Trudel et al., 1983. be authorized, no guidance is given as to application rates or effects of conditions such as weather, spill size, and of! condition. Comparison of Decision-Making Diagrams The four decision-making diagrams shown in Figures 6-1 through 6-4 are compared in Table 6-1. From the comparison, it appears that
TECHNICAL BASIS OF DECISION MAKING Dispersant Use Decision Matrix. (The Following Questions Must be Answered Before Deciding to Use Dispersants.) OIL MOVING ONSHORE OR INTO CRITICAL AREA / \ YES \`NO IS MECHANICAL CONTROL AND IS ACTION REQUIRED RECOVERY FEASIBLE? ~ . YES ~ OR DESIRED? ~ \ NO ~ YES \ MONITOR MOVEMENTS IMPLEMENT \ ~ NO ARE CONTROL/RECOVERY ACTIONS ADEQUATE? YES / CONTINUE ACTIONS NO, OR PARTIALLY \` CAN OIL TYPE AND CONDITION BE CHEMICALLY DISPERSED? YES/ \ NO IS A DISPERSION NO OPERATION POSSIBLE? - ~ TREAT ONSHORE ~ YES WILL ENVIRONMENTAL IMPACTS ASSOCIATED WITH CHEMICAL DISPERSION BE LESS THAN THOSE OCCURRING WITHOUT CHEMICAL DISPERSION? REQUEST APPROVAL FOR / USE OF DISPERSANTS USING ATTACHED PROCEDURES YE~\NO \ TREAT ONSHORE NOTE: Immediate threat to life PREEMPTS the necessity to use this matrix. 251 FIGURE ~4 State of Alaska dispersant sloe decision matrix. Source: Regional Response Learn Working Group, 1986.
252 USING OIL SPILL DISPERSANTS ON THE SEA TABLE 6-1 Comparison of Decision-Making Diagrams Factor EPA, API, SLY, Alaska, Figure 6-1 Figure 6-2 Figure 6-3 Figure 6-4 Surveillance Personnel hazards Danger to sensitive areas Is natural removal appropriate? In oil thick enough to be a concern? Spill size Is mechanical recovery feasible? Is mechanical recovery effective? Is the oil dispersible? Are dispersant resources available and effective? Need to obtain approval Are environmental impacts of dispersed oil less than those of untreated oil? Is dispersant use effective? Application rates l 1 2 2 1 1 1 2 2 2 1 2 1 1 1 1 2 1 2 l 1 1 1 1 l 2 1 1 2 1 1 1 2 2 KEY: 1 = Primary consideration or guidance is given; 2 = Included only indirectly or by inference. the U.S. EPA Oil Spill Response Decision Tree (Figure 6-1) is more complete and detailed than the others. It was developed as an overall too] to guide response to an oil spin. As reported by Flaherty et al. (1987), a user can reach a decision within a few minutes, providing the data are available. The speed of use of this process results in part from its having been programmed for a personal computer. which makes it particularly suited for training purposes. The AP] decision diagram (Figure 6-2) emphasizes the need for dispersant use as the only really feasible means of responding to spins that exceed the capabilities of available booms and skimmers. In many cases mechanical cleanup capabilities may be only on the order of i,000 bb} per day. Figure 6-2 points out the serious limi- tations to mechanical containment and recovery for extremely large spills (over 1,000 bb} per day). The concepts embodied in the APT .r 7
TECHNICAL BASIS OF DECISION MAKING 253 decision diagram could be effectively incorporated into the EPA com- puter program, which would be especially useful for training response personnel. The SER decision-making method (Figure 6-3) addresses almost exclusively the question of biological trade-offs. It is relatively unique in its approach to comparing the environmental (biological) effects of dispersed oil with the those of untreated oil. This methodology appears to be needed in order to make the judgments called for both in the U.S. EPA computerized Oil Spill Response Decision Tree and the Alaska decision matrix, which is designed as a means of regulating and controlling dispersant use.
