A Review of Dispersant Use in Response to the British Petroleum Deepwater Horizon Oil Spill

PAGE 1

A Review of Dispersant Use in Response to the British Petroleum Deepwater Horizon Oil Spill By Bryant Turffs A Thesis Submitted to the Division of Natural Sciences New College of Florida In partial fulfillment of the requirements for the degree Bachelor of Arts Under the sponsorship of Dr. Diana Weber Sarasota, FL May, 2011

PAGE 2

A Review of Dispersant Use in Response to the British Petroleum Deepwater Horizon Oil Spill by Bryant Turffs Copyright 2011 ii

PAGE 3

Acknowledgements I owe many people a great deal of thanks for assistance thro ughout this project and for helping me to arrive at this point in my life. I would like to thank my Sponsor, Dr. Diana Weber, for taking me on as a thesis student. Her guidance and editing greatly improved this thesis. I would like to thank my committee, Dr. Paul Scudder, Dr. Frank Alcock, and Julie Morris for their help and comments. Particularly, I thank Dr. Alcock for taking the time to read my thesis and partake in my baccalaureate exam while on leave in New Zealand. I would also like to thank Dr. Elzie McCord and Prof. Arden Calvert who offered me assistance in creating this project. I thank my family for their unconditional support and keeping me on track. I thank Carly for her invaluable support and assistance during the past ye ar, which helped me realize my full potential. I thank all my friends for their support and helping me relax during this proce ss. I owe particular thanks to Katie Dean and Katie Smith for thei r assistance in formatting this thesis and for the chocolate mousse. I thank to the mari ne lab staff for going above and beyond in helping me with the early stages of my thes is, particularly to Karen for wading through toxic muck in search of fish. Finally, I would like to acknowledge the Nalco Company, which was remarkably forthcoming and helpful to me given the nature of this project. iii

PAGE 4

Table of Contents List of Tables................................................................................................................. ....vi List of Figures...................................................................................................................vii Abstract....................................................................................................................... .......ix CHAPTER 1...........................................................................................................................1 INTRODUCTION: THE DEEPWATER HORIZON BLOWOUT......................................................1 The Deepwater Horizon Accident...............................................................................2 The Response...............................................................................................................6 The First Deepwater Spill..........................................................................................12 Research Aims...........................................................................................................14 References.................................................................................................................15 CHAPTER 2.........................................................................................................................24 PHYSICAL AND CHEMICAL EFFECTS OF OIL DISPERSANTS...................................................24 The Composition of Crude Oil..................................................................................25 Physical and Chemical Pr operties of Petroleum........................................................28 Oil Fate......................................................................................................................3 2 Transport....................................................................................................................33 Weathering.................................................................................................................36 Differences in the Fate of Deep Water Spills............................................................39 Oil Dispersants..........................................................................................................42 Effects of Dispersants on Deep water Releases.........................................................48 Conclusion.................................................................................................................50 References.................................................................................................................52 CHAPTER 3.........................................................................................................................66 iv

PAGE 5

THE POTENTIAL BIOLOGICAL AND ENVIRONM ENTAL EFFECTS OF OIL DISPERSANT USE......66 Introduction...............................................................................................................67 Toxicity: The Effects of Dispersants on Bi ological Systems....................................67 Bioavailability and Bioaccumulation........................................................................80 Biodegradation...........................................................................................................81 Plankton.....................................................................................................................85 Microbes....................................................................................................................86 Benthos......................................................................................................................87 Unique Communities and Characteristics.................................................................88 Deepwater Horizon....................................................................................................90 References.................................................................................................................94 CHAPTER 4.......................................................................................................................104 THE POLITICAL FRAMEWORK THAT SURROUNDED THE APPLICATION OF DISPERSANTS IN THE DEEPWATER HORIZON OIL SPILL......................................................................................104 Introduction: The Parties Involved..........................................................................105 Legal Framework.....................................................................................................110 What Scientific Input Pertains?...............................................................................115 Strength of Positions................................................................................................116 Control of Information.............................................................................................118 Risk Assessment, Future Directions and Conclusions............................................119 References...............................................................................................................125 CHAPTER 5.......................................................................................................................130 CONCLUSIONS ABOUT THE USE OF CHEM ICAL DISPERSANTS IN MARINE SPILLS................130 References...............................................................................................................138 APPENDIX A.....................................................................................................................149 v

PAGE 6

List of Tables Chapter 1 1. Decisions Leading to the Macondo Well Blowout................................18 Chapter 2 1. Components of Corexit Dispersants......................................................55 2. Properties of EPA Standard Crude Oils ................................................56 Chapter 3 1. NCP Toxicity Data for Select Dispersants...........................................100 1. EPA Independent Toxicity Data..........................................................101 Chapter 5 1. Notable Global Marine Blowouts pre-2010.........................................141 2. Notable US Gulf of Mexico Blowouts 1947-2009 .............................142 3. Deepwater Horizon Oil Budget...........................................................143 4. Gulf of Mexico Petroleum Operations as of April 27, 2011................144 vi

PAGE 7

List of Figures Chapter 1 1. Diagram of a Deepwater Drilling Rig....................................................19 2. Well Finishing Techniques....................................................................20 3. The Deepwater Horizon on Fire............................................................21 4. Theoretical Diagram of Deepwa ter Dispersant Application..................22 5. Diagram of the Macondo Wellhead Post Blowout................................23 Chapter 2 1. Diagram of Oil Fate Processes...............................................................57 2. Emulsified Oil........................................................................................58 3. A Tar Ball..............................................................................................59 4. Diagram of Deepwater Oil Transport....................................................60 5. Simplified Diagram of Di spersant-Oil Interaction................................61 6. Detail Diagram of Disp ersant-Oil Interaction........................................62 7. Chemical Structure of Se lect Corexit Ingredients ................................53 8. Diagram of Swirling Flask Test.............................................................65 Chapter 3 1. Pathways of Photo-Enhanced Toxicity................................................102 2. Mechanisms of Biological Hydrocarbon Degradation........................103 Chapter 4 1. Dispersant Pre-Authorization Map......................................................128 2. BP Dispersant Decision Tree...............................................................129 vii

PAGE 8

Chapter 5 1. Relative Location of the Ixtoc I and Macondo Wells..........................145 2. Extent of Oil on the Surface from the Macondo Blowout...................146 3. Theoretical Extent of the Deepwater Horizon Discharge ...................147 4. Gulf of Mexico Oil and Gas Leases by Depth.....................................148 viii

PAGE 9

ix A Review of Dispersant Use in Response to the British Petroleum Deepwater Horizon Oil Spill Bryant Turffs New College of Florida, 2011 Abstract The British Petroleum Deepwater Horizon oil spill was an unprecedented event due to the depth at which it occurred and the amount of oil discharged. Chemical dispersants, which break oil into droplets a nd facilitate diffusion into the water column, were applied to the oil spill to mitigate the damage caused to shoreline environments and surface dwelling organisms. Dispersants may al so increase the rate at which oil is consumed by microbes. Due to the location of the oil discharge in the deep ocean, dispersants were also applied at the sea fl oor to enhance efficacy. The biological effects of chemically dispersed oil are poorly understood especially in the deep sea. Dispersants are thought to decrease the harmful effects of oil to shorelines and surface communities, but may increase harm to the communities of the water column. The scale of this spill and its location made responding to the spill difficult and highlighted insufficient industry oversight and contingency planning. Dispersants were used in accordance with existing US laws. The biological and environmental impacts have yet to be fully understood. Further research on the effect s of this spill must be conducted because the offshore drilling industry is growing, increasing the chan ces of a similar occurrence in the future. _________________________ Dr. Diana Weber Division of Natural Sciences

PAGE 11

CHAPTER 1 INTRODUCTION: THE DEEPWATER HORIZON BLOWOUT 1

PAGE 12

The Deepwater Horizon Accident In March 2010, President Obama announced a new federal program to encourage offshore oil and gas production (Johnson and Torrice 2010). As part of this effort, a drilling rig, owned/operated by Transocean, but contracted by British Petroleum (BP) was finalizing the drilling of a well in the no rthern Gulf of Mexico, was located in the Mississippi Canyon (Reed and F itzgerld 2011; Steffy 2011). This canyon is located 40 miles off Louisiana, in an area where deep water drilling has been ongoing since 1979. The well in this canyon was named Macondo, after a fictional town in a novel, One Hundred Years of Solitude, by Gabriel Garcia Marquez in which the town is destroyed by a hurricane. In actuality, th e drilling of the Macondo well ha d already been interrupted by a hurricane. The associated offshore rig, Deepwater Horizon (DWH), was drilling in the northern Gulf of Mexico in approx imately 1500 m (5000 ft) of water on April 20, 2010. The DWH drilling platform stood twenty stor ies above the surface of the ocean, had a crew of nearly 130 persons (Steffy 2011), a nd was capable of drilling wells in 3000 meters (10,000 ft) of water and 9000 meters (30,000 ft) below the sea floor (Figure 1). Offshore petroleum production has led to the drilling of 50,000 wells within the Gulf of Mexico since 1950. More than 700 of these wells have been drilled in deep water, between 150 and 1500 meters (500 and 5000 ft ), and in ultra-deep water below 1500 meters (5,000 ft) recently. Drilling deeper was risky, but held big rewards and BP was the leader in deep water drilling in the Gulf of Mexico (Reed a nd Fitzgerald 2011). BP began experiencing safety problems long before the DWH began drilling the Macondo well (Steffy 2011). In the mid 1990s John Browne, the CEO of BP, began a 2

PAGE 13

campaign to cut costs and bring BP back as a major player in the oil industry. This campaign was later characterized as an exch ange of engineers for accountants. Browne successfully bolstered the profits of BP but in doing so cr eated a culture within the company that put profits before safety. As a result, BP experienced many safety problems throughout its branches, includi ng a petroleum leak in its Alaskan pipeline and several fatal accidents at its refinery in Texas City, Texas. The successor to Browne, Tony Haywood, enacted changes meant to improve sa fety but his countermeasures proved too weak on more than one occasion. In early 2010, the DWH took over drilling the Macondo we ll from the Marianas, a drilling rig that had been experiencing technical problems causing BP to be behind schedule (Steffy 2011). Because drilling had been costing BP over $500,000 per day, the mangers of BP pressured its employees and contractors (Transocean and Halliburton) to complete the well quickly. The DWH had recently finished drilling the deepest well in history, the Tiber, which wa s in 1200 m (4000 ft) of water and a total drilling depth [water and below the sediment] of more than 10,600 m (35,000 ft) (R eed and Fitzgerald 2011). The Macondo well was in slightly d eeper water of 1500 m (5000 ft), but was considered a modest well of 5600 m (18,360 ft) deep, a routine depth for the DWH. Shortly after the Horizon began drilling, it bega n to experience difficulties, such as a loss of drilling mud when the rock formation be ing drilled cracked (Reed and Fitzgerald 2011; Steffy 2011). This can be extremely dange rous because natural gas is released from the rock, but engineers can usually fix the problem in less than a day by pumping cement into the hole. However, it took Halliburton co ntractors, who were in charge of cementing the well, ten days to fix the problem. 3

PAGE 14

A month later, the drilling crew had a w ell control situation, which occurs when the upward pressure in the well is at dangerously high levels (Reed and Fitzgerald 2011). The petroleum within a well is buried deep within the surface of the earth under the immense pressure of the rocks above and heated up to 250 F (Lehner and Deans 2010). Drilling into these reservoirs releases pressu re and therefore, is very dangerous because without proper control the petroleum will surg e up the well. This occurred two days later; the crew experienced a kick, which is when the pressurized petroleum in the well exerts more upward pressure than the drill crew can exert downward with drilling mud. The upward pressure broke the drill shaft and result ed in a subsequent re lease of natural gas (Reed and Fitzgerald 2011). The lost drill shaft and bit represen ted a loss of $25 million dollars worth of equipment. Engineers formulated a new plan to drill around the broken drill shaft and were finally able to reach the desired depth. When finishing the well, the DWH crew had two options: a long string or a tie back (Figure 2) (Steffy 2011). The former is more cost effective a nd potentially longerlived but provides less security against a blow out because it uses one cement seal rather than the two used in a tie back. BP changed its plans three times, all of which were rapidly approved by the Bureau of Ocean Energy Management, Regulation and Enforcement (BOEMRE) [Formerly the Minera ls Management Service (MMS) prior to June 21, 2010] (Table 1). This highlighted another problem leading to the disaster. BOEMRE, the branch of the US government responsible for overseeing the deep water drilling industry was lenient towards industry in enforcing regulations (Lehner and Deans 2010). BP chose to finish the well with the long string option as it is cost effective. BP again failed to take adequate precaution when it realized it had few centralizers to 4

PAGE 15

complete the proposed plan. Centralizers are devi ces needed to secure the drilling shaft in the center of the well and ensure a solid cement job. BP completed the well with the centralizers on hand rather th an waiting to properly complete the well (Steffy 2011). When finishing the well Halliburton us ed light cement, which put less stress on the fragile rock formation (Steffy 2011). The light cement was less likely to crack the geological formation and compromise the well integrity, but was also less likely to stand up to the forces of a blowout. BP did not c onduct the usual test to determine integrity following cementing (Steffy 2011). On April 20, 2010, as the drill crew removed the heavy drilling mud from the well, the upward force exerted by the pressurized petroleum in the well caused the cement seal in th e well to fail, a blowout. A 15 m high blowout preventer situated on the sea floor at the t op of the well was the last protection against this blowout. This piece of equipment consisted of valves and shearing rams, blades that cut through the drill pipe, which together are designed to close the well and separate the rig from the well in the event of a blowout. Pr ior to the blowout, workers had at least two indications of problems with th e blowout preventer, but did not take care of these (Reed and Fitzgerald 2011). The resulting blowout sent natural gas shooting to the surface, which then filled the DWH and caught fire cau sing the rig to sink and the loss of 11 lives (Figure 3). The pressure inside the well for ced petroleum to spew from the damaged well and would eventually lead to the worst oil discharge in the history of the United States (Jernelov 2010). 5

PAGE 16

The Response Immediately after the accide nt all parties involved began a response to mitigate the damage caused by the blow out (Reed and Fitzgerald 2011; Steffy 2011). The incident response plan for this offshore site drafte d by BP in the event of an oil spill was ultimately ineffective and unrealistic. The plan failed to address potential impacts on key Gulf of Mexico species, and focused on spec ies not found in the Gulf i.e., walrus (Reed and Fitzgerald 2011l; Lehner and Deans 2010); s uggesting BP had used portions of a plan from another of their installations, i.e. north ern Alaska. The difficulty of working in the deep sea, the lack of knowledge about spills in the d eep sea, and the apparent lenient regulation negatively affected the response efforts. BP and the US government had two goals for the response to the well blowout. First, the flow of oil needed to be stoppe d and second, the damage caused by the released oil needed to be minimized. To accomplish the first goal, BP engineers tried several techniques to close the well before it was ul timately killed by drilling a relief well. Second, the damage from the released oil needed to be mitigated. Prior studies and experience have provided the industry with n eeded responses for a su rface oil spill (NRC 1989, 2003, 2005). As a result of this, ultimatel y, most available tools used were in response to the oil at the surface and on th e shore (National Commission on the BP Deepwater Horizon Oil Spill and Offshore Dr illing 2011), not in the deep sea. BP improvised and developed new techniques to respond to the challenges of a deep-water blowout. BP tried several methods to stop the flow of oil and accomplish its primary response goal. The only sure way to kill a blowout is by drilling a reli ef well to intersect 6

PAGE 17

with the original well, thus diverting the pr essure away in a controlled manner (Steffy 2011). This process can take months, as demons trated by the drilling of the original well. BP began drilling a relief we ll shortly after the blowout and additionally, tried other techniques to stop the oil flow. The oil indu stry has experience with blowouts but not at depths of a mile below the su rface, which make closing a well more difficult. Initially, but without success, BP tried to close th e failed blowout preventer using an ROV (National Commission on the BP Deepwate r Horizon Oil Spill and Offshore Drilling 2011). Following that, engineers constructe d a novel containment dome to siphon off the oil being discharged, but th e collection tube was clogged by the formation of methane crystals, called hydrates (Tor rice 2010a). BP tried, without success, a top kill where heavy drilling fluid is pumped back into the well. BP did achieve success using an insertion tube in the broken riser, which allowed them to siphon oil and gas and eventually cap the well (Steffy 2011). After 87 days, BP capped the well and the oil flow stopped on July 15, 2010. BP then permanently killed the well with the completion of a relief well on September 19, 2010 (Nationa l Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling 2011). The second goal was to limit the damage caused by the discharged oil. A variety of techniques were available to do this, including natural removal, in situ burning, mechanical removal and chemical counter measures including dispersants (NRC 1989, 2003, 2005). Each of these techniques will be explained but it is important to understand that each has limitations and its utility must be weighed against its efficacy, availability, and local environmental variables. The pot ential harm of each technique must be considered against its efficacy. Planning and responding to any spill is difficult because 7

PAGE 18

each spill varies in location, conditions, t ype of oil, and the environment and biota present (NRC 2005; Schooner 2010). The need for careful and thorough decision-making must be weighed against the fact that re sponse technology is most effective when deployed rapidly due to weathering and spread ing. Ultimately, the response efforts to the Macondo well blowout used all of the aforemen tioned techniques to li mit the effects of the oil. Natural removal is the process of allowing nature to dilute and degrade oil without human intervention (NRC 1989). This is the simplest form of oil spill response and therefore, to some extent is part of every response, including the DWH spill. The process of natural oil removal is referred to as fate an d is generally divided into transport and weathering. Transport processes remove oil from the local environment. Weathering processes change the chemical co mposition of oil and include evaporation, photooxidation, dissolution, physical dispersi on, sedimentation, and biodegradation. These processes ultimately result in the rest oration of the natural environment (NRC 1989). Natural removal is a slow er process than using techno logy to expedite recovery, but it does not introduce the po ssibility of further enviro nmental harm from response activities (NRC 2005). All other spill mitigation techniques attempt to expedite the natural processes by removing oil from th e environment or altering its fate. In situ burning removes oil from the surface enhancing evaporation (NRC 2005) without the need for recovery and disposal (Zurer 2003). The disadvantage is that it creates smoke that contains harmful chemicals, such as dioxins. Burning cannot be used near shore because of the creation of sm oke and the problem of fire spreading. Limitations on burning in deep water include the need for calm conditions, a minimum 8

PAGE 19

surface thickness of the oil, and a lower efficacy in emulsified oil (NRC 2003, 2005). These are complicated by a diffu se slick and early emulsion formation created by a deep water blowout. Ultimately, five percent of the oil from the DWH spill was removed through burning (National Commission on th e BP Deepwater Horizon Oil Spill and Offshore Drilling 2011). Mechanical oil collection technologies include the co ntainment of oil through barriers and the collection of oil via skim mers and sorbents (NRC 1989; Shirbon and Foster 2010). Barriers are used at the surf ace to contain oil slicks. Skimmers are machines that separate oil from water, wh ich can completely remove the oil from the marine environment. Sorbents are materials that absorb and adsorb petroleum from the water (U.S. Environmental Protection Agency 2003). Mechanic al collection is ideal for its ability to remove the oil from the environment, but is limited by the collection capacity of responders and the need for di sposal (NRC 1989, 2005). Furthermore, these techniques are not viable in seas of great er than one meter. Traditional mechanical techniques used at the surface were able to co llect three percent of the oil released from the Macondo wellhead (Nati onal Commission on the BP De epwater Horizon Oil Spill and Offshore Drilling 2011). A further 17% of the oil discharged was collected by the newly improvised technique of using an insertion tube at the wellhead to siphon oil (National Commission on the BP Deepwate r Horizon Oil Spill and Offshore Drilling 2011; Shirbon and Foster 2010). The fourth type of technology used in response to the DWH accident was chemical dispersants, which are chemical form ulations used to cha nge the distribution of oil in the environment (NRC 1989, 2005). They act similar to dish soap by breaking 9

PAGE 20

down the oil into small droplets that dissolve into the water column. Dissolving oil into the water column can have several benefits: i ) it can prevent oil from stranding onshore, ii) it lowers the concentration at the surface, which benefits surface dwelling animals like sea turtles, sea birds, and marine mammals, and iii ) it may enhance the biological consumption of crude oil (Mulkins-Phillip s and Stewart 1974), though this fact is disputed (Literathy et al. 1989; NRC 2005). Al ong with potential benefits, the use of dispersants causes reasonable concern, as mo re chemicals are added to an already disrupted environment. Removing oil from the surface and preventing it from reaching the shore may actually harm communities in the water column, whic h are subsequently exposed to more oil. The application of dispersants must be approved by the US Coast guard (USCG), Environmental Protection Agency (EPA), Depart ment of the Interior and Department of Commerce these agencies have created a preapproval process (U SCG 2011; NRC 2005). This allowed BP to immediately be able to use a suite of approved products at their discretion. Traditionally, dispersants are applie d to oil slicks at the surface by spraying the chemical from aircrafts or boats. The novel conditions of the deep-water blowout allowed BP to consider the use of new techniques in applying disper sants (Figure 4). On May 15, 2010, EPA approved a new application technique for the use of dispersants, which allowed BP to use underwater remotely operated vehicles (ROVs) to directly inject dispersant into the well discharge 1500 m (5000 ft) below the surface (Torrice 2010b; Torrice and Voith 2010). The theory was that the turbulence at th e wellhead would mix dispersant with freshly discharged oil before it reached the surface enhancing dispersant efficacy and ultimately resulting in the use of less chemical dispersant. The scientific 10

PAGE 21

community and the public became increasingly concerned about the us e of dispersants in the untested deep-water applica tion, especially as the total amount of dispersant applied at the site reached unprecedented levels (Torrice 2010b; Lehner and Deans 2010). The use of dispersants became one of the most contentious issues of the spill (Johnson and Torrice 2010; Lehner and De ans 2010; Torrice 2010b). Conflicting information about the toxicity and the effects of dispersant s on the environment and also human health was available during the time of the spill (Lehner and Deans 2010; Torrice 2010ab). Some scientists expressed concern th at as the dispersant s were intended to accelerate biodegradation, the bacteria consum ing the oil could cause oxygen depletion, which would have damaging effects (Torrice 2010a). The benefits of dispersants have been br ought into question, especially as the role and efficiency of biodegradation is poorly understood, as is the potential problem of hypoxia (Johnson and Torrice, 2010; Joye and MacDonald 2010). The use of dispersants may enhance the oil exposure of organisms in the water column (NRC 2005); however not using dispersants would allow oil to reach shore, causing coastal environmental and economic damage. BP applied in excess of 2.1 million gallons of dispersant, i.e., Corexit (Nalco Company), to the spill area, 1.4 m illion gallons were applied at the surface application, and another 0.77 million gallon s were administered at the wellhead (Kujawinski et al. 2011). These amounts a nd the novel subsea appl ication technique caused concern among scientists and the public (Lehner and Deans 2010). Preliminary results suggest that up to 16 percent of the oil discharged by the Macondo well was chemically dispersed (N ational Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling 2011). An early independent study showed that 11

PAGE 22

components of the dispersant injected at de pth were associated with deep-water oil plumes and that the oil and dispersant mix persisted for up to 60 days above detectable limits (Kujawinski et al. 2011). Hazen et al. (2010) found that these deep-water plumes stimulated the growth of oil-eating bact eria though the long-term effects remain unknown. It is too soon to determine the efficacy of dispersants used in this spill. The First Deepwater Spill From the beginning the petroleum (crude oil) release cause d by the Macondo well blowout was different than any prior incident (Figure 5). It was not the largest marine oil release, this statisti c belonging to the intentional spills in the Arabian Gulf during the 1991 conflict between Iraq and Kuwait (NRC 2003). It was not even the first major subsea blowout in the Gulf of Mexico, as a similar incident occurred in 50 meters of water in the Bay of Campeche, Mexico in 1979 (Jernelv 2010). That blowout, known as Ixtoc 1, was the largest accidental mari ne oil spill until the DWH. The DWH was different because it was the first blowout to occur in deep water, the well was located approximately 1500 meters (5000 ft) below the surface (Reed and Fitzgerald 2011). The depth of the water in which the Macondo wellhead was located was important because the deep sea is an ex treme environment of high pressure, low temperature, and low light, making it an extr emely difficult place to conduct research or work (Johansen 2003; NRC 2003). As a result, relatively little is known about the deepsea environment or about the behavior and impact of oil discharges emanating from these depths. Though the study of deep-water blowouts began only recently (NRC, 2003), the behavior of oil released in deep water is know n to differ from the behavior of oil released 12

PAGE 23

at the surface. The enhanced pressure at depth causes more petroleum to disperse and dissolve in the water column when compared to oil at the surface. This enhanced dissolution of oil may result in the formati on persistent deep sea oil plumes. Subsea currents also partition the oil by buoyancy re sulting in the creation of surface slicks that cover huge areas. Finally, the oil may emulsify or mix with water altering its composition prior to reaching the surface, which increases the persistence and lowers the efficacy of traditional response technology. Our knowledge on the effects of petroleu m discharge on deep-water biological communities is minimal to non-existent even compared to our knowledge of oil behavior. Oil is generally assumed to have similar eff ects on deep-sea organisms, as it would have on analogous communities on th e continental shelf (NRC 20 03), though this may not be the case. It is important to note that deep -sea organisms are special ly adapted to high pressure and low light conditions present in the deep sea and are gene rally not exposed to pollution making them potentially very sens itive to disturbances (NRC 2003). Initially, BP pointed to the fact that natural deep wa ter seeps of petroleum exist in the Gulf so impacts on deep-water organisms should be minimal (National Commission on the BP Deepwater Horizon Oil Spill and Offshore Dr illing 2011). This is refuted by most scientists due the local nature of these natu ral seeps and the small amount of petroleum that is emitted in comparison (NRC 2003; Thibodeaux 2010; Joye and MacDonald 2011). The full environmental effects of the DWH discharge remain unknown and so it remains uncertain what response technology was most ef fective in this unprecedented oil spill. 13

PAGE 24

Research Aims In the aftermath of the disaster, th e amount of conflic ting and confusing information about the use of dispersants, par ticularly in relation to subsea application, made it difficult to understand thei r effects and determine if th e use was beneficial in the Gulf of Mexico. This thesis reviews the com position, effects, and utility of oil dispersants used in marine oil spills. To accomplish this objective, I will address four broad questions, with the first two questions also addressing if there may be differences between shallow and deep-water applications. First, what are the physical and chemical effects of oil dispersants used on oil spills? Second, what are the potential biological and environmental effects of using oil disper sants on an oil spill? The third question investigates the political framew ork that surrounded the applic ation of dispersants in the DWH accident and if lessons have been learne d so that changes can be made. Finally, I will ascertain what, if any conclusions can be drawn about the use of chemical dispersants in marine spills and deep water oil drilling. 14

PAGE 25

References Freudenburg WR, Gramling R. 2011. Blowout in the gulf: The BP oil spill disaster and the future of energy in America. Cambridge, Mass.: MIT Press, p. 254. Hazen TC. 2010. Deep-sea oil plume enriches in digenous oil-degrading bacteria. Science. 330 (6001): 204-208. Jernelv A. 2010. The threats from oil spills: No w, then, and in the future. Ambio. 39 (6): 353-366. Johansen 2003. Development and verifica tion of deep-water blowout models. Mar Pollut Bull. 47 (9-12): 360-368. Johnson J, Torrice M. 2010. BP's ever-growing oil spill. Chem Eng News. 88 (24): 1524. Joye S, MacDonald I. 2010. Offshore oceanic impacts from the BP oil spill. Nat Geosci. 3 (7): 446. Kujawinski EB, Kido Soule MC, Valentine DL, Boysen AK, Longnecker K, Redmond MC. 2011. Fate of dispersants associated with the deep water horizon oil spill. Envir Sci Technol. 45 (4): 1296-1308. Lehner P, Deans B. 2010. In deep water: The an atomy of a disaster, th e fate of the gulf, and ending our oil addiction. Ne w York: The Experiment. 154 p. Literathy P, Haider S, Samhan O, Morel G. 1989. Experimental st udies on biological and chemical oxidation of dispersed oil in seawater. Water Sci Technol. 21: 845-856 Mulkins Phillips GJ, Stewart JE. 1974. Effect of four dispersants on biodegradation and growth of bacteria on crude oil. Appl Microbiol. 28(4): 547-552. National Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling. 2011. Deepwater: The gulf oil disaster and the future of offshore drilling. Washington, D.C.: Commission. National Research Council (U.S.) Committee on Effectiveness of Oil Spill Dispersants. 1989. Using oil spill dispersants on the sea. Washington DC: National Academy Press, p. 335. 15

PAGE 26

National Research Council (U .S.) Committee on Oil in the Sea: Inputs, Fates, and Effects. 2003. Oil in the s ea III: Inputs, fates, and effects. Washington DC: National Academy Press, p. 265. National Research Council (U .S.) Committee on Understandi ng Oil Spill Dispersants: Efficacy and Effects. 2005. Oil spill dispersants: Efficacy and effects. Washington, D.C.: National Academies Press, p.377. Progressive Management. 2011. US departme nt of the interior deepwater horizon response and restoration. Restori ng the gulf [CD-ROM]. United States: Progressive Management. Reed S, Fitzgerald A. 2011. In too deep: BP and the drilling race that took it down. Hoboken NJ: Bloomberg Press, p. 226. Schooner JL. 2010. The gulf oil spill. Envir Sci Technol. 44 (13): 4833. Shirbon E, Foster K. 2010. BP chief hopes cap will capture most of gulf oil (Online). United Kingdom: Rueters. Available: h ttp://uk.reuters.com/article/2010/06/06/oilspill-hayward-idUKLDE65503H20100606. [Accessed: 28 Apr 2011]. Steffy LC. 2011. Drowning in oil: BP and the reckless pursuit of profit. New York: McGraw-Hill, p. 285. Thibodeaux LJ. 2011. Marine oil fate: Knowledge gaps, basic research, and development needs. A perspective based on the deepwa ter horizon spill. Environ Eng Sci. 28 (2): 87-93. Torrice M. 2010a. Cleaning up the gulf spill. Chem Eng News. 88 (20): 36-37. Torrice M. 2010b. Dispersed oil raises concerns. Chem Eng News. 88 (21). Torrice M, Voith M. 7 May 2010. Ne w oil clean-up technique tested (Online). United States: Chemical and Engi neering News. Available: http://pubs.acs.org/cen/news/88/i1 9/8819notw1.html. [Accessed: 24 Apr 2011]. USCG. Vessel response plan program. Un ited States: United States Coast Guard. Available: https://homeport.uscg.mil/m ycg/portal/ep/channelView.do?channelId=30095&channelPage=%252Fep%252Fchannel% 252Fdefault.jsp&pageTypeId=13489. [Accessed 27 Apr 2011]. 16

PAGE 27

U.S. Environmental Protection Agency. 2003. 40 CFR 300.910 920. Available: http://www.epa.gov/oem/docs/oil/cfr/900_920.pdf. [Accessed 15 May 15, 2011] Zurer PS. 2003. Countering oil spills. Chem Eng News. 81 (14): 32-33. 17

PAGE 28

Table 1. The Series of decisions that led to the Macondo well Blowout. The decisions were made to save time and money, but ul timately led to disaster. (From: National Commission on the BP Deepwater Horiz on Oil Spill and Offshore Drilling 2011) 18

PAGE 29

Figure 1. A simplified diagram of an offshore deep water oil well, including the drilling rig and detail of the blowout prevente r. (From Progressive Mangament 2011) 19

PAGE 30

Figure 2. A simplified diagram of a deep wate r well. The well on the left was finished with the long string technique that failed at the Macondo well, while the well on the right shows the more costly, but safer tie back me thod. Note the two black cement seals for the tie back method. (From: National Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling 2011) 20

PAGE 31

Figure 3. The Deepwater Horizon drilling rig bu rning before sinking as a result of the well blowout. (From: National Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling 2011) 21

PAGE 32

Figure 4: Theoretical purpose of deep-water di spersant application (F rom: Kujawinski et al. 2011). 22

PAGE 33

Figure 5. A diagram of broken riser and sour ce of oil discharge. Dispersants were eventually directly injected into area labl ed leak after May 1. (From: Progressive Manaegement 2011). 23

PAGE 34

CHAPTER 2 PHYSICAL AND CHEMICAL EFFE CTS OF OIL DISPERSANTS 24

PAGE 35

To begin a discussion of oil dispersants it is important to understand what they are and how they work. I will address in this ch apter, the physical and chemical effects of dispersant on oil and how this affects the behavior of oil spills. I will start with a discussion of the composition of oil to facilitate an understand ing of the next section on the properties of oil, and then explain the composition, effects, and purpose of dispersants. I will finish with a discussion of how oil behaves in the marine environment and how its behavior is a ffected by dispersants. The Composition of Crude Oil To understand the behavior of oil in the envi ronment and the effects of dispersants, it is important to understand what oil is. Petroleu m is a gaseous, liquid, or solid mixture of hydrocarbons that occurs natu rally on earth (Speight 1980). The terms petroleum and crude oil are equivalent and describe the natural state (native) of oil. These terms will be used interchangeably in this thesis and alternate terms will be used for refined oils and petroleum products. Petroleum is composed primarily of hydrogen and carbon, up to 97 percent, in varying forms (Speight 1980). Petroleum may also contain other organic and metallic constituents. The molecular contents of crude oil are relatively consistent despite a variation in properties from heavy to light crudes (Spe ight 1980). Rather than a variation in molecular content, the differing properties of cr ude oils are accounted for by the variation in the structur e of the hydrocarbon constituents (NRC 2003). The structure and resulting properties of hydrocarbons affect their behavior once released into the marine environment. 25

PAGE 36

The hydrocarbons found in petroleum can be divided into three broad classes based on structure: saturates, nonsaturates and aromatics (NRC 2003). Each of these classes contains further delineations. Saturated compounds are often referred to as paraffins within the petroleum industry and as alka nes in chemistry (Speight 1980). They are straight and branched chains of hydro carbons bound by single car bon to carbon bonds. The lightest paraffins are gases while heavie r ones are highly viscous liquids at room temperature. Paraffins are non-polar, relativ ely chemically inert, and decrease in abundance with increasing molecular weight of oil. Napthenes are the second type of saturate compounds and are also known as cycloparaffins or cycloalkanes (Speight 1980). These are saturated hydrocarbons containing ring structures Napthenes are non-polar and generally constitute a significant fraction of petroleum (NRC 2003). Unsaturated hydrocarbons occur in severa l different forms. Straight chain and branched molecules containing one carbon-car bon double bond are referred to as olefins and molecules with two double bonds are dienes, three double bonds are trienes and so forth, but outside the petrol eum industry these molecules are referred to as alkenes (Speight 1980). They are not found in large concentrations in crude oil because the double bond is more reactive than a singl e bond but are more common in refined petroleum products (NRC 2003). Olefins are generally non-polar but may combine with other elements that alter their properties. Aromatics are another class of unsaturated hydrocarbons, such as benzene and napthalene (Speight 1980). The resonance structure of these cyclic aromatics makes them stable and thus they occu r in relatively high quantities, between 10 and 50 percent in cr ude oil (NRC 2003). Aromatics are the most toxic components of crude and can be separa ted into further categories. BTEX molecules 26

PAGE 37

are a group of the most volatile components of crude and are si ngle ring structures including benzene. Polycyclic aromatic hydrocarbons (PAHs) c ontain at least two aromatic rings (NRC 2003). PAHs are ofte n grouped with other polycyclic compounds containing non-hydrocarbons and considered the most e nvironmentally hazardous components of crude oil. The primary non-hydrocarbon organic compounds contain sulfur, oxygen, or nitrogen (Speight 1980). These elements may be pres ent in amounts from 0.1 up to 6.0 percent of the elemental composition of petroleum (NRC 2003). Sulfur is presen t in all crude oils and generally occurs at higher concentrati ons in higher density oils. Oxygen generally occurs at low levels within petroleum but exposure to the atmosphere can increase the oxygen content of petroleum. Nitrogen is the final organic constituent that is commonly found in petroleum. The effects of all organi c constituents on the properties of oil are relatively mild but their presence may affect polarity and volatility. The metallic constituents of petroleum generally occur in very small amounts. Metallic compounds too can be divi ded into two classes, the first contains zinc, titanium, calcium, and magnesium and occur in petr oleum as organo-metallic soaps (Speight 1980). That is, the metallic compounds are polar and thus miscible with water but also contain non-polar hydrocarbon constituents. This property, known as bipolarity, gives these compounds surface-active properties, wh ich allow them to aggregate at the interface between oil and water and chemically stabilize the oil and water. The remaining metals that occur in petroleum at appreciab le amounts are vanadium, copper, and nickel (Speight 1980). These metals are generall y present as oil soluble compounds, in 27

PAGE 38

elemental and ionic (chemical salt) form and can complex with certain organic molecules. Physical and Chemical Properties of Petroleum The physical and chemical properties of petroleum define how an oil or oil product will behave once it has entered into the environment. Properties that affect oil behavior include density, polarity, solubilit y, viscosity, interfacial and surface tension, and volatility. These properties along with othe r environmental variables can also govern the successful application of cleanup technologies. Thes e properties are important because dispersants seek to chemically change them to alter the beha vior of oil. Despite the range of contents most crude oils behave similarly with certain properties varying by fraction (Speight 1980). Density is a measure of mass per unit volume. It is a critical property of petroleum in understanding oil spills beca use it is the primary property affecting the distribution of oil in the water. Density is measured in grams per cubic centimeter or as specific gravity (Speight 1980). Most petroleum ranges from 0.7 to 0.99 g/cm3 (specific gravity 0.8 to 1.0) (Speight 1980; NRC 2003). The density of seawater is 1.03g/cm3 (specific gravity 1.028) and therefore, most petr oleum fractions will float in the marine environment. Density is determined by the relative amounts of heavy and light constituents (Rogowska and Namiesnik 2010). For instance, aromatic components are dense, whereas paraffins are relatively light. Density can be used to assess persistence because the lightest compounds are removed from the environmen t faster than heavier ones (NRC 2003). 28

PAGE 39

Polarity is a measure of a molecules electric dipole. Molecules of similar polarity interact favorably, chemically stabilizing on e another, while molecules of differing polarity repel. Petroleum is generally a s ubstance of low polarity (NRC 2003). This is because pure hydrocarbons, the main constituen ts of oil, are non-polar. Therefore, oil and water typically do not interact favorably. Oil is not perfectly non-polar because it contains other atoms, organic and metallic, that affect its polarity. The presence of certain metallic constituents and organic molecules can increase the solubility of oil (Speight 1980). Examples of polar compounds in petroleum are resins, which are small compounds and asphaltenes, larger compounds that can dramatically increase the viscosity of oil (NRC 2003). Oi l dispersant surfactants contain polar and non-polar sections that are capable of favorably inte racting with water and oil simultaneously. Solubility is the measure of the amount of oil that will dissolve into the water column (NRC 2003). The dissolution of oil is driven by chemically favorable interactions between water molecules and oil constituents. Petroleum solubility is generally very low because oil and water do not favorably intera ct due to their differing polarity. Petroleum solubility in the marine environment is us ually less than 100 parts per million. Because this amount is so low, dissolution plays a minor role in the natural be havior of oil in the marine environment. Oil dispersants act betw een the boundary of oil a nd water to create a more favorable chemical interaction between the liquids (NRC 1989, 2005). This surfactant action breaks oil into small drops that dissolve into the water column, effectively enhancing the solubility of oil. Th is property is important to consider because water-soluble fractions of oil are often the most toxic to marine life (NRC 2003). 29

PAGE 40

Removing oil from the surface by dissolution is the purpose of dispersants, but in doing so they enhance toxicity to marine life in the water column. While oil and water are immiscible, under cert ain conditions they can be forced to mix (NRC, 2003; Rogowska and Namiesnik 2010) Mixtures of two immiscible liquids such as oil and water are referred to as emul sions. In the case of water and oil two types of emulsions can form: oil in water emulsions and water in oil emulsions. Oil in water emulsions occur when droplets of chemically unstable oil become entrained in the water due to neutral buoyancy. Oil in water emulsi ons can be referred to as physically or naturally dispersed oil and are beneficial in increasing the rate of oil fate. Water in oil emulsions form when water droplets become entrained in a petroleum plume or slick. Water in oil emulsions can stabilize petrol eum, increasing viscos ity and persistence making water in oil emulsion prevention a conc ern to responders. Water in oil emulsions are often referred to as tar balls. Interfacial tension is the measure of the forces ac ting at the boundary of two immiscible liquids such as oil and water (Spe ight 1980). This property is directly related to polarity and solubility. The tendency for two liquids to mix is affected by their polarity difference. In immiscible liqui ds this difference creates inte rfacial tension at the boundary of the two liquids. The unfavorable forces at the boundary cause the liquids to aggregate amongst themselves decreasing solubility. Interfacial tension increases directly proportional to temperature and molecular we ight. Dissolved gases in the immiscible liquids decrease this property. A rise in pH will also decrease interfacial tension. Interfacial tension is important in relation to dispersants because it is lowered by the bipolar nature of surfactants allowing for mo re favorable oil and water interactions. 30

PAGE 41

Surface tension is the measure of the for ces acting at the boundary of two phases of matter and is closely rela ted to interfacial tension (S peight 1980). Surface tension comes into play, for example, in the interac tions between oil and air, such as at the surface of a slick and in between oil and solids, such as sand or sediment. Surface tension is related to viscosity and increases in a directly pr oportional relationship. Increases in molecular weight raise this property in petroleum. Elevated temperatures and the presence of dissolved gases lower this pr operty of oil. Surface tension is important because it can affect the tendency of a liquid to flow in the e nvironment (Rogowska and Namiesnik 2010). Viscosity is the measure of a fluids inte rnal resistance to flow as a result of cohesion between the molecules of that fluid (Speight 1980). This is the most important physical characteristic governing the motion of petroleum and varies by oil composition and environmental characteristics. The viscosity of petroleum is affected by the proportion of heavy and light fractions (NRC 2003). Petroleum that contains primarily light fractions like saturates flows readily, but heavier fractions containing compounds, such as asphaltenes, are resistant to flow The viscosity of oil is highly temperature dependent and increases in cold er temperatures, therefore oil spilled in colder climates will spread more slowly (Speight 1980). Oil viscosity also changes over time due to weathering (NRC 2003). As lighter fractions evaporate the remaining compounds are more resistant to flow and the weathering process slows. The remaining heavy compounds are persistent and are more likely to form emulsions. Early application of dispersants is important because they are most effective on the lighter fractions of oil and can help prevent the formation of viscous em ulsions and tar balls. Viscosity of petroleum 31

PAGE 42

also increases under pressure (Speight 1980) This contributes to altered transport behavior of petroleum in deep-sea releases. Volatility is the tendency of a liquid to evaporate (NRC 2003). The majority of crude oil components are liquids under norma l conditions, but these liquid components vary in volatility (Spe ight 1980). The lighter petroleum fractions evaporate more readily. Gasoline will evaporate completely in hour s, whereas no more than 10% of heavy distillates will evaporate. Volatility increas es with higher temperature. Volatility may also greatly vary between refined fractions of oil (NRC 2003). Volatility is important because it removes oil form the immediate environment and because the loss of light fractions change the characteristics of oil. The remaining heavy fractions are more likely to form water in oil emulsions. Dispersant s must be applied qu ickly to prevent the formation of these emulsions. Oil Fate Oil fate is the collection of processes that oil undergoes once it has entered the marine environment resulting in change of composition, transportation and eventually, incorporation into the surface environment of the earth (NRC 2003). Oil fate can be divided in two categories: transport and weathering. Transport is the series of processes that physically move the oil and weathering is the series of physical and chemical changes that oil undergoes. Weathering proc esses include, both biotic and abiotic reactions (Thibodeaux 2011). The collective fate processes are very complex and interdependent; they vary by oil composition, source of spi ll, and local environmental characteristics (Wang et al. 1999). All released oil is subject to the same fate process 32

PAGE 43

upon reaching the surface. Oil released in deep water is subject to further processes that alter its fate and environmental effects. Transport Once oil has been released into the marine environment, its motion is governed by a broad suite of transport processes (Figure 1). These include phase and fraction distribution, dispersion, dissolution, spreadin g, advection, Langmuir circulation, sedimentation, over washing, and st randing (NRC 2003 1989 2005: Rogowska and Namiesnik 2010). Surface and shallow water oil releases have been the most heavily studied because deep water drilling is a relatively new process. Most transport processes are shared regardless of where the oil is rele ased because most oil will eventually reach and reside on the surface. Deep-water spills differ in the behavior of phase distribution, the amount of dissolution that occurs, and in subsea transport and emulsification that affect the distribution and behavior of oil once it has reached the surface (NRC 2003; Thibodeaux 2011). Phase and fraction distribution are among the first processes to affect the distribution of oil (Gossen 2006; Thibodeaux 20 11). Petroleum contains different phases from gases to liquids to solids. Once oil has been released each phase behaves differently. Gases will dissolve in the water or diffuse into the atmosphere. In the case of a deepwater release, most gases will dissolve and re main at depth (Kujawinski et al. 2011; NRC 2003). The liquid components of oil will par tition by fraction (Gossen 2006). Up to sixty percent of the oil will form a liquid film on th e surface. Another thirty percent of oil with higher density and solubility will reside in th e near surface layer. Ten percent or less will 33

PAGE 44

be dissolved or become entrained and dist ributed throughout the water column as a physically dispersed oil in water emulsion. Finally, the heaviest fractions and solids will sink and interact with the sediment. Phase a nd fraction distribution occurs quickly after release and the remaining processes govern the subsequent behavior. Advection and spreading are the primary processes affecting horizontal transport of oil at the surface (Rogowska and Namies nik 2010). Advection is the bulk motion of a fluid relative to its surroundi ngs. It can be driven by wind or current or even buoyancy in the case of deep water plumes. Advection does not increase the area of an oil slick and therefore, advection can be considered horizontal movement without spreading. Spreading is the horizontal growth or increase in surface area of an oil slick. As a slicks surface area increases its thic kness decreases. This is cau sed by the interplay between gravity, viscosity and surface tension. Because spreading is related to the viscosity and surface tension of oil, it is affected by the ambient temperature. Spreading is thus, expected to be affected by the type oil and the ambient temperature. Spreading is important because responses such as burni ng are less effective on thin oil slicks. Dispersion is the mixing of oil and water driven by external mechanical energy (Rogowska and Namiesnik 2010). Natural (physical) dispersion is the formation of oil in water emulsions mentioned above. This process is chemically unfavorable because the oil and water are not miscible; it relies inst ead on energy input from waves, wind and current. Dispersion forms many small droplets of oil, which are distributed in the water column. If sufficiently buoyant the droplets w ill rise again, but if not they remain entrained in the water column. Dispersed o il is not chemically stable and will reaggregate if given the opportunity. Natural di spersion is a relatively minor process, which 34

PAGE 45

oil dispersants seek to enhance by incr easing the solubility of oil (NRC 1989). The dynamics of dispersion are different in a deep -water spill because th e oil must travel a long distance in the water column to the surface. The added pressure decreases the buoyancy and results in more dispersed oil becoming entrained in the water column (NRC 2003). Dissolution is the chemica lly driven process of oil spreading throughout the water column (NRC 2003). This process differs from dispersion because it is driven by chemical energy rather than mechanical. In dissolution, solubl e components of oil interact favorably with water and mix into the water column. Dissolution is a minor process because most fractions of oil are not soluble in water, affecting no more than ten percent of released oil. Dissolution can be enhanced by oxidation of oil by light or microbes, which results in products that ar e water soluble (Page et al. 2000; Rogowska and Namiesnik 2010). Dissolution is also enhanced in deep-wat er spills by tw o processes: the oil is under greater pressure at depth e nhancing solubility and is exposed to more water in its travel to the surface (NRC 1989). Langmuir circulation, sedimentation, a nd over-washing are further mechanisms that affect the vertical di stribution of oil in the water column (NRC 2003). Langmuir circulation is a complex wind driven process that results in water transport. This process can move oil, but it is poor ly understood in this contex t. Langmuir circulation may enhance the efficacy of dispersants, which de pend on energy input to facilitate dispersion. Sedimentation is the adherence of oil to particulate matter in the water column via adsorption or absorption. The particulate matter eventually settles out of the water column where the oil then interacts with the benthos. The effects of dispersants on 35

PAGE 46

sedimentation are poorly understood (NRC 2005) A variety of processes affect the sedimentation of oil in the presence of particulate matter including, the size of agglomerations of oil and the amount of se diment (Guyomarch et al. 2002). Generally, it assumed that dispersants enhance sediment ation by spreading the oil into the water column to enhance the oils interaction w ith sediment (NRC 2005). Over-washing is the process of water being driven over floating oil by wave s (NRC 2003). This process can enhance dispersion and dissolution, but is generally considered to be a temporary redistribution of the oil. Ove r-washing can increase the disper sion of oil in the presence of dispersants. Oil stranding is the process by which oil interacts with sediments on a shoreline (NRC 2003). Stranding results in the trapping of oil and a lteration of other transport mechanisms. The type of sediment found on a beach influences the behavior of the oil. (Defeo et al. 2009; Owens et al. 2008; NRC 200 3). Course sediment is more permeable to oil; this makes oil highly persistent, when it st rands on these types of beaches. Oil that is trapped in sediments is a problem because it is at risk of being moved on and offshore by wave action and settling further into the interstitial space. Weathering Weathering is the host of physical and chemical changes oil undergoes in the marine environment (Figure 2). It is important to keep in mind that the weathering of oil is affected by the transport pr ocesses mentioned above and th at it in turn can affect transport. The first mechanism that I will consider is evaporat ion, or volatilization. Evaporation is a very important weathering process because it results in the removal of 36

PAGE 47

oil from the local environment and can aff ect large amounts of oil (NRC 2003). A spill of light oil may lose as much as 75% of its init ial volume within a few days, whereas a spill of heavy oil may not lose only 10% during this same time period. Regardless of the amount of oil removed evaporation changes the composition of a spill. Evaporation removes the lightest components of the spill (often early in th e timeline) leaving heavier, persistent, and viscous components in the water. This change in composition makes the oil more likely to emulsify. Dispersants and other mitigation techniques have lower efficacy on emulsified oil necessitating rapid response. Oxidation reactions are important in breaking down petroleum in the sea. Oxidation reactions change the structure of the hydrocarbons in oil by oxidizing them to alcohols, ketones, and organic acids (NRC 2003) In photo-oxidation oil reacts with light to form the aforementioned products. The pr oducts of photo-oxida tion are more water soluble than reactants and thus, photo-oxidation tends to increase the process of dissolution. The products however, are mo re toxic to marine life (Rogowska and Namiesnik 2010). These light-dependent reacti ons only occur close to the surface and are dependent on the amount of UV radiation an d are therefore, most prevalent at low latitudes. Oil dispersants can lower the impact of photoo xidation on weathering oil because once the oil dissolves into the water column less light reaches it (NRC 2003, 2005). The second type of oxidation reaction to occur with oil is caused by microbial degradation. Bacteria, fungi, and heterotrophic plankton carry out aerobic reactions to break down the oil. Other anaerobic bacter ia that reduce sulfate and iron may also degrade the oil. In microbial oxidation, oil co mponents are degraded preferentially, with 37

PAGE 48

alkanes being broken down first, followed by aromatics, and finally polar compounds containing sulfur and oxygen (Price 1993; N RC 2003). This hierarchy may result from the difficulty microbes with lipophilic cell membranes have interacting with polar compounds (NRC 2003, 2005). These reactions use oxygen from the surrounding environment and are sensitive to oxygen depletion, which can slow the rate at which they occur (Thibodeaux 2011). Oxygen depletion is also at problem for surrounding marine fauna and flora. Oil dispersants are generally assumed to increase the rate of microbial oxidation by breaking oil slicks and plumes into small droplets, thereby increasing the surface area available to the bacteria (N RC 2005). The exact effects of this process remain controversial. The final weathering process considered he re is the formation of oil and water emulsions. Emulsions are formed when water droplets become entrained in the oil and are often referred to as mousse due to their appearance (NRC 2003; Rogowska and Namiesnik 2010). Water in oil emulsions usua lly contains 20 to 80 percent water by volume. The incorporation of water affects th e oil by increasing the viscosity and surface tension and expanding its volume, resulting in a stable state. Emulsions generally float, though emulsions of heavy fractions ma y sink (NRC 2003). The increased surface tension and viscosity make em ulsions unlikely to spread and highly pers istent. Their persistence and tendency to fl oat puts emulsified oil at hi gh risk for stranding. Stranded emulsions are often referred to as tar balls (Figure 3). O il is more likely to form emulsions under low temperatures due to increased viscosity. Dispersants have less effect on emulsified oil, and must be applied quick ly after oil is spilled. If dispersants are 38

PAGE 49

applied in time they can prevent emulsions a nd thus, decrease the pe rsistence of the oil and its likelihood of washing up on shore. Differences in the Fate of Deep Water Spills Subsurface oil releases differ from surface spills in a variety of ways. Subsurface spills are typically caused by blowouts, shipwrecks and dama ged pipelines. Subsurface spills can be divided into two categories by dept h: shallow releases that occur at less than 200 m and deep releases that occur at greater than 200 m depths (NRC 2003). The distinction is made because pressure caused by deeper spills changes the behavior of the oil enough to alter fate processes. Transport and dissolution processes in subsurface spills are different than for surface spills (NRC 2003). When oil is released underwater, it is transported by currents more readily which results in a surface slick that behaves differently than a slick beginning at the surf ace (Figure 4). The second difference is that dissolution of oil is greatly enhanced in subs urface spills because th e oil is exposed to more water over longer periods of time under increased pressure. Shallow water subsurface spills begin in what is known as a jet phase, provided the oil is under pressu re (NRC 2003). Oil is released from the source under pressure, as an expanding cone of high velocity oil, which dissipates quickly. Once oil has traveled beyond the jet, it enters plume phase. Oil in pl ume phase begins to uniformly rise to the surface driven by the buoyancy of contained gases and liquid. The more buoyant the oil, the faster the plume rises. The buoyancy of the plume acts to pull non-buoyant fractions of oil, gas and water with it, though some of these fractions may not be buoyant. Plumes pull large quantities of adjacent seawater to the surface with them. Undersea currents also 39

PAGE 50

act to spread the oil as it rises and partition it by buoya ncy. More buoyant fractions undergo less horizontal transport because they rise faster and are subject to horizontal transport for less time. This horizontal tran sport acts to fractiona te the slick by buoyancy (affected by the density and size of oil drople t) with the most buoyant fractions surfacing closest to the source. Once at the surface the oil forms the slick, which expands due to radial outflow and gravitational forces. The behavior of subsurface releases d eeper than 200 m is more complex than those under this depth (Johansen 2003; N RC 2003). The jet phase is not appreciably different to the analogous shallow subsurface release phase mentioned above. Plume behavior is more complex at depth. Deep pl umes entrain more dense water at depth and can form suspensions that are neutrally buoyant and cease to rise. Eventually lighter and heavier components may separate and allow portions of the plume to rise again until it entrains more dense water in a repetitive process known as peeling. Plumes will eventually separate and cease to rise cohesively at their term inal phase, instead allowing droplets to rise under their own buoyancy. Cross current effects on deep plumes are similar to those described above except that th ey can have an even greater effect, as the oil takes longer to reach the surface. Gas behavior at depth differs drastically from behavior in shallow water (Johansen 2003) Gas compression and thus buoyancy is not affected linearly at depth. At deeper than 300 m natural gases are likely to mix with water and create solids known as hydrates. Hydrates are less buoyant than gases and thus slow the rise of plumes. Eventually the hydrates will decompose and most gas released at depth will dissolve into the water column. 40

PAGE 51

The majority of oil released in deep water will eventually reach the surface (Johansen 2003; NRC 2003; Thibodeaux 2010). Under ideal conditio ns no more than 10% of liquid oil will dissolve in the water column. The surface slick formed will be thin and diffuse due to the fractionation and unders ea transport of the oil. Most gases will have dissolved before reaching the surface. Deep oil releases become entrained with water while rising and are like ly to contain significant amou nts of emulsified oil by the time they reach the surface. Upon reaching the surface, oil released in deep water is subject to the same fate processes as other oils released at the surface (NRC 2003; Thibodeaux 2011). The major differences lie in the fractionation of the surface slick and the amount of oil and gas that dissolves or remain in the de ep water column. Deep oil is also subject to sedimentation (Thibodeaux 2011). Oil that remains in or settles out of the water column is subject to biodegradation, bu t oil that remains at depth may be more persistent than surface oil because it is not subject to the full spectrum of fates, notable exceptions include evaporation and photo-oxidation, which have marked effects on surface oil. Another important difference between shallo w and deep water releases is the risk of oil emulsifying before it reaches the surface (Daling et al. 2003, NRC 2003). Because oil is exposed to more water during transport to the surface it is not only subject to more dissolution, but also to entrai nment. Oil droplets may become entrained in the water, not chemically stable but neutrally buoyant. Mo re importantly, water may become entrained in the oil, causing water in oil emulsion be fore the oil reaches the surface. If oil emulsifies before reaching the surface it sign ificantly decreases the efficacy of response and increases the longevity of the oil in the environment. 41

PAGE 52

Oil Dispersants Dispersants are chemical formulations designed to be a tool in combating oil spills by changing the behavior of the oil. Oil and water are by nature largely immiscible and most liquid fractions of crude oil ar e buoyant in the marine environment (NRC 2003). As a result of these properties, oil spill s create large surface oil slicks. Oil slicks are unsightly, harmful to organisms that depe nd on surface waters, and at risk of being transported onto shore. Oil disp ersants were designed to facilitate the dissolution of oil off the surface and into the water column (NRC 1989, 2003). Dispersants do not change the composition of the oil or remove it from the environment, but rather act in a similar manner to dish soap breaking the oil into small droplets and acting at the oil-water interface to allow the oil to dissolve into the water column (Figures 5 and 6). Dispersants are dependent on external energy in the form of waves and mixing to facilitate breaking the oil into droplets. In theory, this has ma ny benefits. When applied to the leading edge of an oil slick, the dispersant can dissolve the slick and prev ent it from traveling towards the shoreline. Removing oil from the surface can lower the exposure of surface dwellers, like sea turtles and marine mammals to oil. Oi l dispersants may also have benefits that are harder to see, i.e. increasing the surface area of the spilled oil by breaking it into droplets allowing microorganisms greater access to the oil and speeding its consumption. All oil dispersant products are based on the same gene ral principles, though their specific compositions vary greatly. The key ingr edients in any dispersant are one or more surfactants and a solvent (NRC, 1989, 2005). Surfactants, al so known as surface-active agents or detergents, are the active ingredie nts of dispersant. Th ese molecules are bi42

PAGE 53

polar, meaning that they contain water soluble portions (hydro philic) and oil soluble parts (lipophilic or hydrophobic). Surfactants are often defined by their hydrophilic-lipophilic balance (HLB), which is measured on a scale from 1 to 20 with 1 being the most oil soluble and 20 being the most water soluble (Figure 6). Adjusting the HLB can increase the ability of a surfactant to stabilize oil. A highly oil soluble surfactant, low on the HLB scale, will stabilize a solution of oil and wa ter that is predominantly oil. Similarly, a surfactant high on the HLB scale will stabilize oil and water solution composed primarily of water. Surfactants are also affected by salinity because lipophili c portions tend to be less soluble in higher salinity. This is of concern for dispersants used in the marine environment, which need to be more hydrophilic to avoid being salted out. Surfactants are generall y grouped by their molecular charge type. Molecular charge determines polarity and influences HL B. Surfactants may be either ionic or nonionic (NRC 1989). Ionic surfactants can be further delineated into: i ) anionic molecules that have a negative molecular charge and include compounds like sodium dioctyl sulfosuccinate; ii) cationic surfactants contain a pos itive charge, for instance ammonium salts, but the toxicity of these molecules gene rally precludes their us e in dispersants; and iii ) zwitterionic molecules, also known as amphoteric, which contain both a positive and negative charge and are not used in disper sants. Non-ionic molecules are the most common surfactants used in oil dispersants, i.e. ethoxylated sorbitan mono-oleate. The concentration of surfactants is an important property in determining the efficacy of a dispersant alongside the HLB (NRC 1989). Surfactants dissolve oil by creating micelles, ordered groupings of molecules with th e hydrophilic portions of the surfactants facing outward and th e lipophilic portions interacting with oil in the center. In 43

PAGE 54

this way, the hydrophobic oil does not come in contact with th e surrounding water. Generally, higher concentrations of surfactant are more effective, to a point. When surfactant concentration becomes too high, a point called the critical micelle concentration (CMC) is reached (Chatterjee 2001). At this point surfactant molecules become ordered into micelles without oil in their center such that their hydrophilic portions face out and the lipophil ic portions face in preventing interactions with oil. This property makes the dispersant to oil applicat ion ratio important be cause dispersants are most effective when applied at a specific concentration. Solvents are the second major co mponent of dispersants (NRC 1989, 2005). Solvents reduce the viscosity of surfactants and facilitate dispersal by creating a homogeneous mixture and increasin g the solubility of surfactant s in the oil. Solvents are usually water, water miscible compounds or hydrocarbons. Solvents can be important in determining the efficacy of dispersants in a particular environment as well. Aqueous solvents are less effective in arctic conditions because they tend to freeze in application nozzles (NRC 1989). Most dispersant solven ts are hydrocarbon based (Exxon Valdez Oil Spill Trustee Council 1994; Zurer 2003). This fact adds to the controversy behind dispersant use because using dispersant s adds more hydrocarbons by weight to a hydrocarbon spill, which seems counterproductiv e. This problem led to a discontinuation in the use of hydrocarbon based solvents on s horelines. The Corexit dispersants used in the DWH contain hydro-treated light petrol eum distillates, such as kerosene MSDS (Nalco 2005, 2008; Progressive Management 2011b). Early oil dispersant formulations were highly toxic and had low efficacy in the environment (NRC 1989). These dispersant s were based upon ship engine room 44

PAGE 55

degreasers and were toxic due to their surfactants and solvents, which were based on toxic aromatic hydrocarbons. Modern dispersants designed specifically for use in the environment have changed to less toxic surfactants and removed aromatic hydrocarbons from their solvents. Dispersants have also been engineered to enhance efficacy. Early dispersants contained surfactants with the op timum HLB for their intended use. Modern dispersants are sophisticated containing se veral surfactants of different HLB which average to the ideal combined HLB increas ing their efficacy in the real world. The most common oil spill dispersants ar e the Corexit line of products produced by Nalco (NRC 2005). Nalcos two products, Corexit 9500A and Corexit 9527A, were exclusively used in the Gulf of Mexico responding to the Deep water Horizon (Torrice 2010). Deepwater application was limited to the newer product, 9500A. Both Corexit products use a mix of surfactants includi ng the anionic surfact ant dioctyl sodium sulfosuccinate (DOSS) and the non-ionic surfactants tween 80, tween 85 and span 80 (Progressive Management 2011b). Both products contain hydro-treated light petroleum distillates, propylene glycol and 2-propanol as solvents Corexit 9527A also contains 2butoxy-ethanol, which has adverse health effect s in humans and as a result the use of 9500A was preferred in this oil spill. The sp ecific ingredients of both Corexit products are listed in Table 1, relative amounts are proprietary (NRC 2005). Chemical Structures for select components can be found in Figure 7. To secure product approval, EPA require s the manufacturers of dispersants to submit standardized toxicity and efficacy data (NRC 2005; U.S. Environmental Protection Agency 2010a). Efficacy data is base d on a swirling flask test (Blondina et al. 1997; U.S. Environmental Protection Agency 2003). In this test, oil and dispersant are 45

PAGE 56

mixed at a 10:1 ratio and mechanically swirle d in a modified Erlenmeyer flask for 20 minutes (Figure 8). Afterwards the solution is allowed to sit for 10 minutes, a sample is taken from the bottom of the flask, and an alyzed for oil content with UV absorption spectroscopy using three visible wavelengths 340, 370, and 400 nm. Results are displayed in terms of percent efficacy in which the calculated mean value for total mass of dispersed oil is divided by the total ma ss of oil in the system multiplied by 100: %EFF=(Cmean/Ctotal)*100 (U.S. Environmental Protecti on Agency 2003). Each dispersant is tested on two standardized crude oils, a Prudhoe Bay Cr ude Oil and a South Louisiana Crude Oil. Effectiveness is provided for each type of crude and an average is given (Table 2). The data for each registered di spersant can be found in the EPA National Contingency Plan (NCP) Product Schedule (U.S. Environmental Protection Agency 2010ab). The dispersants found on the NCP product schedule vary from 50 percent up to 70 percent average efficacy (Appendix A) (U.S Environmental Protection Agency 2011). Some dispersants are relatively consistent be tween the two types of oil, but others are significantly more effective on one type of oil versus the othe r. An example is the NEOS AB3000 dispersant, which is less than 20% e ffective on Prudhoe Bay Crude Oil but 90% effective on South Louisiana crude. The Corexit dispersants are on th e low end of the list in terms of average efficacy. Corexit EC9500A has a 45.30 percent efficacy on Prudhoe Bay oil and a 54.70 percent efficacy on Sout h Louisiana oil. Corexit EC9527A has a Prudhoe Bay Crude efficacy of 37.40 and a Sout h Louisiana Crude efficacy of 63.40 percent. The averages for these two disper sants are relatively low, 50 and 50.4 percent respectively but their percent efficacy per indivi dual oil is about averag e. This is because 46

PAGE 57

most dispersants with a high average efficacy are characterized by high efficacy in one oil and low for the other. The motive behind the choice by BP to use the Corexit products in response to the Deepwater Horizon spill is unknown (Voith 2010a). Other dispersants with higher efficacy on Louisiana crude oil exist, but ma y not have been available in sufficiently large quantities (Progressive Manageme nt 2011b). Corexit produ cts are the most common oil dispersants and have been tested in natural environments many times (NRC 2005). Furthermore, the information that can be drawn from EPA test data is limited. EPA tests do not account for varied applicat ion ratio, variant temperatures, and the weathering of crude oil in th e marine environment. The choice between the two Corexit products is more clear (NRC 2005). Core xit 9500A was favored and used more frequently than 9527A because it had less adverse human health effects and is more effective on weathered oil. Theoretically, dispersants could be ma tched to the type of oil and the environmental conditions to ensure optimum efficacy (NRC 1989, 2005). In recent years, more effort has been put into field testing dispersants and accurately monitoring the effects of dispersants on oil sp ills. As of yet, it is not possible to accurately choose one product over another based on the type of oil spilled. Oil type and particularly the relative fractions present in oil are important in determining toxicity and the relative properties of that oil and its behavior in the environment. 47

PAGE 58

Effects of Dispersants on Deep water Releases Upon reaching the surface, oil released at depth can theoretically be treated with any traditional response method. The tendency for oil to be transpor ted and emulsify on the way to the surface can decrease the efficacy of any response technique on the surface. BP chose to add dispersants at depth, some thing that had not been done before. Their justification was that it could enhance the e fficacy of the dispersant and overall response by dispersing oil before the formation of em ulsions and possibly se questering the oil in the deep, where it would not impact surface fauna and shorelines (Torrice and Voith 2010). BP used ROVs to directly inject disper sant into the oil being released from the wellhead, relying on the turbulent jet phase to mix the dispersant with the oil. Because the Macondo well blowout was the first instance of deep water dispersant application, the early studies of this event provide the best evidence as to the eff ects of dispersants at depth. At the time of this writing, few publicat ions had been released, making conjecture beyond these early results purely speculation. Early studies showed that oil releas ed from the Deepwater Horizon blowout aggregated in massive deep sea plumes (Thibodeaux 2011). The size of the plumes, up to ten miles long, three miles wide, and 120 m (400 ft) in depth, led scientists to conclude that they could not be comprised solely of oil, but rather of oil droplets in greater quantities of water. The presence of oil droplets is promising suggesting that the dispersants had served their intended purpose. Another study showed that the surfactant Dioctly Sodium Sulfosuccinate (DOSS), a component of both Corexit products, had indeed been associated with underwater pl umes (Kujawinski et al. 2011). This study concluded that it was too early to determine if the dispersants had served their intended 48

PAGE 59

role, or if the surfact ants had instead been associated with other components of the deep water plume (such as dissolved methane and methane hydrates) and thereby, been rendered unavailable to the oil. Regardless of whether or not the dispersants served their intended role, Kujawinski et al. ( 2011) tracked the underwater plume and dispersant component for 64 days, 300 km from the wellhead. This suggests that deep water plum es were both long lived and subject to deep water transport. The concentrations found in this study suggest that the dilution of the plume was due to transpor t rather than micr obial oxidation. This result is in line with the Thibodeaux et al. (2011) paper, which suggests that subsea plumes weather slowly and that the ultimate fate is unknown. The results of these early studies are consistent with the speculation by other scientists involved (Joye a nd MacDonald 2010). The deep water plumes are shown to travel within the water column with ocean currents. The lack of photo-oxidation and evaporation at depth, make these plumes re latively long lived b ecause they weather slowly. Without these two mechanisms at play the ultimate fate of the oil is dependent on transport through dilution and advection, sedimentation, and microbial oxidation. The water column distributio n of the subsea plumes may be pa rticularly import ant in the long term impact of oil according to a modeling study (Maltrud et al. 2010), which suggested that oil remaining deeper than 800 m would be subject to relatively little transport and remain in the Gulf of Mexico. Any oil th at traveled further towards the surface but remained in the water column would be su bject to significant transport, eventually reaching the Atlantic within six months. The co sts and benefits of either situation remain unclear. Oil that remains in the deep may ha ve more narrow impacts, but those impacts 49

PAGE 60

could be devastating to the deep sea harboring unknown repe rcussions. Oil closer to the surface and subject to more transport could ha ve far reaching impacts, but could also be diluted and subject to other fate processes th at prevent significant environmental harm far from the release. Conclusion The purpose of this chapter has been to assess the chemical and physical effects of dispersants on marine oil spills and to consider the differences that may exist in applying dispersants to deep water. Oil itself is a complex mixt ure of hydrocarbons and other elements, which has some basic principle properties. The transport and compositional changes that oil undergoes in the environment, referred to as fate, are highly complex (Rogowska and Namiesnik 2010; NRC 2003). These processes ar e affected by specificities of the local environment including temper ature, currents, upwelling, wave energy, light, and biota present. These in tu rn are affected by lo cal geography, latitude, climate, the time of year, and short term w eather conditions. The colle ctive fate processes through time remove spilled oil from local environments and work to break down the oil and incorporate it into the ecosystem. Oil dispersants are a tool designed to alte r fate processes. Oil dispersants dissolve oil into the water column by lowering the interfacial tension between water and oil caused by the difference in their polarity. Removing oil from th e surface can limit transport towards sensitive coastal habitats, lower the exposure of su rface dwellers to the oil, enhance biodegradation, and prevent emulsi fication that increases the longevity of oil in the environment. While dispersants have drawbacks, such as adding more chemical 50

PAGE 61

mass to an oil spill, they are a useful t ool for the response te am when used in a conventional manner (Daling et al. 2002). Deep-water spills differ from conventional sp ills in several ways. Oil released at depth increases the amount of oil dissolved in the water column, is subject to transport and partitioning by subsea current, and oil ma y emulsify before reaching the surface (NRC 2003). The resulting oil slick at the surface is likely to be diffuse and to contain emulsified oil, both of which decrease the efficacy of response countermeasures including dispersants. Applyi ng dispersants at depth would theoretically increase the efficacy of dispersant application by allowing more time and water exposure for dispersion. Additionally, it woul d reduce the amount of dispersant needed by reaching the oil before emulsion formation. Dispersing the o il in the water column could also allow it to become sequestered in the deep sea, wher e it could not reach other sensitive habitats. The problems of sequestering oil in the deep sea are i ) the effects of oil on deep sea habitats, ii ) reducing the effects of important fa te processes including photo-oxidation and evaporation, and iii ) extending the time oil remains in the environment. Early studies indicate that dispersants may have been effective in the creation of large deep-sea oil plumes near the Mac ondo wellhead (Kujawinksi et al. 2011; Thibodeaux 2010). These plumes have so far been shown to be persiste nt over a period of months and are subject to subs ea transport. Dispersing this oil may have prevented more oil from washing ashore, causing devastati ng environmental and economic side effects (Schooner 2010). The oil and dispersants may be very harmful on the other hand and affect the Gulf for decades to come. In this way the decision to apply dispersants subsea and in such large quantities may yet pr ove to be brilliant or devastating. 51

PAGE 62

References Blondina GJ, Sowby ML, Ouano MT, Singe r MM, Tjeerdema RS 1997. A modified swirling flask efficacy test for oil spill dispersants. Spill Sci Technol B. 4 (3): 177-185. Chatterjee A. 2001. Thermodynamics of micelle fo rmation of ionic surf actants: A critical assessment for sodium dodecyl sulfate, cetyl pyridinium chloride and dioctyl sulfosuccinate (na salt) by microcalorimetric, conductometric, and tensiometric measurements. J Phys Chem A. 105 (51): 12823-12831. Daling PS, Moldestad M, Johansen Lewis A, Rdal J. 2003. Norwegian testing of emulsion properties at SeaThe importanc e of oil type and release conditions. Spill Sci Technol B. 8 (2): 123-136. Defeo O, McLachlan A, Schoeman DS, Schlach er TA, Dugan J, Jones A, Lastra M, Scapini F. 2009. Threats to sandy beach ecosystems: A review. Estuar Coast Shelf Sci. 81 (1): 1-12. Exxon Valdez Oil Spill Trustee Council, Loughlin TR. 1994. Marine mammals and the exxon valdez. San Diego: Academic Press, p. 395. Guyomarch J, Le Floch S, Merlin F. 2002. Effect of suspended mineral load, water salinity and oil type on the size of OilMi neral aggregates in the presence of chemical dispersant. Spill Sci Technol B. 8 (1): 95-100. Johansen 2003. Development and verifica tion of deep-water blowout models. Mar Pollut Bull. 47 (9-12): 360-368. Joye S, MacDonald I. 2010. Offshore oceanic impacts from the BP oil spill. Nat Geosci. 3 (7): 446. Kujawinski EB, Kido Soule MC, Valentine DL, Boysen AK, Longnecker K, Redmond MC. 2011. Fate of dispersants associated with the deepwater horizon oil spill. Envir Sci Technol. 45 (4): 1296-1308. Maltrud M, Peacock S, Visbeck M. 2010. On th e possible long-term fate of oil released in the deepwater horizon incident, esti mated using ensemble of dye release simulations. Environ Re s Lett. 5 (3): 1-7. Nalco. 14 Jun 2005. Safety data sheet: Product corexit EC9500A (Online). United States: Nalco. Available: http://www.lm rk.org/corexit_9500_uscueg.539287.pdf. [Accessed: 30 Jan 2011]. 52

PAGE 63

Nalco. 15 October 2008. Safety data sheet: Produc t corexit EC9527A (Online) United States: Nalco. Available: http://www.doh.state.fl.us/chd/bay/Do cuments/Oilspill/Master_EC9527A_MSDS _539295.pdf. [Accessed: 30 Jan 2011]. National Research Council (U S) Committee on Effectiveness of Oil Spill Dispersants. 1989. Using oil spill dispersants on the sea. Washington DC: National Academy Press, p. 335. National Research Council (U .S.) Committee on Oil in the Sea: Inputs, Fates, and Effects. 2003. Oil in the s ea III : Inputs, fates, and effects. Washington DC: National Academy Press, p. 265. National Research Council (U .S.) Committee on Understandi ng Oil Spill Dispersants: Efficacy and Effects. 2005. Oil spill di spersants : Efficacy and effects. Washington, D.C.: National Academies Press, p.377. Owens EH, Taylor E, Humphrey B. 2008. The persistence and charac ter of stranded oil on coarse-sediment beaches. Mar Pollut Bull. 56 (1): 14-26. Page CA, Bonner JS, Sumner PL, Autenrieth RL. 2000. Solubility of petroleum hydrocarbons in oil/water systems. Mar Chem. 70 (1-3): 79-87. Price RC. 1993. Petroleum Spill Bioremediation in the Marine Environment. Crc Cr Rev Microbiol. 19(4): 217-242 Progressive Management. 2011a. US departme nt of the interior deepwater horizon response and restoration. Restori ng the gulf [CD-ROM]. United States: Progressive Management. Progressive Management. 2011b. EP A response to the BP spill. Restoring the gulf [CDROM]. United States: Pr ogressive Management. Schooner JL. 2010. The gulf oil spill. Envir Sci Technol. 44 (13): 4833. Rogowska J, Namiesnik J. 2010. Environmental implications of oil spills from shipping accidents. Environ Contam Toxicol. 206: 95-114. Speight JG. 1980. The chemistry and technolog y of petroleum. New York: M.Dekker, p. 498. Thibodeaux LJ. 2011. Marine oil fate: Knowledge gaps, basic research, and development needs A perspective based on the deepwate r horizon spill. Environ Eng Sci. 28 (2): 87-93. 53

PAGE 64

U.S. Environmental Protection Agenc y. 2003. 40 CFR 300 Appendix C. Available: http://www.epa.gov/oem/docs/oil/cfr/a ppendix_c.pdf. [Accessed 15 May 2011] US Environmental Protection Agency. 2011. National contingency plan product schedule toxicity and effectiveness summaries (Online). United States: United States Environmental Protection Agency. Available: http://www.epa.gov/osweroe1/content/ncp /tox_tables.htm. [Accessed: 28 Apr 2011]. U.S. Environmental Protection Agency. 2010a. Guide to using the NCP product schedule notebook. Washington DC (US): United States Environmental Protection Agency. Voith M. 2010a. Oil spill leads to fame and fury for makers of dispersant chemicals. Chem Eng News. 88 (24): 22-23. Wang Z, Fingas M, Page DS. 1999. Oil spill identification. J Chromatography A. 843 (12): 369-411. Zurer PS. 2003. Countering oil spills. Chem Eng News. 81 (14): 32-33. 54

PAGE 65

Table 1. The Components of Corexit EC9500A and EC9527A (Nalco 2005, 2008; Progressive Management 2011b) CAS Registry Number Chemical Name Common name Purpose 57-55-6 1,2-Propanediol Propylene Glycol Water Soluble Solvent 111-76-2 Ethanol, 2-Butoxy-* N /A Water Soluble Solvent 577-11-7 Butanedioic acid, 2sulfo, 1, 4-bis(2ethylhexyl) ester, sodium salt (1:1) Dioctyl Sodium Sulfosuccinate (DOSS) Anionic Surfactant 1338-43-8 Sorbitan, mono-(9Z)9-octandecenoate Span 80 N onionic Surfactant 9005-65-6 Sorbitan, mono-(9Z)9-octandecenoate, p oly(oxy-1,2ethanediyl) derivs Tween 80 N onionic Surfactant 9005-70-3 Sorbitan, mono-(9Z)9-octadecenpate, p oly(oxy-1,2ethanediyl) derivs Tween 85 N onionic Surfactant 29911-28-2 2-propanol, 1-(2butoxy-1methylethoxy)N /A 64742-47-8 Distillates (petroleum), hydrotreated light N /A Oil Soluble Solvent *denotes that this chemical is not a component of Corexit 9500A 55

PAGE 66

Table 2. Characteristics of EPAs standard cr ude oils for efficacy testing. (From: U.S. Environmental Protection Agency 2003) 1 At 150 -Celsius 2 Not Calculable when viscosity at 1000 Celsius is less than 2.0 API (American Petroleum Institute) Gravity An inverse measure of the relative density of petroleum liquid and the density of wate r. Greater API gravity corresponds to lower density. API gravity has no units mathemati cally, but is referred to in degrees. 56

PAGE 67

Figure 1. A diagram of oil fate processes (From: NRC 2005) 57

PAGE 68

Figure 2. A weathered oil slick (Fro m: Progressive Management 2011a) 58

PAGE 69

Figure 3. A tar ball (From: Progressive Management 2011b) 59

PAGE 70

Figure 4. A diagram of the subsurface transpor t processes for oil released in deep-water (From: NRC 2003) 60

PAGE 71

Figures 5. Simplified diagram of dispersants acting on an oil slick and breaking it into droplets which can spread throughout th e water column. (From: NRC 2005) 61

PAGE 72

Figure 6. A detailed diagram of surfactant mol ecules interacting with oil. Surfactant A is sorbitan monooleate (a.k.a., Span 80; HLB 4.3); surfactant B is ethoxylated sorbitan monooleate (a.k.a., Tween 80; HLB 15). (From: NRC 2005) 62

PAGE 73

Figure 7. The chemical structure of select Corexit components. Tween 80: Sorbitan, mono-(9Z)-9-octandecenoate, poly(oxy-1,2-ethanediyl): Tween 85: Sorbitan, mono-(9Z)-9-oct adecenpate, poly(oxy-1,2-ethanediyl): Span 80: Sorbitan, mono-(9Z)-9-octandecenoate: Dioctyl Sodium Sulfosuccinate (DOSS): 2-Propanol, 1-(2-butoxy-1-methylethoxy)-: 63

PAGE 74

Propylene Glycol: 1,2-Propanediol: Ethanol, 2-Butoxy-*: denotes component of Corexit EX9527A only 64

PAGE 75

Figure 8. Flask for dispersant efficacy testi ng. Oil and dispersant are mixed for a set period of time and then allowed to settle fo r a set period of time. Afterwards water is collected from the side spout and analy zed for oil content. (From: NRC 2005) 65

PAGE 76

CHAPTER 3 THE POTENTIAL BIOLOGICAL AND ENVIRONMENTAL EFFECTS OF OIL DISPERSANT USE 66

PAGE 77

Introduction Dispersants redistribute oil in potentially useful ways, i.e., preventing oil from stranding on shorelines, dispersing slicks on the surface into the water column protecting surface dwellers, and breaking oil into smaller droplets potentially accel erating the rate of microbial oxidation (NRC 1989, 2003). Organism s living in the water column may be affected by higher concentrations of oil and dispersant and the di spersants may change the bio-availability of the oil. Additionally, dispersants may impact certain communities more than others, some of which little is known, and the relationship between dispersants and the risk of oxygen depletion has not su fficiently been studie d (Joy and MacDonald 2010; NRC 2003). Overall, environmental effe cts of dispersants are characterized by uncertainty. It is important to understand the biological and envir onmental effects caused by the use of oil dispersants in response to marine oil spills. To investigate this topic, I review aspects of dispersant toxicity, the affects on biodegradation, the impacts on communities that are comparable from shallow to deep water, the differences of the deep water environment, and the early data from the DWH spill. Toxicity: The Effects of Disp ersants on Biological Systems The most direct way of assessing the impact of oil and oil dispersants on biological systems is measuring the toxicity of the substance(s) of concern on a single organism. In a general sense, toxicity is the measure of harm a substance imparts on a biological system (Klassen 2001). In the case of an oil spill the toxica nt, the substance of concern, can be any phase or fraction of th e hydrocarbon, dispersant or combination that comes in contact with an organism (NRC 1989, 2005). Marine organisms can be exposed 67

PAGE 78

to oil through ingestion, filtration, inhala tion, absorption, and fouling from contact (National Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling 2011). The impacts of oil and dispersants on biological systems are highly complex and can be affected by a variety of biotic and abiotic factors (NRC 1989, 2005). Toxicity assessments for oil and oil dispersants can be conducted in the laboratory or in the field and each method has benefits and drawbacks. Toxicity is easiest to measure in the laboratory because researchers are able to control for complex variables and then discer n the specific affects of a toxicant (Klaassen 2001). Laboratory testing cannot account for all f actors that may affect the toxicity in the natural environment. Therefore, laboratory te sts are primarily useful as reference points. The toxicity testing of a subs tance attempts to quantify leth al and sub-lethal levels of effects and the minimum exposure that resu lts in mortality. Duration of exposure is important in the toxic effects of a chemical because high levels over a short time may cause death and low levels of exposure ove r a longer period of time may also cause death. This is the premise for tests that are separated into trials to measure acute effects (short term usually 24-96 hours) or chroni c effects (longer term, often weeks) (NRC 2005). The most common toxicity tests conducte d for oil dispersants are acute lethal toxicity tests, which contain a fixed quan tifiable endpoint and can be assessed over a short period of time to obtain a reliable estimate of the maximum allowable exposure (U.S. Environmental Protection Agency 2002; Klemm 2004; NRC 2005). The results of acute toxicity tests are e xpressed in terms of an LC50, or concentration that is lethal to 50 percent of the test organisms over a gi ven period of time (Klaassen 2001). The LC50 68

PAGE 79

concentration is statistically determined from test data using one of two dose-response paradigms. A standard dose response model curve assumes that there is a threshold toxicant concentration below wh ich there is no response. The level at which there is no observable response is referred to as the no observable adverse eff ect level (NOEL). A no-threshold model assumes that there is alwa ys a response to a toxi cant; it is just not measurable below a certain level. The EPA st andard testing procedures have begun to use the no-threshold model because it can expre ss a higher degree of sensitivity (U.S. Environmental Protection Agency 2003). To register a dispersant pr oduct so that it may be used in response to an oil spill, EPA requires that manufacturers submit standard ized acute lethal toxicity data on two species along with efficacy data for their pr oduct (U.S. Environmental Protection Agency 2003). The required tests are a 48 hour LC50 for the mysid shrimp ( Mysidopsis bahia ) and 96 hour LC50 for the inland silverside (Menidia beryllina) The tests are conducted on larval stage organisms, as this stage is t ypically the most sensitive to toxicants. Two methods exist for conducting LC50 tests on marine organisms, a static non-renewal test and a flow-through test. (U.S. Environmen tal Protection Agency 2002; Klemm 2004; Aurand and Coelho 2005). For a static-exposure test, organisms are housed in the same water without circulation or re placement for the duration of th e test and the toxicant is added only once, at the beginning. Static tests are typically not preferred because the concentration of the chemical may decline from chemical degradation, biodegradation, evaporation, and adsorption to the tank walls. In addition, waste can accumulate over time confounding the toxic eff ects. The preferred toxicity testing method is a flowthrough test, in which a constant known quantity of toxicant is added to the system along 69

PAGE 80

with fresh water. An equal amount of water is then removed, ensuring that the toxicant concentration is constant and there is wa ste filtration. The flow-through test is the standard but is more costly to conduct. Historically, there have been many con cerns about dispersants because early ones were much higher in toxicity relative to m odern formulations. They were derived from industrial strength degreasing agents (NRC 1989) and containe d high quantities of toxic aromatic hydrocarbons (Nelson-Smith 72). Surfact ants in early formulations were also more toxic than those found in modern disp ersants (Abel 74, NRC 89) Factors affecting surfactant toxicity ar e complex but are known to vary based on the hydrophilic-lipophilic balance (HLB) and the ionic stat e of the surfactant. Ionic surf actants are more toxic than nonionic ones and early dispersa nts contained high concentra tions of nonionic surfactant that enhanced their toxicity. A second concern about dispersant toxicity is that early testing overe stimated the toxicity of chemically dispersed oil (NRC 1989). These LC50 tests evaluated the relative toxicity of oil versus chemically dispersed oil and based toxicity on the nominal concentration of oil in the sy stem, which is a major flaw. Nominal concentration is the weight of oil per unit volume of water in th e experimental system. Because oil and water are non-miscible, the organisms in the tests are really subject to the concentration of oil dissolved or dispersed in the water colu mn. The fraction of the oil dissolved and dispersed in the water column increases with the application of chemical dispersants. This is logical as it is the function of the disp ersants. Thus, the organisms in this test are effectively exposed to higher con centrations of oil. More r ecent toxicity testing based on the concentration of oil in the water column has shown that the enhanced toxicity of the 70

PAGE 81

chemically dispersed oil can primarily be attr ibuted to the relative level of hydrocarbons dissolved in the water column (NRC 1989, 2005). Despite limitations of extrapolating laboratory results to environmental effects, EPA requires manufacturers of dispersants to submit results from several standardized laboratory LC50 tests to secure product approval (U.S. Environmental Protection Agency 2003). Once a product is approved, it is added to the EPA product schedule on the NCP, which details proper use. The EPA NCP product schedule contained 14 dispersant products at the time of the Macondo well bl owout (Appendix A) (U.S. Environmental Protection Agency 2010b). One parameter of the EPA product schedule is an LC50 test of a 1:10 ratio of dispersant and No. 2 fuel oi l, a heavy refined product. The most widely used Corexit dispersant product, EC9500A, when mixed with No. 2 fuel oil has a LC50 value for inland silversides of 2.61 parts per million (ppm) and a LC50 of 3.40 ppm for mysid shrimp (U.S. Environmental Protection Agency 2010b). The second Corexit product, EC9527A, was used in this oil spill to lesser degree and limited only to surface applications. When EC9527A is mixed with No. 2 fuel oil, it has a LC50 value of 4.49 ppm for inland silversides and 6.60 for mysid shrimp. Compared to other dispersant products listed, the Corexit produ cts appear relatively toxic. The dispersant, Nokomis 3F4 (Mar-Len Supply, Inc), combined with No. 2 fuel oil has a LC50 of 100 ppm for inland silverside and 58.40 ppm for mysid shrimp (U.S. Environmental Protection Agency 2010b). The most toxic products listed are Corexit EC9500A on the inland silverside and Zi-400 (Z.I. Chemicals) on the mysid shrimp (U.S. Environmental Protection agency 2010b). Though testing showed EC9527A to be the least toxic of the Corexit products, EC9500A may have been preferably used becau se it does not contain 2-butoxyethanol as 71

PAGE 82

does EC9527A (Nalco 2005, 2008). The compound 2-Butoxyethanol is a known to be a mammalian carcinogen and a strong irritant, whic h may pose a risk to response workers. In December of 2010, EPA released an updated guide to its NCP product schedule (U.S. Environmental Protection Agency 2010a). This guide contained more detailed information than that on the original product summary, calling into question the reliability of the data presented above. Because each manufacturer conducted their own toxicity tests, each was required to submit an LC50 for both species exposed to dispersant only, No.2 fuel oil only, a 1:10 mixture of disper sant to No. 2 fuel oil, and a reference toxicant dioctyl sodium sulfosuccinate [D SS or DOSS, a common anionic surfactant]. The results presented in this report highlight inconsistencies between different tests that confound the direct extrapolati on of information from the af orementioned tests (Table 1) (U.S. Environmental Protection 2010a). Nokomis 3-F4 mixed with No. 2 fuel oil has an LC50 of 100 ppm for inland silverside, which was far less toxic than Corexit EC9500A with a LC50 of 2.61 ppm. However, the inland silverside LC50 values for No.2 fuel oil were only 100 ppm for Nokomis and 10.72 ppm for both Corexit products. This difference highlights inconsistencies in the methodology fo r these tests, which limits the inferences that can be drawn from these results. Similar inconsistencies were present for the reference toxicant. Another difficulty in analyzing the data from the NCP is that there are not clear patterns between the toxicity of dispersant and the toxicity of the combined dispersant and oil (U.S. Environmental Protection Ag ency 2010a). While Corexit EC9527 and No.2 oil mixture was less toxic to both test spec ies than the combination Corexit EC9500A and No. 2 fuel oil, EC9500A by itself was less toxi c to both species. This difference may be 72

PAGE 83

accounted for by inaccuracies in mixing, resulting in different effectiv e concentrations of the test substance within the water column, which has been a problem in past tests (NRC 2005). In addition, Nokomis 3-F4 showed a similar level of toxicity as Corexit EC9500A in the dispersant only test, but Nokomis 3F4 significantly decreased toxicity for the dispersant and oil test. Nokomis 3-F4 had an LC50 of 29.80 ppm for the inland silverside alone and an LC50 of 100 ppm for dispersant a nd oil. Corexit EC9500 had an LC50 of 25.20 ppm for inland silverside and an LC50 of 2.61 when combined with No.2 fuel oil. The toxicity of combined oil and dispersant is generally believed to be additive or synergistic, as is the case with the Corexit test (NRC 2005). The Nokomis 3-F4 data, which is consistent in pattern across test sp ecies, shows what must be a flaw, inaccuracy or inconsistency in the testi ng procedure for these dispersant s. During the response to the DWH incident, BP and the US Government were subject to intense criticism for the use of dispersants, which were widely believed to enhance the toxicity of the spill (Johnson and Torrice 2010). In response, EPA conducted its own independe nt toxicity tests of oil dispersants and released two pha ses of testing in the summer of 2010. In the first phase of testing, LC50 tests were conducted on the two refe rence species using dispersant only (Hemmer et al. 2011). Results across the ei ght dispersants tested were relatively consistent with the values listed on the NCP (Table 2) (Hemmer et al 2011, Judson et al 2011; U.S. Environmental Protection Agency 2010a). A notable exception was Corexit EC9500A, which was less toxic to both test sp ecies than shown in independent testing and considered almost non-toxic to inland silverside, as it had an LC50 value of 130ppm, much lower than the original 25.2ppm value. A second round of testing considered cytotoxicity and the potential of these eight dispersants to act as endocrine disrupting 73

PAGE 84

compounds (Judson et al. 2010). The test conclu ded that none of these eight dispersants tested showed any significant endocrine disrupting activity, including Corexit EC9500A. Endocrine disrupting compounds interfere with cellular communication, including hormones and can cause various sub-acute effects that may eventually lead to death (Klaassen 2001). The test also showed low pot ential for cytotoxicty, which is the toxicity of these chemicals to individual cells. An independent 2004 study considered the toxicity of Corexit EC500A and oil on several species, including the standard EPA test organisms, M. beryllena and M. bahia (Fuller et al. 2004). The LC50 results for dispersant alone an d dispersant and oil together for the inland silverside and mysid shrimp were similar to those on the NCP and the independent EPA testing. The researchers also ran tests of decreas ing concentration to simulate dilution and degradation in the natural environment. The LC50 values obtained from these tests were far higher (less toxic) than those obtained from the standard constant concentration test. For this test, researchers took care to analyze the total petroleum hydrocarbon concentrat ions that organisms were exposed to, removing the bias of early tests, which considered only the nominal concentration. They concluded that dispersant and oil is equal to or less toxic than oil alone and that dispersant contribution to toxicity under the parameters of thes e tests was minimal (Fuller et al. 2003). The results of this test are similar to EPA re sults, though this study found dispersants were generally slightly less toxic than the EPA test. It is wo rth noting that this study was funded in part by the American Petroleum Institute, the industries main lobbyist. That said, the comparable EPA product schedule tests were also conducted by industry researchers. 74

PAGE 85

Conclusions that are able to be drawn from this laboratory testing are limited because the tests do not account for variations and complexities that exist in the natural environment and there were inconsistencie s in the NCP product schedule. However, recent toxicity testing indicated that newer dispersants are significantly less toxic than earlier formulations because of the removal of the aromatic constituents and toxic surfactants (NRC 1989). The dispersant by itse lf is generally less toxic than the oil (Fuller et al. 2004; U.S. Environmental Pr otection Agency 2010b). Though the toxicity of dispersant and oil appears to additive in many tests, the oil is the main contributing factor to this toxicity and enhanced toxicity may be from the increas ed solubility of chemically dispersed oil (NRC 2005; U.S. Environmenta l Protection Agency 2010a). Interestingly, each dispersant product was also tested for heavy metal concentration, as this was an environmental concern following the DWH bl owout (Gertz 2010; U.S. Environmental Protection Agency 2010a). For Corexit EC9500A all heavy metals were at very low concentrations (less than 1ppm). Our understanding of both acute and chroni c toxicity will benefit from additional field and laboratory studies (NRC 2005). Desp ite advances, laboratory studies can only give so much information about the behavior and toxicity of oil a nd dispersants in the environment. Factors that may affect the to xicity of chemically dispersed oil in the environment include efficacy of dispersion, type of dispersant and oil, weathering processes including emulsification and ev aporation, temperature, salinity, photoenhanced toxicity, sedimentation, duration, and concentration of exposure (NRC 2005). Laboratory tests cannot account for the relative importance of exposure routes and shortterm tests often fail to account fo r metabolism and bio-accumulation. 75

PAGE 86

As evidenced by the standardized toxicity tests presented above, mixtures of oil and modern dispersants owe their toxicity more to the oil than the dispersant itself (U.S. Environmental Protection Agency 2010a, NRC 1989). Dispersants act to change the concentrations of oil at the water surface by dissolving and dispersing oil into the water column. In theory and in laboratory testing, e nhancing concentrations of oil in the water column will increase the acute toxicity to subsurface organisms by increasing their exposure (NRC 1989). Whether or not this effect will happen in the ocean is questioned (NRC, 1989). In the open ocean, chemically di spersed oil will be greatly diluted by the volume of water present. This limits th e acute biological effects because the concentration and duration of exposure will be low compared to those that show acute effects in the laborator y. Without the potential to cause acute effects in the open ocean, the use of dispersants should decrease th e overall damage to ecosystems because dispersing oil will limit acute effects to surface dwelling organisms. This fact may not be true in sheltered and confined waters with poor circulation (NRC, 1989). Oil that accumulates in shallow sheltered bodies of wate r has the potential to cause acute effects. One place that the effects of chemically disp ersed oil have been heavily studied is on shallow water coral reefs (NRC 2005). Chemi cally dispersed oil has been shown to enhance toxicity to corals when dispersa nts are applied above a reef however, this application limits damage to coastal plants like mangroves. Laboratory tests are typically r un under standard conditions (25 C) that do not account for the effects of salinity and te mperature (U.S. Environmental Protection Agency 2002; Klemm et al. 2004). Increased salin ity can change the toxicity and efficacy of dispersants by salting out surfactants and changing the behavior of oil (NRC 2005). 76

PAGE 87

The biological effects of this mechanis m are not well understood. In cold waters dispersants become less toxic, but in warm wa ters dispersant toxic ity can be enhanced (NRC, 2005). Biodiversity and activity levels can be higher in certain warm water areas and increased contact with oil leads to higher levels of biol ogical uptake owing to the enhanced toxicity. These are important points considering the warm conditions found in the Gulf of Mexico during the time of the DWH discha rge (National Commission on the BP Deepwater Horizon Oil Spill an d Offshore Drilling 2011; NRC 1989). Standardized laboratory testing has historically failed to account for photoenhanced toxicity, which can contribute significantly to th e toxicity of dispersed oil (Landrum et al. 1987; Ankley 1994, Boese 1997; Pelletier 1997; NRC 2005). Polycyclic aromatic hydrocarbons (PAHs), a toxic compon ent of oil, react with UV light in two ways, photo-modification and photosensitizati on (Figure 1). In photo-modification, free radicals are created from the reaction of PAH compounds and UV light, which then oxidize into more toxic compounds (NRC 2003, 2005). In photo-se nsitization, PAH compounds are excited by UV light and transf er energy to dissolved oxygen forming free radicals. Free radicals cause cel lular damage, but typically requi re that a large quantity of oil be accumulated within tissue to create da maging concentrations of free radicals. Typical dispersant toxicity test s do not measure this effect, but can be adapted to simulate it (NRC 2003, 2005. Despite the ability to mode l photo-enhanced toxicity, laboratory tests are not the most accurate way to determ ine effects from varying light intensity and penetration in the environment. Laboratory testing can provide insight a bout the complexities of oil dispersant toxicity in natural environments (Aurand and Coelho 2005; NRC 2005). The Ecological 77

PAGE 88

Research Forum (2003) suggested that a series of changes be made to the standardized LC50 dispersant and dispersed oil testing pr ocedures. At the top of this list was determining an accepted methodology for th e ongoing problem of quantifying water column exposure to oil and dispersed oil (Ecologi cal Research Forum 2003). Furthermore, the standard toxicity tests need to be expanded to incl ude investigations of photo-enhanced toxicity by exposure to UV li ght. Finally, tests s hould account for the evaporation of certain critic al hydrocarbon constituents of pe troleum to more accurately model dispersants in the environment. Once an organism has come in contact w ith chemically dispersed oil, the effects incurred can vary by species, life history, physiology, time exposed to oil, and etc (NRC 1989, 2005). Species vary in susceptibility to di spersant and oil toxic ity as demonstrated by mysid shrimp and inland silverside speci es (Environmental Protection Agency 2010a). Similar species may show varying degrees of su sceptibility to disper sant and oil toxicity, i.e., inland silverside and sim ilarly-sized sheepshead minnow ( Cyprinodon variegatus ) (Fuller et al. 2004). Generally, smaller species ha ve a greater risk from dispersant and oil toxicity and this susceptibili ty raises concerns (NRC 1989, 20 05). Another concern is the effect oil and dispersants may have in deep ocean and its inhabitants. Deepwater oil discharges may have large impacts on the highly specialized deep-water communities (NRC 2003). Species that use bioluminescen ce for communication or food attainment may be affected by deep-water plumes of oil and dispersant (National Commission on the BP Deepwater Horizon Oil Spill and Of fshore Drilling 2011; NRC 2003). Larval and embryonic stages of organisms are usually more sensitive to toxicants (U.S. Environmental Protection Agency 2003 ;N RC 2005). Individual physiology may be 78

PAGE 89

affected by oil toxicity, as f eeding and activity may expose an organism to higher levels of toxicant. Organisms can come in contact with chem ically dispersed oil in various forms, dissolved materials, either gases or water soluble hydrocarbons, dispersed oil droplets, hydrocarbon solids, or a combination of thes e via several routes of exposure (NRC 2005), i.e., dermal contact, ingestion, respira tion and filtration (National Oceanic and Atmospheric Administration 2010). The relative importance of the state of the chemically dispersed oil and route of exposure ar e not well understood (NRC 2005). Organism response is complex being dependent on level of contact, uptake and internal storage, the toxic action of the substance, detoxifica tion, and depuration processes. Generalized biological uptake of dissol ved phase exposures to chem ically dispersed oil is characterized by interruption of receptor-mediated pathways leading to narcosis (NRC 2005). This narcotic effect can be traced to both the components of dispersants and oil, particularly PAHs (Di toro 2000). Narcosis can be a predictor of mortality from acute toxicity (NRC 2005). (Barron et al. 2004). T oxic effect can be observed when early life stages are exposed and can result in deformities and cardiovascular dysfunction (Barron et al.2004). Edema, abnormalities in circulat ion leading to swelling, may also occur from exposure to dissolved fractions. Contact with chemically dispersed oil droplets and dispersant components may disrupt membrane integrity, smother organisms or gills, and can cause internal damage through inges tion (NRC 1989, 2005). Crus tacean gills are hydrophobic, which attract and lead to damage from surfactant and hydrocarbon exposure (Granmo and Kolberg 76; NRC 1898, 2005). Fish gills are mucous coated and are less hydrophobic. Surfactant packing at the surface may lead to asphyxiation in fishes and 79

PAGE 90

lead to changes in gill membrane perm eability. The hydrophilic-lipophilic balance of dispersants can affect these and other bi ological interactions depending on the hydrophilic and lipophilic nature of biological membranes (NRC 2005). Bioavailability and Bioaccumulation The bioavailability of a substance is a m easure of how readily a biological system accumulates and metabolizes it and increased bioavailability can lead to enhanced toxicity because of higher exposure (N RC 2003, 2005). Toxicity of hydrocarbons is related to their solubility in water because these dissolved components can diffuse across exposed membranes such as gills. Dispersant s may enhance the bioa vailability of oil by enhancing interactions at biological membranes (Wol fe et al. 1998; Wolfe 1998b,c; Wolfe et al. 2000, 2001; NRC 2005). Dispersants may also affect membrane permeability through interactions with proteins and ion chan nels. Differences in th e bioavailability of chemically dispersed oil and physically disp ersed oil are unknown, but may differ in the total amount of oil dispersed in the water column. Regardle ss of the relative effects of dispersed oil, enhanced uptake through memb ranes should make chemically dispersed oil more toxic. Studies have s hown that dispersants enhanced the initial uptake of oil components by microalgae (NRC 2005). Bioaccumulation is the tendency for a materi al to build up over time in the tissues of an organism (National Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling 2011). Bioaccumulation depends on the concentration of a substance in water and food and the duration and concentr ation of exposures. Fa tty tissues like the brain, liver, kidneys and ovaries bioaccumulate more than other tissues. Higher molecular 80

PAGE 91

weight components of oil are the most pers istent in organisms (NRC 2003). Chemically dispersed oil may enhance bioaccumulation thro ugh greater uptake but do not directly result in higher levels of accumulation or persistence (Wolf et al. 1998; Wolfe 1998bc; Wolfe et al. 2000, 2001; NRC 2005). Biodegradation Biodegradation of petroleum is the pr ocess in which microbes consume oil, breaking down hydrocarbons into common, non-toxi c molecules that can be incorporated in to the environment (Figure 2) (NRC 1989, 2005). Theoretically, oil dispersants should increase the rate of biodegrad ation by breaking the oil into droplets and spreading the oil over a larger volume, effectively incr easing the surface area and exposure of hydrocarbons to bacteria capable of c onsuming oil (NRC 1989). Despite intensive testing, laboratory and field studies have not confirmed th is hypothesis (NRC, 2005). In testing all possible outcomes have been reac hed, with some studies showing enhanced degradation, others no effect and still ot hers inhibition (Linds trom and Braddock 2002; Literathy et al. 1989; Mulkins-Phillips and Stewart 1974). These effects do not appear to be systematically affected by the dispersant and surfactant, the o il substrate, or the microbial communities present. Reasons for this confusion may be based on testing methodologies and variation in the e ffects of different surfactants. Researchers have used multiple testin g methodologies to measure the rate of hydrocarbon biodegradation including the monitoring of oxygen uptake, growth of microbes and oil composition. Conclusions ba sed on indirect evidence have caused confusion about biodegradation (NRC 1989, 2005) Some studies have monitored oxygen 81

PAGE 92

uptake and cited an increase in uptake as in direct evidence of bi odegradation (NRC 2005; Hazen 2010). Many bacteria strains capable of decomposing petroleum hydrocarbons are aerobic and require oxygen in the reactions th at degrade the oil. The flaws in this methodology are that: i ) it does not account for abioti c reactions, which break down hydrocarbons and utilize oxygen, and ii) the failure to include the effects of chemosynthetic bacteria that do not require oxygen in consuming petroleum. Measuring the growth of microbial communities is another indirect indicator that has been used tp measure the decomposition of chemically dispersed oil (NRC 2005). This method may overestimate the rate of oil biodegradation if microbes use th e dispersants as a preferred carbon source or if the oil and dispersants are toxic to bacteriavores on a short timescale (NRC, 2005). Measuring microbi al growth may also underestimate biodegradation of chemically dispersed oil if microbes are lim ited by nutrients other than oil. A final and more direct measure of biodegradation is monitoring oil composition with ion trap mass spectrometry or gas chromatograph mass spec trometry (NRC, 2005). This too is limited though, by the unknown mechanism of oil degrad ation, which may be physical, chemical, or biological. A solution for future study may be to use multiple indicators to extrapolate information. Dispersants and the surfactants contai ned therein have been shown to have varying effects on the rate of biodegradation wh en other test paramete rs are held constant (Foght et al. 1987). Ionic surfact ants have been shown to inhibit biodegradation more than nonionic surfactants (B ruheim et al. 1999). Other st udies have shown that the surfactant hydrophilic-li pophilic balance (HLB) affects the rate of biodegradation (Van Hamme and Ward 1999; Varadaraj 1995). Van Hamme and Ward (1999) concluded that 82

PAGE 93

an HLB between 12 and 14 enhanced biodegrad ation, whereas, Varadaraj et al. (1995) concluded that an HLB of 8 was optimal. Surf actants have varying effects because they may enhance or inhibit interactions at the cell surface, which affect the permeability of the biological membrane (Bruheim et al.1999; NRC 2005). Surfactant interactions w ith the cellular surface may affect microbial uptake by three mechanisms: transport of aqueous phase substrate, direct c ontact with non-aqueous phase liquids or solids, and uptake of hydrocar bons present in surf actant micelles (Singer and Finnerty 1984; Watkinson and Morgan 1990; NRC 2005). The uptake of aqueous dissolved hydrocarbons should not markedly be affected by surfactants because these components of oil are the most water soluble and will have less interaction with dispersants. The direct attachme nt of bacteria to non-aqueous oil substrate may be highly affected by surfactants because both the bacter ia and surfactants act at the surface of the oil (Watkinson and Morgan 1990). Bacteria th at consume non-aqueous phase substrate are typically hydrophobic and the presence of hydrophilic surfactants at the interface between oil and water may limit interacti ons. The final mechanism, the uptake of micelles is considered unimportant because its occurrence is unlikely (Schippers 2000; Garcia 2001). While bacteria might uptake su rfactant micelles containing oil, actual surfactant concentrations are likely to be well below the critical micelle count making their formation unlikely. The overall effects of surfactants on biodegradation are complex because they consist of an interplay between various mode s of uptake. The effects of surfactant HLB are most likely to change the interactions of bact eria with non-aqueous phase hydrocarbons. Long-term effects will also vary because oil may become detached from the surfactant over time (NRC 2005). 83

PAGE 94

Biological Effects of Using Dispersants in the Deep Sea The DWH was the first ultra-deep blowout and the first instance of subsea dispersant application (Na tional Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling 2011). Scientific knowledge of the deep sea, where drilling is now a common occurrence, is limited. The potential fo r discharges of such great magnitude and the application of large quantities of dispersant in this environment, as in the case of the DWH, necessitates the study of these comm unities to analyze the potential harm and risks. Deep-water petroleum release and disper sant application is lik ely to affect deepwater biological communities differently than communities at the surface. The first reason for the difference is that deep-water communities have adapted to different conditions that surface communities. Second, oil and dispersant released at depth potentially behave differently due to the unique parameters of this environment. There is limited information in the literature about the effects of large quanti ties of petroleum on deep-water biological communitie s and little research has be en done on the effects of dispersants and oil/dispersant mixtures on deep-sea organisms This section of the paper will attempt to characterize the unique biological risks posed by chemically dispersed oil in the d eep sea. Analogous shallow water organisms, unique deep-water organisms, and the early data from the DWH will be considered. Assessing probable impacts to deep-sea communitie s of flora and fauna is difficult due to the relative lack of knowledge about these organisms. It is problematic to predict or study impacts without basic knowledge of commun ity ecology (such as rates and mechanisms for population control), biologi cal interactions, predation, recruitment and rate and potential for recovery (NRC 2003). Furtherm ore, little knowledge exists about the 84

PAGE 95

feeding type, reproduction, life sp an, growth rate, and sensi tivity to contaminants for many deep sea organisms. Lack of knowledge is primarily due to the difficulty of studying organisms in such extreme environm ents. Deep-sea communities face a range of physical and environmental differences, i.e., pressure and light, from shallow water communities. Despite inherent differences, ge neral predictions can be made about the effects of oil and dispersants on some deep-sea communities based on information drawn from analogous continental shelf organisms, such as plankton, microbes, and the benthos. Plankton Plankton communities consist of any freely drifting plants, animals, and microbes inhabiting the open ocean. The effects of o il and dispersants on plankton communities are important because plankton make up the ba sis of the food chain. In addition, many diverse species, including commercially importa nt fish species, have planktonic larval stages (NRC 1989; Bond 1996). Unfortunatel y, studies on the effects of oil and dispersant on these communities are nonexist ent. A survey of studies by the National Research Council (1989) found that dispersed oil is more toxic than oil alone to phytoplankton in four out of 11 studies. Studi es comparing the toxicity of oil for zooplankton are largely inconclusive. Rogerson and Berger (1981) suggested that zooplankton communities grew better in disperse d oil than oil alone. Studies that focused on commercially important ichtyoplankton have largely been based on nominal concentrations and concluded th at chemically dispersed oil is more toxic than oil alone (Linden 1975; NRC 1989). A more recent review of studies suggested that when toxicity was based on total petroleum hydrocarbon cont ent rather than nominal concentration, 85

PAGE 96

chemically and physically dispersed oil were of comparable toxicity (NRC 2005). The major concern for plankton communities when us ing dispersants is that these organisms will be exposed to more dispersed oil in the water column, enhancing toxicity. The ability of oil to be passed up th e food chain is a second concern when considering the effects of oil on plankton. Disp ersant surfactants break oil into droplets, which overlap in size with the preferred size range of food consumed by many species of zooplankton (NRC 2005). Oil consumed by plankton may be transferred up the food chain when larger organisms consume the pl ankton. Wolf et al. (1998) demonstrated the ability of a planktavore to accumulate hydrocarbons from the consumption of plankton, e.g., 20 to 45 percent of oil found in the ro tifers was from consumption of plankton. Oil that was originally consumed by plankton move up the food chain to fish and beyond. Microbes Though some microbes are capable of consuming petroleum hydrocarbons, dispersants and oil can be toxic to ot hers (NRC 1989, 2005). The microbial communities present in continental shelf waters, the deep -sea, and different ge ographic regions vary, but the effects of oil and dispersants are largely analogous (NRC 2003, 2005). The majority of studies investigating the relati onships of oil, dispersants and microbes are focused on biodegradation and shed little light on overall effects, including toxicity (Watkinson and Morgan 1990; Bruheim et al. 1999; NRC 2005; Hazen et al. 2010). Furthermore, very few of these studies focus on natural marine populations. Biodegradation studies often rely on rate of hydrocarbon uptake or total bacterial community growth for results. Growth in a part icular strain of bacteria will mask the 86

PAGE 97

toxic effects of oil and dispersant mixtures to other members of the community (Linden 1987, NRC 2005). Dispersants may also alter the ability of bacteria to uptake hydrocarbons and can result in increased growth of some strains and toxicity for others (Zhang 1994). Ultimately, input of chemically dispersed oil is likely to decrease localized diversity of microbes (Zhang 1994; Hazen et al. 2010). Growth of hydrocarbon consuming microbes may also lead to oxyge n depletion in certain environments. Benthos The benthos of the ocean consists of communities developed to feed on detritus deposited from the water column (NRC 2003). These communities are generally negatively affected by the input of oil, as it is toxic to many members of the benthic community and heavy fractions of petroleum may sink and smother these communities. Oil adheres to suspended sediments and the detritus they consume, which eventually settles out of the water column. Dispersants are most likely to affect this community by altering oil distribution and the interactions between oil and suspended particulate matter (NRC 2005). In shallow water, chemically disperse d oil is more likely to come in contact with benthic communities by removal from th e surface. In the deep ocean, dispersants may actually decrease the rate of sediment ation by decreasing oil interaction with suspended particulate matter. Certain benthi c residents, i.e. polychaete worms, are tolerant of oil and are the first organisms to colonize oil contaminated benthos (NRC, 2005). These are benthic communities that col onize soft substrates and those found in hard substrates (Little 2000; Nybakken and Bertness 2005). Hard bottom communities are typically genetically and ecologically diverse (Nyabakken and Bertness 2005). These 87

PAGE 98

communities are afforded protection from drilling in the Gulf of Mexico but can be impacted by large-scale oil discharges as can soft bottom communities (NRC 2003). Unique Communities and Characteristics Certain characteristics of deep-sea communities may make them more susceptible to oil releases, for instance minimal exposur e to pollution (NRC 2003). Diversity in deepsea communities is augmented by the presence of microhabitats. These communities include those surrounding chemical seeps and those that inhabit localized hard substrate. These habitats differ in light, temperature, pressure, oxygen availability, food availability, and production of nutrients. Even small-scale discharges may have large impacts on the overall diversity of an area by affecting one or more of these habitats. A rare advantage of deep-sea communities is that the cold temperatures of deep water may lower the toxicity of chemically dispersed oil (NRC 1989) One distinctive feat ure of the deep-sea is that it is a low to no light environment. Deep-sea plumes of oil, as seen after the DWH, will increase the turbidity of the deep sea, possibly impacting the organisms in unforeseen ways (NRC 2003; National Commis sion on the BP Deepwater Horizon Oil Spill and Offshore Drilling 2011; Kujawinski et al. 2011). The deep ocean is characterized by large slow moving currents that interact in a limited manner with the faster moving stream s of water in the upper water column (NRC 2003; Joye and MacDonald 2010; Joye et al. 201 1). This low level of mixing in the deep ocean results in slow oxygen exchange. Chemic al changes in the deep sea may affect the water chemistry. Of particular concern with oi l, is that chemical and more importantly, biological degradation of hydrocarbons, utiliz es oxygen directly in reactions (NRC 2003). 88

PAGE 99

These reactions lower oxygen leve ls locally and result in hyp oxia and anoxia (Joye et al. 2011). This is of particular concern in th e deep ocean environment where oxygen levels are already low and are minimally at most replaced by mixing (NRC 2003). Additionally, dissolved oxygen in the deep ocean is not replaced by in situ photosynthesis (Joye and Macdonald 2010). Thus, depletion of deep-sea oxygen may be prolonged and detrimental to the communities. The organisms that inhabit the water and benthos immediately surrounding deepwater chemical seeps comprise distinct deep-water communities (NRC 2003). These microhabitats, known as chemosynthetic seep communities are common in the Gulf of Mexico between 300 and 1000m deep (NRC 2003; Joye and MacDonald 2010). These organisms depend on extremes in their environmen ts to survive, but are very sensitive to changes in the chemistry of their environm ent. Chemical seeps, including petroleum seeps, are often characterized by low local di versity due to lack of competition (Davis and Spies 1980). Petroleum seeps are characte rized by mats of chemosynthetic bacteria of the genus Beggiatoa These bacteria synthesize carbohydrates from carbon dioxide using inorganic sources of energy, such as sulfide oxidation through hydrogen sulfide. Many organisms present in deep-sea communitie s are long lived, this combined with low levels of diversity may make recovery from damage slow or non-existent. The advantage chemosynthetic communities have is a tolera nce of low oxygen concentrations, but the sheer volume of pollution and chemical flux brought about by a well blowout can harm even the communities of natural petroleum seeps (Joye and Macdonald 2010). The negative environmental effects of subs ea dispersant are likely to be driven primarily by changes in oil behavior rather than dispersant toxicity (Johnson and Torrice 89

PAGE 100

2010). Dispersants may enhance the exposure of plankton communities to oil and subsequently affect the other levels of the food chain (NRC 1989). Chemically dispersed oil will likely impact some microbial communities decreasing diversity and others may thrive (NRC 2005). Dispersant application to oil will vary in its effects to benthic communities, though deep-water communities may benefit from lower sedimentation. Highly adapted deep-water communities are likely to be negatively impacted from changes to the chemical and physical e nvironment, toxicity, and oxygen depletion. Despite the potential negative impacts to d eep-sea communities, another consideration of chemically dispersing oil in the deep ocean, is that oil sequestered in the deep-sea is subject to less transport than oil closer to the surface (Maltrud et al. 2010). Oil sequestered in the deep-sea will impact fewe r communities than oil subject to surface transport, but the impacts may be proportiona tely greater due to dilution. The ultimate effects and tradeoffs of chemically disper sing oil in the deep ocean remain to be determined. Deepwater Horizon The Macondo well blowout is the only ultr a-deep sea blowout and therefore, the first occurrence to investigate the environmen tal and biological eff ects from a deep sea oil discharge and subsequently subsea appl ication of chemical dispersants (National Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling 2011). The effects should be easy to discern though, as th is was the largest accid ental marine spill in the history of the United States (National Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling 2011; Federal Interagency Solutions Group 2010). The 90

PAGE 101

blowout was unprecedented in size, the locati on of the wellhead, and the duration of the spill. Deep-water ecosystems were exposed to oil and dispersants for prolonged periods of time, but science cannot accu rately predict the consequences. The knowledge of deepsea gulf ecosystems is mostly limited to specialized communities, such as those around natural seeps and localized ha rd substrate. Scientific knowledge has not advanced with the spread of deep-sea drilling. A significant portion of the oil di scharged from the Macondo well was sequestered in deep-sea plumes between 1000 and 1400 meters (Camilli 2010; Dierks 2010; Hazen et al. 2010; Valentine 2010). H ydrocarbon concentration in these plumes was high enough to cause acute toxicity and the plumes were shown to contain high levels of toxic PAHs. Up to 15 percent of th e oil discharged from the well was physically dispersed, so plumes would have formed without dispersant application (Federal Interagency Solutions Group 2010), though disper sant application likely enhanced the plumes. Dispersant surfactants were detectable within the oil plumes but this does not infer that dispersants were the cause behind th e creation of the plumes (Kujawinski et al. 2011). The concentrations of hydrocarbons enco untered in the plume were diluted with distance from the well, but were persistent for up to 60 days and 300km of transport. Oil deposits on the sea floor have also been reported (Burdeau and Borenstein 2010). The effects of the spill in the deep ocean are still generally unknown, but scientists have monitored both oxygen deple tion and microbial degradation. Prominent scientists were initially concerned with the potential of the spill to cause widespread oxygen depletion (Joye and MacDonald 2010; Jo ye et al. 2011). However, monitoring suggests that though depleted oxygen levels were found within plumes and low levels 91

PAGE 102

occurred around plumes, the decrease in oxygen levels were generally localized to the vicinity of the undersea plumes (Nationa l Incident Command Joint Analysis Group 2010). Widespread oxygen depletion may have been prevented by curre nts that supplied sufficient mixing. Still, the concentration of hydrocarbons encountered in deep-sea plumes and the presence of dispersants was su fficient to cause toxic effects to exposed organisms. Hazen et al. (2010) attempted to analyze the level of biodegradation in the deep-sea plumes and while drastic oxygen depletion has not been seen, the plumes stimulated the growth of -proteobacteria, which are closely related to known hydrocarbon degrading bacteria. Their grow th suggested faster than expected biodegradation at the low temperatures encoun tered in the deep sea. The density of all bacteria was found to be higher within the plume, but lower in taxonomic richness. This is consistent with studies that show oil and dispersants stimulate growth of select communities while causing toxic ity in others (NRC 2005). Generally, the Macondo well discharg e may have significant ongoing effects on many important gulf species. Fortunately, the majority of the oil discharged was kept both away from shorelines by prevailing winds and out of the loop current by a circling eddy (Hamacher 2010). The oil stil l impacted a very rich a nd diverse environment. Of particular concern is that the blowout was located in the mi ddle of the Atlantic bluefin tuna spawning ground and may have negatively a nd drastically affected the larvae of this already endangered species (National Comm ission on the BP Deepwater Horizon Oil Spill and Offshore Drilling 2011). The spill ma y also have long lasting impacts on gulf oysters, which were impacted coastally and co incidentally they spaw n in the late spring. The oil eventually covered 40 percent of th e offshore area known to be used by oyster 92

PAGE 103

larvae (National Commission on the BP D eepwater Horizon Oil Spill and Offshore Drilling 2011). The effects of this spill are likely to be ongoing for years (Schooner 2010). The exact effects of the use of dispersants in the deep sea ar e unknown as of now. Ultimately, the verdict on the use of disper sants in subsea application for biological benefits is still undetermined (Schooner 2010). 93

PAGE 104

References Abel PD. 1974. Toxicity of synthetic detergents to fish and a quatic invertebrates. J Fish Biol. 6 (3): 279-298. Ankley GT. 1994. Influence of ultraviolet light on the toxicity of sediments contaminated with polycyclic aromatic hydrocarbons Environl Toxicol Chem. 13 (11): 17911796. Aurand D, Coelho G, editors. 2005. Cooperative aquatic toxicity testing of dispersed oil and the "chemical response to oil spi lls: Ecological effects research forum (CROSERF).". Lusby MD: Ecosystem Ma nagement & Associates Inc, p. 105. Bond, CE. 1996. Biology of Fishes. 2nd edition. Fort Worth. Saunders College Pub. P. 750. Barron MG, Carls MG, Heintz R, Rice SD. 2004. Evaluation of fish early life-stage toxicity models of chronic embryonic expos ures to complex polycyclic aromatic hydrocarbon mixtures. Toxico l Sci. 78 (1): 60-67. Boese BL. 1997. Photoinduced toxicity of fluoranthene to seven marine benthic crustaceans. Arch Environ C ontam Toxicol. 32 (4): 389-393. Bruheim P, Bredholt HE, Eimhjellen K. 1999. Ef fects of surfactant mixtures, including corexit 9527, on bacterial oxidation of acet ate and alkanes in crude oil. Appl Environ Microb. 65 (4): 1658-1661. Burdeau C, Borenstein S. 13 Sep 2010. Wheres the oil? On the Gulf floor, scientists say (Online). United States: US News. Available: http://www.usnews.com/science/articles/2010/09/14/wheres-the-oil-on-the-gulffloor-scientists-say. [Accessed: 29 Apr 2011]. Camilli R. 2010. Tracking hydrocarbon plume transport and biodegradation at deepwater horizon. Science. 330 (6001): 201-204. Davis PH, Spies RB. 1980. Infaunal benthos of a natural petroleum seep: Study of community structure. Mar Biol. 59: 31-41. Diercks AR. 2010. Characterization of subsurface polycyclic aromatic hydrocarbons at the deepwater horizon site. Geophys Res Lett. 37: 1-6. Federal Interagency Solutions Group, Oil Budget Science and Engineering Team, Oil Budget Calculator Technical Docume ntation (November 2010). Available: http://www.noaanews.noaa.gov/stories 2010/PDFs/DeepwaterHorizonOilBudget2 0100801.pdf. [Accessed: 27 April 2011] 94

PAGE 105

Foght JM, Fairbain NJ, Westlake DWS. 1987. Effect of oil dispersants on microbially,mediated processes in freshwater systems. In: JH Vandermeulen and SE Hrudley, editors. Oil in freshwater: Chemistry, biology, countermeasure technology. Oxford, England: Pergamon Press, p. 252-263. Fuller C, Bonner J, Page C, Ernest A, McDonald T, McDonald S. 2004. Comparative toxicity of oil, dispersant, and oil plus dispersant to several marine species. Environ Toxicol Chem. 23 (12): 2941-2949. Garcia JM. 2001. Influence of the nonionic surf actant brij 35 on the bioavailability of solid and sorbed dibenzofuran. Environ Sci Technol. 35 (10): 2033-2039. Gertz E. 24 Jun 2010. Marine toxicologist susan shaw dives into gulf spill, talks dispersants and food web damage (Online). United States: A survival guide for the planet on earth. Available: http://www.onearth.org/bl og/marine-toxicologistsusan-shaw-dives-into-gulf-spill-t alks-dispersants-and-food-web-dam [Accessed: 28 Apr 2011]. Granmo Kollberg S. 1976. Uptake pathways and elimination of a nonionic surfactant in cod (gadus morrhua L.). Water Res. 10 (3): 189-194. Hamacher B. 19 Jul 2010. Wind Keeps Oil From Loop Current & Away From Florida Shores (Online). United States: NBC Miami. Available: http://www.nbcmiami.com/news/local/Wind-Keeps-Oil-From-Loop-CurrentAway-From-Florida-Shores-98757749.html. [Accessed: 29 Apr 2011]. Hazen TC. 2010. Deep-sea oil plume enriches in digenous oil-degrading bacteria. Science. 330 (6001): 204-208. Hemmer MJ, Barron MG, Greene RM. 2010. Coope rative toxicity testing of eight dispersant products on two Gulf of Mexico aquatic test species. Washington DC: US Environmental Protection Agency Office of Research and Development, p.1. Johnson J, Torrice M. 2010. BP's ever-growing oil spill. Chem Eng News. 88 (24): 1524. Joye S, MacDonald I. 2010. Offshore oceanic impacts from the BP oil spill. Nat Geosci. 3 (7): 446. Joye SB, MacDonald IR, Leifer I, Asper V. 2011. Magnitude and oxidation potential of hydrocarbon gases released from the BP oil well blowout. Nat Geosci. 4: 160164. Judson R, Linnenbrink M, Ka vlock R, Dix D. 2010. US EPA's oil spill dispersant screening results: Rapid te sting for potential endocrine related activity & cytotoxicity. Washington, DC: US Environmental Protection Agency, p.1. 95

PAGE 106

Klaassen CD. 2001. Casarett and doull's toxicol ogy: The basic science of poisons. 6th ed. New York: McGraw-Hill Professionals, p. 1236. Klemm DJ, Morrison GE, Norberg-King TJ P, W.H., Heber MA, editors. 2004. Shortterm methods for estimating the chronic toxicicty of efflue nts and recieving waters to marine life and estuarine orga nisms. Cincinnati: U.S. Environmental Protection Agency, p. 470. Kujawinski EB, Kido Soule MC, Valentine DL, Boysen AK, Longnecker K, Redmond MC. 2011. Fate of dispersants associated with the deepwater horizon oil spill. Envir Sci Technol. 45 (4): 1296-1308. Landrum PF, Geisy JP, Oris JT, Allred PM 1987. Photoinduced toxicity of polycyclic aromatic hydrocarbons to aquatic organisms. New York: Pergamon, p. 15. Linden O, 1975 Acute effects of oil and oil/dispersant on mixture on larvae of Baltic herring. Ann Zool Fenn 11: 141-148 Linden O. 1987. Effects of oil and oil dispersant on an enclosed marine ecosystem. Environ Sci Technol. 21 (4): 374-382. Lindstrom JE, Braddock JF. 2002. Biodegradation of petroleum hydrocarbons at low temperature in the presence of the di spersant Corexit 9500. Mar Pollut Bull. 44(8): 739-747. Literathy P, Haider S, Samhan O, Morel G. 1989. Experimental st udies on biological and chemical oxidation of dispersed oil in seawater. Water Sci Technol. 21: 845-856 Little C. 2000. The biology of soft shores and estuaries. The Biology of Habitats Series Oxford University Press: Oxford. 252 pp. Maltrud M, Peacock S, Visbeck M. 2010. On th e possible long-term fate of oil released in the deepwater horizon incident, esti mated using ensemble of dye release simulations. Environ Re s Lett. 5 (3): 1-7. Mulkins Phillips GJ, Stewart JE. 1974. Effect of four dispersants on biodegradation and growth of bacteria on crude oil. Appl Microbiol. 28(4): 547-552. Naclo. 14 Jun 2005. Safety data sheet: Product corexit EC9500A (Online). United States: Nalco. Available: http://www.lmrk.org/corexit_9500_uscueg.539287.pdf [Accessed: 30 Jan 2011]. Nalco. 15 October 2008. Safety data sheet: Produc t corexit EC9527A (Online) United States: Nalco. Available: 96

PAGE 107

http://www.doh.state.fl.us/chd/bay/Do cuments/Oilspill/Master_EC9527A_MSDS _539295.pdf. [Accessed: 30 Jan 2011]. National Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling. 2011. Deepwater: The gulf oil disaster and the future of offshore drilling. Washington, D.C.: Commission. National Incident Command Joint Analysis Group, Review of Preliminary Data to Examine Oxygen Levels In the Vicini ty of MC252#1: May 8 to August 9, 2010 (August 16, 2010). National Oceanic and Atmospheric Administra tion: Office of Response and Restoration. 2010. Bioaccumulation of oil chem icals in seafood. Available: http://www.doh.state.fl.us/chdokaloosa/Deepwater_Horizon/Media/May05/Bioac cumulationDWH1Pager.pdf. [Accessed: 27 April 2011]. National Research Council (U S) Committee on Effectiveness of Oil Spill Dispersants. 1989. Using oil spill dispersants on the sea. Washington DC: National Academy Press, p. 335. National Research Council (U .S.) Committee on Oil in the Sea: Inputs, Fates, and Effects. 2003. Oil in the s ea III : Inputs, fates, and effects. Washington DC: National Academy Press, p. 265. National Research Council (U .S.) Committee on Understandi ng Oil Spill Dispersants: Efficacy and Effects. 2005. Oil spill di spersants : Efficacy and effects. Washington, D.C.: National Academies Press, p.377. Nelson-Smith A. 1972. Oil pollution and marine ecology. London: Paul Elek Scientific Books Limited, p. 260. Nybakken JW, Bertness MD. Marine Bi ology: An Ecological Approach. San Francisco: Pearson/Benjamin Cummings. P. 570 Pelletier MC. 1997. Phototoxicity of individual polycyclic aromatic hydrocarbons and petroleum to marine invertebrate larvae and juveniles. Environ Toxicol Chem. 16 (10): 2190. Rogerson A, Berger J. 1981. The toxicity of the dispersant corexit 9527 and oildispersant mixtures to ciliate pr otozoa. Chemosphere. 10 (1): 33-9. Schippers C. 2000. Microbial degradation of phenanthrene by addition of a sophorolipid mixture. J Biotechnol. 83 (3): 189. Schooner JL. 2010. The gulf oil spill. Envir Sci Technol. 44 (13): 4833. 97

PAGE 108

Singer ME, Finnerty WR. 1984. Microbial metabolism of straight-chain and branched alkanes. New York: R.M. Atlas, Ed. Macmillan, p. 1-59. U.S. Environmental Protection Agenc y. 2003. 40 CFR 300 Appendix C. Available: http://www.epa.gov/oem/docs/oil/cfr/a ppendix_c.pdf. [Accessed 15 May 2011] US Environmental Protection Agency. 30 Mar 2011. National contingency plan product schedule toxicity and effectiveness summaries (Online). United States: United States Environmental Protection Agency. Available: http://www.epa.gov/osweroe1/content/ncp/tox_tables.htm [Accessed: 28 Apr 2011]. U.S. Environmental Protection Agency. 2010a. Guide to using the NCP product schedule notebook. Washington DC (US): United States Environmental Protection Agency. U.S. Environmental Protection Agenc y. 2010b. National contingency plan product schedule. Washington DC (US): United St ates Environmental Protection Agency, p. 1-25. U.S. Environmental Protection Agency. 2002. Me thods for measuring the acute toxicity of effluents and receiving waters to freshwater and marine organisms. Washington DC (US): United States E nvironmental Protection Agency, p. 1-275. Valentine DL. 2010. Propane respiration jump-starts microbial response to a deep oil spill. Science. 330 (6001): 208-211. Van Hamme JD, Ward OP. 1999. Influence of ch emical surfactants on the biodegradation of crude oil by a mixed bacterial cultur e. Canadian journal of microbiology 45.2 (1999):130. Varadaraj R, Robbins ML, Bock J, Pace S, MacDonald D. 1995. Dispersion and biodegradation of oil spills on water (Online). Available: http://www.iosc.org/papers/00544.pdf [Accessed 27 Apr 2011]. Watkinson RJ, Morgan P. 1990. Physiol ogy of alipathic hydrocarbon-degrading microorganisms. Biodegradation. 1: 79-91. Wolfe MF. 1998a. Effects of salin ity and temperature on the bi oavailability of dispersed petroleum hydrocarbons to the golden-brown algae, isochrysis galbana. Arch Environ Contam Toxicol. 35 (2): 268-273. Wolfe MF. 1998b. Influence of disp ersants on the bioava ilability of naphthalene from the water-accommodated fraction crude oil to the golden-brown algae, isochrysis galbana. Arch Environ Contam Toxicol. 35 (2): 274-280. 98

PAGE 109

Wolfe MF, Schwartz GJB, Singaram S, Mi elbrecht EE, Tjeerdema RS, Sowby ML. 2001. Influence of dispersants on the bioavailabi lity and trophic tran sfer of petroleum hydrocarbons to larval topsme lt (atherinops affinis). A quat Toxicol. 52 (1): 49-60. Wolfe MF, Schwartz GJB, Singaram S, Mi elbrecht EE, Tjeerdema RS, Sowby ML. 2000. Influence of dispersants on the bioa vailability and trophic transfer of phenanthrene to algae and rotifers. Aquat T oxicol. 48 (1): 13-24. Wolfe MF, Schlosser JA, Schwartz GJB, Singaram S, Mielbrecht EE, Tjeerdema RS, Sowby ML. 1998. Influence of dispersant s on the bioavailability and trophic transfer of petroleum hydr ocarbons to primary levels of a marine food chain. Aquat Toxicol. 42 (3): 211-27. Zhang Y. 1994. Effect of a pseudomonas rhamnolipid biosurfactant on cell hydrophobicity and biodegradation of octadecane. Appl Environ Microbiol. 60 (6): 2101-2106. 99

PAGE 110

Table 1. Toxicity data for select dispersa nts (From: U.S. Envi ronmental Protection Agency 2010a) 100

PAGE 111

Table 2. Results of EPA independent toxicity te sting of eight oil dispersants. Data for the inland silverside ( M. Beryllina ) compared to NCP information. (From Hemmer et al. 2010) 101

PAGE 112

Figure 1. The two pathways of photo-enhan ced toxicity for polycyclic aromatic hydrocarbons (From: NRC, 2005) 102

PAGE 113

Figure 2. Mechanisms of biological hydr ocarbon degradation (From: NRC 2003) 103

PAGE 114

CHAPTER 4 THE POLITICAL FRAMEWORK THAT SURROU NDED THE APPLICATION OF DISPERSANTS IN THE DEEPWATER HORIZON OIL SPILL 104

PAGE 115

Introduction: The Parties Involved The BP oil spill involved a wide array of actors both in terms of primary parties involved in the spill response and in terms of secondary parties affected by the spills broad reaching impacts (National Commissi on on the BP Deepwater Horizon Oil Spill and Offshore Drilling 2011). BP is the obvious primary responsible party. BP owned the rights to the well and was in charge of overs eeing drilling by its co ntractors and its own employees. Despite this, BP was not the only responsible indus try actor. The DWH drilling rig was owned and operated by Tr ansocean Limited, a drilling contractor. Contractors from Halliburton Corporation were in charge of drilling cement during drilling operations alongside workers from at least two other cont racting companies. The majority of the employees on the drilling rig were Transocean workers (National Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling, 2011). These employees worked alongside BP employees on the rig, and took orders from BP. Because Transocean was responsible for day-to-day operation on the rig, BP tried to pass the blame for the blowout onto them (Steffy 2011). BP claimed that Transocean had not followed proper drilling pr ocedures, which resulted in the blowout. BP was known to have a corporate culture th at did not properly value safety, which draws questions to its claims against Transocean. Transocean employees also testified that BP had ordered them to forgo typical safety measures to save money and time (Freudenburg and Gramling 2011). Ultimately, BP as the leas eholder was responsible for overseeing all safety and maintenance procedures in the well-drilling process, lending little credibility to their attempt to shift blame (Steffy 2011). 105

PAGE 116

BP also tried to shift the blame for th e blowout to Halliburton, the other major contractor on board the DWH (Steffy 2011) Halliburton was in charge of drilling cement. It was ultimately up to Halliburton to se al the well, and it is likely that a failure of this cementing resulted in the fatal blowout (National Commission on the BP Deepwater Horizon Oil Spill and Offshore Dr illing 2011). That said, a string of bad decisions from BP may have been responsible for the failure of the cement job. It was a BP engineer who ultimately decided to use t oo few centralizers, devices that hold the pipe in the center of the casing ensuring a sound cement job (Steffy 2011). BP laid blame on its two major contractors at the Macondo well and this shifting of blame became something of a pattern. The manufacturer of the failed blow out preventer, Cameron International Corporation, was also named as a respons ible party (National Commission on the BP Deepwater Horizon Oil Spill and Offshore Dr illing, 2011). Whether the design of the blowout preventer was flawed or if maintena nce prevented it from functioning properly remains the subject of uncertainty (Nati onal Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling 2011; Ste ffy 2011). Regardless of why, this final protection against blowout failed alongside ev ery other safety measure. As was the case with Transocean and Halliburton, Cameron International also found itself in legal action following the disaster (Bergin 2011). In the aftermath of the blowout, the shif ting of blame among industry actors was more consistent than any admissions of res ponsibility. The shifting of blame actually resembled the internal operations of BP to so me degree. In the two decades leading up to the blowout, BP had adopted a policy of shif ting managers frequently to avoid one person 106

PAGE 117

being labeled as responsible in the event of an accident (Steffy 2011). Similarly, it was very unclear just who was responsible fo r the Macondo blowout. A full year after the spill, BP is filing lawsuits against Transocean, Halliburton and Cameron International for their roles in the spil l (Bergin 2011). This latest action is an attempt by BP to make other responsible parties share the co st for the spill rather than a direct attempt to continue shifting the blame. Ultimately, regardless of any carelessness on the part of other industry actors, BP had the ultimate responsibility as the owner of the leas e under the provisions of the Oil Pollution Act of 1990. Under this act BP was also responsible for mitigating the damage of the spill and was therefore c ontinually involved in the decisions to use dispersants in the aftermath of the spill. Under the Oil Pollution act of 1990, the Un ited States government was required to be involved in and oversee the mitigation of this oil discharge (National Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling, 2011). A variety of government agencies became primary actors in the spill as a result. BOEMRE, the Department of the Interior agency in charge of offshor e drilling was responsible for overseeing and approving drilling operations and contingency planning (Steffy 2011). Suggestions of corruption within BOEMRE and close ties to industry were cited as contributing factors for the spill. Once the blow out occurred, other agen cies had to step in to assist in the mitigation process. The Framework of the Oil Pollution Act of 1990 establishes a hierarchical system of response teams (National Research Council (U.S.) Committee on Understanding Oil Spill Dispersants: Efficacy and Effects, 2005). At the top of this hierarchy is the National Response Team (NRT), which is co-chaired by the USCG and the EPA. This group is 107

PAGE 118

tasked with providing the framework for the NCP, which outli nes the appropriate response to oil spills. U nder the NRT, a Regional Response Team (RRT) provides regional planning for federal incidents and this team has authority to authorize the use of oil dispersants. Under the RRT, area committees provide more site specific planning. The decision to use dispersants must also be approved by the fede ral natural resource trustee agencies, the Department of Commerce and the Department of the Interior. The Oil Pollution Act (1990) is highly complex a nd involves many government actors coming into play. The NCP contains provisions to shor tcut the dispersant a pproval process, thus limiting the interactions of the ag encies to expedite decisions. In addition to the primary actors from industry and government, the far-reaching effects of the DWH accident brought many s econdary actors into play. The American Petroleum Institute (API) received scrutiny in the National Commission on the BP Deepwater Horizon Oil spill final report (2011). The API develops safety standards and procedures for the oil and gas industry in the United States. The Department of the Interior has historically adopted these pr ocedures as formal regulations. The API, however, is also the primary lobbyist for the oil industry, calling into question its ability to generate reliable safety sta ndards regardless of expense. The joint response to the oil spill involved the use of dispersants alongside many other oil spill response tec hniques. The dispersants BP chose to use were manufactured by the chemical engineering company, Nalco (Voith 2010). Nalco was not a primary actor in the spill response because it supplied BP with its products and their recommended use without directly influencin g their use in the spill. Nalco, like other actors from industry came under pressure from the public and government during the 108

PAGE 119

spill. EPA eventually forced the company to disclose the composition of its products (Johnson and Torrice 2010). Nalco fought back and attempted to prevent itself from becoming framed as a dispersant company, as dispersants account for a small portion of their product line (Voith 2010). The unprecedented amount of dispersants used in response to the DWH spill only boosted Nalcos profits by 1 percent. Private industries in the Gulf and the general public living along the Gulf Coast were heavily impacted by the BP DWH discharge (BBC World Service 2011a; BBC World Service 2011b). The fishing and tourism i ndustries in the Gulf were the hardest hit and tourism is only beginning to recover a year after the spill, but the total impact on the fishing industry has yet to be realized. Ma ny economically important Gulf species, such as oysters, may have been impacted in term s of their reproductio n prolonging the affects of the spill (National Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling 2011). Under th e Oil Pollution Act of 1990 BP is responsible for compensating all parties affected by the oil discharge. The extent of liability was questioned and policy makers in Washington sought to raise the cap set by the Oil Pollution Act (1990), which had not been updated since 1990. The impacts of the spill are still affecting communities a year after the spill and its damage is likely to be felt for some time (Bresenahan 2011). The affects of the spill on the health of re sponse workers and local populations who live and work in the effected areas are also questioned. Some workers directly blame the dispersants for impacting their health, but ther e is no direct evidence that the effects of the dispersants on their own caused health problems (National Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling 2011). 109

PAGE 120

Legal Framework The legal framework for the application of dispersants to oil spills can be split into response planning and emergency respons e. Both categories fa ll under the umbrella of the Oil Pollution Act of 1990. In res ponse planning, BOEMRE, a subset of the Department of the Interior, is responsible for overseeing industry in planning and preparedness (National Commission on th e BP Deepwater Horizon Oil Spill and Offshore Drilling, 2011). BOEMRE both approves drilling plans and emergency response plans, which industry is required to subm it (Steffy 2011). Response plans are evaluated on several parameters (National Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling 2011). First, th e action specified in the pl an must be effective and reasonable and include plans for dispersant use and burning. Industry is required to show proof of response equipment inventory sufficient to carry out th e actions specifi ed in their plan. If the industry actor itse lf requires assistance in ca rrying out effective response, contractual agreements must be shown. The re sponse plan must also include a calculation of the worst-case scenario possible from each well drilled. Under the Oil Pollution Act (1990), indus try is responsible for containing and mitigating any harm caused as a result of th eir activities (National Commission on the BP Deepwater Horizon Oil Spill and Offshore Dr illing 2011; NRC 2005). The act also states that government should be capable of handl ing emergency scenarios. This creates a difficult situation because the government is dependent on private industry for its technology, experience, and resources. Th e framework created under the Oil Pollution Act (1990) provides a network of appropriate response options to assist industry in 110

PAGE 121

mitigating spills. The framework also provides government with a plan of action should it need to assist. Applying dispersants to any oil spill begi ns with the hierarchical framework mentioned above involving the National a nd Regional Response Teams and the area committees (NRC 2005). In the event of a sp ill, an onsite USCG official will be designated the Federal on Scene Coordinator (F OSC), whose job is to ensure a safe and effective response. In order to apply disp ersants, the FOSC must deem chemical dispersion an appropriate response. To jus tify use, the FOSC mu st consider if the application of dispersants will have the desired effect, if dispersants can be affectively applied, and if the environmental tradeoffs of dispersant use are in favor of their application (NRC 1989). If the FOSC is in favor of applying disp ersants, the official must seek the approval of all part ies in the response framework. The USCG and EPA co-chairs of the NRT, the state representative of th e RRT and the federal resource trustees, the Department of Commerce, and th e Department of the Interior all must concur that the plan submitted by the FOSC is sound. Consulting each of the necessary parties is a time consuming process, which can hinder the ability of responders to enact an effective and rapid response (U.S. Environmental Protection Agency 2003; NRC 2005). The NRT revised the NCP to require the RRTs and Area committees to provide plans for preauthorization of dispersant use. The revised plan allowed for three modes of dispersa nt authorization. In waters meeting certain requirements, most frequently three nautical miles (nm) seaward and greater than 10 meters in depth, dispersant application has been preapproved (Figure 1). The dispersant products au thorized are limited to thos e approved by EPA and listed 111

PAGE 122

on the NCP Product Schedule (U.S. Environmen tal Protection Agency 2010). If a spill occurs in waters in which dispersants ha ve been preauthorized, the FOSC may allow application of dispersants immediately followi ng a spill without fu rther authorization. In situations that do not meet the requi rements of preapproval, two further options for dispersant application exist (U.S. Environmental Protection Agency 2003; NRC 2005). In some predetermined wate rs closer to shore than three nm and others that do not meet the requirements of pre approval, an ex pedited approval option exists. In this case the FOSC can receive approval from the NRT co-chairs and the federal resource trustee agencies but submitting a request of limited information. In areas that are not approved for pre authorization or expedi ted approval, the FOSC request for dispersant use must be approved through the entire command chain of NRT, RRT, and federal resource trustees on a case-by-case basis. In all cases of dispersant use, responde rs are limited to a number of preauthorized products listed on under the NCP, wh ich have been approved by EPA (U.S. Environmental Protection Agency 2003; NRC 2005). At the time of the DWH spill 14 dispersants were listed on the product schedule. In order to be listed, the ma nufacturers of products must submit standardized efficacy and toxicity data to EPA for approval. Among the products approved at the time of the Macondo blowout were Corexit EC9500A and EC9527A. These two products were selected by BP and applied under preauthorization until the NRT sought to limit th e application of di spersants (National Commission on the BP Deepwater Horiz on Oil Spill and Offshore Drilling 2011). In the Gulf of Mexico at the ti me of the DWH discharge two similar preauthorized zones existed in the Gulf of Mexico (United States Coast Guard 2011; 112

PAGE 123

NRC 2005). In Florida, Mississipp i, and Alabama, dispersant s were preauthorized three nm seaward to the extent of the exclusive eco nomic zone and in waters deeper than 10 meters. Case-by-case approval was needed for waters closer to shore or less than 10 m of depth. Case-by-case approval was also re quired for State and Federal special management areas including national marine sanctuaries, national or state wildlife refuges and national parks. Additionally, th e Florida west coast required case-by-case approval of dispersant shoreward of nine nm. Texas and Louisiana had similar, though slightly altered regulations. Di spersants were preauthorized in waters seaward of three nm to the exclusive economic zone or in wate rs of 10 m or more in depth, whichever was further from shore. Dispersant use was limite d to daylight hours and was to be based on weather conditions, sea state, th e type of oil, history of th e spill, and the risks involved. Furthermore, dispersants were always to be applied in the deepest possible waters. The one restriction for case-by-case approval in these states was application of dispersants within the Flower Gardens National Marine Sanctuary. Response to the Deepwater Horizon oil sp ill was flawed from the start (National Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling 2011). BP did not show sufficient attention to detail in their emergency planning, which Boemre approved without question. The response plan written up by BP was ineffective. It also underestimated the worst-case scenario, furt her undermining the planned response. This plan also drew heavy criticism for its poor analysis of environmental impact (Reed and Fitzgerald 2011). It did not express concer n or contingency plan ning for effects on endangered Gulf species, instead addressing impacts to animals including walrus, which do not inhabit the Gulf of Mexico. 113

PAGE 124

Dispersants were applied on the DW H spill beginning April 22, 2010 under preauthorization (National Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling 2011). Because the spill was located in a preauthorized zone, responders we legally allowed to use any EP A approved product without concern for the size of the treated area, volume of product us ed, or the duration of use (Figure 2). The application of dispersants sub-sea was somethi ng that had not been considered at the time of the Macondo blowout and was not covered by the NCP. Shortly after the blowout, BP began asking for approval to use dispersants at the well head. Scientists and federal responders were concerned with the unknown eff ects of dispersants at depth. Because the NCP did not address the method of dispersant application, respond ers were confused about the legality of sub-sea application. EPA claimed that s ub-sea use of dispersants was not covered by the NCP and that authoriza tion would require high-level approval. Communication problems within EPA slowed the response aut horization of sub-sea use, which angered the USCG and National Oc eanic and Atmospheric Administration (NOAA) that saw sub-sea appl ication as a useful tool. While awaiting EPA approval for subsea dispersant use between April 30th and May 10th, on the scene scientists developed a protocol for monitoring the success of subsea dispersant application (National Commi ssion on the BP Deepwater Horizon Oil Spill and Offshore Drilling 2011). On May 10th, EPA adopted the testing protocol and began allowing dispersants to be appl ied at depth. The application technique that BP developed was limited to the application of 15,000 gall ons of dispersants per day. Responders monitored toxicity during the su b-sea application and the toxi city levels never exceeded 114

PAGE 125

the guidelines set forth by EPA. Thus, sub-sea dispersant use was allowed to continue for the duration of the spill. An interesting twist in the story of dispersant use during the DWH spill was the conflict that it created amongst response au thorities (National Commission on the BP Deepwater Horizon Oil Spill and Offshore Dr illing 2011). By the end of May, EPA had become concerned with the quantity of dispersant being used in the Gulf of Mexico and ordered BP to curb their use by up to 75% (Voith 2010). The USCG, co-chair to the EPA on the NRT, and BP resisted the EPA direc tive on the grounds that dispersants were a highly effective tool that could not be removed from use (National Commission on the BP Deepwater Horizon Oil Spill and Offshor e Drilling 2011). BP began applying for approval on a case-by-case basis so frequently th at EPA threatened to stop dispersant use. This provoked a backlash from the USCG who complained that the EPA was revoking the rules it had created in the fi rst place. This situation was a bout to come to a head when the capping of the well diffused it. What Scientific Input Pertains? The creation of an effective policy re garding oil dispersants is dependent on expert input from a variety of fields. To begin, the fate of oil in the environment must be understood (NRC 1989, 2003, 2005). This process re quires oceanographers to understand the behavior of oil, chemists to underst and its breakdown, and biologists to understand the organisms that interact with oil. The dispersants themselves must be created and optimized by engineers. An understanding of how the product works and how it is best used must also come from engineers. Finally, oceanographers, toxicologists and 115

PAGE 126

ecologists are necessary to understand the effect s of the dispersant on the oil spill and its effects on the surrounding environment. The DWH spill was unprecedented in its application of dispersants by amount and location (Johnson and Torrice 2010). The policy guidelines for the legal application of dispersants before the spill were not sufficient and required modification. Subsequently, scientists from government, industry, and academia have responded to provide informa tion as to the effects of the spill and the effects of dispersants. Toxicologists worked during the spill to determine the effects of dispersants on wildlife (Hemmer et al. 2010; Judson et al. 2010). Oceanographers too have begun to unravel the affects of the disp ersant and consider how their application may have affected the environments dama ged by the spill (Kujawinski et al. 2011; Thibodeaux 2011). Strength of Positions The science and opinion behind the use of oil dispersants is split into two viewpoints. On one side, industry, the government and a wide array of scientists believe that dispersants are an effective tool when used properly (Johnson and Torrice 2010; National Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling, 2011; NRC 1989, 2005). Most actors will not cl aim that dispersants are a good thing; after all they are used in response to oil sp ills, which are ecologica l disasters. Rather, those in favor of dispersant use claim that they are the lesser of two evils and can help expedite recovery after an oi l spill. Some scientists and environmentalists on the other hand consider dispersants to be highly t oxic and a bad option in any circumstance (Torrice 2010b; Gertz 2010). Toxicologist Susan Sh aw even went as far as to claim that 116

PAGE 127

the use of dispersants was not a consideration of environmental tr adeoffs, but rather a tool to improve industry public relations by hiding the extent of the oil (Gertz 2010). The problem in reconciling the opinion of th e two viewpoints is that we are still missing a large amount of information on th e effects of dispersants (NRC 1989, 2005). The environmental, chemical, and biological variables effecting the application and effects of dispersants are so complex that it is impossible to achieve a thorough understanding of all variables i nvolved. The parties in favor of dispersant use could argue on the precautionary principle that the prevention of damage to coastal and surface communities without their use merits applicatio n despite the tradeoffs. The parties against dispersant use might use the precautionary pr inciple to claim that dispersants should not be used in any circumstances based on the in creased toxicity of dispersed oil and the unknown effects. In the end, it is clear that usin g dispersants is the choice between the lesser of two evils. That said, substantially mo re research has shown dispersants to be an effective tool despite their drawbacks and th at the potential increase in toxicity is outweighed by their benefits in most circumstances (Fuller et al. 2004; Judson et al. 2010; NRC 2005). The best source of information regardi ng oil dispersants a nd the policy behind their application comes from two Nati onal Research Council texts (NRC 1989, 2005). These texts thoroughly review the effects of dispersants on oil sp ills, their efficacy, toxicological and environmenta l effects. The 2005 text also contains a thorough analysis of the necessary decision making process in pr eparation for the applic ation of dispersants. Both texts analyze the tradeoffs and provide a fair and unbiased analysis of dispersant use. These sources are the closest one can co me to finding an Honest Broker opinion 117

PAGE 128

that considers tradeoffs and effects in the use of dispersants. Many academic peer reviewed papers and government reports provide impartial arbiter stance on the toxicity and utility of dispersants, but do not go as far as these NRC reports in providing a complete picture (Fuller et al. 2004 ; Hemmer Barron and Greene 2010). Control of Information Despite the reliable sources of information on dispersant use, actors involved in the BP DWH spill attempted to spin information in their favor. BP repeatedly used its scientists and engineers to downplay the extent of the spill (National Commission on the BP Deepwater Horizon Oil Spill and Offshor e Drilling 2011). Follo wing the spill, BP also attempted to claim that the Gulf would recover quickly w ithout properly weighing the extent of the possible impacts of the spill (BBC World Service 2011b). The time passed since this spill has been too short to allow the direct convers ion of science into policy, scientification of policy, but exam ples baring similarity may be found. When the EPA considered revoking the rights for BP to use dispersants, both BP and the USCG argued on the grounds that science had led to the current policy and that it should be followed (National Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling 2011). This approach attempted to bypass caution by EPA due to the potential unknown impacts of the unprecedented dispersant use. Beyond spinning science to benefit parties involved, there were also attempts by both BP and the federal government to limit information during the response. In the days following the blowout, BP sought to downplay the significance of th e spill consistently providing low flow rate estimates (Nati onal Commission on the BP Deepwater Horizon 118

PAGE 129

Oil Spill and Offshore Drilling 2011). In limiting the informati on about actual flow rate BP sought to limit damage to its public relati ons. Instead, this effort undermined attempts by BP to save its image and undermined early spill response (Shogren 2011). The US government was also complicit in limiting the fl ow of information. Independent scientists complained of having samples seized and barre d from working in certain areas after the spill (Flatow 2010; Hooper-Bui 2010). Supposedly this only occurred for scientists who had not been cleared by unified command to wo rk in certain areas, but it caused backlash within the scientific community and fueled wider distrust. Risk Assessment, Future Directions and Conclusions When considering the full effects of the spill in hindsight, it remains difficult to assess the impact and merit of using oil dispersants in unprecedented amounts and through new application types (National Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling 2011; Schooner 2010). Si gnificant risk assessment went into the creation of the Oil Pollution act of 1990 an d into the NCP created by EPA (National Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling 2011; U.S. Environmental Protection Agency 2011a; United States Coast Guard 2011). These analyses attempted to provide a reasonable framework for making the decisions between the tradeoffs of dispersant use. Very early in the spill, it became apparent that risk assessment had not kept pace with the deve lopment of drilling technology. As a result, responders lacked sufficiently developed plans for dispersant use and effective containment technology. Failures in risk as sessment began with the failure by BOEMRE to update analysis schemes to account for the advances in drilling technology. Without 119

PAGE 130

appropriate regulation, indus try preparedness also falte red. In its preparation of contingency plans, BP relied on the flawed and inadequate analyses provided by BOEMRE. BOEMRE was the government agency with the closest ti es to industry and should have been the most pr epared but as this agency fell behind, National Response planning lagged even further. The failure of risk assessment to keep pace with technology resulted in a need for emergency responders to adapt and perform in situ analyses during the response to the Macondo Blowout. Major changes need to be made in nati onal contingency planning and industry oversight to prevent another similar disaster from occurring. BOEMRE needs to provide better oversight of industry to ensure that gaps between contingency planning and technological advancement do no t occur to such a substan tial degree in the future (National Commission on the BP Deepwate r Horizon Oil Spill and Offshore Drilling 2011). The National Commission report on the BP oil spill suggests that an offshore authority be created as a new department within BOEMRE. This department would be solely responsible for ove rseeing the offshore drillin g industry providing focused oversight. The offshore authority should be responsible for upda ting and periodically reviewing risk analyses and the state of emergency preparedness. This enhanced regulation could prevent the occurrence of a similar incident. The offshore authority should also be responsible fo r creating a unified government response, in which it could lead and organize response from the EPA, the USCG, and NOAA. The lead of the offshore authority should incorporate the expe rtise of each agency, while providing the framework for a more unified and prepared response. Along with unified response, an interagency approach to risk assessment should be adopted to analyze oil spill risks, 120

PAGE 131

issues, and impacts. This unified analysis mu st be periodically updated and incorporate environmental reviews and outside consultations. Major changes also need to occur to the NCP to provide a better national response. EPA should create a plan for the handing of any oi l spills of national importance that differs from response to minor spills (Nationa l Commission on the BP Deepwater Horizon Oil Spill and Offshore Drill ing 2011). This plan should increase the level of government involvement in any spill that is a threat the public health and welfare. The increased level of government involvement should come from the National and RRTs. Clear plans need to set forth to provide easy clear communication with and enhanced roles for high level government officials with EPA and other involved agencies. This change could prevent confusi on over established polices as occurred over the preapproval of dispersants and the legality of their use in deep water. Enhanced communication with high-level officials will also provide expedited decision making on important matters. The NCP should also be modi fied to provide analysis of human health impacts during response. Considering the poten tial adverse health effects of burning oil and applying chemical dispersants in deci sion making could result in a safer and more appropriate spill response. The framework of the National Envir onmental Policy Act (NEPA) should be updated to include planning and analyses of the offshore dr illing industry and potential accidents (National Commission on the BP D eepwater Horizon Oil Spill and Offshore Drilling 2011). NEPA requires that the federa l government consider the impacts of its proposed operations on the environment, thus incorporating environmental values (U.S. Environmental Protection Agency 2010b). Adding provisions for the o ffshore oil industry 121

PAGE 132

could lower the risk in the i ndustry. It would also require th at regulators provide stronger emphasis on the environmental effects of a sp ill and its response. A stronger emphasis on environmental protection during a spill respons e can help stem some negative backlash and balance the needs of diverse interests. Changes need to begin being made in di spersant policies whilst larger changes occur. The EPA needs to periodically review its protocols for product listing on the NCP to ensure that responders have sufficient information to make decisions surrounding dispersants use (National Commission on th e BP Deepwater Horizon Oil Spill and Offshore Drilling 2011). Further changes need to be made to the NCP to include limiting guidelines on dispersant use including, maxi mum application time, area, and volume before further review is required. The NCP pr oduct schedule could th en be modified to include these guidelines for each product based on other parameters such as toxicity. To improve the efficacy of these changes further research needs to be conducted on dispersants and the effects of their use on e nvironmental tradeoffs, including their effects on bioremediation with special attention pa id to deep-water application (National Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling 2011; NRC 2005). The EPA and NOAA should have a role in this research, but independent scientists should continue to be involved. To ensure that adequate resear ch is conducted congress should also establish a fund for oil spill research. Improvement cannot come solely from government, especially when the government is reliant on industry for technological advances (National Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling 2011). Improveme nt in industry contingency planning and safety regulations together with gove rnment planning will help 122

PAGE 133

prevent another such accident and prepare a more appropriate response plan in the event that another unexpected accident occurs. Two in itiatives have come from industry so far to improve the safety of deep water drilli ng. Four out of the five major oil and gas corporations operating in th e Gulf of Mexico came toge ther to create a non-profit organization called The Marine Well Containm ent Company. BP has subsequently stated that it intends to join. The company will provide containment equipment capable of operating in up to 3000 m (10,000 ft) of wate r and collecting 100,000 ba rrels of oil per day. This equipment would be capable of be ing operational within 24 hours of a blowout. The startup members would have free access to the capability of the company as a result of a combined one billion dollars start-up cost. Others could gain access to this technology on a contractual basi s. A second more modest but less expensive proposal has been proposed by the existing Helix Corporation. Their proposal is to modify the equipment designed to contain the Macondo blowout for a more general use. Both of these proposals are promising. Each needs to ensure that its structure can adapt to respond to future technological changes with in the drilling industr y to guarantee longterm efficacy. In conclusion the established framework for oil spill response, including dispersant application, was not sufficient at the outset of the Macondo blowout (National Commission on the BP Deepwater Horizon Oil Spill and Offshor e Drilling 2011). The reason for this was that industry oversight had not kept pace with drilling advances. Lack of sufficient framework made on-site risk an alysis a necessity. Responders made the best decisions possible considering the tradeoffs in dispersant applica tion, including use at depth. The results of these acti ons are still not fully elucid ated. Continued monitoring of 123

PAGE 134

the effects of this spill and i nput from a variety of players in the scientific community can provide necessary information for future pol icy. Future policy needs to consider new scientific input alongsid e political and social tradeoffs to ensure that the best decisions are made in future accidents. A joint approach from governme nt working with industry is necessary to accomplish these goals. 124

PAGE 135

References BBC World Service. 18 Apr 2011. BP oil spill: Gulf flight marks year since disaster (Online). Available: http://www.bbc.co.uk/programmes/p00fwm6f. [Accessed: 18 Apr 2011]. BBC World Service. 2 Feb 2011. Gulf of mexico to recover from BP spill by end 2012 (Online). United Kingdom: BBC. Availa ble: http://www.bbc.co.uk/news/worldeurope-12352051. [Accessed: 28 Apr 2011]. Bergin T. 21 Apr 2011. Lawsuits fly in BP's gulf spill blame game (Online). United States: Reuters. Available: http:// www.reuters.com/artic le/2011/04/21/us-bphalliburton-idUSTRE73K1B820110 421. [Accessed: 21 Apr 2011]. Bresenahan R. 18 Apr 2011. Louisiana deepwaterBP gulf of mexico oil spill (Online). United Kingdom: BBC. Available: http ://www.bbc.co.uk/programmes/p00fwm6f. [Accessed: 18 Apr 2011]. Freudenburg WR, Gramling R. 2011. Blowout in the gulf : The BP oil spill disaster and the future of energy in america. Cambridge, Mass.: MIT Press, p. 254. Flatow I. 20 Aug 2010. Science in the gulf (Online). United States: Everyday is science Friday. Available: http://www.sciencefriday.com/program/archives/201008202 [Accessed: 10 Aug 2010]. Fuller C, Bonner J, Page C, Ernest A, McDonald T, McDonald S. 2004. Comparative toxicity of oil, dispersant, and oil plus dispersant to several marine species. Environ Toxicol Chem. 23 (12): 2941-2949. Gertz E. 24 Jun 2010. Marine toxicologist susan shaw dives into gulf spill, talks dispersants and food web damage (Online). United States: A survival guide for the planet on earth. Available: http://www.onearth.or g/blog/marine-toxicologistsusan-shaw-dives-into-gulf-spill-talks-d ispersants-and-food-web-dam. [Accessed: 28 Apr 2011]. Hooper-Bui L. 5 Aug 2010. Opinion: The oil's stain on science (Online). United States: The Scientist. Available: http://www.thescientist.com/templates /trackable/display/news.js ptype=news&id=57610&o_url= news/display/57610. [Accessed: 28 Apr 2011]. Hemmer MJ, Barron MG, Greene RM. 2010. Coope rative toxicity testing of eight dispersant products on two Gulf of Mexico aquatic test species. Washington DC: US Environmental Protection Agency Office of Research and Development, p.1. Johnson J, Torrice M. 2010. BP's ever-growing oil spill. Chem Eng News. 88 (24): 1524. 125

PAGE 136

Judson R, Linnenbrink M, Ka vlock R, Dix D. 2010. US EPA's oil spill dispersant screening results: Rapid te sting for potential endocrine related activity & cytotoxicity. Washington, DC: US Environmental Protection Agency, p.1. Kujawinski EB, Kido Soule MC, Valentine DL, Boysen AK, Longnecker K, Redmond MC. 2011. Fate of dispersants associated with the deepwater horizon oil spill. Envir Sci Technol. 45 (4): 1296-1308. National Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling. 2011. Deepwater: The gulf oil disaster and the future of offshore drilling. Washington, D.C.: Commission. National Research Council (U S) Committee on Effectiveness of Oil Spill Dispersants. 1989. Using oil spill dispersants on the sea. Washington DC: National Academy Press, p. 335. National Research Council (U .S.) Committee on Oil in the Sea: Inputs, Fates, and Effects. 2003. Oil in the s ea III : Inputs, fates, and effects. Washington DC: National Academy Press, p. 265. National Research Council (U .S.) Committee on Understandi ng Oil Spill Dispersants: Efficacy and Effects. 2005. Oil spill di spersants : Efficacy and effects. Washington, D.C.: National Academies Press, p.377. Reed S, Fitzgerald A. 2011. In too deep : BP and the drilling race that took it down. Hoboken NJ: Bloomberg Press, p. 226. Progressive Management. 2011a. US departme nt of the interior deepwater horizon response and restoration. Restori ng the gulf [CD-ROM]. United States: Progressive Management. Steffy LC. 2011. Drowning in oil: BP and the reckless pursuit of profit. New York: McGraw-Hill, p. 285. Schooner JL. 2010. The gulf oil spill. Envir Sci Technol. 44 (13): 4833. Shogren E. 21 Apr 2011. BP: A textbook example of how not to handle PR (Online). United States: National Public Radio. Available: http://www.npr.org/2011/04/21/135575238/bp-a-textbook-example-of-how-notto-handle-pr. [Accessed: 21 Apr 2011]. Thibodeaux LJ. 2011. Marine oil fate: Knowledge gaps, basic research, and development needs A perspective based on the deepwate r horizon spill. Environ Eng Sci. 28 (2): 87-93. 126

PAGE 137

Torrice M, Voith M. 7 May 2010. New oil clean-up technique tested (Online). United States: Chemical and Engi neering News. Available: http://pubs.acs.org/cen/news/88/i1 9/8819notw1.html. [Accessed: 24 Apr 2011]. U.S. Environmental Protection Agency 2003. 40 CFR 300.910 920. Available: http://www.epa.gov/oem/docs/oil/cfr/900_920.pdf. [Accessed 15 May 15, 2011] U.S. Environmental Protection Agency. 2010a. Guide to using the NCP product schedule notebook. Washington DC (US): United States Environmental Protection Agency. U.S. Environmental Protection Agenc y. 2010b. National contingency plan product schedule. Washington DC (US): United St ates Environmental Protection Agency, p. 1-25. USCG. Vessel response plan program (Online). United States: United States Coast Gaurd. Available: https://homeport.uscg.mil/mycg/porta l/ep/channelView.do?channelId=30095&channelPage=%252Fep%252Fchannel%252Fdefault.jsp&pageTypeId=13 489. [Accessed 27 Apr 2011]. 127

PAGE 138

Figure 1. Dispersant pre-authorization map: US Coastlines. Note preapproval of dispersants in the Gulf of Mexico 3 nautical miles from coasts in waters greater than 10 meters (From: NRC, 2005). 128

PAGE 139

Figure 2. BP dispersant use decision tree. This diagram was part of the BP incident response plan submitted to BOEMRE (From: Progressive Mangament 2011a). 129

PAGE 140

CHAPTER 5 CONCLUSIONS ABOUT THE USE OF CHEMICAL DISPERSANTS IN MARINE SPILLS 130

PAGE 141

The first large scale attempt at using oil dispersants was in 1967 (NRC 1989). A Liberian tanker named the Torrey Canyon ra n aground on the Western Cornish Coast of England releasing roughly one million barrels of crude oil into the English Channel. This event was one of the first large scale mari ne oil spills and impacted over 200 km of English coast and 80 km of French coastl ine (Rogowska and Namiesnik 2010). Over 10,000 barrels of various dispersants were a pplied to aid in br eaking up the massive slicks of oil (NRC 1989). The dispersants applied had low success, as the solvents quickly evaporated and the su rfactants remained. These did not mix well with seawater and instead they formed stable and persistent emulsions of surfactan t, oil and water. The dispersants used were highly toxic to marine life, and the most toxic dispersants formed the most persistent emulsions. Subsea biolog ical impacts were not well studied at this time, but impacts to the coast line were highly visible, in cluding the presence of dead limpets and empty barnacle and mussel shells fo r miles. The toxicity of the dispersants was attributed to the aromatic hydrocarbons in the solvents and alkyphenol surfactants (NRC 1989). This first major application of di spersants gave them a bad reputation that continues up to the present. Further damage to the environment occurred when the British Air Force unsuccessfully attempted to burn th e oil by bombing oil slicks with napalm (Jenelov 2010). Tanker spills were of great conc ern during the second half of the 20th century (Rogowska and Namiesnik 2010). One of the most famous incidents was the highly publicized Exxon Valdez spill in Prince William Sound, Alaska, (Exxon Valdez Oil Spill Trustee Council 1994). This spill is now the second largest accident al spill in United States history, following the DWH (Nationa l Commission on the BP Deepwater Horizon 131

PAGE 142

Oil Spill and Offshore Drilling 2011). Th is spill had profound and long lasting environmental effects that have been we ll documented elsewhere (Jenelov 2010; Exxon Valdez Oil Spill Trustee Council 1994). Oil disper sants were not used to a large degree in this oil spill. Poor weather conditions preven ted the application until it was deemed that dispersants would not be usef ul due to oil weathering. Disp ersants were used in the process of cleaning rocky shorelines. Two Corexit products were used on beaches, the dispersant 7764 and the surface-washing ag ent 9580. It was believed that shoreline application would remove oil from coated beaches by facilitation dissolution into the surf. Dispersants were not highly successful in this applicati on and caused further concern due to the large amounts of kerosene solvents they contained, which effectively increased the oil load on beaches. This effect led to the discontinued use of Corexit dispersant products in shoreline application. Today, the use of these dispersants is discouraged within three mile s of the shore (Voith 2010). As of the late 20th century, tanker spills trended downwards as improved regulations enhanced safety r ecords (Jernelov 2010). Ruptures in pipelines and in marine drilling accidents began to tr end upwards during this same period as subsea drilling began in earnest in the late 1950s (Jenelov 2010; Steffy 2011). The most comparable oil spill to the DWH, is the 1979 Ixtoc I blowout in the Bay of Campeche, Mexico (Figure 1) (Waldichuk 1980). The Ixtoc I well was located in 50 meters of water 80 km from the Yucatan Peninsula on the southern Gulf of Mexico continental shelf. The blowout initially discharged 30,000 barrels of oil pe r day and after months of flow over three million barrels of crude oil were released before a relief well finally killed the blowout. This was the largest oil spill in history at that time (Lee et al.1980). 132

PAGE 143

Response to the Ixtoc appears to be co mparable to the DWH incident, which was caused by a similar event (Jernelov 2010). Ma ny failed measures were tried before a relief well finally stopped the discharge (W aldichuk 1980). Responders attempted to top kill the well with drilling mud and even added tennis-ball sized lead spheres to the drilling mud at one point, which managed to slow the flow. Another measure that slowed the flow was the placement of a 3010 ton cont ainment dome referred to as a sombrero over the wellhead. Similar techniques were used in responding to the Macondo Blowout, but also failed (National Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling 2011). The oil discharged from the Ixtoc blowout was a relatively light crude (Jernelov 1981). Despite this fact, and th e relatively shallow depth of the well, the oil that reached the surface was found to be pa rtially emulsified. This emulsification is a major concern in deepwater blowouts becau se it lowers the efficacy of response. Responders used most available methods to clean up after the Ix toc spill including burning, containment booms to protect shore lines and lagoons, and dispersants. Over 9000 metric tons (60,000 barrels) of dispersant were used including 6750 tons of Corexit EC9527A (Jernelov 1979). Response saw some success, but the magnitude of this spill meant environmental damage would occur. The spill threatened Gulf of Mexico ecosystems from the continental shelf, including flat uniform sediment bottoms inha bited by shrimp, invertebrates, mollusks and fish, to the pelagic deep-water environmen t, which is among the most productive in the world due to upwelling and nutrient inflow fr om rivers (Jernelov 1979). Coastlines also contained coral reefs and mangrove habitats. Th e oil released had acute toxic effects on many inhabitants and the shrimp, fish and oc topus harvest dropped for many years after 133

PAGE 144

the spill. Studies found low evidence of b acterial degradation following the spill (Atwood 1982; Lizarraga Partida 1982). The microbial community present ha d the capability to consume oil, but was limited by other nutrients and sediment inflow from rivers. In the years following the spill, la rge bacterial blooms were not ed along the coast (Jernelov 1979). Zooplankton were shown to decrease in magnitude for at least three years following the spill and population structures we re also dramatically altered (Guzman del Proo 1986). Though this spill occurred on the continental shelf, its impacts are the best basis for understanding the long-term consequences of the DWH spill. Despite the upward trend in accidents subsea drilling has had a relatively good safety record (Steffy 2011; Jernelov 2010). Ov er the last half cen tury more than 50,000 subsea wells have been drilled in the US Gulf of Mexico, including over 700 in deepwater below 1500 meters (500 ft). Far mo re wells have been drilled worldwide, including in third world count ries with poor safety record s and yet few major accidents have occurred. Aside from the major Ixtoc I blowout of 475,000 metric tons and the Macondo blowout of 672,000 metric tons, only one other blowout has resulted in a spill of over 100,000 metric tons the Nowruz well in 1983 (Table 1) (Federal Interagency Solutions Group 2010; Jernelov 2010). That blowout; however, is an anomaly as it occurred in Iranian waters after Iraqi airc raft attacked the platform. Following these spills, the other major subsea blowouts we re all below 30,000 metric tons of oil and occurred at shallow water depths (Jernelov 2010). Blowouts of note from this list include a Norwegian Spill in 1977, a Nigerian spil l in 1980 and an Australian spill in 2009. Within the US Gulf of Mexico there were only five blowouts that exceed 1,000 barrels of oil spilled between 1964 and 2009 and the most recent occurred in 1970 (Table 2) 134

PAGE 145

(Progressive Management 2011). With so few accidents to study, most scientific knowledge about deep-sea spills is theoretical or has come from field tests (Johansen 2003; NRC 2003). One study in the North Sea simulated blowouts and pipeline leaks in 840 meters of water, which helped scientists to better understand the behavior of oil in deep water, including slick and emulsion formation (Johansen 2003). For now, the DWH is the only example of an ultra deep-water well blowout and the largest accidental marine oil discharge in history at nearly five million barrels (Table 3; Figure 2) (Federal Inte ragency Solutions Group 2010; National Commission on the BP Deepwater Horizon Oil Spill and Offshore Dril ling 2011). For scale, five million barrels is 210 million gallons. This spill is also the only example of subsea dispersant use and the use of dispersants on such a large scale (Kujawinski 2011; National Commission on the BP Deepwater Horizon Oil Spill and Offshor e Drilling 2011). In total, 43,900 barrels or about two million gallons of di spersants were applied. It is still too early to understand the full environmental and biological impacts of the DWH spill. The government oil spill commission seems to believe that many of th e worst feared impacts have not come to fruition, most of the oil stay ed out of the loop current. Th ere is little evidence for widespread oxygen depletion in deep water and most of the oil has disappeared from the surface and the mind of the public after the spill. Independent sc ientists are more hesitant to express optimism before we have all the facts (Joye and M acdonald 2010; Schooner 2010). The effects of dispersant use remains to be fully understood, but for now it appears that they served their intended purpose in removing oil from the surface and sequestering it in the deep sea, possibly e nhancing microbial degr adation (Hazen 2010; Schooner 2010; Kujawinski 2011). Whether thes e assumptions are accurate or not, the 135

PAGE 146

spill will impact biological comm unities for some time to come and we still do not know to what extent, especially for deep sea comm unities. The oil, provide d it is not too dilute or degraded, may yet reach European shores from transport in the gulfstream (Figure 3) (Jernelov 2010; Maltrud 2010). Our current state of knowledge about oil di spersants leads to the conclusion that they are useful tools in appr opriate circumstances, but that their use contains intrinsic tradeoffs (NRC 1989, 2005). Dispersant will always add chemical weight to a spill. Chemically-dispersed oil toxicity is prim arily driven by the co mponents of oil. The dispersant affects the distribution of oil, which may impact certain communities more than others. Highly complex and precise ecol ogical studies need to be conducted on the impacts of dispersants and the decision to use these chemicals in re sponse to oil spills. Otherwise, we will rely on assumptions, the value of protecting shorelines, and highly visible animals such as marine mammals and seabirds over inhabita nts of water column. The question of enhanced petroleum bioremed iation through dispersant use also remains a point of uncertainty (NRC 2005). A better understanding of biodegradation needs to come from studies that examine multiple parameters of oil decomposition, including oxygen depletion, the variability in biological communities, and chemical composition of oil. Despite our relatively paltry knowledge of oil spill mitigation, some experts believe that research and development on mitigation techniques is money and time wasted (Jernelov 2010). The magnitude of oil that can be released in these events is so great that any technology will unlikely be able to prevent severe environmental damage. Due to the potentially catastrophic nature of subsea blowouts, some scien tists believe that research 136

PAGE 147

and development priorities should be in the prevention of blowouts and the development of technology for capping them, if they occur. Regardless of safety concerns, drilling in deep water is on the rise throughout the world (Minerals Management Service 2004). W ith global energy use continuing to grow, oil drilling is likely to increas e and venture into even deep er waters. In the Gulf of Mexico, deep-water production has increased an average of 16 percent per year since 1985 (BOEMRE 2011b). As of April 27, 2011 ther e were over 500 l eases and 22 active drilling and production sites in th e Gulf deeper than 300 meters (1000 ft) (Table 4; Figure 4) (BOEMRE 2010, 2011a). The only way to prevent environmental damage from oil spills is to prevent them fr om occurring. The only way to ensure prevention is to stop drilling. Drilling will not stop until the world is willing to find other sources of energy production, pay more for sustaina ble energy, or oil production becomes as costly as other forms of energy. 137

PAGE 148

References Atwood DK. 1982. An example study of the weathering of spilled petroleum in a tropical marine environment: IXTOC-1. Bulletin of marine science 32(1) (1982):p.1. BOEMRE. 25 Apr 2011a. Current deepwater activity (Online). United St ates: Bureau of Ocean Energy Management, Regulation and Enforcement. Available: http://www.gomr.boemre.gov/homepg/offshor e/deepwatr/Current_Deepwater_Ac tivity.pdf. [Accessed: 27 Apr 2011]. BOEMRE. 28 Apr 2011b. Deepwater production summary by year (Online). United States: Bureau of Ocean Energy Management, Regulation and Enforcement. Available: http://www.gomr.boemre.gov/homepg/ offshore/deepwatr/summary.asp. [Accessed: 28 Apr 2011]. BOEMRE. 9 Sep, 2010. Deepwater natural gas and oil field discoveries (Online). United States: Bureau of Ocean Energy Management, Regulation and Enforcement. Available: http://www.gomr.boemre.gov/homepg/o ffshore/deepwatr/deeptbl2.html. [Accessed: 27 Apr 2011]. Exxon Valdez Oil Spill Trustee Council, Loughlin TR. 1994. Marine mammals and the exxon Valdez. San Diego: Academic Press, p. 395. Federal Interagency Solutions Group, Oil Budget Science and Engineering Team, Oil Budget Calculator Technical Docume ntation (November 2010). Available: http://www.noaanews.noaa.gov/stories 2010/PDFs/DeepwaterHorizonOilBudget2 0100801.pdf. [Accessed: 27 April 2011] Guzman del Proo SA, Chavez EA, Alatriste EA, de la Campa S, De la Cruz G, Gomez L, Guadarrama R, Cuerra A, Mille S, Torruco D. 1986. The impact of the Ixtoc-1 oil spill on zooplankton. J Plankton Res. 8(3): 557-581 Hazen TC. 2010. Deep-sea oil plume enriches in digenous oil-degrading bacteria. Science. 330 (6001): 204-208. Jernelv A. 1981. Ixtoc I: a case study of the world's largest oil sp ill. Ambio 10(6) p.299. Jernelv A. 2010. The threats from oil spills: No w, then, and in the future. Ambio. 39 (6): 353-366. Johansen 2003. Development and verifica tion of deep-water blowout models. Mar Pollut Bull. 47 (9-12): 360-368. 138

PAGE 149

Joye S, MacDonald I. 2010. Offshore oceanic impacts from the BP oil spill. Nat Geosci. 3 (7): 446. Lizarraga Partida ML. 1982. E ffects of the Ixtoc I blowout on heterotrophic bacteria. Marine pollution bulletin 13(2) p.67. Lee WY, Morris A, Boatwrite D. 1980 Mexican oil spill: a toxi city study of oil accommodated in seawater on marine invertebrates. Marine pollution bulletin 11(8) p.231. Maltrud M, Peacock S, Visbeck M. 2010. On th e possible long-term fate of oil released in the deepwater horizon incident, esti mated using ensemble of dye release simulations. Environ Re s Lett. 5 (3): 1-7. Minerals Management Service. 2004. Deep water: Where the energy is (Online). United States: Bureau of Ocean Energy Management, Regulation, and Enforcement. Available: http://www.boemre.gov/Assets/Press Conference11152004/MSGlossySingle_1104 04.pdf. [Accessed: 29 Apr 2011]. National Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling. 2011. Deepwater: The gulf oil disaster and the future of offshore drilling. Washington, D.C.: Commission. National Research Council (U S) Committee on Effectiveness of Oil Spill Dispersants. 1989. Using oil spill dispersants on the sea. Washington DC: National Academy Press, p. 335. National Research Council (U .S.) Committee on Oil in the Sea: Inputs, Fates, and Effects. 2003. Oil in the s ea III : Inputs, fates, and effects. Washington DC: National Academy Press, p. 265. National Research Council (U .S.) Committee on Understandi ng Oil Spill Dispersants: Efficacy and Effects. 2005. Oil spill di spersants : Efficacy and effects. Washington, D.C.: National Academies Press, p.377. Progressive Management. 2011. US departme nt of the interior deepwater horizon response and restoration. Restori ng the gulf [CD-ROM]. United States: Progressive Management. Rogowska J, Namiesnik J. 2010. Environmental implications of oil spills from shipping accidents. Environ Contam Toxicol. 206: 95-114. Schooner JL. 2010. The gulf oil spill. Envir Sci Technol. 44 (13): 4833. 139

PAGE 150

Steffy LC. 2011. Drowning in oil : BP and the reckless pursuit of profit. New York: McGraw-Hill, p. 285. Voith M. 2010a. Oil spill leads to fame and fury for makers of dispersant chemicals. Chem Eng News. 88 (24): 22-23. Waldichuk, M. 1980. Retrospect of the Ixtoc I blowout. Marine pollu tion bulletin 11 (7): 184. 140

PAGE 151

Table 1. Notable global blowouts prior to th e Deepwater Horizon (From: Jernelov 2010). 141

PAGE 152

Table 2. Notable blowouts in the U.S. Gu lf of Mexico 1947-2009 (From: Progressive Management 2011). 142

PAGE 153

Table 3. Deepwater Horizon oil budget (Fro m: Federal Interagency Solutions Group 2010) 143

PAGE 154

Table 4. Petroleum drilling and production operati ons in the Gulf of Mexico as of April 27, 2011 (From: BOEMRE 2011a) 144

PAGE 155

Figure 1. The locations of the two major Gulf of Mexico blowouts (From: Jernelov 2010) 145

PAGE 156

Figure 2. The actual surface extent of the Deepwater Horizon spill (From: National Commission on the BP Deepwater Horiz on oil spill and offshore drilling 2011) 146

PAGE 157

Figure 3. Theoretical extent spilled oil by de pth six months after the blowout, based on dye modeling. (a) dye release at 0 m depth; (b) dye release at 20 m depth; (c) dye release at 210 m depth; (d) dye release at 800 m depth. Color represents dilution factor on a logarithmic s cale. (From: Maltrud et al. 2010) D C B A 147

PAGE 158

Figure 4. Oil and gas leases in the Gulf of Mexico as of May 12, 2010 (From: Progressive Management 2010a) 148

PAGE 159

APPENDIX A 149

PAGE 160

The toxicity and efficacy summary for each of the 14 dispersant products listed on the EPA National Contingency Plan (From: U.S. Environmental Protection Agency 2011) Toxicity(LC50 values in ppm) Effectiveness (%) Product (1:10 Product-to-No. 2 Fuel Oil ratio) Menidia (96hr) Mysidopsis (48hr) Prudhoe Bay Crude Oil South Louisiana Crude Oil Average of Crude Oils BIODISPERS 5.95 2.66 51.00 63.00 57.00 COREXIT EC9500A 2.61 3.40 45.30 54.70 50.00 COREXIT EC9527A 4.49 6.60 37.40 63.40 50.40 DISPERSIT SPC 1000 7.90 8.20 40.00 100.00 73.00 FINASOL OSR 52 5.40 2.37 32.50 71.60 52.10 JD 109 3.84 3.51 26.00 91.00 58.50 JD 2000 3.59 2.19 60.40 77.80 69.10 MARE CLEAN 200 42.00 9.84 63.97 84.14 74.06 NEOS AB3000 57.00 25.00 19.70 89.80 54.80 NOKOMIS 3 AA 7.03 5.56 63.20 65.70 64.50 NOKOMIS 3 F4 100 58.40 62.20 64.90 63.55 SAFRON GOLD 9.25 3.04 84.80 53.80 69.30 SEA BRAT #4 23.00 18.00 53.55 60.65 57.10 SEACARE ECOSPERSE 52 (see FINASOL OSR 52) 5.40 2.37 32.50 71.60 52.10 SEACARE E.P.A. (see DISPERSIT SPC 7.90 8.20 40.00 100.00 73.00 1000) SF GOLD DISPERSANT (see SAF RON 9.25 3.04 84.80 53.80 69.30 GOLD) SUPERSPERSE WAO2500 3.70 2.53 77.84 87.56 82.70 ZI 400 8.35 1.77 50.10 89.80 69.90 150

PAGE 161

151 Toxicity(LC50 values in ppm) Effectiveness (%) Product (1:10 Product-to-No. 2 Fuel Oil ratio) Menidia (96hr) Mysidopsis (48hr) Prudhoe Bay Crude Oil South Louisiana Crude Oil Average of Crude Oils ZI 400 OIL SPILL DISPERSANT (see ZI 400) 8.35 1.77 50.10 89.80 69.90