Past Oil Spills in Buzzards Bay

Bouchard No. 120 compared to other Buzzards Bay spills

By Dr. Joe Costa

The recent Bouchard No. 120 oil spill was the largest oil spill in Buzzards Bay in nearly 35 years, and the second largest recorded. It is also the largest spill of No. 6 Fuel, which can result in high bird mortality depending upon conditions and species affected.

Buzzards Bay is a major transit route for small tanker and barge traffic transporting heating and industrial oil and gasoline into Sandwich, greater Boston and northern New England markets. Nearly 1.6 billion gallons of oil pass through the canal annually with additional deliveries made to New Bedford. Buzzards Bay has been the site of several catastrophic oil spills.

The largest spill occurred on September 16, 1969 when approximately 189,000 gallons of #2 fuel oil spilled when the barge Florida, which ran aground off West Falmouth. In recent years, improvements to navigation and more rigorous pilotage requirements are believed to be minimizing risks of future spills in Buzzards Bay. Nonetheless, smaller spills from barge and vessel groundings in the Bay have continued during the 1980s and 1990s. Other notable groundings in Buzzards Bay include the grounding of the Bermuda Star off Cleveland Ledge in 1990, releasing 7,500 gallons of No. 6 fuel. There was also a 50 gallon spill from an empty fuel tank when the QEII grounded off Sow and Pigs Reef near Cuttyhunk in 1992. A January 1996 grounding of the barge North Cape off Moonstone Beach in Rhode Island releasing 880,000 gallons of Number 2 fuel raised concerns of local officials about oil preparedness.

While the Bouchard No. 120 spill remains relatively small compared to the multimillion gallon oil spill disasters like the 1989 Exxon Valdez, Buzzards Bay is a relatively small body of water, with many sensitive resources. The conspicuous nature of this oil, and the complex craggy coastline will make cleanup a challenge.

In the Table below we list known major oil spills for Buzzards Bay. It has been updated with what we believe is the more accurate information and it will continue to be updated as new information is gathered. More detailed information and citations will be posted later.

Note: It is reported on some other websites that the oil tanker Argo Merchant disaster of 1976 occurred in Buzzards Bay. This is incorrect. The Argo Merchant broke up southeast of Nantucket.

Date	Location	Type	Volume (gallons)	Comments
1940s	Western Buzzards Bay, Westport (at Hen and Chicks?)	No. 2 Fuel Oil	unknown
1963	Near Nyes Neck, North Falmouth	No. 2 Fuel Oil	unknown	came ashore during the winter
16-Sep-69	Fassets Point, West Falmouth	No. 2 Fuel Oil	189,000	*Florida* Fuel Barge. Final estimate was 4,500 barrels spilled.
9-Oct-74	Cleveland Ledge (near canal entrance)	No. 2 Fuel Oil	11,000 to 37,000 [under review]	*Bouchard 65* barge grounded. Was towed to an anchorage off Wings Neck. Oil came ashore in North Falmouth and Bourne.
28-Jan-77	Cleveland Ledge	No. 2 Fuel Oil	81,144	*Bouchard 65* barge grounded, oil on iced over bay, some burned. Final estimate was 1,932 barrels spilled.
10-Jun-90	Cleveland Ledge	No. 6 Fuel Oil	7,500	*Bermuda Star* cruise ship went aground, impacts to Naushon
18-Jun-90	Cleveland Ledge	Diesel oil or heating oil	100 or 200	*Bouchard 145* fuel barge
7-Aug-92	Sow and Pigs Reef, Cuttyhunk	No. 6? Fuel Oil	50	*Queen Elizabeth II* cruise ship. Residual from empty fuel tank that was ruptured.
27-Apr-03	Press reports: Hen and Chicks Reef, Westport	No. 6 Fuel Oil	98,000 (best estimate final)	*Bouchard No. 120* fuel barge
Smaller spills of gasoline and fuel oil have occurred every few years in the Bay or in the Cape Cod Canal.

Map of Buzzards Bay oil spill locations.

Past oil spills in Buzzards Bay: Just How many gallons were really spilled?

Bouchard No. 120 is really the second largest spill for Buzzard Bay

Do you think it is difficult to estimate the volume of an oil spill? Well in the case of the 1969 Florida barge oil spill, and the Bouchard 65 spill in 1974; keeping straight the final estimated volume of oil spilled has proved next to impossible. The official final estimates for both these spills are different than what has been widely reported. An upshot of this is that the recent Bouchard 120 spill is in fact the second largest spill since the 1969 Florida spill, not the third largest as previously reported.

Microfiche copy of 1969 Standard Times Newspaper.

Microfiche copy of 1969 Standard Times Newspaper.

The Buzzards Bay NEP first reported the Florida oil spill as 185,000 gallons, based on our 1991 Management Plan, and then revised it to 175,000 in the interim, because numerous other sources reported the smaller quantity as the volume of the spill. However, the actual volume of the spill appears to have been 189,000 gallons as described below. In the case of the first Bouchard 65 spill in 1974, (there was a second Bouchard 65 spill in 1977), the volume spilled was likely to have been between 11,000 and 37,000 gallons (we are still trying to get a good source of information), but far less than the 165,000 gallons previously posted.

Errors in reporting of oil spills

When oil is spilled at sea, the responsible party, with guidance and review of the US Coast Guard, estimates the volume of oil spilled. Generally, this calculation is based on the difference between what the ship was known to have carried, and what was measured in the vessel after the accident (so-called ullage measurements). Ullage measurements involve the height of oil (or more accurately, the height of the air space) in each compartment, and the depth of any water, if any, in each of those compartments. Estimates are also made by taking account the initial transported volume, and measurements of the total volume offloaded, either by emergency craft, or offloaded at an oil terminal. It is not calculated by measuring recovered oil, because typically less than 10% of spilled oil is recovered. Many factors are used to refine the estimate of a spill including oil level gauges in the barge tanks, the degree of oil and water mixing in ruptured tanks, and even the temperature of the oil. If the amount of oil spilled is less than 1% of the total initial volume, many errors in can be introduced into the calculation and can effect -or manipulate-- final estimates of oil spill volumes.

After this evaluation and review, the Coast Guard makes a final determination on the volume of the spill. This value represents a best estimate, and is usually included in a brief report to the National Transportation Safety Board (NTSB). This final estimated volume of oil represents the "official" government estimate for the spill. Theoretically, everyone should cite this volume.

However, in the case of the 1969 and 1974 Buzzards Bay spills, apparently, the final accident reports were not widely distributed or publicized by the press, and original reports about these spills have proven very hard to find. Unsurprisingly, later investigators may recall or can find only the first press reports, or rely on subsequent scientific publications.

In the case of the Florida spill, there was considerable inconsistency in scientific publications about the volume of the spill. However, press reports two days after the spill gave the preliminary estimate of the spill as 4000 barrels, or 168,000 gallons. This preliminary estimate apparently was revised upward somewhat, because we found a 1975 Congressional Report from the Office of Technology Assessment that clearly identifies the Florida spill as 4500 barrels (189,000 gallons). These two values appear to explain the range reported in the scientific literature.

For the Bouchard 65 1974 spill, virtually nothing was published on the volume of the spill. In our 1991 management plan, we provided no estimate of the spilled volume. Complicating matters, this barge spilled oil again in 1977, and sometimes the two Bouchard 65 spills are confused. Even worse, a 1999 Coast Guard document summarizing oil spills lists the 1974 spill as 36,650 barrels (more than 1.5 million gallons), an unrealistic total based on contemporary newspaper reports.

Contemporary newspaper reports of the 1977 Bouchard 65 spill suggest that the 1974 spill was smaller than the 81,000 gallons released in 1977. The Cape Cod Times reviewed their archives and determined the 1974 spill was initially reported at 25,000 gallons. The Bouchard 65 held more than 3 million gallons of oil. In the 1974 spill, 3 of the 10 tanks ruptured. In the 1977 spill, 4 of 10 tanks ruptured. Oil floats on water, so only oil near the depths of the fracture tend to escape. The 1974 press accounts describe a sheen only one mile long and 200 yards wide.

In contrast to the 1974 accident, the 1977 Bouchard 65 spill was well reported because it occurred 1 month after the Argo Merchant disaster off Nantucket, and because upper Buzzards Bay was locked with sea ice. The spill is memorable because some of the oil trapped on the ice was ignited on fire and newspapers carried dramatic photos of plumes of black smoke rising over Buzzards Bay toward Cape Cod. Equally amazing, clean crews walked onto Buzzards Bay off of Wings Point with 5000 foot long hoses to vacuum off pools of oil on the ice. Apparently, the first estimate of the spill was 500,000 gallons, but this was revised by the third day to 100,000 gallons as reported in Standard Times. Eleven days after the spill, the final volume of oil was provided in newspaper accounts as "about 80,000 gallons," which coincides with the 1,932 barrels (81,140 gallons) reported in subsequent government reports about the attempted oil burning. Only around 2,000 gallons were presumed to be burned.

Technical information for students of oil spills

Barrels and Gallons and Tons

We found many discrepancies among the technical reports we had on file about the volume of oil spilled in the Florida accident. Part of the problem was that in scientific reports, the exact volume of the spill was of incidental significance, or rounding errors were often introduced in the calculations. Adding to the problem was the fact many units of measure are used to report oil spills, a problem common in technical literature on the subject.

Most often errors arise from conversion errors, rounding errors, or incorrect assumptions about earlier reported units of measure or oil densities. The petroleum industry tends to measure volumes of product in barrels, a US measure dating back to the mid 1800s, and equivalent to 42 gallons. This is a different volume than say, a barrel of wine or olive oil. Today, the government tends to report spills in gallons, but in the 1960s and 1970s, reporting was usually in barrels. On the other hand, shippers tend to report tonnage, something easy to measure by the displacement of water on a vessel (n.b. hash marks on the side of the vessel). Of course, in 1969, that would have been US tons or "short tons", equal to 2000 pounds. However, scientific publications report in metric tons, which is equal to 2205 pounds, or in liters.

To further, complicate matters, a ton of No. 2 oil does not have the same volume of No. 6 oil or water. On the east coast of the US, No. 2 fuel oil often has a specific gravity around 0.86, whereas No. 6 oil has a specific gravity of nearly 1.0, close to the value of fresh water. So one short ton of No. 2 fuel has 280 gallons but one short ton of No. 6 has 240 gallons. However, for metric tons, the values are 309 and 265 gallons respectively. For comparison, there are 1000 liters (264 gallons) of fresh water weighs a metric ton (at 10 degrees Celsius).

These values are actually valid only for the identified specific gravities. In reality, crude oil can range from 250 gallons to 300 gallons per metric ton. Today, the average often given for crude oil is 7.33 barrels or 308 gallons per metric ton (6.65 barrels or 279 gallons per short ton), but in the past, different averages and volumes were used based on "typical" grades of crude oil on the market. Even No. 2 fuel oil can range from 302 to 316 gallons per metric ton. For every spill, tonnage to gallon conversions must be based on the precise characteristics of that oil.

Of course, the preceding discussion ignores temperature, and oil expands when warmed. Some of the petroleum tables report volumes at 68 degrees Fahrenheit, other report volumes at 86 degrees. Finally, the petroleum industry does not even report oil density using the traditional measure of specific gravity used by scientists. They use a measure a measure called "degrees API" which must be converted.

Simple? Turn back the clocks, take away calculators, add rounding errors, and introduce press reports with incorrect first estimates of the spill, and you have a reporting disaster of oil spill proportions.

Post Mortem of an oil spill: Misreporting of the 1969 Florida spill

The Sanders et al. (1980) and Sanders et al. (1981) reports on the Florida spill stated the values "650,000 to 700,00 liters" (172,000 to 185,000 gallons) as the volume of oil spilled. These were the principal references used when the Buzzards Bay CCMP was written. The fact that Sanders et al. gave a range of values for the Florida spill was unusual. There is a strong tendency in government reports to settle on a single value as a best estimate for spills, and not give ranges. A single number makes it far easier to calculate and publish statistics about spills. Was this based on a government report, or was it based on other estimates? We began to review reports we had on file of the Florida spill.

As it turns out, the Sanders estimates were originally given by Blumer et al. in 1970. There was actually an earlier report by Hampson and Sanders (1969), just one month after the spill, which cited a release of "250,000 to 280,000 liters (60,000 to 70,000 gallons)". Besides the evident rounding errors in converting liters to gallons, this estimate may have been based on incorrect preliminary information or a calculation error by the authors.

Interestingly, an article in the newspaper the New Bedford Standard Times, just two days after the spill, had already quoted a Coast Guard official that the spill was "probably 168,000 gallons." This equals precisely 4,000 barrels, which was the usual unit of measure for reporting oil spills in government documents at the time. Sanders also subsequently reported the 4,000-barrel total in a 1974 report. Blumer and Sass in 1972 reported "600 metric tons" of oil was spilled, which at a typical 309 gallons per metric ton for East Coast No. 2, appears to coincide with the Coast Guards revised estimate and explains the 185,400 gallons in the upper range reported by Sanders. However, Blumer et al. (1971) reported 650-700 "tons" were spilled. If these values are in metric tons, for typical No. 2 oil, this equals his equals 215,000 to 231,000 gallons, if these are short (American) tons, which appears to be the case, this equals 182,000 to 196,000 gallons (280 gallons per short ton for No. 2). However, what appeared to happen in this latter report was that the authors incorrectly used the water conversion of 264 gallons per short ton, which for the range gives the range of 172,000 to 185,000 gallons, which precisely matches the values repeatedly later cited (650,000 to 700,000 liters).

This same error can be deduced from Wertnbaker (1973, as reprinted in Wertenbaker, 1974), who in a detailed lay article in the New Yorker Magazine title "Anatomy of an Oil Spill" only reports the total oil cargo on the barge as 2,500 tons. Blumer et at. 1970 report the cargo as "14,000 barrels (588,000 gallons or 2,220,000 liters)." The 14,000-barrel cargo volume coincides with newspaper accounts at the time. The barrel to liter conversion is correct, but the cargo would have to be denser than water for 14,000 barrels to be equal to the reported 2,500 tons, so presumably the 2,500-ton cargo volume is another published error.

The lessons from these reports are clear. Be very careful in applying the correct conversion factors when calculating oil volumes or weights.

Retrospective on the Florida spill

On September 16, 1969, the oil tank barge Florida came aground off West Falmouth Harbor, on a point of land. The vessel, pulled by the tugboat Evening Star was headed to the petroleum facility serving the power plant on the Cape Cod Canal (now called the Mirant Canal Station electricity generating facility). The accident is believed to have occurred after midnight, and there is some evidence that the spill might have occurred before midnight on September 15.

In the 1991 Buzzards Bay Comprehensive Conservation Plan prepared by the Buzzards Bay National Estuary Program reported that 185,000 gallons were spilled in the accident. This was close, but the actual total adopted in a 1975 Congressional report on oil spills, was that 4,500 barrels or 189,000 gallons were spilled in the Florida accident. We recommend that this volume be used in subsequent publications and reports about the Florida oil spill.

Excerpt of 1975 Office of Technology Assessment Report

WEST FALMOUTH SPILL
On September 16, 1969, the oil barge Florida, on the way to a power plant on the Cape Cod Canal, came ashore off Bassets Point in Buzzards Bay, near the entrance to West Falmouth Harbor, Massachusetts. Nearly 4,500 barrels of No. 2 fuel oil were released into these coastal waters.

Immediately after the spill, massive destruction of marine life occurred offshore. Extensive trawling and dredging showed that a wide range of fish shellfish, worms, crabs and other invertebrates were affected. Trawls made in 10 feet of water soon after the spill showed that 95 percent of the animals collected were dead. The bottom muds contained many dead snails, clams, and crustaceans. Similar mortality occurred in the tidal rivers and marshes into which the oil had moved under the combined influence of tide and wind.

Eight months after the spill, the pollution covered an area of approximately 5,000 acres offshore and 500 acres of marshes and tidal rivers-about eleven times the area initially affected. Secondary pollution from heavily affected areas continued after the accident. In heavily polluted marshes, oil penetrated to a depth of at least one to two feet, and in these areas vital bacterial degradation was almost negligible eight months after the spill. Wherever the oil spread, there was concomitant animal mortality, and after nine months the affected areas had not repopulated. A study conducted four years after the spill indicates that some effects still persist.

References Cited

Blumer, M, G. Souza, and J. Sass. 1970. Hydrocarbon Pollution of edible shellfish by an oil spill. Mar, Biol. 5:195-202.

Blumer, M. and J. Sass. 1972. Oil Pollution: persistence and Degradation of Spilled Fuel Oil. Science. 176:1120-1122.

Blumer, M, H. L. Sanders, J. F Grassle, and G. R. Hampson. 1971.

Office of Technology Assessment (US Congress). 1975. Oil Transportation by Tankers: An Analysis of Marine Pollution and Safety Measures July 1975 NTIS order #PB-244457
pdf file excerpt at Princeton University Website

Sanders, H. L, J.F. Grassle, G. R. Hampson, L.S. Morse, S. Garner-Price and C. C. Jones. 1980. Anatomy of an oil spill: long term effects from the barge Florida off West Falmouth. J. Mar. Res. 38:265-380.

Sanders, H. L, J.F. Grassle, G. R. Hampson, L.S. Morse, S. Garner-Price and C. C. Jones. 1981. Long Term Effects of the Barge Florida oil Spill. EPA-600/2-81-012. January 1981. 217 pp.

Other Useful Conversions

These are averages compiled and calculated using data from various sources including published API gravities on Material Safety Data Sheets, and other data. Values are for 15 degrees C (=60 degrees F). Note that a short ton (= "US Ton") is 2,000 pounds, which differs from the metric ton (2,205 pounds =1,000 kilograms), and the English Long Ton (2,240 pounds). Volume per ton increases with increasing temperature.

Gallons per Short Ton of Fuel Oil #1: 280-290
Gallons per Short Ton of Fuel Oil #2: 267-280
Gallons per Short Ton of Fuel Oil #4: 263-265
Gallons per Short Ton of Fuel Oil #5: 253-260
Gallons per Short Ton of Fuel Oil #6: 246-250

Gallons per metric ton "crude oil, US": 308

Buzzards Bay National Estuary Program