National Academies Press: OpenBook

Oil in the Sea: Inputs, Fates, and Effects (1985)

Chapter: 2. INPUTS

« Previous: 1. CHEMICAL COMPOSITION OF PETROLEUM HYDROCARBON SOURCES
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 43
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 44
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 45
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 46
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 47
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 48
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 49
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 50
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 51
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 52
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 53
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 54
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 55
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 56
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 57
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 58
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 59
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 60
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 61
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 62
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 63
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 64
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 65
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 66
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 67
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 68
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 69
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 70
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 71
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 72
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 73
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 74
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 75
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 76
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 77
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 78
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 79
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 80
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 81
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 82
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 83
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 84
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 85
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 86
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 87
Suggested Citation:"2. INPUTS." National Research Council. 1985. Oil in the Sea: Inputs, Fates, and Effects. Washington, DC: The National Academies Press. doi: 10.17226/314.
×
Page 88

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

J Inputs Petroleum hydrocarbons (PHC) sources. Estimates of these sources are interrelated and INTRODUCTION enter the marine environment from many PHC inputs remain uncertain because the ~ available data are minimal. Figure 2-1 shows the international flow of petroleum. The width of that flow is representative of the amount of petroleum being transported along these routes. This pattern of flow may change significantly in future years, particularly in arctic areas where petroleum production is increasing. A major fraction of the world's petroleum continues to be produced and transported from countries different from those in which the petrol- eum is refined and consumed. During the past decade the quantity of petroleum transported by sea, as well as the number and tonnage of ships in operation, has increased significantly (British Petroleum Company, Ltd., 1980; Lloyd's Register of Shipping, 1980~. This increase is shown in Table 2-1. Sources of PHC into the marine environment considered in this report include natural sources; offshore oil production; marine transportion (operational discharges, drydocking, marine terminals, bunker operations, bilge and fuel oil transfer, and accidental spillages); the atmosphere; coastal, municipal, and industrial wastes and runoff; and ocean dumping. Each source type will be addressed in the following sections. NATURAL SOURCES The direct input of PHC from natural sources is estimated to be 0.025-2.5 million metric tons per annum (mta), the best estimate being 0.25 mta. Natural seeps contribute the major fraction of this total. A minor contribution is estimated to come from erosional processes. These consensus estimates, developed at the 1981 workshop, are based on geological and geochemical principles, many of which were described by Wilson et al. (1973~. In this report on natural sources, hydrocarbons of a petroleum origin are the only ones considered. Biogenically produced hydrocar 43

Ad\ - ( arm. ~ :. ~ ~- ~ ~- c . &~ . ~ v - . - . , t........................................... , ~ , \< . ~,, \ .~ > ). I _ ~ - ~ ~ aim, . W. , ~ - ~\ · Ji. ~ \ : \ #, ~. . Rae ~ ~ \ ~ I, ~, ~ . _ ~- ~ I : \ W .. ....22.2'.'.\ · -~ t~ ........... .\ \ 2 ·\ t................ . J I............ [ ' I.......................................... :'.~) \ ~I'"''''2''22'''22'"'''2''} ~- <~\ ~N ~/;~ 22/h / ~ ~4 \ ; t ~/ ~p , ~\ ~ ~ ,lll / / ~ A ~ ~ \ \ D [] ~. ~: / I.' .:22.2 ~\ ~ i'.! ~' '.2 2 :,.,.,,' I' [., 2 ~ ~ ~ I' an 2.2 2 2 : ' ' I] I,, ~I] I. ..... 2. 2'.2.: 2 ". ~I ! t ~\ A .. {} t ~\ \ ' ' ' /,' ,,., : ~\ ' ,, ~ ..... ~\ ~: 22 2 ~// : :. . H ~ ~ . ,.f:, .................... ~2' 2 ' ........... \ ~ ~, ' .~/ & ~\ \ < 2 2 ' ~ 22 2 2 " ? \ If ~ . ~\\ ~\ v. \.~ p7' ~ V'~' 96 :'~.\~> 'W' ~N'-~, W:,~.: ~-~: 'WN -~ ~: . ' ~\ 3 O O ~ o ~D - U] 3 O O . - O ~C) O U) 3 O S" . - ~n · - · - m w C' H ·e C) ~: o U)

45 TABLE 2-1 Petroleum Tr anspor t at Sea 1971 1980 Ratio. 1980/197 1 O it movement at sea (mta) Crude oil Product oil Total World' s merchant fleet Number of sh ips Gross tonnage Wor Id ' s tanker f feet Number of sh ips Total deadwe igh t tons Average deadweight tons 1,100 255 1,355 55, 041 247,200,000 6, 292 169, 355, 000 26,900 1,319 269 1,588 73,832 419,911,000 7,112 339 ,802,000 47,800 1.19 1.05 1.16 1.34 1.70 1.13 2.01 1.78 loons, some of which have the same chemical structure as some PHC (e.g., n-alkanes and isoprenoid alkanes), are synthesized by marine organisms (see Chapter 3, Chemical Methods section). Petroleum hydrocarbons, considered here as liquid petroleum and tar ~ ,~ ~ _ , _ or more carbon atoms), enter the marine environment naturally by means of two main processes--submar ine seepage and erosion of sedimentary rocks. Estimating the contribution of each of these is a formidable problem for the following reasons: (hydrocarbons and other organic compounds with five 1. Direct observation of submarine seeps is limited because the seeps are not normally visible. This invisibility leads to inaccurate estimates of seepage rates. 2. Submarine seeps flow intermittently, thus complicating both detection and estimation of seepage rates. The estimate is an average over geologic time, and in any particular year seepage events can exceed this estimate by orders of magnitude. 3. The potential area of continental margins where submarine seeps can occur is vast, whereas the areas of individual seepages are usually small, making an adequate inventory impossible with current technology and available monetary resources. In addition, the products of seepage cannot always be distinguished from petroleum pollution. 4. There are no direct measurements of the amount of petroleum entering the oceans by means of erosional processes, thus limiting the accuracy of any estimate. Natural Seeps Wilson et al. (1973) combined seepage rates on land with information on reported marine seeps, then extrapolated the data to the continental

46 margins, which they classified into areas of potentially high, medium, and low seepage. They incorporated tectonic history, earthquake activity, and sediment thickness in their appraisal. Five basic assumptions were used in their estimates: 1. More seeps exist in offshore basins than have been observed. 2. Factors that determine the total seepage in an area {number of seeps per unit area and the daily rate for each seep) are related to the general geologic structure of the area and to the stage of sedimentary basin evolution. 3. Within each structural type, the number of seeps and, to a lesser extent, rate per seep are thought to depend primarily on the area of exposed rock and not on rock volume. This assumption presumes that there is suff icient sediment volume and organic matter for matur at ion and gener at ion of petroleum. 4. Most mar ine seeps are clustered within the continental margins where the thickness of sedimentary rocks, which provides the needed source rocks for the seepage, exceeds a certain minimum. 5. Seepage rates are lognormally distributed. Although the geologic relationships developed by Wilson et al. (1973) that affect seepage rates are reasonable and seem to agree with observations, the statistical arguments of the last assumption may be questionable. On purely abstract grounds, an exponential distribution of seepage rates is more likely than a lognormal distr ibution. While oil field volumes are generally lognormally distributed, the actual volumes of all oil accumulations (most of which are perhaps too small to be produced and thereby cannot be classified as fields) are likely to have an exponential distribution (Harbaugh and Ducastaing, 1981~. The volumes of natural seepages are probably statistically distr ibuted in a manner similar to the volumes of oil accumulations in general, because seeps do not necessarily need sources as large as oil f ields . Consider ing the cliff iculties encompassed in the other assumptions, however, the form of the frequency distr ibution may be a minor matter. Since Wilson et al. (1973, 1974) made their estimate, little new information has become available that would alter their worldwide estimates of marine seepage rates. Their compilations of 190 reported submarine seeps were derived mostly from Johnson (1971) and Landes (1973) and can be augmented by four newly identif fed seep areas (Scott Inlet, Canada; Buchan Gulf, Canada; Australian North Coast; and Laguna de Tamiahua, Mexico) . All identif fed submar ine seep areas are shown in F igure 2-2; 54 individual submar ine seeps are represented by one dot of f the Cal if ornia coast, and another 28 are so represented in the Gulf of Alaska. Of the four recent reports (Levy, 1978; Levy and Ehrhardt, 1981; McKirdy and Horvath, 1976; and Geyer and Giammona, 1980), none estimates rates of seepage. The estimates available to Wilson et al. `1973) for Coal Oil Point (Santa Barbara Channel) and Santa Monica Bay ranged from 0.0007 to 0.05 mta. The more recent estimate of Fischer (1978) for the entire Santa Barbara Channel ranges from 0.002 to 0.03 mta, a span of values not greatly different from earlier estimates.

9~-~\ Sod\. i - .~: bait ~ If. 05 is. :.~`' 'I ~ .' ..N . ._ -~_ t;- i ~: ~ J i; , 1 up 1 Q. 1 ~ 1 e) 1 CQ 1 1 c: 1 .,, 1 so 1 t15 1 1 1 1 1 ~ Q 1 1 1 - 1 ~ o o . - C) o Cad 1 Cad :D C' H SO ~ S ' me to CO - ~ O · - V _ ~ V - a 1 dJ · ~ O o ~ ~ ~5 · - · - as: ·e u) ' c) :r; Go o ~ u] -

48 TABLE 2-2 Petroleum Resource Estimates Source Amount Mill ions of Metr ic Tons Billions of Barrels Reference Offshore oil "tar" sands Total offshore petroleum resources Total petroleum resources 1,000,000 in place Proven oil reserves and potential resources offshore Total proven reserves and potential resources onshore and offshore World exploitable oil (discovered 163,000 mt; undiscovered 141,000 mt) Large ~ tar n depos its 30,000 350, 000 14, 000 200 2,500 7, 200 100 86, 000630 304,0002,200 320, 0002, 100 Weeks (1965) Weeks ~ 1965) Hunt (1972) Albers et al. (1973} Albers et al. (1973) Halbouty and Moody ( 1980 Dema ison (1977 ~ Geolog ical Impl icat ions of Seepage Rates A compar ison of estimated seepage rates with the amount available for seepage can be used to assess the maximum geologic time during which seepages could be sustained. Table 2-2 lists petroleum resource estimates for several categories of petroleum and illustrates the wide range of resource estimates that have been calculated. Table 2-3 illustrates the compar ison. The wide range of assumed seepage rates includes the estimates of Wilson et al. ( 1973, 1974 ~ but extends downward to 0 .02 mta and upward to 10 mta. At the low end, 10 ,000 mt is near the value of 14 ,000 mt of total proven reserves and potential resources offshore as estimated by Albers et al. (1973) . At the other end of the scale, Wilson et al. (1973, 1974) accepted an estimate of 300,000 mt in place. Because this amount may represent only l% of the petroleum mobilized from source beds, they assumed that the amount available for seepage may be as much as 30 ,000 ,000 mt. This scale available for seepage has been augmented to 100,000,000 mt to attempt to include unknowns with regard to the amount of petroleum that would have been available for seepage during geologic time and will become available in the future dur ing the 1 if etimes of the seepage. Table 2-3 shows that to maintain petroleum seepage for a span of geologic time of at least 50 million years (most of the Tertiary period) requires that seepage rates be equal to or less than 2 mta, and

49 TABLE 2-3 Maximum Lifetimes (Million Years) of World Oil Deposits O il Available Assumed Seepage Rates (mta) forSeepage (mt) 0.02 0.040.20 0.60 1.0 2.0 6.0 10 10,000 0.5 0.25.05 0.02 0.02 0.005 0.002 0.001 30,000 1.5 0.75.15 0.05 0.03 0.015 0.005 0.003 100,000 5 2.S0.5 0.2 O.1 0.05 0.02 0.01 300,000 15 7.51.5 0.5 0.3 0.15 0.05 0.03 1,000,000 50 25 5 2 1 0.5 0.2 0.1 3,000,000 150 75 15 5 3 1.5 0.5 0.3 10,000,000 500 250 50 20 10 5 2 1 30,000,000 1,500 750 150 50 30 15 5 3 100,000,000 5,000 2,500 500 200 100 50 20 10 at the same time the amount available for seepage must be equal to or greater than 1,000,000 mt. If seepage is maintained for 500 million years (most of the Phanerozoic), then seepage rates must be equal to or less than 0.02 mta and the amount available for seepage must be equal to or greater than 10,000,000 mt. The petroleum seepage rate that best seems to accommodate the requirements of reasonable geologic time and reasonable assumptions concerning availability for seepage is 0.2 mta with an uncertainty both upward and downward of an order of magnitude. Thus, the conclusion is reached that the average rate of petroleum seepage over time ranges from 0.02 to 2.0 mta, with a best estimate of 0.2 mta. This value is lower by a factor of 3 than the best estimate of Wilson et al. (1973, 1974) of 0.6 mta. Erosional Inputs of Petroleum The amount of petroleum that enters the marine environment by erosional processes has not been estimated before. Previous work by Wilson et al. (1973, 1974) considered only the marine input from natural seeps. Any estimation of erosional input of petroleum into the oceans can only be approximate. There are at least three places where the erosional input of petroleum into the oceans could be studied in detail. Weaver (1969) showed examples of petroleum seeps at the beach and in the sea cliffs facing the Santa Barbara Channel, where erosion presently is taking place. Giammona (1980) described the Laguna de Tamiahua area where there are onshore and offshore seeps. The Marine Pollution Subcom- mittee of the Br itish National Committee on Oceanic Research (1980) identif fed the Dorset coast of southern England as another place where petroleum source rocks as well as petroleum-containing reservoir rocks are exposed. They suggested this area for the study of natural seeps and erosional processes affecting the distribution of petroleum in the marine environment. Because no direct information is available on erosional inputs of petroleum into the oceans, an indirect approach must be taken. This

so approach assumes that a portion of the organic carbon transported by all r ivers is petroleum. Estimates of the organic carbon input into the oceans by r ivers vary by nearly 2 orders of magnitude, from 30 to 1,000 mba, as summarized by Schlesinger and Clack (1981~. They concluded , however, on the teas is of two approaches , that the amount o f organic carbon transported by r ivers is 370-410 mta. Independently, Meybeck (1981) reached a similar estimate of 400 mta. In estimations of the organic carbon content of r ivers, no distinc- tion has been made between carbon from modern biological sources, carbon from pollution, and ancient carbon indigenous to the eroded sediment being car r fed by the r ivers . This latter class of carbon is of interest in estimating the eras tonal input of petroleum into the oceans . The total organic matter content of rivers is divided about equally between dissolved organic carbon and particulate organic carbon (Ileybeck, 1981) . Meybeck further estimated that of the approximately 179 mta of particulate organic carbon that is transported by r ivers, about 88 mta is ancient organic carbon. This ancient organic carbon is f inely dispersed in Plastic and carbonate rock particles, eroded from sedimentary rock formations on the continents (Ronov, 19761. In ancient sedimentary rocks the amount of extractable organic matter constitutes, on the average, about 6% of the total organic matter tHunt, 19791. If the extractable fraction in sediment par ticles in rivers is the same as that of source rocks, the amount in particu- lates in r ivers would be 10.6 mta. Most of the extractable organic mater Hal is d isper sed in sedimentary rocks, but O. .596 of this mater ial is petroleum tHunt, 1972) . I f this factor is applied to the extractable organic matter of sediment par- ticles in r ivers, then the amount of petroleum transported from eroding outcrops by r ivers to oceans is about O. .05 mta. This estimate may be high, because loss of organic carbon by oxidation during river transport and by sedimentation in estuar ies leading to the oceans was not con- sidered because of lack of data on these processes. Because of the numerous assumptions used to obtain this estimate, the uncertainty is at least an order of magnitude. In estimating rates of seepage of petroleum into the marine environment, these rates have been compared with the amount assumed to be available for seepage over geologic time (Table 2-3 ~ . Th is same petroleum would be available for erosional processes over geologic tune. The amount available is suff icient to sustain the estimated rates of natural seepage as well as rates of erosion of petroleum for an amount of time equivalent to the Tertiary per lad and probably longer . OFFSHORE PRODUCTI ON The amount of petroleum enter ing the mar ine environment from offshore petroleum production is estimated to be from 0.04 to 0.07 mta. Of these totals, ma jor spills (>7 metr ic tons) from platforms contr ibute 0.03-0.05 mta, minor spills (<7 metric tons) 0.003-0.004 mta, and operational discharges O. . 007-0 .011 mta.

51 TABLE 2-4 Offshore Petroleum Production, 1979 Country Pr oduct ion Rate . x 109 bbl/year x 106 mta Saudi Arabia 1.03 147 United Kingdom 0 .57 81 United States 0.39 56 Venezuela 0.38 54 Other countries 2.24 320 TOTAL 4.61 658 These estimates for the release of petroleum into the marine environment are lower, by about 30-50%, than the estimates generated earlier (NRC, 1975~. Better data are available for operations, and major spill incidents in the United States have been more comprehen- sively documented since the earlier estimates were made. The available international data suggest that reductions have also been experienced outside the United States. As reported by Burnet (1980), worldwide offshore petroleum produc- tion totaled approximately 658 mta in 1979. Over 50% of the production came from four countries: Saudi Arabia, the United Kingdom, the United States, and Venezuela. These data, which are the latest available published information, are summarized in Table 2-4. They are the basis for all subsequent calculations of the current petroleum input to the oceans from offshore petroleum production operations. Operational (Produced Water) Discharges In the United States, offshore produced water is normally discharged into the ocean after being processed to minimize the entrained petroleum content. Actual rates of discharge for produced water are not currently available. However, until 1976, the U.S Geological Survey (USGS) maintained records on these discharges from outer continental shelf operations in the Gulf of Mexico. At that tine, 0.8 barrels of water were produced with every barrel of crude oil. This ratio is assumed to be still valid, and the same ratio is assumed to apply to all U.S. offshore production. This estimate is believed to be conservative, because recent Bureau of Land Management (BLM) environmen- tal impact statements for outer continental shelf (OCS) lease sales assume a 0.6 water-to-crude-oil production ratio. The Department of Environment, U.K. (1976} report concerning discharges from offshore operations in U.K. waters stated: "The proportion of production water in crude oil will initially be less than 1 percent but will increase to

52 some 30 percent as the reservoir becomes depleted, particularly when water in jection is used on an increasing scale. n This study assumed a proportion of 10% averaged over the life of the U.K. fields. For other countries an assumption of 30% was used. It should be noted that no water is produced off the Saudi Arabian shore. Produced water regulations in the United States require that the daily maximum petroleum content not exceed 72 mg/L and that the monthly average be less than 48 mg/L. The Environmental Protection Agency (EPA, 1976) Development Document on which these guidelines are based also includes the results of an in-depth statistical analysis of all available data, which indicates that facilities meeting the above limitations will achieve a long term average petroleum content of 25 mg/L or less. These figures, however, do not include the C6-C14 "volatile liquid" hydrocarbons, which are not determined by the solvent extraction technique used for "oil and grease" analysis. Therefore, a somewhat higher estimate of 35 mg/L hydrocarbons entering the oceans from U.S. produced water discharges was used. Because this regime does not include (1) upset and bypass situations in which higher discharge 1 evels are probably experienced and (2) the fact that state-of-the-art equipment is not installed at all locations, a high estimated average Is believed to be twice this level, 70 mg/L. A reasonable best estimate is 50:20 mg/L volatile liquid" hydrocarbons and 30 mg/L higher-molecular-weight hydrocarbons (>C14~. Similar arguments for the U.K. offshore operations and those in other countries (Table 2-5) lead to a range of 50-70 mg/L for the estimated hydrocarbon content of produced waters. Based on these assumptions, the volume of petroleum enter ing the world's oceans from offshore produced water discharges is calculated to be between 0.0075 and 0.0115 mta, with a best estimate of 0.0095 mba (Table 2-6 ) . Specific estimates were not made for deck drainage, drilling fluid discharges , and other minor sources of petroleum (Schreiner, 1980 ~ . These sources are probably accounted for within the 1 imits of conf idence of the above number s . Minor Spills S ince 1971 the USGS has maintained a computer ized OCS events f ile for Gulf of Mexico oil and gas operations (Danenberger, 19761. Included are data on all crude oil spills . The USGS classif ies spills as minor (<7 metric tons or 50 barrels) or major (>7 metric tons or 50 barrels) . Table 2-7 summer izes the minor spi Its in the Gulf of Mexico OCS for the 8-year per lad 1971-1978 . The average spillage rate for this period was 0.000249 of total crude oil produced. The record for minor spills in offshore Alaska is better. The Lower Cook Inlet spillage rate for all spills from 1971 to 1980 is 0.0001% of total crude oil produced (Wondzell, 19811. Similar data for operations in other U.S. areas and outside the United States are not readily available. Offshore operations are moving into more severe environments, such as the arctic regions.

53 TABLE 2-5 Offshore Produced Water Effluent Limitations Country Oil and Grease Content Limit (mu/L) Average Maximum Abu Dhabi - 15 Australia 30 50 Denmark 40 Egypt - 60 (Mediterranean) 15 (Red Sea) France - 20 Indonesia 30 _ Malaysia 100 (offshore) - 10 (coastal) - Netherlands Nigeria Norway Spain - 75 25-30 40 (Atlantic) 60 (Mediterranean) 100 40 100 - Trinidad 50 _ United Kingdom 40 {large facilities) no more than 4% 50 (small facilities) greater than 100 mg/L United States 48 (monthly) 72 25 (long term) Venezuela 35 _ NOTE: Limitations shown here are from various sources. They are either existing government regulations, proposed government regulations (which could change), or limitations imposed by authorities for installations in operation in countries without regulations. However, to balance this effect, there have been significant techno- logical advances (such as warning systems and improved blowout preventers) that are reducing the occurrence of spills of all sizes. Average experience for all U.S. offshore operations probably is comparable to the Gulf of Mexico average, so a range of 0.00021-0.00030% is used for the United States. Another assumption is made that the worldwide percentage is about twice that of the United States, or 0.00042-0.00060 (Table 2-8~. Clearly, there is uncertainty associated with this assumption. With these data a range of 0.0027-0.0038 mta has been calculated as the estimate of petroleum entering the marine environment from minor spillage from offshore drilling and production activities worldwide. The best estimate is 0.003 mta, which is lower than the earlier NRC (1975) estimate of 0.01 mta.

54 s .,. x tn s oo .,. a a, JJ 3 3 o ~4 P4 o S tQ o o I:: oc C ~: S V o o 1 v ~Q m o 4 S C> · - ~ O C] 4 ~ ~: · - O O - Q. - S ~ _ {> _~ u} ~: Q -~' ~ S:: Q O Q ,`' _ 4 o o o o ~ o 0 o ln 0 0 U ~CO _ o o o o o o o o o ~e o u) ~ O ~. o o o O t~- ~(D_ U~ o o oo ~ o o oo o ~ 3 ~o ~ o _' ~ C: Q Q o c ~ o O C: .~:- _I ~t~ _ a, ~ O ~U ~U. o O C~)- ~U~ o 3 o 3 _ Q Q o O o _ 61 o m 3 m - O = - O Q 0 -53 14 ~ O V 3 o. ~ O O P4 o o o o o o o ~r u, a, oo . · ~ o o o o o o o o o CO r ~ U' ~o U] O `:: 3 ~ U] · - :~: ~E~ O `: O E~ . - o o ~q Q ~n o m ~5 ~: 3 o ~n D ~q . - 3 p

55 TABLE 2-7 Minor Oil Spills, Gulf of Mexico Outer Continental Shelf, 1971-1978 Oil Number Volume Percentage of Production of Spilled Production Year (1,000 bbl) Spills (bbl) Spilled 1971 386,400 1,245 1,500 0.00039 1972 391,000 1,159 1,000 0.00026 1973 375,300 1,171 900 0.00024 1974 343,900 1,129 700 0.00020 1975 316,000 1,126 700 0.00022 1975 303,100 948 500 0.00016 1977 293,000 864 600 0.00020 1978 282,500 873 600 0.00021 TOTAL 2,691,200 8,515 6,500 0.00024 (avg.) (-384 mta) (-0.0093 mta) NOTE: Minor oil spills are defined as <7 metric tons or 50 barrels. Major Spills As was mentioned previously, the USGS has maintained a computerized OCS events file for Gulf of Mexico oil and gas operations (Danenberger, 1976~. The history of major oil spills (<7 metric tons or 50 barrels) from U.S. Gulf of Mexico operations for the period 1971-1978 is sum- marized in Table 2-9. Because of the statistical distribution of spills, one large incident in a particular year greatly influences the annual figure. However, the Gulf of Mexico 8-year average is fairly representative of any current year, i.e., the oil spillage rate from major incidents is 0.002% of the oil produced. A similar average may apply nationwide. Accurate worldwide information on major spills is often difficult to obtain. Since 1979 the Oil Shill Thrill ia~n~- R - nary `1979, 1980' has attempted to provide annual summer ies of all incidents involving more than 20,000 gallons (or 68 metric tons) of oil. However, these surveys are admittedly incomplete. A single catastrophic incident is usually a major contributor to the annual total, but the probability of such an occurrence on an annual basis is very low and its amount is unpredictable. Currently the world record spill resulted from the Petroleos Mexicanos (Pemex) Ixtoc I well blowout on 3 June 1979. until it was capped on 23 March 1980, a total of 0.44-1.4 mt of crude oil was released. The uncertainty in the amount of crude oil spilled is related to the problems associated with estimating flow from the open hole. Estimates of the amount burned vary from 30% (Ross et al., 1979) to as much as 58% (Program a Coordinada de Estadios Ecologicos en la (Ross et

56 TABl;r 2-8 Oil to the Mar ine Environment From Minor Spills 1979 Oil Product ion Country ( 1, 000 bbl ) Percentage of Production Volume Spilled Spilled (bbl ~- Low H igh Low H igh UnitedStates 389,000 0.00021 0.00030 Other4, 227, 000 0 .00042 0 .00060 TOTAL NOTE:: Minor spills are defined as <7 metric tons or 50 barrels. 820 1, 170 17 ,800 25,400 18 ,620 26,570 (~0.0027 mta) (~0.003 mta) Sonda de Campeche, 19801. No such massive incidents occurred in 1978. Only one major spill from offshore operations was reported that year, a spill of 0.003 mt in Indonesia (Oil Spill Intelligence Report, 1979~. Major oil spills occur sporadically. In order to calculate a meaning- ful annual input, several years of experience have to be averaged. A recent U.K. report on spills from offshore operations indicated that the average total oil spillage rate in U.K. waters for 1975 through 1979 was 0.00068% of production (Royal Commission on Environmental Pollution, 1981~. This spillage rate is lower than that for the United States. We feel, however, that the worldwide spillage rate, excluding the United Kingdom, is probably higher than that of the United States because of less rests ictive regulation of blowout prevention. AS with minor spills, uncertainty is associated with this assumption. Therefore, for purposes of this study, the estimated range of major crude oil spillage outside the United States (and the United Kingdom) is from 2 to 4 times the U.S. rate. The best estimate is 3 times the U.S. rate. The estimate of oil input to the oceans from major accidents during offshore oil and gas operations ranges from 0.025 to 0.05 mta with a best estimate of 0.04 mta. The calculations to obtain these figures are given in Table 2-10. MARINE TRANSPORTATION The estimated range in the amount of PHC discharged into the oceans due to maritime transportation activities is from 1.0 to 2.6 mta, with a best estimate of 1.45 mta. Just under half (about 0.7 mta) of this total is estimated to come from tanker operational discharges. The remainder is distributed among terminals (0.02 mta), dry-docking (0.03 mta), bilges and fuel oil from all ships (0.3 mta), and accidental spillages from tankers and other ships (0.4 mta). The earlier NRC

57 TABLE 2-9 Major Spills, Gulf of Mexico Outer Continental Shelf, 1971-1978 Oil NumberVolume Percentage of Production ofSpilled Production Year (1,000 bbl) Spills(bbl) Spilled 1971 386,400 111,300 0.0007 1972 391,000 2200 0.0003 1973 375,300 422 ,200 0.0062 1974 343,900 822,700 0.0068 1975 316,000 2300 0.0003 1976 303, 100 34, 700 0 .0017 1977 293,000 4700 0.0004 1978 282,500 31,100 0.0006 TOTAL 2,691,200 37 53,200 0.0020 (avg. ) (-386 mt) (~0 .0076 mt) NOTE: Major oil spills are defined as >7 metric tons or 50 barrels. (1975 ~ estimated range for total mar ine transportation losses was 1. 5-2 .8 mta with a best estimate of 2 .1 mta. The quantity of oil discharged from ships depends on how effectively the standards developed for the control of oil pollution from sh ips ar e implemented. In 1973 the applicable rules and standards were the International Convention for the Prevention of Pollution of the Sea by Oil 1954, as amended in 1962 (OILPOL 1954/1962 ~ . The 1969 amendments to OILPOL 1954/1962 were adopted by the International Maritime Organization (IMO), formerly IMCO, assembly in 1973 but did not enter into force until February 1978 . As of 9 November 1981, OILPOL ~ 954/196 9 had been in force for more than 3 years by 66 nations representing approximately 95 % of the wor Id ' s merchant f leet. The requirements of OILPOL 1954/1969 have been considerably strengthened by the adoption of the International Convention for the Prevention of Pollution from Ships, 1973, as modified by the Protocol of 1978 relating thereto (MARPOL 1973/1978) . In particular, the worldwide implementation of the mandatory provision of segregated ballast tanks (SBT), dedicated clean ballast tanks (CBT), and crude oil washing systems (COW) for new and existing oil tankers would result in a s ignif icant reduction of the quantity of oil discharged into the oceans. The United States implemented these regulations 1 June 1981 with respect to U.S. flagships and foreign ships visiting U.S. ports. MARPOL 1973/1978 has been ratif fed by the requisite number of nations and will enter into force on 2 October 1983. The ma jor ity of the requirements of MARPOL 1973/1978 pertaining to oil (Annex I ~ will become effective when the convention enters into force. The remaining

58 TABLE 2-10 Oil to the Mar ine Environment From Ma jor Spills Percentage of 1979 Oil Production Production Spilled _ Volume Spilled (bbl) Country (1,000 bbl) Low High Low High United States389,000 0.0020 0.00207,800 7,800 Other 4 ,227 ,000 0.0040 0.0080169,100 338,200 TOTAL 18 ,620176,900 346,000 (-0.025 mta) (-0.05 mta) NOTE:: Major spills are defined as >7 metric tons or 50 barrels. provisions will become effective no later than 2 October 1986. The present situation should therefore be regarded as a transitional period until MARPOL 1973/1978 is fully implemented. The present estimates are based on a report prepared by a group of experts, consisting of representatives from maritime administrations and oil and shipping industries, under the auspices of the Marine Environment Protection Committee of IMO. This IMO workshop was convened prior to the November 1981 NRC workshop. OPERATIONAL DISCHARGES Crude Oil Dur ing normal operations, oil tankers discharge into the sea a certain amount of oil contained in the ballast and tank washing water. OILPOL 1954/1969 stipulates that instantaneous rates of discharge from cargo tank areas of oil tanker s must not exceed 60 L/mi, and the total quantity of oil discharged dur ing any one ballast voyage must not exceed 1/15,000 of the total cargo carrying capacity (Tc). MARPOL 1973/1978 sets the same discharge standards outside special low pollution areas, but the maximum quantity of oil permitted to be discharged for new oil tankers has been reduced from 1/15,000 to 1/30,000 Tc. In order to comply with the requirements of OILPOL 1954/1969, oil tankers should operate with load-on-top (LOT) procedures. At the time of the 1973 Ned study, 80% of the tanker fleet was assumed to be operating with LOT. Presently all crude oil tanker s engaged on long haul voyages (exceeding 71 hours or 1,200 nautical miles) should operate with LOT, but tankers engaged on short haul voyages may not be able to do so. The International Association of Independent Tanker Owners (INTERTANKO) estimates that long and short haul voyages constitute 85 and 15%, respectively, of the world ' s crude oil movements .

59 During a ballast voyage, discharges of oil into the ocean may occur during two types of operation: discharge of departure (dirty) ballast without adequate separation and discharge of decanted water from slop tanks. After settlement, departure ballast separates into three parts, the largest part of which consists of water with an oil content on the order of 15-50 mg/L. An oil-water interface contains a relatively high oil content (in the order of 100-2,000 mg/L), and the oil layer on the surface is essentially pure oil. By careful operation, the discharge of ballast water can be stopped as soon as the interface level is reached. In adverse sea conditions, the interface may be diffuse, and discharge is stopped when the oil content in ballast water rises above about 50 mg/L. Under less careful operation, the ballast may continue to be discharged for some time after the interface level has been reached. A similar situation might arise with discharge from slop tanks, which would have higher oil content but could be more easily controlled because of the slower pumping rate. Cargo lines that are not thoroughly flushed with water to slop tanks before being flushed to the sea may also cause oil discharges. In the worst cases, the LOT procedure may be completely ignored and the total oil-water mixture will be discharged into the sea The LOT operations emphasize "retention on board" procedures where dirty ballast water, tank washings, and oily residues are held in slop tanks for discharge at terminals. Before LOT procedures were initiated these materials were discharged routinely into the sea. Four oil companies {Socal, Mobil, Exxon, and Texaco) measured the quantities of retained slops for the period 1972-1977 (Gray, 1978~. These data indicate a steady increase in the quantities of retained slops for the period 1972-1975. From 1975 to 1977 the quantities of retained slops leveled off or began to decline. This decline of retained slops for company-owned tankers is probably attributable to the improved efficiency of pumping out cargo oil as well as the increasing use of COW systems, which, when fully imple- mented, reduces retained slops to a lower level. Thus, it is not currently considered appropriate to use slop recovery data as a basis for estimating quantities of oil wastes discharged into the oceans. Smaller quantities of retained slops may not necessarily be an indication of the discharge into the oceans of larger quantities of oil wastes. Tests have been carried out in various countries to evaluate the efficiency of LOT operations. The results of some of these tests are shown in Table 2-11. The above data and experience by major oil companies indicate that perhaps two-thirds of crude oil tankers on long haul voyages already meet the OILPOL 1954/1969 discharge criteria of 1/15,000 TO or better. Because of the dearth of data for oil tankers engaged on trades in which major oil companies are not involved, particularly oil trades on spot market, the assumption is made that half the long-haul-voyage crude oil tanker fleet meet the 1/15,000 TO standard. AS long haul tankers carry 85% of the 1,319.3 mta total, discharges from this source would be 0.037 mta (1,319.3 x 0.85 x 0.5 x 0.0000671.

60 TABLE 2-11 Results of LOT Operation Eff iciency Tests Vessel S ize Country or Oil (thousand ton Deadweight/ Oil Company Tanker deadweight) Discharged Oil Reference Japan Alriyadh 237 53 r°°° IMCO (1981) Norway Berge Pr incess 280 36, 000 Overaas and Solum (1974 ) Mobil Three tankers 50-212 15 ,000 Desel (1972) Exxon Five tankers 52-254 11,000-200,00Oa Gray (1978) Seven out of nine discharged less than 1/30,000 Tc. Scene ctata (INTO, 1981) show that of 650 tankers inspected during the years 1979-1980, approximately 29 err ived at their loading terminals ~ situated all over the world) with no slops at all on board, for unexplained reasons . The data from the four oil companies (Gray, 1978 show s imilar trends . These vessels probably represent performances better than the worldwide average, so the assumption is made that in the worst case, 5% of the long-haul-voyage crude oil tankers would discharge all their ballast, a total quantity of oil wastes equal to 0.4% of Tc. The discharge from this source would therefore be O.224 mta (1,319.3 x 0.85 x 0.05 x 0.004~. The oil discharge for the remaining 45% of the tankers is estimated as follows: 30% would discharge 1/7,500 Tc (i.e., twice the OILPOL 1954/1969 criteria) producing, 0.045 mta, and 15% would discharge oil equal to 0.1% of Tc, producing 0.168 mta. The total discharged by long haul tankers would therefore be 0.474 mta (0.037 + 0.224 + 0.045 + 0 .168 ) . Cr ude o it tanker s engaged in shor t haul voyages may not be able to perform LOT; some of these tankers are provided with SET or similar arrangements to avoid contaminated ballast. These tankers would contribute negligible pollution. Other short haul tankers are engaged in dedicated trades that include arrangements to transfer contaminated ballast to shore reception facilities or to long haul, very large crude carriers (VLCC) from which they take oil cargo. The remaining short haul tankers (estimated to be 50%) discharge into the sea oil amounting to 0.2% of Tc (which corresponds to the total oil content in dirty ballast tanks), or 0.198 mta (1,319.3 x 0.15 x 0.5 x 0.002~. The total annual discharge of crude oil into the sea resulting from the normal operation of crude oil tankers is estimated to be 0.672 mta (0.474 + 0.198). Although there is a degree of uncertainty associated with the numbers, there does seem to be an improvement since the earlier NRC (1915) estimate was made (0.67 versus 1.08 mta), particularly when an increase in the amount of crude oil transported by sea from 1971 to

61 1981 is taken into account. Possible reasons for this improvement are as follows: ~ . In 1973 tankers were allowed to discharge unl imited quantities of oily wastes outside the prohibited zone (which was normally 50 miles from land), whereas such discharge is illegal under OILPOL 1954/1969; 2. There has been significant improvement in the awareness of the master and crew, shipowners and operators, of the existence of and need to observe international rules for oil pollution prevention; 3. Surveillance and control of illegal discharges have been considerably tightened in many countries; 4. Dramatic increases in the price of oil in recent years have resulted in more careful handling of cargo oil at discharge ports, with less oil remaining on board after discharge; 5. Increased use of COW systems has enabled a higher proportion of tankers to comply with the 1/15,000 TO standard; and 6. Inclusion of "cleaner seas" provisions in charter party agree- ments has alleviated the economic disadvantages for operators to retain oil res idues on board. At the same time, certain adverse factors must be borne i n mind, such as the aging of the existing tanker fleet, the lack of well equipped new tankers, the shift of the control of tankers from - experienced to less experienced operators, and the increase in spot market oil trades. Product O,1 Of 269 mta of product oil carried by tankers, one-fourth (67 mta) is estimated to be persistent {lubricating oil, fuel oil) and three- fourths (202 mta) nonpersistent (gasoline, kerosene). The discharge of persistent oil is subject to OILPOL 1954/1969, whereas the discharge of nonpersistent oil is not presently controlled. The NRC (1975) report did not provide specific figures for the operational discharge of product oil. There are no measured data on the quantities of oil residues for persistent oil trades. The operation of tankers carrying persistent product oil is assumed to be similar to that of crude oil tankers on short voyages; namely, 50% are engaged in dedicated trades that include arrangements to transfer contaminated ballast to shore reception facilities or are provided with arrangements to avoid contaminated ballast. These ships will contribute negligibly to pollution. The LOT performance for the remaining 50% of these tankers' compared with crude oil tankers, may be affected by the following: (1) the relatively higher viscosities of persistent product oil may result in higher clingage and {2) the relatively higher density of oil may provide some empty cargo tanks in which ballast water without contami- nation may be carr fed on a subsequent ballast voyage.

62 Consider ing the above factors, the remaining 50% of these tanker s are likely on average to discharge persistent oil equal to 1/2,000 TO, or O .017 mta (67 x 0 .5 x 0 .0005 ~ ~ The discharge of nonpersistent oil is permitted at present under OlLPOL 1954/1969, but the quantity of such product discharged into the sea might be less than expected, because of less cl ingage, easy decanting, and higher rate of evaporation. It is assumed that 50% of tankers carrying nonpersistent oil have means to avoid the discharge of contaminated water; hence, the discharge from this source should be negligible. The remaining 509s disabarge, on average, oil equal to 1/5,000 To, and, hence, the discharge of nonper- sistent oil from this source would be approximately 0.020 mta (202 x 0.5 x 0.0002). Therefore, the annual operational discharge of product oil is estimated as 0.037 mta (persistent 0.017 mta, nonpersistent 0.020 mtal. The sum of crude oil plus product oil discharges is thus estimated at O .71 mta (crude oil 0 .67 , product oil 0.041. DRY-DOCKING The NRC (1975) report estimated that half the tankers would arrive for dry-docking at average intervals of 18 months without tank washing residue. Since then the situation has changed considerably, including longer dry-docking intervals (2 years on average), the increased availability of reception facilities in repair ports, the reduction of sludge or slop due to more efficient stripping and COW systems, the increased degree of enforcement of OILPOL 1954/1969, and the increase in the value of crude oil. On the basis of the above factors, it is estimated that of the world tanker fleet of 340 million deadweight tons (dwt), 5% of tankers dis- charge into the sea sludge or slop amounting to 0.4% of dwt prior to dry-docking at intervals of 2 years. The annual estimated loss to the sea then becomes 0.034 mta (340 x 0.5 x 0.05 x 0.004~. MARINE TERMINALS INCLUDING BUNKER OPERATIONS The N~C (1975) study estimated discharges of 0.003 mta during terminal operations . This result was attr ibuted to spillages that resulted from human error, such as over f ill ing tanks and disconnecting hoses without adequate drainage. There are other causes of spillages, including line or hose failures, submarine pipeline ruptures, or storage tank ruptures. Discharges under this category include spillages occurring dur ing bunkering operations (filling the ship's fuel compartments) either at a terminal or from a bunker ing barge. The U.S. Coast Guard ( 1976 , 1977 , 1979 , 1980 ~ keeps statistics on marine terminal spillages from all types of incidents (such as hose breaks, tanker overfilling, line fractures, shore tank ruptures). The average marine terminal spillage for 1976, 1977, and 1979 was 0.0025 mta (the 1918 figure was not included in the average because it included

63 a large tank rupture that was not a marine spill). As approximately 25% of the world oil movement by sea is around the U.S. coast and the rate of spillage is not greatly different elsewhere, an estimate can be made for marine terminal discharges of 0.010 mta (0.0025 x 4~. Major accidental spills in marine terminals, such as submarine pipeline and storage tank ruptures, although occurring rarely, may be the major causes of oil losses under this category (sometimes over 0.010 mt). However, no worldwide statistical data are available. For the present estimate, an average 0.010 mta of oil is assumed to be spilled into the oceans due to such accidents. Thus, total spillage {discharges plus spills) from marine terminals is estimated at 0.020 mta (0.010 + 0.010~. BILGE AND FUEL OI L Discharges under this category can be divided into three types: machinery space bilges, fuel oil sludge, and oily ballast from fuel tanks. Machine Space Bilges Steam tankers generate approximately 5 gal of bilge oil per day, while motor tankers generate about 15 gal per day. As there are about equal numbers of steam and motor tankers and tankers may operate some 300 days per year, the average quantity of bilge oil generated in a tanker per year is about 10 gal/42 x 7 x 300, or 10.2 metric tons. The majority of the 7,100 world tankers retain such bilge oil in slop tanks for cargo oil or discharge it to shore reception facilities. Assuming that 10% of the total bilge oil generated in machinery spaces of tankers may be discharged into the sea, the annual discharge of bilge oil from tankers is estimated to be 10.2 tons x 7,100 x 0.1 = 7,242 tons, or 0.007 mta. With a similar approach, the average quantity of bilge produced in cargo ships can be also estimated. The average size of tankers is approximately 25,000 gross tonnage (GT), with an average size of propul- sion machinery of 20,000 horsepower (HP); in comparison, the average size of nontankers is approximately 3,700 GT, with an average size of propulsion machinery of 4,000 HP. Almost all nontankers are motor ships. Therefore, the slop oil generated in each nontanker would on average be 3.0 gal per day (15 gal x 4,000/20,000), or 3.1 metric tons per year (3 gal/42 x 7 x 300~. For the world's nontanker fleet of 66,700 the amount of total bilge oil produced would be 3.1 tons x 66,700 per year, or 0.207 mta. The quantity of this bilge oil discharged into the ocean would depend on whether the ships are fitted with oily-water separators and on the availability and use of shore reception facilities. About half these ships are fitted with separators.

64 Assuming that ships with separators would discharge into the sea 10% of bilge oil and ships without separators tw~thirds of bilge oil, the quantity of oil from bilge discharged into the sea per year is estimated to be 0.079 mta {with separators: 0.207 x 0.5 x 0.1 = 0.01 mta; without separators: 0.207 x 0.5 x O.67 = 0.069 mta). Thus, the rate of PHC input into the oceans from bilge discharges is 0.086 mta {tankers 0.007 mta, nontankers 0.079 mta). Fuel Oil Sludge Worldwide annual use of heavy residual bunker fuel for mar ine applica- tion is 108 mt. Tankers use 44 mt of bunker fuel, while nontankers use 64 mt of bunker fuel as well as 18 mt of gas oil. Before being used in diesel engines, residual bun Her fuel is puri- f fed to remove such impur ities as sludge and water e The average sludge content in heavy fuel for marine application is 0.5%. For the present estimate, 0.3% of the quantity of heavy fuel oil used for diesel engines is assumed to be disposed of. In the case of steam tankers, the usual practice is to retain sludge in cofferdams or slop tanks for cargo oil for eventual disposal to shore facilities. In the case of nontankers, the capacity of the sludge holding tank on board may not be sufficient to retain the sludge until the ship arrives at port. There would then be no alternative but to dispose of it into the sea. I f 20% of the sludge for motor tankers and 90% of the sludge for nontankers is discharged into the ocean, then the annual quantity of sludge discharged is estimated to be 0 .186 mta (tankers 44 x 0 .5 x 0.003 x 0.2 = 0.013 mta, nontankers 64 x 0.003 x 0.9 = 0.173 mta). Oily Ballast From Fuel Oil Tanks Water ballast for tankers is carried in cargo oil tanks or SET, and no contamination of water ballast with fuel oil should occur. However, nontankers, which have to carry large quantities of water ballast for safety reasons, particularly fishing vessels, may have to carry water ballast in fuel oil tanks. It is estimated that 2% of nontankers carry ballast water in fuel oil tanks, with an average clingage of 0.8% (including heavy and light f Hell, a quarter of which will be discharged into the sea. Nontankers use some 64 mt of residual bunker fuel plus 18 mt of gas oil. Thus, the annual quantity of such oily ballast discharges is estimated to be 0.003 mta (82 x 0.02 x 0.008 x 0.251. The total for the bilge and fuel oil inputs is 0.28 mta (bilge 0.086 mta, fuel oz1 sludge 0.186 mta, oily ballast 0.003 mta).

65 ACCIDENTAL SPILLAGES Tanker Accidents Var ious sources of data on tanker accidents producing oil pollution have been available, including data from the International Tanker Owners Pollution Federation Ltd. (ITOPF, 1981) and the French Institute of Petroleum (IFP) (Bertrand, 1979), as shown in Table 2-12. Nontanker ACC idents IFP {1981) average of annual oil spillages from nontanker accidents over the years 1974-1979 is 0.017 mta. Therefore, the total quantity of oil discharges due to mar itime accidents is estimated to be 0 .41 mta (tanker 0.39, nontanker 0.02~. The estimated range for quantity of oil discharged annually into the sea from transportation activities is 1.0-2.6 mta. This compares with the earlier NRC (1975) range of 1.5-2.8 mta. Table 2-13 shows the estimated range and best estimate of PHC discharged into the sea from each category of transportation losses. In general a +100% range was considered realistic, but each category has been reviewed and slight adjustments have been made. Tanker accident data were considered the most reliable and were therefore assigned a +10% range. Although not addressed in this report in detail, spills caused through acts of war should be considered, where appropriate, in future discussions of inputs, particularly in areas such as the Persian Gulf. TABLE 2-12 Annual Quantity of Oil Spills Due to Tanker Accidents Quantity lmt) Year ITOPF IFP 1975 0.368 0.362 1976 0.456 0.364 1977 0.316 0.297 1978 0.388 0.487 1979 0.760 0.649 1980 0.187 TOTAL 2.731 (1974-1980) 2.374 (1974-1979) Average 0.390 mta (1974-1980) 0.396 mta (1974-1979) NOTE: The annual f igure var. ies considerably, influenced pr imar fly by a few catastrophic incidents. For the purpose of the present estimate, the aver age f igur e of 0 . 39 mta is appropr late .

66 TABLE 2-13 Summary of Transportation Losses (mta ~ Type of Loss Range Best Estimate Tanker operations 0.44-1.45 0.71 Dry-docking 0.02-0.05 0.03 Mar ine terminal s 0 . 01-0 . 0 3 0 . 0 2 Bilge and fuel oil 0.16-0.60 0.28 Tanker accidents 0.35-0.43 0.39 Nontanker accidents 0 . 02-0 . 04 0 . 02 TOTAL 1.00 - 2.60 1.45 ATMOSPHERE The estimated range of atmospheric input of PHC into the marine environment is 0.05-0.5 mta. The workshop panel working on atmospheric input agreed that they could not provide a "bests estimate because of the great uncertainty associated with their estimate. The primary pathway for this input appears to be removal by rain of particulate material. Secondary pathways involve dry deposition of atmospheric particulate matter, precipitation scavenging of trace gases, and direct gas exchange with the ocean. Less is known about the global sources, distribution, and fluxes of organic matter than any other major class of chemical substances in the atmosphere. Aside from methane and certain halocarbons in the vapor phase, very few measurements of gaseous or particulate organic matter are available outside urban areas. Recent reviews by Duce (1978) and Simoneit and Mazurek (1981) have attempted to synthesize the available data and summarize our knowledge. The situation is complicated, of course, by the fact that there are probably thousands of different organic compounds emitted to the atmosphere from natural and pollution sources, and many other compounds are produced from atmospheric, particularly photochemically induced, reactions. Each of these substances has its own characteristic chemical and physical properties and associated atmospheric sources, residence times, and sinks. In terms of these pathways, petroleum entering the sea via the atmosphere must first evaporate or be emitted into the atmosphere. National emission inventories identify vehicle exhaust and evaporation losses as the greatest source, followed by industr ial losses through evaporation, particularly oil industry operations. Rough estimates of the input of PHC to the ocean from the atmosphere have been made in the past. The Study of Critical Environmental Properties (SCEP, 1970} estimated that 9 mta of PHC entered the ocean from the atmosphere at that time and suggested this number could double by 1980. NRC (1975) estimated that the atmospheric input of PHC was much lower, about 0.6 mta. This latter estimate was not based on any

67 measurements over the ocean, but simply on the total quantity of PHC injected into the atmosphere, its assumed reactivity in the atmosphere, and the general distribution of particles and the patterns of rainfall over the sea and the land. The actual atmospheric input of petroleum to the ocean sur face is very difficult to ascertain for several reasons. Petroleum is a complex mixture of many classes of compounds whose components have different reactivities and solubil ities. For example, low-molecular-weight polynuclear aromatic hydrocarbon (PAM) and n-alkane reactivities with the OH radical span 5 orders of magnitude (Darrell et al ., 1976 ~ . Chameides and Cicerone (1978) suggested that the photochemical lifetime of atmospheric ethane is about 25 days, while that of propane, butane, and pentane may be about 5 days. Zimmerman et al. (1978) and Hanst et al . (1980 ~ have pointed out the potential importance of the photo- oxidation of nonmethane hydrocarbons as a source for atmospheric CO. Gas to particle conversion also occurs for organic mater ial (Simoneit and Mazurek, 1981; Duce, 19781. During transport from continents to the sea via the atmosphere, particle fractionation may occur. Hence, if the organic composition is different for various particle size classes, the overall atmospheric particulate organic composition will change as a function of transport distance and time. Thus the organic chemical composition of petroleum-derived substances in the remote marine atmosphere may bear very little resemblance to what was emitted into continental air masses several thousand kilometers away. Many individual compounds in petroleum are also produced from other natural sources, such as n-C15 and n-C17 alkanes from marine phytoplankton, pristane from zooplankton, and n-C27 and n-C29 alkanes from land plants (see Chapter 3, Chemical Methods section). In several areas of the ocean, such as upwelling zones or downwind of ma jor forests , these compounds may make up a significant portion of the hydrocarbons in the atmosphere. Finally, there is a paucity of data on petroleum organic compounds in rain, vapor, and particulate samples from open ocean areas, thus requiring a large number of simplifying assumptions to be made in any estimate of air to sea transport. Taking into consideration the problems discussed above, the workshop panel on atmospheric input decided to concentrate on the n-alkane com- ponents of petroleum. The alkanes constitute approximately 30% of petroleum and some data are available, albeit very limited, to undertake estimates of their atmospheric input to the open ocean (Ketseridis and Eichmann, 1978; Eichmann et al., 1979, 1980; Hahn, 1981; Gagosian et al., 1981a; Gagosian et al. , 1982; Atlas and Giam, 1981~. Data for PAH are fewer and not sufficient to estimate their atmospheric input. Aerosol samples that had an oceanic origin were collected from coastal Norway by Bjorseth et al. (19791. The total PAH concentration averaged 1.6 ng/m3 (nanogram per cubic meter). Hahn (1980) found PAH to be 80% of the n-alkane concentration for aerosol samples from the southern North Atlantic. His PAH values averaged 11 ng/m3. However , no single PAH was detected in greater than 5-pg/m3 (picogram per cubic meter ~ air concentration for particles at Enewetak Atoll, Marshall Islands ~ in the central North Pacif ic (Gagosian et al ., 1981a) . No PAH vapor or rain data have been reported from the open ocean.

68 As stated earlier, many of the n-alkanes produced by marine plank- ton and land plants are the same as those in petroleum. It is difficult to subtract out this ~biogenic. component from many of the data sets available without introducing a preconceived bias. Rather than do this for selected samples, we did not do it for any of them. Atmospheric inputs to the ocean are thus derived for n-alkanes as a class of organic substances. The fraction of these n-alkanes that are of petroleum origin is uncertain. Thus the fluxes obtained represent maximum values relative to petroleum n-alkane input into the ocean from the atmosphere. There are few data for n-alkanes over the ocean, limited basically to those of three research groups: the Ketseridis-Eichmann-Hahn group, the Gagosian-Duce-Zafiriou group, and the Atlas-Giam group. Data from over the open North Atlantic Ocean, from the Irish coast, and from Cape Grim, Tasmania, in the Indian Ocean are from the Ketseridis-Eichmann- Hahn group {Ketseridis and Eichmann, 1978; Eichmann et al., 1979, 1980; Hahn, 19811. Data from Enewetak Atoll, Marshall Islands, in the tropical North Pacific, have been obtained by the Gagos ian-Duce- Zafiriou and Atlas-Giam groups (Gagosian et al., 1981, 1982; Zafiriou et al., 1982; Atlas and Giam, 19811. The observed concentrations of particulate and vapor phase alkanes in the marine atmosphere are presented in Tables 2-14 and 2-15. Data from three North Atlantic sites are presented. Loop Head is on a peninsula on the west coast of Ireland at about 52°30'~, 9°50'W. Samples were collected from a cliff about 70 m above sea level. Samples collected only when the wind was from the ocean are reported here. Samples were also collected from a ship at the Joint Air/Sea Interaction (JASIN) site, located between Iceland and Scotland (60°N, 13°W). The tropical North Atlantic samples were also collected from a ship, in this case operating in the North Atlantic trade wind regime at approximately 15°N between Africa and the Caribbean Sea. The German data from Cape Grim, Tasmania, were obtained from the Australian Baseline Atmospheric Monitoring Station located on the northwest tip of Tasmania (40°41'S, 144°40'E). Samples were collected on a cliff 90 m above sea level . The samples obtained by the Gagosian-Duce-Zafiriou and Atlas-Giam groups were collected from a 20 m tower located on the windward coast of Bokandretok Island, Enewetak Atoll, Marshall Islands (11°20'N, 162°20'E). Sample collection on Enewetak was controlled automatically by wind speed and direction as well as atmospheric particle counts to avoid local contamination. Efforts were made In all studies to avoid local contamination. Concerning the analytical methodology of measur ing hydrocarbons in atmospheric samples, a recent review by Simoneit and Mazurek (1981) and reports by Ketseridis et al. (1976) and Gagosian et al. (1981a) discuss the necessity of ultraclean samplers and sampling conditions. The need to separate the hydrocarbon classes from other organic compound classes (usually by liquid chromatography) before gas chromatography {GC) and GC/mass spectrometry for quantitative analyses and structural determina- tion was stressed by Gagosian et al. (1981a). The use of high resolu- tion glass capillary GC for analysis is needed. These hydrocarbon measurements must be made in conjunction with micrometeorolog~cal studies of the sampling site and long range transport studies to

69 TABLE 2-14 Particulate n-Alkane Concentrations in the Mar ine Atmospher e ~ ng/m3 STP ~ Loop Head, Tropical Republ ic of JASIN Cape Nor th _-Alkane Ireland Siteb Gr imp Atlantic Enewetakp n-cl5 0.06 0.10 0.12 4.1 n~l6 0 .13 0 . 12 0 .27 3 . 2 n~l7 0.08 0.26 0.32 4.2 n~l8 0 .19 0 . 17 0 .13 2 . 5 nail' O .16 0 .20 0.13 8 . 2 n'20 0 .33 0 .24 0 .17 1. 3 n'21 0.22 0.26 0.35 1.1 0.0017 n~22 0.26 0.28 0.11 3.0 0.0020 n'23 0.31 0.29 O.15 2.0 0.0023 n'24 0.45 0.21 0.18 0.4 0.0021 n'25 0.37 0 33 0.18 0.6 0.0030 n'26 0 .27 0 .28 0 .20 0 .5 0 . 0022 _~27 0 .23 0 .37 0 .52 0.6 0 . 0067 n'28 0.22 0.19 0.40 0.3 0.0037 n'29 0 . 0170 n'30 0 . 0033 Eichmann et al. (1979) and Hahn (19813. ~ akin (l9Bl). Hahn (1981) and Eichmann et al. (1980~. Ketseridis and Eichmann (19781. Gagosian et al. (1981a, 19821; average of six samples. ascertain the sources and transport pathways involved. Clearly, data are also needed on other anthropogenic compounds, such as chlorinated hydrocarbons, phthalate esters, and trace metals, alone with source marker Information suan as ~-~P~ ana 613C to interpret the hydrocarbon data more fully. Table 2-14 lists the particulate n-alkane data from C15-C30, and Table 2-15 presents the vapor phase n-alkane data for C10-C30. Particulate n-alkane data for C1O-Cl4 were not presented, since these compounds cannot be quantitatively recovered during the extraction of the filter with organic solvent, the solvent evaporation, and the liquid chromatography steps in the analytical scheme (Mackay and Wolkoff, 19731. As might be expected for such different oceanic regions, the measured concentrations of particulate and vapor phase n-alkanes wer e quite different at several of these locations. Data for particulate n-alkanes (Table 2-14 ~ from the Treland, JASIN, and Cape Gr im s ites ar e all rather similar, generally within a few tenths of a ng/m3. All

70 TABLE 2-15 Vapor Phase n-Alkanes in the Mar ine Atmosphere (ng/m3 STP ) Loop Head, Enewetakd Enewetal~ Republ ic of JASINCape Polyurethane Polyurethane _-Alkane Ireland] SitebGr imp Florosil Plugs P1U9S n-C1O 12 1521 null 14 92 n~l2 11 58 n~l3 9 8 0. 23 n~l4 9 36 0.19 n~l5 14 611 0. 66 n~l6 8 48 0.13 n~l7 10 59 0. 55 n~l8 12 58 0. 07 n{:l9 10 510 0. 07 n'20 18 69 0.07 n'21 14 611 0.07 n~C22 20 54 0. 07 not 32 56 0.08 0 .11 - 23 n'24 22 36 0.09 0 .14 0 . 032 n'25 16 36 0.10 0.14 0. 095 n'26 9 25 0 .08 0 .10 0 . 088 n'27 7 23 0.06 0.08 0.055 n~C28 6 12 0.06 0.024 n'29 0 .006 0 .019 -~30 0.013 ~Eichmann et al. (1979) and Hahn (1981). tHahn (1981), Eichmann et al. (1980) and Hahn (1981). Atlas and Giam (1981). eZaf ir iou et al . ( 1982 ) . are much lower, however, than the data from the tropical North Atlantic. The reason for this difference is unclear. The latter sampling area is in the region of the Sahara dust plume, which car r ies large quantities of sand and soil-der ived materials to the tropical North Atlantic in the northeast trade winds. However, the west coast of Africa would not be expected to be a significant source of petroleum- derived atmospheric n-alkanes, even though there is extensive tanker traffic along that coast. Viewed In the context of the other data presented in Table 2-14, it is tentatively concluded that the tropical Nortn Atlantic data reported are not representative of that region. Clearly additional measurements to evaluate these data are needed. The JASIN, Ireland, and Cape Grim data are probably most representative of concentrations over the North Atlantic and in the coastal regions of the other oceans. Particulate n-alkane concentrations from Enewetak Atoll are considerably lower than those over the North Atlantic or at Cape Grim. These data are probably more nearly representative of mid-North Pacific . . .

71 TABLE 2-16 Summary of Atmospheric Inputs of n-Alkanes Into the Ocean (mta) Mechan ism Case A Case B Rain scavenging of particles 0.023-0.230.0013-0.013 Rain scavenging of gases 1 x 10-7-0 .031 x 10-7-0 .00 2 Dry deposition of particles 0.0048-0.0480.00022-0.0022 Direct gas exchange 0-0.020-0.0004 TOTAL 0.28-0.320.0015-0.018 GRAND TOTAL 0.03_0.30.03_0.3 Ocean, South Pacific Ocean, South Atlantic Ocean, and Indian Ocean regions far from continental influences. Vapor phase n-alkane concentrations are presented in Table 2-15. Again, the Ireland, JASIN, and Cape Grim data are quite similar, but they are considerably higher than the Enewetak data, generally by a factor of 50-100. Note that there is, in general, good agreement between the Gagosian-Duce-Zafiriou and Atlas-Giam data at Enewetak for n-C24 to n-C29 alkanes. There are no vapor phase n-alkane data available from the tropical North Atlantic region. Again, the Ireland' JASIN, and Cape Grim data appear to be most representative of concentrations over the North Atlantic Ocean and in coastal regions, while the Enewetak data may be more representative of concentrations over the Indian, South Atlantic, South Pacific, and mid-North Pacific Oceans. As can be seen from the data presented in Tables 2-14 and 2-15, the geographical coverage for atmospheric n-alkanes is very sparse. However, from this limited data, mean atmospheric particulate and vapor phase n-alkane concentrations over the world ocean have been derived, and input of this class of hydrocarbons into the oceans estimated. Details of the approach used are given by Duce and Gagosian (19823. Table 2-16 presents a summary of calculations of the input of n-alkanes into the ocean. The total estimated input of atmospheric - n-alkanes is 0.03-0.3 mta. The primary ~nput mechanism clearly is via rain scavenging of n-alkanes on particles. However, better solubility data for n-C;0 to n-C30 alkanes are needed before the importance of rain scavenging of gases and direct gas exchange in the deposition of n-alkanes to the sea surface can be fully assessed. The estimates of the input of n-alkanes into the ocean via rain could be evaluated relatively easily by making measurements of the n-alkane concentrations in rain from samples collected, for example, in open North Atlantic and North Pacific regions--the regions in which most of the atmospheric petroleum hydrocarbons are apparently entering the oceans. Such rain measurements are strongly recommended.

72 As stated earlier, n-alkanes constitute approximately 30% of the organic components of petroleum. Cycloalkanes, PAH, and heteroatomic (nitrogen, sulfur, and oxygen) organic compounds make up the remainder. No data are available for the cycloalkane and heteroatomic compounds. Only a few numbers are available for PAH. Measurement of these other organic constituents of petroleum in vapor, aerosol, and rain samples are also strongly recommended. The approach taken in using n-alkanes to estimate the input of petroleum into the ocean via the atmosphere is problematic. On one hand, using n-alkanes may give a maximum value of petroleum hydrocarbon atmospheric input, because many natural marine and terrestrially derived n-alkanes are included in the overall n-alkane deposition value. On the other hand, many organic components of petroleum, such as branched alkanes, cycloalkanes, and alkylated aromatics--the latter of which react very fast with OH radical to produce oxygenated species that fall to the ocean surface--are not included in the approach presented here. This suggests that using n-alkanes as an atmospheric input marker for the petroleum underestimates the input. On the basis of these facts, the estimate of PHC input is increased about two-thirds over that of the total n-alkane input. The range estimate for PHC input is thus about 0.05-0.5 mta. More precise estimates of the atmospheric input of petroleum to the ocean will have to await information on the inputs of the various components of petroleum into the sea surface and further understanding of the reaction products, pathways, and rates of transformation of these compounds in the atmosphere. COASTAL, MUNICIPAL, AND INDUSTRIAL WASTES AND RUNOFF The estimated range of the input of PHC into the mar ine environment from municipal and industrial wastewaters, urban and river runoff, and ocean dumping is from 0.6 to 3 .1 mta, with a best estimate of 1.2 mta (Table 2-17~ . Municipal wastewater appears as the largest contributor, followed by industrial discharges and urban runoff. The earlier NRC (1975) study did not estimate a range of inputs of PHC from these sources, but made only a best estimate of 2.7 total mta. Many more data on these inputs have been accumulated over the past 7 years, so the lower estimates may be due in major part to better predictive capability and not necessarily to lower actual inputs. Municipal Wastewaters In 1979, Eganhouse and Kaplan (1981, 1982) analyzed 38 samples of treated municipal wastewater from five major wastewater pollution control plants in Southern California as reported by the Southern California Coastal Water Research Project (SCCWRP, 1980~. The workshop panel decided to use four of these discharges in making their estimates for facilities serving approximately 9.8 million people in 1979.

73 TABLE 2-17 NRC Estimates of Hydrocarbons to World Ocean From Municipal and Industrial Wastes and Runoff (mta) 1981 NRC Workshop Source NRC (1975) Most Probable Likely Range Municipal wastewater 0.3 0.75 0.4-1~5 Industrial Nonrefinery 0.3 0.2 0.1-0.3 Refinery 0.2 0.1 0.06-0.6 Urban runoff 0.3 0.12 0.1-0.2 River discharges 1.6 0.04 0.01-0.45 Ocean dumping ~0.014 0.005-0.02 TOTAL 2.7 1.2 0.6-3.1 Knot estimated. The wastewater samples were analyzed for total extractable organics and for total hydrocarbons (THC). The results of these analyses were compared with reported concentrations of oil and grease from the routine monitoring done by the was tewater management agencies as reported by SCCWRP (1980~. Regression analysis indicates that THC accounts for approximately 38% of the oil and grease discharged from these treatment plants. The total mass emission from the four discharges is estimated to be approximately 43 mta in 1979, resulting in an overall contribution of oil and grease of about 12 grams per capita per day (g/cap/d). These results can be used to calculate that the total per capita contr ibution of THC from the Southern California outfalls in 1979 was 38% of 12 g/d or 4 . ~ g/d. The type and level of treatment a iven to the wastewater will affect J - san itarY end ineer inq the amounts of THC discharged. Based on general _ experience with the removal of oil and grease in municipal wastewater, it is reasonable to assume an average removal of about one-third of the PHC in primary treatment and about 40% in secondary treatment. These removals can vary widely from plant to plant, depending on the plant design and operation. AS most of the effluents in the Southern California wastewaters had been given primary treatment, the THC load in the untreated wastewater would be about 6.8 g/cap/d from municipal wastewaters {4.5 divided by 0.67~. In 1978, 120 million people lived within 50 miles of the coasts of the United States (U.S. Census Bureau, 1978~. About 30% of this population lived on the West Coast of the continental united States, and 70% on the Gulf and East coasts. Assuming that the bulk of the wastewaters on the West Coast are g iven primary treatment , and those in

74 the remainder of the country receive secondary treatment, the THC d ischarged to the U . S . coastal water s would be [120 x 106 x 0.3 x 6.8 (10-0.33) x 10-6 x 365] ~ [120 x 106 x 0.7 x 6.8 (1.00-0.40) x 106 x 365] = 185,000 ta, or 0.19 men This calculation assumes that the oil and grease values reported by Eganhouse and Kaplan (1981, 1982) for the Los Angeles area are repre- sentative of discharges throughout the United States. Evidence that this estimate of 0.19 mta for the entire United States is not too far out of line comes from Connell (1983~. He reported that about 0.012 mta of petroleum hydrocarbons are going into the Hudson- Raritan Estuary from sewage discharge. This represents about 6-7% of the overall U.S. estimate for sewage discharges. The calculated per capita THC discharge rate of 6.8 g/d cannot be used for other areas of the world because of the widely varying usage of petroleum products. In 1980, the United States used 18.3 million barrels per day (bbl/d) of petroleum products (International Petroleum Encyclopedia, 1980), and the estimated discharge of THC to the coastal waters if the wastewaters were untreated would have been about 298,000 ta (120 x 106 x 6.8 x 10-6 x 365), or about 16.1 ta for each 1,000 bbl/d consumed. This factor and an estimate of the extent of wastewater treatment in various areas of the world are used to estimate a global discharge, as shown in Table 2-18. The estimated global discharge of 0.75 mta is based on a series of assumptions that are supported by few data. Note that one of these assumptions is the equivalency of THC with PHC. There is no doubt that PHC makes up a ma jor fraction of THC, but the exact percentage is not known. However, the calculations do provide a rationale for the estimation procedure and show the areas in which measurements and data are required. Nonref inery Industrial Wastes A sizable fraction of nonrefinery industr ial waste discharges into municipal wastewater systems and its PHC have been accounted for in the previous section. However, there is a quantity of PHC that goes more or less directly into the marine environment through coastal nonrefinery effluent discharges. Extremely limited quantitation of this source has been made, and even less information is published for reasons of confidentiality. Previous estimates have been made by the NRC (1975) of 0.3 mta, and the Royal Commission on Environmental Pollution (1981) of 0.150 mta. Therefore, the estimate of this input of PHC is put at 0.2 mta, with full realization that the confidence in this number is quite limited.

75 TABLE 2-18 Global Discharge of Hydrocarbons Into Municipal Wastewaters 1980 Petroleum Estimated Percent Residual Consumptiont Untreated THC THC (= PHC) (millions THC Loa ~ Removed by Discharged Area of bbl/d) (mta) Treatment (mta) North America United States 18.3 0.30 Sac 0.19 Canada 1.8 0.03 38c 0.02 Latin America 4.2 0.07 0 0.07 Asia and Pacific 9.1 0.15 0 0.15 China 1.7 0.03 0 0.03 Middle East 2.0 0.03 0 0.03 USSR and Eastern Europe 10.5 0.17 30d 0.12 Western Europe 10.5 0.17 30d 0.12 Africa 1.2 0.02 0 0.02 TOTAL 63 .1 0 .97 0 .7 5 ~Source: International Petroleum Encyclopedia (1980) "Assuming 16.1 ta of THC per 1,000 bbl/d consumed. c1 _ 185, 000/298, coo = 0 · 38 ~Assumed. Industrial Wastes From Refineries This category of refinery discharges includes only those refineries that discharge PHC from their own wastewater facilities. Other refineries that do not have their own facilities are assumed to discharge their wastewater into municipal wastewater facilities. Recently, estimates were made of the amount of PHC discharged with refinery industry effluents (National Petroleum Council, 1981). The National Petroleum Council (NPC) determined during 1977-1979 that for those refineries that treat and discharge their own wastewater, 0.002-0.004 mta of PHC were discharged. These values were based entirely on oil and grease analyses (one can assume that volatile liquid hydrocarbons were not analyzed). The NPC related PHC discharge rates to total operating capacity of U.S. discharge refineries. It is estimated that 0.0025-0.005 kg of PHC is discharged annually for each 103 kg/yr of operating capacity. This value can be compared to a 1977 European value of 0.04 kg/103 kg (Royal Commission on Environmental Pollution, 19811. It also can be compared to the 1967 U.S. refinery survey value of 0.075 kg/103 ky (National Petroleum Council, 1981) .

76 Selection of PHC discharge rates for the world is difficult. The United States seems to be unique in its rate estimate. Either its rate of 0.005 kg/103 kg (upper value of U.S. range) or the European rate of 0.04 kg/103 kg could be applied to the Canadian refinery rate. For these calculations, the higher rate was used for Canada. For the rest of the world, the PHC discharge rate is assumed to be no better than that which was measured in the 1967 U.S. refinery survey. Estimating the fraction of PHC that reaches the ocean from all worldwide refinery sources is also difficult. In the United States, the Environmental Protection Agency (EPA) has determined the amount of refinery wastewater discharged directly into receiving bodies and indirectly into publicly owned treatment plants (Environmental Protection Agency, 1978, 1979~. The following percentages are based on the processing capacity of refineries, not numbers of refineries. Approximately 81% of all refineries discharge wastewater directly into receiving bodies; another 14% discharge indirectly to publicly owned treatment plants; the remaining 5% have no wastewater discharges. In California, 64% of all refineries discharge directly into receiving bodies; the rest into publicly owned treatment plants. In the con- tinental United States, approximately 50% of all refineries discharging directly into receiving bodies are near the coast (an additional 7% occurs outside the continental United States). Of all the refineries discharging into public treatment plants, 63% are near the coast. For the United States, the fraction of PHC directly reaching the world oceans from refinery sources is estimated to be 0.5. The fraction of PHC reaching the world oceans from other locations of the wor Id is a rough est. imate based on 1 imited data . The new values of 0 .005 kg PRC/103 kg production for the United States, 0.04 for Canada and Europe, and 0.075 for the rest of the world have been used to obtain a ref inery PHC global discharge estimate of 0.10 mta (Table 2-197. The Royal Commission on Environmental Pollution ( 1981 ) estimates a total global discharge from ref iner ies to the sea of 0.06 mta. Urban Runoff The global input of petroleum hydrocarbons to coastal waters from urban runoff was estimated by NRC (1975) to be 0.3 mta. The value was based in part on the assumption that urban runoff contributed about half the amount of PHC contributed by municipal and nonref~nery wastewaters. The crudeness of this estimate was unavoidable because of the lack of measurements of PHC in urban runoff. The situation 8 years later is only slightly better because most of the studies undertaken in the intervening years have focused on analytical methods of characterizing the PHC fractions rather than on mass contributions of PHC. Part of this dilemma may be due to the difficulty of representative sampling of the r unoff. Other problems are the determination of mean PHC concentra- t~ons and the volume of runoff, which permit accurate estimation of

77 TABLE 2-19 Estimated Global Discharge of Petroleum Hydrocarbons in Refinery Wastewaters Total PHC Crude Oil Estimated Discharge PHC Input Refinery Hydrocarbon in Refinery Estimated Into Geographic Capacity Loss Wastewaters Fraction to Ocean Area 10~ bbl/d mta (kg/103 kg) (mta) World Ocean (mta) North America United States18.4960 0.005 0.0048 1/2 0.002 Canada2.2115 0.04 0.0046 1/5 0.001 Latin America8.1420 0.075 0.0315 4/5 0.025 Asia-Pacific10.6550 0.075 0.0413 1/2 0.021 China1.890 0.075 0.0068 1/4 0.002 Middle East3.7190 0.075 0.0143 1/3 0.005 USSR and Eastern Europe14.2740 0.075 0.0555 1/3 0.018 Western Europe20.221,050 0.04 0.0420 1/3 0.014 Africa1.790 0.075 0.0068 1/2 0.003 TOTAL80.94,205 0.1946 0.10 (best estimate) NOTE: Conversion is 1 mta = 19,000 barrels per calendar day capacity. mass PHC contributions. Estimates of PHC in runoff should be based on factors such as runoff area, watershed characteristics, PHC usage, and population density. Recent papers that have r epor ted on the character ization of PHC in urban runoff appear in Table 2-20 . Most of the studies in Table 2-20, except that of Hoffman et al. (1982), do not provide sufficient data for the estimation of PHC contributions from urban runoff based on watershed characteristics. There have not been enough studies reporting watershed characteristics to permit rational estimation. Recognizing the cliff iculties of quantifying the mass of PHC contr ibuted and consider ing hydrological, physical, and land use var. iat~ons in urban areas (as well as the def inition of urban), we conclude that the best estimate of urban PHC r unof f must be based on estimates of per capita contr ibutions . Population is one of the principal generating factors of urban PHC runoff for a given petroleum consumption level. Table 2-21 shows data for per capita estimates of PHC contributions from several locations . Despite the gross variation in per capita PHC contribution, it is believed to be the most accurate basis for current estimation of urban PHC runoff. A per capita PHC contribution of 1.0 g/cap/d is probably the most reliable estimate that can be made from present information. Employing the unit per capita contribution of 1.0 g/cap/d per day and a coastal population of about 120 million, one can estimate the urban r unof f contr ibution of the United States to be about 0.04 mta. Assuming the United States uses about 0.3 of the world' s hydrocarbons, one can estimate the world urban runoff PHC contribution to the world ocean to be about O .12 mta, which is about one-third of the contr ibution e stimated by NRC ( 1975 ~ .

78 TABLE 2-20 Selected Urban Runoff Studies Location Drainage Area (ha) Number of Storms Studied Reference Seattle-- Wakeham (1917) North Philadelphia616 Hunter et al. (1979) North Philadelphia6165 Hunter et al. (1979) North Philadelphia616 Whipple and Hunter ( 19 7 9 Trenton A82 Wh ipple and Hunter (1979) Trenton B823 Tipple and Hunter ( 1979 Los Angeles210,0001 Eganhouse and Kaplan (1981, 1982) Leon County, Fla. 357 1 Byrne et al . (1980 Narragansett Bay, R.T . 167,000 21 Hoffman et al. (1982) River Discharges Reexamination of the global input of hydrocarbons to the oceans indi- cates that the inclusion of a separate category for river discharges may be improper because of double accounting of hydrocarbon input. The ma jor sources of hydrocarbons in rivers are the untreated and treated wastewater discharges, runoff {both urban and rural) , and spills. All these sources are quantified and reported separately for coastal areas. If an additional 110 million people discharge PHC into the interior r ivers of the United States (at a rate of 6 .8 g/cap/d) and if 5% of these PHCs eventually reach the oceans, then this yields an annual flux of PHC from rivers to U.S. coastal waters of 0.013 mta. Assuming this amount is one-third of the world total, the river discharge of PHC to the ocean would be 0.04 mta. OCEAN DUMPING Some hydrocarbons are discharged into U.S. and world coastal regions in association with municipal wastewater treatment plant sludge/underflow. The sludge is generally discharged from dumping by barge or by dis- charges through pipelines. In the United States, this sludge is discharged by dumping in the New York Bight and by Pipeline on the West Coast. In the New York Bight ~ approximately 7 x 10 wet tons of sludge are discharged per year. This material contains approximately 2,000 ppm of oil and grease, of which about 4096 are hydrocarbons. This

79 TABLE 2-21 Per Capita Estimates of PHC Contr ibutions in Urban Runof f On it PHC Contr ibution Location (g/cap/d) Reference Ph iladelph ia 0 . 0 3 ~ ipple and and Trenton Hunter (1979 Narragansett Bay 2.7 Hoffman et al. (1982 ~ Los Angelesa 1.9 Eganhouse and Kaplan (1981) Seattle 0.3 Wakeham (1977) Swede ~0.3 NRC (1975) BSingle storm extrapolated to annual runoff by author. bTypical urban area (0.2 parking, 0.3 multifamily, and 0.6 single family). amounts to 0.006 mta of THC. In addition, the Los Angeles pipeline discharges about 2,4SO tons of oil and grease annually through the 7 mile sludge outfall (Eganhouse and Kaplan, 1981~. This is estimated to be 0.001 mta of TIC. The annual worldwide discharge of wastewater sludge into the oceans is approximately 16 million tons. Thus, applying a similar ratio to that used for the United States, the total amount of hydrocarbons discharged worldwide by ocean dumping is about O .02 mta. Hydrocarbons are also released to the oceans from the dumping of dredge spoils. Dredge spoils are river and channel sediments that have been relocated by dredging and dumping operations. The hydrocarbons that accompany these spoils are accounted for in other sections of this report and are not included in the ocean dumping category. GEOGRAPHICAL DISTRIBUI ION OF INPUTS The input of petroleum hydrocarbons into the ocean is certainly not distr ibuted evenly. The geographical distr ibution of the inputs from each source is discussed below. · Mar ine transportation (1.5 mta) . The input of PHC from this source is concentrated along the pr incipal transportation routes and in harbors and ports where oil tankers or other vessels are loaded or unloaded. About half the transportation total is der ived from tanker operations (0.7 mta). Most of this loss is probably at sea along the prominent tanker routes from the Middle East to Europe, the American

80 continents, or the Far East. Another major source in this category is tanker accidents (0.4 mta). These also tend to occur along the tanker routes, but in more congested areas near ports or in narrow straits. The third major source, that of bilge and fuel oils (0.3 mta), probably follows a similar distribution pattern to that of the tanker operations. · Offshore oil production (0.05 mta). This relatively minor input occurs at offshore oil production facilities, and these tend to be near coastlines. The largest offshore producing areas are the Arabian Gulf, the North Sea, the Gulf of Mexico, offshore California, offshore Malaysia and Indonesia, and the west coast of Africa. · Refineries (0.1 mta). This input of PHC into the sea is con- centrated near the coasts of countries that do most of the refining of petroleum (e.g., the United States, Great Britain, Germany, France, Japan, Canada, Mexico, Kuwait, and Saudi Arabia). · Nonrefinery wastes (0.2 mta). This input into the sea is concentrated near the coasts of the more industrialized nations in the world, such as the United States. the northern EuroDean countries and Japan. , _ , _ _ , · Municipal wastes (0.75 mta). This input of PHC is distributed in much the same way as the nonrefinery industrial wastes. It would be concentrated near the coasts of the more highly industrialized and heavily populated nations. Best examples would again be the united States, the northern European countries, and Japan. · Urban runoff (0.12 mta). This input of PHC closely follows the input from municipal wastes. The input would be primarily into coastal areas of counts ies with high industrialization and large populations. · River runoff (0. 04 mta). This input is in coastal areas near the mouths of large rivers, such as the MiSSiSSippi, the Rhine, the Danube, the Saint Lawrence, and the Elbe. · Natural sources (0.3 mta). Submarine seeps, at least those identified thus far, seem to be associated with tectonically active regions of the world and are usually near the coasts of continents. Such areas are offshore California and Alaska, the Arabian Gulf and the Red Sea, the northeast coast of South America, and the South China Sea. · Atmosphere (0.05-0.5 mta). This input of PHC into the seas would be primarily downwind of heavily industr ialized areas. Again, the inputs are greatest near the coastlines, with concentrations decreasing away from the coasts. The northwest Atlantic, the North Sea, and the northwest Pacific (near Japan) would probably have typically large atmospheric inputs of PHC. Data are not available to estimate total PHC input by region except in an extremely qualitative manner. If one looks at information on the geographical distribution of each input, then one can say, qualitatively, that coastal areas off the United States, Europe, and Japan and the Arabian Gulf would probably have greater inputs.

81 SUMMARY AND RECOMMENDATIONS The estimated range for total input of petroleum from all sources is 1.7-8.8 million mta. The best single-number estimate of total input is 3.2 mta. We believe that the range is a more accurate summary of the state of knowledge than a single-number best estimate. Uncertainties are particularly evident with certain sources, i.e., natural inputs (seeps and erosion), transportation, municipal/industrial runoff, and atmospheric inputs. There are also wide geographical gaps in informa- tion on sources, especially in the southern hemisphere. Table 2-22 presents sources, probable ranges, and best estimates for sources. The spread in probable range about the best estimate is a qualitative measure of the faith in the best estimate. For example, the tanker accident probable range is narrow (0.3-0.4 mta), so the best estimate is probably good . On the other hand, the mar ine seep probable r ange is wide (0.02-2.0 mta), indicating small reliability in the best estimate. The 1975 NRC report gave only a single-number estimate of total input of petroleum, namely, 6.1 mta. No range was given. This number falls within the current estimated range of 1.7-8.8 mta. The difference in the two single-number estimates, 6.1 mta in 1975 and the current 3.2 mta, does not necessarily reflect a significant decline in input but indicates better estimation of individual inputs. Although the amount of petroleum and petroleum products transported by sea, as well as crude oil produced offshore, has increased during the past 8 years, PHC input into the marine environment estimated at the 1981 NRC workshop does not appear to have followed this trend. This may be for the following reasons: (1) the individual input estimates are more accurate due to improved analytical data on PHC concentrations in effluent streams, (2) positive steps have been taken to reduce operational and accidental release of petroleum into the sea, and (3) double accounting of PHC inputs from sources has been reduced. Double accounting arises when it becomes difficult to distinguish PHC inputs from closely related sources (e.g., urban runoff, river runoff, industrial and municipal wastes). Thus, there may be the tendency to count the same PHC inputs twice or more times under different sources. One source of PHC into the marine environment that was not estimated was PHC released from pleasure craft, primarily in near-coastal marine waters. Pleasure craft are primarily small inboard or outboard motor- boats. While inputs from pleasure craft may be locally significant, we believe that the total amount of this input would not be on the same scale with the other inputs considered. Major problems still remain in the estimation of PHC inputs into the marine environment. Certainly, significant improvements have been made in recent years in obtaining better analytical data on concentra- tions of PHC entering the marine environment from varied sources. However, additional work is still needed, particularly in the acqui- sition of improved data on PHC inputs from the atmosphere, from municipal and industrial waste sources, and from natural sources such as marine seeps and erosion of terrestrial sediments. Following is a list of recommended research programs or projects that would address these problems:

82 TABLE 2-22 Input of Petroleum Hydrocarbons Into the Mar ine Environment (mta ~ Sour ce Probable Range Best Estimatea Natur al sources Plar ine seeps 0 . 02-2 . O O .2 Sediment erosion 0 . 005-0 .5 0 .0 5 (Total natural sources) (0.025) - (2.5) (0.25) Offshore production 0.04-0-.06 0.05 Tr anspor tat ion Tanker operations 0.4-1.5 0.7 Dry-docking 0.02-0.05 0.03 Mar ine terminals 0 . 01-0 .03 0 .02 Bilge and fuel oils 0 . 2-0 .6 0 .3 Tanker acc idents 0 . 3-0 . 4 0 . 4 Nontanker accidents 0.02-0.04 0.02 (Total transportation) (0.95) - (2.62) (1.47) Atmospher e 0 . 0 5-0 . 5 0 . 3 Municipal and industr ial wastes and runoff Municipal wastes 0 . 4-1 .5 0 .7 Refineries 0.06-0.6 0.1 Nonr ef ining industr ial wastes 0 .1-0 .3 0 .2 Urban runoff 0.01-0.2 0.12 - River runoff 0.01-0.5 0.04 Ocean dumping 0 . 005-0 . 02 0 .0 2 (Total wastes and runoff) (0 .585) - (3.12) (1.18) TOTAL 1. 7-8 . 8 3 . 2 Ache total best estimate, 3.2 mta, is a sum of the individual best estimates. A value of 0.3 was used for the atmospheric inputs to obtain the total, although we well realize that this best estimate is only a center point between the range limits and cannot be supported r igorously by the data and calculations used for estimation of this input. 1. Improved methods should be developed for large scale, areal documentation of the continental margins to determine the extent of submarine seepages of petroleum. A program should be undertaken to gauge accurately flow rates for seeps of significantly different sizes, including probable microseeps. 2. There should be continued monitor ing of all facilities dis- charg~ng low levels of petroleum hydrocarbons dispersed or dispersed in

83 aqueous effluents (e.g., offshore platforms, refineries, and other industrial plants and transportation units such as tankers and terminals). 3. Rain samples collected from several locations on the ocean and near sea coasts should be analyzed for PHC content. This work is important since rain scavenging of atmospheric particles is believed to be the major pathway for petroleum into the ocean from the atmosphere. It is also necessary to determine reactions of, and changes occurring in, various petroleum components as they are transported from sources through the atmosphere across and into the oceans. 4. More applied investigations, including accurate measurements of PHC, are needed to better define municipal, industrial, and runoff inputs to the oceans. This is particularly needed in southern hemis- phere countries. These investigations may lead to quantitative methods for distinguishing petroleum hydrocarbons from oil and grease and natural hydrocarbons found in municipal and industrial waste as well as samples of runoff. 5. Data should be collected on the C2-C1O aliphatic hydrocarbons in vapor, particulate, and rain samples from over the oceans, to relate these to the distributions of other classes of organic compounds present in petroleum. 6. Better solubility data are needed for n-alkanes and polynuclear aromatic hydrocarbons to better ascertain the importance of rain scavenging of gases and air-sea gas exchange processes to the contribu- tion of the flux of atmospheric petroleum hydrocarbons to the ocean. 7. There is a need to determine the reactions and organic compound class distributional changes that occur for the various organic com- pounds in petroleum, as this material is transported from its source through the atmosphere across the oceans. 8. Better solubility data are needed for n-alkanes, polynuclear aromatic hydrocarbons, etc., to better ascertain the importance of rain scavenging of gases and a~r-sea gas exchange processes to the contribu- tion of the flux of atmospheric petroleum hydrocarbons to the ocean. REFERENCES Albers, J.P., M.D. Carter, A.L. Clark, O.B. Coury, and S.P. Schweinfurth. 1973. Summary of petroleum and selected mineral statistics for 120 countries, including offshore areas. Professional Paper 817. U.S. Geological Survey, Washington, D.C. 149 pp. Atlas, E., and C.S. Giam. 1981. n-Alkane atmospheric input into the tropic North Pacific Ocean. Unpublished manuscript. Texas A&M university. Bertrand, A.R.V. 1979. Les principaux accidents de diversements patrollers en mer et la banque de donnees de l' Institute Francais du Petrole sur les accidents de navires (1955-1979~. Rev. Inst. Francais Petrole 34:3-7. Bjorseth, A., G. Lunde, and A. Lindskog. 1979. Long range transport of polycyclic aromatic hydrocarbons. Atmos. Environ. 13:34-53.

84 British Petroleum Co., Ltd. 1980. BP statistical review of the world o ~1 industry . London . Burnet, B. 1980. Worldwide drilling and production. Offshore 40:62-70. Byrne, C., C.R. Donahue , and W.C . Burnett. 1980 . The effect of urban s tormwater r unof f on the water qual ity of Lake Jackson . Unpubl ished manuscr ipt. Flor Ida State University, Tallahassee. Chameides, W.L., and R.J. Cicerone. 1978. Effects of non-methane hydrocarbons in the atmosphere. J. Geophys. Res. 83: 947-952 . Connell, D.W. 1983. Sources and fates of petroleum hydrocarbons in the Hudson-Raritan Estuary. Coastal Ocean Pollut. Assessment News 2~4~:39-39. Danenberger, E.P. 1976. Oil spills, 1971-1975, Gulf of Mexico outer continental shelf. Circular 741. U.S. Geological Survey, Washington, D.C. 47 pp. Darnell, K.R., A.C. Lloyd, A.M. Winer, and J.N. Pitts, Jr. 1976. Reactivity scale for atmospher ic hydrocarbons based on reaction with hydroxyl radical . Environ. Sci. Technol. 10: 692-696 . Demaison, G.J . 1977 . Tar sands and super giant oilf ields . Am. Assoc . Petrol. Geol. Bull. 61 :1950-1961. Department of the Environment. 1976. The Separation of Oil From Water for North Sea Oil Operations. Her Majesty's Stationery Office, London. 29 pp. Desel, R.F. 1972. Improving the ocean environment through tanker operating techniques, pp. 81-91. In Proceedings of the Seventeenth API Tanker Conference. American Petroleum Institute, Washington, D.C. Duce, R.A. 1978. Speculations on the budget of particulate and vapor phase non-methane organic carbon in the global troposphere . Pur e Appl . Geophys. 116: 244-273 . Duce, R.A., and R.B. Gagosian. 1982. The input of atmospheric n-C1O to n-C30 alkanes to the ocean. J. Geophys. Res., in press. Eganhouse, R.P ., and I .R. Kaplan. 1981. Extractable organic matter in urban stormwater runoff. I. Transport dynamics and mass emission rates to the ocean. Environ. Sci. Technol. 16 :180-186. Eganhouse, R.P., and I .R. Kaplan. 1982 . Extractable organ~c matter in municipal wastewaters. 1. Petroleum hydrocarbons: temporal variations and mass emission rates to the ocean. En~riron. Sci. Technol. 16 :180-186. Eichmann, R., P. Neoling, G. Ketseridis, J. Hahn, R. Jaenicke, and C. Junge. 1979. n-Alkane studies in the troposphere. I. Gas and particulate concentrations in north Atlantic air. Atmos. Environ. 13: 587-599 . E ichmann, R., G . Ketser idis , G . Schebeske , R. Jaenicke, J . Hahn, P. Warneck, and C. Junge. 1980. n-Alkane studies in the troposphere II. Gas and particulate concentration in the Indian Ocean air . Atmos. Environ. 14: 695-703 . Fischer, P.J. 1978. Natural gas and oi1 seeps, Santa Barbara Basin, pp. 1-62 . In The State Lands Comnission 1977, California Gas, Oil, and Tar Seeps . Sacramento, Cal if .

85 Gagosian, R.B., E.T. Peltzer, and O.C. Zafiriou. 1981a. Atmospheric transport of continentally der ived 1 ipids to the tropical North Pacific. Nature 290: 312-314. Gagosian, R.B., E.T. Peltzer, and O.C. zafiriou. 1981b. SEAREX News 4~2~:31-35. Gagosian, R.B., O.C. Zafiriou, E.T. Peltzer, and J.B. Alford. 1982. Lipids in aerosols from the tropical North Pacif ic: temporal variability. J. Geophys. Res., in press. Geyer, R.A., and C.P. Giammona. 1980. Naturally occurring hydrocarbons in the Gulf of Mexico and Caribbean Sea, pp. 37-106. In R.A. Geyer, ed. Marine Environmental Pollution. Elsevier, Amsterdam. Giammona, C.P. 1980. Biota near natural marine hydrocarbon seep in the western Gulf of Mexico, pp. 207-228. In R.A. Geyer, ed. Marine Environmental Pollution. Elsevier, Amsterdam. Gray, W.O. 1978. Oil Tanker Pollution: Proceedings of Hearings Before a Subcommittee of the Committee on Government Operations, House of Representatives, Ninety-fifth Congress, Second Session, July 18-20. Washington, D.C. Pp. 94-111. Hahn, J. 1980. Organic constituents in natural aerosols, pp. 359-376 . In T.J . Kneip and P.J. Lioy, eds. Aerosols: Anthropogenic and Natural Sources and Transport. Annals N.Y. Acad. Sci. 338. Hahn, J. 1981. n-Alkane atmospheric input into the open North Atlantic Ocean, near the Irish coast, and Indian Ocean. unpublished manuscr ipt. Max-Planck Institute. Halbouty, M.T., and J.D. Moody. 1980. World ultimate reserves of crude oil, pp. 291-302. In Proceedings of the Tenth World Petroleum Congress. Vol. II, Exploration Supply and Demand. Heyden and Son Ltd., London. Hanst, P.L., J.W. Spence, and E.O. Edney. 1980. Carbon monoxide production in photooxidation of organic molecules in the air. Atmos. Env~ron. 14 :1077-1088. Harbaugh, J.W., and M. Ducastaing. 1981. Historical changes in oilfield populations as a method of forecasting f ield sites of undiscovered populat ions: a comparison of Kansas, Wyoming, and California. Subsurface Geology Series 5. Kansas Geological Survey, Lawrence, Kansas. 56 pp. Harvey, G.R., A.G. Requejo, P.A. McGillivary, and J.M. Tokar. 1979. Observation of a subsurface oil-rich layer in the open ocean. Science 295:999-1001. Hoffman, E.J., J.S. Latimer' G.L. Mills, and J.G. Quinn. 1982. Petroleum hydrocarbons in urban runoff from a commercial land use area. J. Water Pollut. Control Fed., in press. Hunt, J .M. 1972 . Distr ibution of carbon in the crust of the earth. Am. Assoc. Petrol. Geol . Bull . 56: 2273-2277 . Hunt, J.M. 1979. Petroleum Geochemistry and Geology. W.H. Freeman and Sons, San Francisco, Calif. 167 pp. Hunter, J.V., T. Sabatino, R. Gomperts, and M.J. ~lacKenzie. 1979 . Contr ibution of urban runoff to hydrocarbon pollution . J . Water Pollut. Control Fed. 51:2129-2138. Inter-Governmental Maritime Consultative Organization. 1981. Estimates on inputs of petroleum hydrocarbons into the oceans due to mar itime

86 transportation activities. Special Report from Meeting of Experts, convened by IMCO on 26-29 May 1981. London. 19 pp. International Petroleum Encyclopedia. 1980. Volume 13. Penn Well Publishing, Tulsa, Okla. 464 pp. International Tanker Owners Pollution Federation. 1981. Computer Data Base on Tanker Accidents Involving Oil Pollution. London. Johnson, T.C. 1971. Natural oil seeps in or near the marine environment: a literature survey. Report Project No. 714141/002. U.S. Coast Guard, Washington, D.C. 30 pp. Ketser idis , G ., J . Hahn, R. Jaenicke , and C . Junge . 1976 . The organic constituents of atmospher ic particulate matter . AtmOS. Environ. 10: 603-610. Ketser idis , G ., and R. E ichmann . 1978 . Organic compounds in aerosol samples . Pur e Appl. Geophys. 116:274-282. Andes, K.K. 1973. Mother nature as an oil polluter . Am. ASSOC. Petrol. Geol . Bull . 57: 637-641 . Levy, E.M. 1978 . Visual and chemical evidence for a natural seep at Scott Inlet, Baffin Island District of Franklin, pp. 21-25. Current Research Part B. Paper 78-1B. Geological Survey of Canada. Levy, E.M., and M. Ehrhardt. 1981. Natural seepage of petroleum at Buchan Gulf, Baffin Island. Mar. Chem. 10:355-364. Lloyd's Register of Shipping. 1980. Statistical Tables. London. Mackay, D., and A.W. Wolkoff. 1973. Rate of evaporation of low solubility contaminants from water bodies to atmosphere. Environ. Sci. Technol. 7:611-614. Marine Pollution Subcommittee of the Br itish National Committee on Oceanic Research. 1980. The Ef fects of Oil Pollution: Some Research Needs. London. 108 pp. McKirdy, D.~., and Z. Horvath. 1976. Geochemistry and significance of coastal bitumen from southern and northern Australia. APEA J. 16:123-135. Meybeck, M. 1981. River transport of organic carbon to the oceans, pp. 219-269. Carbon Dioxide Effects Research and Assessment Program Report 016-8009140. U.S. Department of Energy, Office of Energy Research, Washington, D.C. National Oceanic and Atmospheric Administration. 1980. Proceedings of a Symposium on Preliminary Results from the September 1979 Researcher/Pierce Ixtoc I Cruise. U.S. Department of Commerce, Office of Mar ine Pollution Assessment, Washington, D.C . National Petroleum Council. 1981. Environmental Conservation in the Oil and Gas Industry. National Petroleum Council, Washington, D.C. 80 PPe National Research Council. 1975. Petroleum in the Marine Environment. National Academy of Sciences, Washington, D.C. 107 pp. Oil Spill Intell igence Repor t. 1979 . Oil spills in 1978--an international summary and review 2 (121 :1-20. Oil Spill Intelligence Report. 1980. Oil spills in 1979--an international suImnary and review 3 (21) :1-32. Over aas, S., and E . Solum. 1974 . The load-on-top system for crude oil tanker--exper fence and possible design improvements, pp. 415-426. In Transactions of the 82nd Annual Meeting. Society of Naval Architects and Marine Engineers, New York.

87 Program a Coordinado de Estudios Ecologicos en la Sonda de Campeche. 1980. Report on work performed to control the Ixtoc I well, to combat the oil spill, and to determine its effects on the marine environment. Petroleos Mexicanos, Mexico City. Ronov, A.B. 1976. Global carbon geochemistry, volcanism, carbonate accumulation, and life. Geochem. Int. 13:172-195. Ross, S.~., C.W. Ross, F. Lepine, and E.K. Langtry. 1979. Ixtoc I oil blowout. Spill Technol. Newsl. 4:245-256. Royal Commission on Environmental Pollution. 1981. Oil Pollution of the Sea. London. 307 pp. Schlesinger , W.H. , and J.M. Melack. 1981. Transport of organic carbon in the world's rivers. Tellus 33:172-187. Schreiner, O. 1980. Discharge of oil-bearing waste water from the production of petroleum on the Norwegian continental shelf, pp. 137-153. In C.S. Johnston and R.J. Morris, eds. Oil Water Discharges . Appl fed Science Publishers, London. Simoneit, B.R.T., and M.A. Mazurek. 1981. Air pollution: the organic components . Cr it. Rev . Environ. Control, in press. Southern California Coastal Water Research Project. 1980. Biennial Repor t 1979-1980 . Long Beach. Study of Cr ltical and Environmental Properties. 1970 . Man' s Impact on the Global Environment--Assessment and Reco~mnendations for Actions. MIT Press, Cambr idge, Mass . 319 pp. U.S. Census Bureau. 1978. Current population reports. U.S. Coast Guard, Department of Transportation. 1976. Polluting incidents in and around U.S. waters--Calendar year 1975. Washington, D.C. 24 pp. rJ.S. Coast Guard, Department of Transportation. 1977. Polluting incidents in and around U.S. waters--Calendar year 1976. Washington, D.C. 31 pp. U.S. Coast Guard, Department of Transportation. 1979. Polluting incidents in and around U.S. waters--Calendar years 1977 and 1978. Washington, D.C. 44 pp. rJ.s. Coast Guard, Department of Transportation. 1980. Polluting inc~dents in and around U.S. waters--Calendar years 1978 and 1979. Washington, D.C. 44 pp. United States Environmental Protection Agency. 1976. Development Document for Interim Final Effluent Limitations Guidelines and Proposed New Source Performance Standards for the Oil and Gas Extraction Point Source Category, pp. 126-133. Washington, D.C. U.~. Environmental Protection Agency. 1978. Draft Development Document Including the Data Base for the Review of Effluent LimitatiOn Guidelines (BATEA). New Source Performance Standards, and Pretreatment Standards for the Petroleum Refinery Point Source Category. Washington, D.C. U.S. Environmental Protection Agency. 1979. Economic Analysis of Proposed Revised Effluents Standards and Limitations for the Petroleum Refinery Industry. EPA-400/2-79-027. Washington, D.C. Wakeham, S.G. 1977. A character ization of the sources of petroleum hydrocarbons in Lake Washington. J. Water Pollut. Control Fed . 49: 1680-1687 .

88 Weaver, D.W. 1969. Geology of the northern Channel Islands. Pacific Sections AAPG and SEPM Special Publication. Thousand Oaks, Cal if 200 pp. Weeks, L.G. 1965. World offshore petroleum resources. Am. Assoc. Petrol. Geol. Bull. 49:1680-1693. Wh~pple, W., and J.V. Hunter. 1979. Petroleum hydrocarbons in urban runoff. Water Resources Bull. 15:1096-1100. Wilson, R.D., P.H. Monaghan, A. Osanik, L.C. Price, and M.A. Rogers. 1973. Estimate of annual input of petroleum to the marine environment from natural marine seepage. Trans. Gulf Coast ASSOC. Geol. Soc. 23:182-193. Wilson, R.D., P.H. Monaghan, A. Osanik, L.C. Price, and M.A. Rogers. 1974. Natural marine oil seepage. Science 184:857-865. Wondzell, B.E. 1981. Crude oil production and oilspills in Cook Inlet offshore. Letter dated October 22. State of Alaska Oil and Gas Conservation Commission, Fairbanks. Zafiriou, O.C., R.B. Gagosian, E.T. Peltzer, and J.B. Alford. 1982. Atmospheric transformations and fluxes of lipids at Enewetak Atoll. Unpublished manuscript. Zimmerman, P.R., R.B. Chatfield, J. Fishman, P.J. Crutzen, and P.L. Hanst. 1978. Estimates on the production of CO and H2 from the oxidation of hydrocarbon emissions from vegetation. Geophys. Res. Lett. 5:679-681. .

Next: 3. CHEMICAL AND BIOLOGICAL METHODS »
Oil in the Sea: Inputs, Fates, and Effects Get This Book
×
Buy Paperback | $150.00
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

This comprehensive volume follows up and expands on an earlier National Academy of Sciences book. It is the result of an intensive multidisciplinary effort to assess the problems relating to petroleum-derived hydrocarbons in the marine environment. Specifically, it examines the inputs, analytical methods, fates, and effects of petroleum in the marine environment. The section on effects has been expanded significantly, reflecting the extensive scientific effort put forth in determining the effects of petroleum on marine organisms. Other topics discussed include petroleum contamination in specific geographical areas, the potential hazards of this contamination to human health, the impact of oil-related activities in the northern Gulf of Mexico, and the potential impact of petroleum on fisheries.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!