[From the U.S. Government Printing Office, www.gpo.gov]
Coastal Zone Information Center U.S. DEPARTMENT OF COMMERCE NOAA COASTAL SERVICES CENTER 8234 SOUTH HOBSON AVE CHARLESTON, SC 29405-2413 THE LIKELIHOOD OF SPILLS REACHING LONG ISLAND FROM JAN 13 1998 HYPOTHETICAL OFFSHORE FINDS OVER THE DEVELOPMENT'S LIFE by J. W. Devanney III Robert J. Stewart Massachusetts Institute of Technology Report to Regional Marine Resources Council Nassau-Suffolk Regional Planning Board February 1975 (revised and expanded, June 1975) TD 427 .P4 Property of CSC Library D47 1975 1. RESULTS FOR INDIVIDUAL LANDING PROBABILITY CONTOURS in an earlier study, Stewart (1974) estimated the probability that an individual spill on the continental shelf off of Long Island would reach Long Island as a function of location of the spill and season. The results of this study were contours of equal probability of individual spills landing such as those shown in Figures 1 and 2. still earlier, Devanney et al. (1974), using Bayesian techniques and historical spill data, estimated the probability density of the number of spills over a thousand' barrels which will occur over the life of hypothetical 125 million barrel, 500 million barrel, and two billion barrel recoverable finds. These estimates were made for platform spills, for pipeline spills assuming the find is brought ashore by pipeline, and for tanker spills assuming the find is brought ashore by tanker. The results of this study are summarized in Figures 3 through 11. Note that the probabilities apply only to sizable spills (over a thousand-barrels) -An actual find would invol*ve hundreds to thousands of smaller spills. In general, these smaller spills will be identifiable as slicks for at most a day or two and hence, with typical times to shore of ten days or more for spills which occur twenty miles or more offshore, have been ignored in this paper. INN 72*W 70OW -LCCK 1. . ........ CAPE COD @-j GAY H@- .4 k1 .2 .2 '05 1.2 EVNANTUCKET A. 40ON (n X .05 .05 Ae FIGURE CONTOURS OF PROBABILITY THAT A SPILL RELEASED IN THE OFFSHORE REGW DU(-\ING SUMMER WILL IMPACT LONG ISLAND F THE 10 MILE SEA BREEZE ASSUMPTION.. JCCAPE C 7 2'o V I 70'oW ion m 720W 70OW J,@ML!CCK 1. sC)b,,N@,D CAPE COD Hs C@oc -0/v 71,4 tji@ F NANTUCKET PROBABILITIES IN THIS -'-ION ARE UNIFORMLY SMALL, 40" N RrL k:, ON THE ORDER OF .01 TO .0 2. co FIGURE CONTOURS OF PROBABILITY THAT A SPILL RELEASED IN WINTER WILL IMPACT LO.NG ISLAND FOR THE .10 MILE SEA BREEZE ASSUMPTION. 720'11 70OW 4 Once one has for a particular spill location and season the probability that an individualspill occurring at this location will reach shore, s, and one has the probability of n such spills occurring over the life of.a specified find, p(n), for all possible n from 0 on up, then an obvious question to ask is: what is the probability, A, that at least one spill reaches shore from a find df specified size at a specified location over this development's life? If one is willing to assume that spill occurrences are independent, then the probability of at least one spill reaching shore, A, given p(n) and S N A I p(n)(1 - (1 _ s)n) n=O where N is large enough so that the probability of more than N spills is small enough to be neglected. MIT has constructed a little program which combines a given spill incidence density, p(n), and'a given s according to the above expression. It is clear that for a given p (n) , A will be the same along any contour of equal s in Figures 1 and 2. The program computes A for the following seven contours: s .01, s = .02, s'= .05, s = .10, s = .20, s = .40, s = .60. Therefore, in interpreting the results, it will be necessary to continually refer back to Figures 1 and 2 to see what hypothetical spill locations correspond to the s being analyzed. We have exercised t his program on the nine different spill incidence densities shown in,iFiqures 3 through 11. The results are tabulated in ITable 1. 5 FIGURE DENSITY OF NO. OF PLATFORM SPILLS 0.75@17 OVER 42,000 GALLONS SMALL FIND, FIELD LIFE Based on all known USA platform spills over 42,000 gallons observed in the period 1964 0.50- through 1972. inclusive. Number observed = 9 CX Exposure observed.= 3,927 MM bbis Exposure contemplated = 122 MM bbls 0.25 MEAN (n) = .28 VAR (n) = .29 L 0 1 2 3 4 0.50- FIGURE 11 DENSITY OF NO. OF PLATFORM SPILLS OVER 42,000 GALLONS C: MEDIUM FIND, FIELD LIFE Exposure contemplated = 567 MM bbls CL - MEAN (n) = 1. 3 0.25 AVAR (n) 1.5 t 0 1 2 3 4 5 6 7 0.50- FIGURE DENSITY OF NO. OF PLATFORM SPILLS OVER. 42,000 GALLONS LARGE FIND, FIELD LIFE Exposure contemplated 2,044 MM.bbIs M E A N (n) = 4.7 0.25- VAR (n) 7 1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 N U M B E ROF SPILLS, n A 6 FIGURE 6 DENSITY OF NO. OF PIPELINE SPILLS 0.75j- OVER 42,OOOGALLONS SMALL FIND, FIELD LIFE Based onall Vno,.-.,n USA offshore pipeline spills over 42,000gallons observed in the period 1967 0.- 50- through 1972 inclusive. Number observed = 8 Exposure observed = 3,169 MM bbIs Exposure contemplated 122 MM bbis 0.25- MEAN (n) = . 31 VAR (n) = .32 0 1 2 3 0.50 FIGURE 7 DENSITY OF NO. OF PIPELINE SPILLS OVER 42,000 GALLONS V MEDIUM FIND, FIELD LIFE Exposure contemplated = 567 MM bbIs 0.2 5 MEAN (n) = 1. 4 VAR (n) 1. 7 0 1 2 3 4 5 6 7 8 0.50- FIGURE DENSITY OF NO. OF PIPELINE SPILLS OVER 42,000 GALLONS LARGE FIND, FIELD LIFE Exposure contemplated 23044 MM bbIs MEAN (n) = 5.2 c,0.25 VAR (n) 8.5 0 1 2 3 4 5 6 T 8 9 1.0 11 12 '13 N U N11 B E ROr- SPILLS, n FIGURE DENSITY OF LARGE TANKER SPILLS SMALL FIND, FIELD LIFE 0.75- Based on all ECO spills on major trade routes over 42,000 gallons. Cn Number observed =99 C Exposure observed =. 29, 326 MM bbls LJ- 0.50 - 0 Exposure contemplated 122 MM bbls >- MEAN (n) .412 VA R (n 414 m .0.25- M CL 0 1 2 3 4 .FIGURE DENSITY OF NO. OF LARGE TANKER SPILLS 0.50- MEDIUM PI ND, FIELD LIFE Exposure contemplated 567 MM bbls MEAN (n) 1.91 0.25- VAR (n) 1.95 0 1 2- -3 4 5 .6 7 0.50 FIGURE DENSITY OF NO. OF LARGE TANKER SPILLS LARGE FIND, FIELD LIFE Exposure contempicted = 2,044 MM bbIs CL 0.25- MEAN (n) 6.9 VAR (n) 7. 4 A t t 0 1 2 3 4 5 6 7 8 9 10 11 12 ',3 14 15 NUMBER CF SPILLS , n TABLE 1 PROBABILITY BY TYPE OF AT LEAST ONE SPILL OVER 1000 BARRELS REACHING LONG ISLAND OVEq THE LIFE OF FIND--FUNCTION OF LOCATIONAL CONTOUR, FIND SIZE, AND SPILL TYPE Contour on which Find is Located .01 ..02 .05 .10 .20 .40 .60 100 MM bbl Platform .003 .006 .014 .027 .054 .105 .153 Produced Pipeline .003 .006 .015 .030 .060 .115 .167 Tanker .006 .013 .031 .062 .120 .226 318 250 MM bbl Platform .013 .026 .063 .121 .226 .397 .526 Produced Pipeline .014 .028 .069 .132 .245 .425 .557 Tanker .014 .028 .069 .133 .248 .434 .575 2000 MM bbl Platform .046 .089 .206 .366 .589 .817 .913 co Produced Pipeline .050 .097 .224 .393 .62.1 .840 .926 Tanker .066 .128 .291 .497 .745 .934 .997 9 The numbers range from .002 for platform spills from a 125 million barrel find on a .01 contour to .982 for tanker spills from a two billion barrel recoverable find located on a .6 contour. In short, the probability of at least one spill reaching shore over the life of a find depends critically on the find's location and size. Not surprisingly, a very large find within the .2 contour has a high probability of at least one spill reaching shore, while the 125 million barrel find has a good chance of not landing a large spill even if it is located rather close to shore. Since any hypothetical find will involve either platform(s) and pipeline(s) or platform(s) and tankers, if we assume that the occurrences of spills of different types are independent we can transform Table 1 into Table 2, which displays the probabilities of at least one spill . of any type reaching shore over the find's life as a function of find size and location. In interpreting these figures, several things should be kept in mind: 1. The spill incidence densities are based on the twin hypotheses that the mean spill incidence rate is proportional to volume of oil produced and that we don't do any better with respect to preventing large spills in the future than we have done over the last ten years. Both these assumptions may be overly pessimistic. For platforms and pipelines all spill data is based TABLE 2 PROBABILITY OF AT LEAST ONE SPILL OF ANY TYPE REACHING LONG ISLAND OVER THE OVER THE LIFE OF A FIND AS A FUNCTION OF LOCATIONAL CONTOUR, FIND SIZE, AND SPILL TYPE Contour on which Find is Located .01 .02 .05 .10 .20 .40 .60 Landed by .005 .010 .027 .052 .1,02 .192 .269 100 MM bbl Pipeline Landed Landed by .009 .019 .043 .084 .158 .295 .401 Tanker Landed by .027 .054 .132 .237 .415 .653 .790 250 MM bbi Pipeline Landed Landed by .027 .054 .132 .238 .417 .660 .799 Tanker Landed by .096 .188 .383 .615 .844 .971 .994 2000 MM bbl Pipeline Landed Landed by Tanker .112 .217 .437 .681 .895 .988 .998 on Gulf of Mexico experience, where small fields by offshore standards, use of now obsolescent technology, and probably unnecessarily low well production limitations combine to produce a considerably higher ratio of wells to production and platforms to production than we would expect to use on the Atlantic outer continental shelf (OCS). Thus, if number of wells or number of platforms is a better measure of activity from the point of view of *Spill incidence than volume handled, then at least our platform spill incidence densities are overly pessimistic. Obviously, any improvements in operating procedures or technology which reduce spill likelihoods from what they have been in the recent past will also make our estimates overly pessimistic. 2. As Figures 1 and 2 indicate, the potential drilling areas on the George s Bank identified by CEQ fall outside the .05 contour. Thus, with respect to spills originating on the Georges Bank, the three leftmost columns of Tables 1 and 2 are of most direct relevence. As Figure 1 indicates, the .6 contour applies only to spills released within twenty -five miles of Long Island during the summer and much less than that during the winter. 12 3. Stewart (1974) estimated the minimum time to Long Island of spills occurring on Nantucket Shoals and eastward was twenty days or more. All but very large spills of medium to heavy crudes are likely to be invisible to the naked eye after twenty days at sea due to weathering. Neither Table 1 nor Table 2 accounts for weathering and the effect that weathering will have-in reducing the impact of the spill if and when it does reach shore. 4. Not all spills need occur at the location of the find. A pipeline spill can, of course, occur anywhere along the pipeline, and a tanker spill anywhere along the tanker route. The above analysis would seem to indicate that from the point of view of Long Island, assuming a find on Nantucket Shoals or eastward, the routing of the trans'port system is more critical than the location of the find'. It is possible that such transport systems could cross relatively high probability contours. Unfortunately, we do not have sufficient data to estimate where along the.route pipeline or tanker spills are likely to occur. What little data we do have seems to indicate, for both modes, that spills tend to occur at either end of che link rather than in the middle. 13 For those situations in which there is a substantial probability of at least one spill reaching shore over the life of a find, then the number of such spills, m, becomes of interest. Consider a two billion barrel find located on a .05 contour landed by pipeline.* According to Table 2, the estimate of the probability that at least one spill reaches shore from such a find is .437. The probability density of the number of such spills is given by where P1 (n) is the density of the number of pipeline spills and P2 (n) is the density of the number of platform spills. Applying the relevant densities for the two billion barrel find and assuming that we are on the .05 contour (s = .05), the program's results are TWO BILLION BARREL FIND, LANDED BY PIPELINE, .05 CONTOUR Number of Spills Larger than 1000 bbl Probability Which Eventually Reach Long Island of m m q(m) 0 .617 1 .294. 2 .075 3 .013 4 .001 5 .0002 6 .00003 14 In this case, the probability that there.will be at least one spill reaching shore is about .38. This .38 is made up of a .29 probability that exactly one spill will reach shore, a .08 probability that exactly two spills reach shore, a probability of about .01 that exactly three spills reach shore, and the probabilities of still higher numbers of spills fall of rapidly. The point is that for the cases toward the upper left of Tables 1 and 2, if a spill reaches shore, it is almost certain there will only be one such spill. For the cases in the bottom left-hand corner of Tables 1 and 2, it is quite likely'that there will be more than one such spill. For the intermediate cases, as we illustrated above, even if a spill does reach shore, it is likely that there will only be one such spill, but multiple landings are not impossible. Once again, these computations do not account for weathering. 15 2. TREATMENT OF SEASONAL VARIATIONS IN LANDING PROBABILITIES All the results of Section 1 were predicated on the assumption that the probability of an individual spill reach- ing shore, s, is constant over the life of the find. Actually, Stewart has shown, as is obvious from Figures 1 and 2, that the probability that an individual spills will reach Long Island from a given release point is sharply seasonally dependent. Consider, for example, a hypothetical find located south of Nantucket at 40*40'N and 700001W. According to Stewart, the probability that an individual spill will reach Long Island from this point in the summer is .05, while in the winter it is about .01. The spring and autumn probabilities fall between these two numbers. As a first approximation to this problem of seasonally varying, landing probabilities, assume that we have a find at this location. Assume further that for six months of the year the individual landing probability is .05 and for the other six months it is .01. Under these assumptions, we desire the probability density of the number of spills which will reach Long Island over the field's life as a function of the size of the find. To this end, M.I.T. undertook the following three step analysis: 1) The negative binomial densities of Devanney et al [19741 were recomputed for the half production life of three hypothetical find sizes: 125 YIM barrels produced, 500 MM barrels, and'12,000' MM barrels produced.* The assumption here is that'half the@oil 16 will be produced in the six months 'summer' and half in the 'winter'. This.was done for platform spills alone, both platform and pipeline spills combined, and both platform and tanker spills combined. This computation resulted in the following nine probability densities. PROBABILITY OF n PLATFORM SPILLS OVER 1,000 BBLS IN HALF LIFE n 250 MM BBL FIND 500 MM BBL FIND 2000 MM BBL FIND-*7. 0 .8675 .5738 .1304 1 .1223 .3091 .2378 2 .0096 .0925 .2407 3 .0005 .0036 .1788 4 .0006 .1086 5 .0572 6 .0270 7 .0117 PROB. OF n PLATFORM & PIPELINE SPILLS, 1000+BBLS in HALF LIFE n 250 MM BBL FIND 500 MM BBL FIND 2000 MM BBL FIND 0 .7344 .3142 .0147 1 .2246 .3516 .0548 2 .0364 .2083 .1086 3 .0041 .0865 .1513 4 .0004 .0286 .1663 5 .0079 .1535 6 0019 .1238 7 :0004 .0843 8 .0589 9 .0360 10 .0206 11 .0111 17 PROB. OF n PLATFORM & TANKER SPILLS, 1000+ BBLS IN'HALF LIFE n 250 MM. BBL FIND 500 MM BBL FIND 2000 MM BBL FIND 0 .6668 .1995 .0018 1 .2697 .3192 .0116 2 .0550 .2577 .0345 3 .0075 .1399 .0718 4 ..0008 .0575 .1130 5 .0191 .1436 6 .0053 .1535 7 .0014 .1418 8 .0003 .1156 9 .0845 10 .0561 11 .0341 12 .0192 13 .0101 14 .0049 2) Next the probability density of the number of spills reaching Long Island in 'summer' and the density of the number of spills reaching Long Island in the 1winter' was computed using Prob(k spills reaching) Pr(n spills occur -n! -s k(1-s n-k i shore in season i nk ing in half life jZ!(n-k)! where s I was taken to be .05 for the summer six months and set equal to .01 for the winter six months. At this point, we are making the erroneus assumption that any pipeline and tanker spills occur in the immediate vicinity of the find. 3) Finally, the probability density of the number of spills reaching Long Island in the 'summer' was convolved with the density of the number of spills reaching Long Island in the 1wihter'. to obtain the density of'the-total numbe.r.,of spills reaching Long Island from a find at 40*40.'N.and 700001w. 18 for the nine different situations under analysis. The results are shown on pages 19 through 21. These tables are taken directly from the computer output. The probabilities are in scientific notation. The number in front of the E is to be multiplied by 10 raised to the power following the E. That is, the number following the E is the exponent. For the negative exponents in these tables, the numbers can be read by simply moving the decimal point to the left in places where m is the exponent. The above analysis is meant to be exemplary only, to outline the steps which must be undertaken in order to estimate the probability of a spill reaching shore from a given site under seasonally varying landing probabilities. An analysis of an actual find would require considerably more detail both with respect to number of 'seasons' (four instead of two) and especially the location of pipeline and tanker spills. 19 V`LNTr-nnI,- (VILY.. SlIALL r-l':LP n K nrle)p, e)c of "OV411ER pfinn, or 11', !WITER mon nc K SPILLS ASP-Inc SPILLS N-11OPE r-. - i I J .0 43 1 111E- V! I .12 ^43 1.5031471E-n1 L '.) v - 1 1 .4 29151 !E- F13 7, -1 ns C!- - -20,?E-n5 1.7 Z742377-)5 1.1215725E 3 .5 .1 -64 B 6 E- 1 .1 1 .1701952E-nIF '795557!:-1" 1 .3101; 1395F-ls) 3.9 - jvi;:)E-l4 n .1 3 U E+ 1) .1 _5 714E-17 7862 PLATMItl mILY, IiEnIUH- FIELn K DrIgi nr- K S1111HER PR11 01: K I.WITER PROR Or- K CDILLS ASHORE '33PILLS A."211ORE SPILLS -%SHORE .1 717172 7 "- ')I 1 '14 2 03 11 JE- )ll MG11504E-11 1 2 7 73 13 r* -) 2 i ". %*1 7 9 1 %"a 5 E - 13 3 .3 1.7972357E-15 u- .100325,"E-',14 0 7.9153,134E-nr 4. )2J1GG7E--),1 :b . Tj 114 IV.,: j.1;545327E-11 7.E1410V17E-1)3 I . 7 7 2 4 5W",1E- 10 3 . r 7 19 5 ^0 V' E - 1 It 5 . fj J 249.3 8 E - I I J.uO3G3OOE+J*) 1.9477711E-12 3.0100011-1-E+1k) 5.1134414E-15 PLWFORM OALY, r) r i-, LARGE P I E L n I S 131 It 1 - R PROH OF K UVITER OR01 Oc K -;PILL3 A:;i;-ORE S!"ILLS ASHORE SPILLS AS1injzF_ :j .773 1,15 j" 1 .1 .72143 7`DE- 7@ 53 'ji:- 3 121 ,, 5-.; 4 .3 1 i,' E - -3 U" 7 A 31 6 F 3 4 03 E - *0 u 4 . 5 4 4 E - 0 it 05 2 . U643,21 7E- 33 1-9197@1;00E-05 4 3 %3 C. 1 It 6 LE 7 1.25587nOE-If.) G-771UPH-07 3 1)" 8 r 9 J124369E-13 7 3. 23421jrmr--i5 S. 6086713E-10 .3 -"V"5 77- 11 1.2463G70E-17 1.3050333@E-11 7 ) - 7' 3. 7 6 2 2 7 1; .1 E- I !) 0 3 E - 13 7. 5 7 3 7 7 4 7 F - i r) 2.5n , - .3 5 US 9 f) 7 7 E-2 -1 4 . 2 "D 2383 E - 15 J 12 7 C - 2 3.8" 5 451CPOE-17 E-2f) 012165 2E- 211 4-693523DE-23 20 p3LAT F OR! I A'11) PIVELVIE, SWALL cIELn nnnrt nF 1, SUMER PRIII OF K IWITER PROB OF K 13 P I L L 3 A S! 10 W: S!3I LLS ASHORE SPILLS I'MIORE J 3453134@-ll a is U."G"GlIGE-11 9.n-l441f)2E-1j A 1.5111405C-102 3.1027109E-30 I . k".31030GI E-02 1 1.2549"737-'14 5 Jc*'p73367E- f'C, 1.7762-in5E-n4 3 3.9517-137@---W 5. r2;)22:.15E- il -1 , -1 -557',103E-nG LIP %-)J'l. 2 J, F - -19 4J L. I 3.7')77U50r-12 5.1072515E-Oj 0 f) n 111011 C+ .1 n 0. ;J30000,0!"7+j%'l 1 .1'527907E-11 11.110j0ojJE+Ua 0.000000p.E+jo I . 622.613!@E-l 4 1.5635144E-17 a.oi), n,)OF)E+n, 00 E+ I 001)(10% la 0.00AMUE+J6 8.685063GE-21 PLATrQR14 AND PIPELINE, MEDIUM FIELD K 11"Till OF K SWIMER PRO!', OF K "WITER P R 03 0 F K 'qPILLS ASHORE SPILLS'o ASHORE SPILLS ASHORE -) .41:) 015 57- )1 4 .8 80 0 G,u9E- al 9.30C,0422E-IDI 5.F!9,)837r--02 1.1820ME-02 6.65502.71E-02 1.770371KE-03 7.4675700E-15 2.4842499E-03 3 3. C*J 063 79 F-- n5 3. 20"37318E-07 6.3878164E-'15 'j.5198844E-07 I .,)97509,.)q-nJ I .2557794E.-@oG 2.7722095E-12. 1.94G7997E-98 7. 166 43 1; 7 E- 11 4.76GO937E-15 2) .'-') 35 84 2GE-1 0 7 J.222734UE-13 4.125092GE-1. ;e.05 J-83 23 E-4 2 4,6nn000QE+0il 1.279G415E-14 J *.J,)-1U0UUE+0,J o.oooooooE+jnl 5 .904730SE-17 jj..j,)nooo(jE+o!) i.1133945E-19 1).GonopooE+6!) 5 .'J7433G2E-2A2 V.0000'100E+ )l 1.2710064E-24 03 0 1) 000 OOE+ .11) 1 . 3 3 5 7 0 - 6 E - 2 7 14 3.,3t294n PLATFORl Ann PIPELIIIE, Lf%RGE FIELD K P-101 OF K SUMMER PR')8 OF K WINTER PROD"*.OF- K - .";PILLS ASIIORE SPILLS ASHORE S P I L L'S'-ASI 10 RE 1 7 . U" 8 2 5 0 0 E I 9.532148voE-31 7.'5137353E-01 A. -I . ur' U"l 9 U U 9 E - n 1 4.5535218E--@02 Z.1337807E-01 2 2 .3277540E-12 1.1510itOtIE-03- 3.1-57-42-83E-02 3 2.145910CE-u") dq-.044G89GE-05 3.2405765E-03 1" 1. 41 6f)5G9E-1)4 2.8575437E-07 2 .589672SE-()4 5 1 .113 3116 E- 1) 6 3.3292739E-l!) 1.7135681E-05 4.07755291:-('.7 3.3317557E-11 9.731;531PE-n7 7 1.77,13449E-11'a 2.90G7959E-13 4.q5l2493E-08 () n n n I 1 .1 E+ I n u.nnnnqonE+qn 1.4945-52SE-09 wl i n i o n!, :+ 1,) 1 . INP 0 a 0 0 1) OE + 1,'- 3. 169 15 8 OE-1 I f)o:)o9oa"+n:% ).jnjonn9E+00 .5 .1 268473F-1 3 L ').09n00l0E+9`l U. 752 G415E-15 12 l.n,1100-10@-+11 n,l0,l0q0,lE+ln 7.5152389E-17 13 -i.jnnnn00E+00 7.11n2352E-19 .111 j.,,,Ojonnn7+jn j.ojnqnl0E+nJ 5.1r.92227E-2.1 21 PLATFORV Vin TA'11'%7'1,t 1.11: IT E. R p r, nn. n prtqrl OF I, -unt-in PPCri 01: K .1 % L f%P ILLS SrILLS 4S,j()nr-. -SPILLr, ASHORE 7 5 I'll 73 NE - r) I 1 72 7 7 77. 4 .13')734f;E-03 2' . ", 7 W) 2 7 3 E - -1 .11 211JG9@- 14 .21 lGr,5SF- If-. 2 . 61! 4 7 E - 4 5 9 L'i 2 "J'E - 13 2 .3 1 7 C "p 17 '") I j. Li 4 7 11 7.1556450E-12 1. .2 '93 1 S"PF- -11, ii ti n fi a rl E + !:,i . 2 "'1 G') 114 E. I j .59, i -jn:)n,1nr+ -Jri .,11) f" f) 1) 0,@ E+ n At) 1,110 r. +.ir C-.2737772E-17 r7 + 1330 Li u PLATFnr,,i vin TANK"R, !jrL-.rIiiji r.IELr, MIMI OF: K SUMMER DR-1 nr K MITER PROFI, OF K SPILLS V3"NORE S.-ILLS ASHORE SPILLS AS110PE i .22, A i - 03"354GE-rii "o' 7 1') 31 IE -Lol 7 .117-1n359@--12 1 .5 "1'19 817 E- -12 '31 3 -15 5 33 E- 12 n, 3, r' - 13 1 14 4 1. 3 2 4 2& 3 5E - n 3 3 4 "1 2 !" ". r-. - 15 7 Irp!25Gi7E-%7 1 . 417746'VE-14 IL . 7" 3 71 1) r, F 2 92 413 8') E 3 4 1".! 3 13 J F , 1 7"l 9 2 5 2"' 5 U 75 1 F 5115 E - n 0. 7 4L 6r L . 2 17 4 2 7 2'-'1 E 7 101 E - n -1 Lr 2 U. 2 2 .742,!I;E-13 3.1,2 )W)(5E-I 0 N, -,10 0 U E 0 o )oonor)nE+.l 0 1.13 7146E-20 000n0oocr'-+on :i .,innnn[)OE+ni 3.22-01107CE-23 I'M!) 91 )d J00J0f,'CE+00p L; . C"371.610E-2G A J -i JJ 2 13000000 E+:)i) 9 .0 75 825 7E-29 I'LATFOR-6"I MID TAAKER, LARC'.E FIELD K P.RUJ OF K SUMMER PROJ OF K 'LINTER PROB OF K S13ILLS ASHURE SPILLS ASHORE SPILLS ASHORE 9 .3 7 0 439 7E- JI G.7748333E-01 I .3,+,j49j7E-Jl U. Gj'2uu47E-j2 2.G332718E-01 z :;.Jz3a;jG9E-j2 1.5315099E-03 5.15-09809E-02 -,.ZJJ4436E-U3 4.3-139347E-05 6.759b659E-03 3.40'7@749E-Uii 7.2JIU631811-07 6.G92204CE-04 :. ..) 3 u U1, ?'@" 0' E - J 5 J . 7517017E-0to 5.327k2JZE-05 I 1. 3 0 ,-,1 G.) 1; E- U 1; 1 . 0 d 3 5 8 5 tjE -10 3.5467765E-06 ;..i,11z1uuLr--Us l.U.')'G7896E-IZ 2.02GG452E-07 J ..i.juujuuu,L-:+.jo j.0000000E+U0 7.b9o"5351E-09 i .).uuuuuo3.r_-+jj U.U.)OOOUUE+UO 2.01'.J"7446E-10 11) '1 . J u u j i i)U E+ 1) 1 J.U.)00JOUE+00 3.946373GE-12 11 J. i ) j 0 a Uu E+ 0 0 -J.-9 '0090UE+00 b.JJ-5G502E-14 13 Jr.;Ujj(),)k)c+U%j J.O.)OOOOOE+UO 7.7259209t-26 1 3 6 ;j'..00UU!)UL)E+.)U 0.J0oo3ooE+00 3.1223975E-13 'I.JJOOJ;)OE+J%') 0 U(1%1)000UE+00 6 . 45009U3 E-20, 22 REFERENCES Stewart, Robert J., and Devanney, J. W. Ill. Probabilistic trajectory assessments for offshore oil spills impacting Long Island. MIT Report to Regional Marine Resources Council, Nassau-Suffolk Regional Planning Board. 15 November 1974. Devanney, J. W., and Stewart,.R. J. Bayesian analysis of oil spill statistics. Primary, physical impacts of offshore petroleum development. MIT Report to Council on Environmental Quality. MITSG-74-20. March 1974. DATE DUE 36 141079782