million tonnes of crude oil. The estimated VOC emissions is 0.2 percent of 3.3 millions tonnes or 6.6 million tonnes per year. This is considered a conservative assumption, as the heavier products will emit significantly lower levels of VOC as compared to crude oils.

Based on the CRUCOGSA program sampling and analysis of the volatile components of approximately 1000 samples of crude oil, the VOC in the hullage space of a tanker is broken down as follows: roughly 7 percent ethane, 45 percent propane, 45 percent butane, and the remaining 3 percent of the components within the VOC emissions are heavier than butane. When released into the atmosphere, the lighter hydrocarbons will not deposit, and while it may be a “greenhouse gas” concern, does not appreciably impact the volume of oil entering the sea. Of the remaining VOC, only a small fraction is likely deposited to the sea.

To assess the potential loading of hydrocarbons from fugitive emissions of VOC, this analysis considers the following conservative calculation. The calculated value assumes that: (1) the 97 percent of the VOC released from tankers consists of light hydrocarbons that do not deposit to the sea surface; (2) the remaining VOC mix (3 percent of total VOC emitted from tankers) has a Henry’s law constant equal to that of decane (which certainly underestimates its volatility and, therefore, overestimates deposition); (3) the released VOC do not react in the atmosphere or in surface waters (which ignores the substantial degradation due to hydroxyl radical attack in the troposphere); and (4) the released VOC rapidly partition between the atmosphere, which is well mixed to 1000 m altitude, and the surface ocean, which is well mixed to 100 m depth.

Assuming 3 percent of the VOC are heavier than pentane, and that 0.2 percent of these heavier compounds are deposited into the oceans, the calculated atmospheric deposition of VOC from tankers worldwide is (6.6 million tonnes) × (.03) × (.002) or approximately 400 tonnes per year. The minimum value is 200 tonnes (50 percent of the best estimate), and the maximum value is 1,000 tonnes (see Table 3-2); (300 percent of the best estimate).8

Within North American waters, the total cargo oil movement in 1999 was approximately 0.9 billion tonnes. In recognition of the shorter voyage lengths within coastal waters and the extensive use of vapor recovery at load ports, a loss rate of 0.01 percent is assumed, resulting in estimated VOC emissions of 90,000 tonnes. The calculated atmospheric deposition of VOC from tankers in North American waters is 5 tonnes per year. The minimum value is 2.5 tonnes (50 percent of the best estimate), and the maximum value is 15 tonnes (300 percent of the best estimate).9 Calculation of atmospheric deposition of VOC was not included in the prior NRC reports. Given the size of the loading they may have an impact on air quality. VOC is a known pollutant and has been the subject of many NRC studies (NRC, 1992, 1995a, 1999).

Coastal Facility Spills (refined products)

Coastal facilities are defined for this report as point sources of spills that are not vessels or oil and gas exploration and production facilities (including crude oil pipelines). Table G-1 of Appendix G lists the types of facilities included in this discussion. The U.S. Coast Guard database of spills greater than 100 gallons for the ten-year period from 1990-1999 was used to estimate the amount of oil spilled from facilities. The U.S. Coast Guard data were sorted geographically to remove spills to inland waters. Also, only spills of refined petroleum products from pipelines in coastal areas were included (so as to exclude the crude oil pipeline spills from the USCG data base that were included in the section on oil and gas exploration and production). Facility spills greater than 100 gallons over the period 1990-1999 account for 9 percent of the number of spills and 98 percent of the spill volume. There was an average of 119 facility spills reported per year, with an average volume of 14.4 tonnes. Two types of facilities were the sources of 66 percent of the oil spilled over the 10-year period: coastal pipelines transporting refined products and marine terminals each were the sources of 33 percent of the spilled oil. Industrial facilities were the next largest source of spilled oil, with 14.4 percent. The pipeline spill volume was dominated by one spill event in 1994 where 5,500 tonnes of gasoline, crude oil, diesel, and jet fuel were spilled (the San Jacinto River spill in Texas). This one spill accounted for 30 percent of all the oil spilled from facilities in the 10-year period. This spill also demonstrates the problem of how to account for oil removal, since a very large fraction of the spilled oil burned.

The computed data from the U.S. Coast Guard data for North American waters resulted in estimating the minimum discharge as 1,700 tonnes per year. The best estimate is 1,900 tonnes per year (minimum + 10 percent), recognizing that spill reporting in the U.S. is very high. The maximum is 2,200 tonnes per year (minimum + 30 percent).10

8  

The 50 percent factor used to develop a minimum estimate and the 300 percent factor applied to develop maximum estimate are somewhat subjective and reflects the committee’s confidence in the data available and the methods and assumptions used to complete the calculation.

9  

The 50 percent factor used to develop a minimum estimate and the 300 percent factor applied to develop maximum estimate are somewhat subjective and reflects the committee’s confidence in the data available and the methods and assumptions used to complete the calculation.

10  

The 10 percent factor used to develop a best estimate and the 30 percent factor applied to develop maximum estimate are somewhat subjective and reflects the committee’s confidence in the data available and the methods and assumptions used to complete the calculation.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement