6,100 tonnes per year of petroleum hydrocarbons are discharged into the world’s oceans by pipeline spills.
Using the estimates cited above, the total amount of petroleum hydrocarbons discharged into the world’s oceans by offshore oil and gas facilities is 6,390 tonnes per year (290.0 tonnes—platforms; 6,100.0 tonnes—pipelines; Table D-3).
During the production, transport, and refining of hydrocarbons, volatile compounds escape to the atmosphere. Some, like methane, are light and rise or degrade rapidly. Heavier compounds, like hexadecane, react or rise more slowly. These heavier hydrocarbons are labeled as volatile organic compounds (VOC) and are defined in the U.S. Clean Air Act to include all volatile hydrocarbons except methane, ethane, a wide range of chlorofluorocarbons (CFC), hydrochloroflourocarbons (HCFC), and a few others, e.g., acetone. A complete listing of the constituents is given in Table D-6. Unfortunately the definition provides no exact ratio of carbon to compound. As a practical definition, the list contains most compounds above propane (C3H8, 82 percent carbon) and below hexadecane (C16H34, 85 percent carbon).
The focus of this section is on estimating the amount of VOC generated from offshore production platforms. Sections B and C include estimates of VOC generated during tanker transport and marine terminal loading. The estimates exclude other potential sources of VOC released from coastal or near-river refineries, storage tanks, etc.
Prior to 1990, VOC from production platforms received little attention and no estimates were provided by NRC (1985, 1975) or GESAMP (1993). One reason is that VOC data just started to appear in the late 1980s. Another reason is that the overall contribution to the sea was thought to be small. That’s in part because the scavenging of VOC from the atmosphere to the sea is inefficient and highly dispersed. At most a few percent of VOC ever make it to the sea, and these are spread over a large area compared to the scale of the generation source.
While data collection on VOC has improved, it remains crude and sparse on a worldwide basis. Regulators have made estimates for the U.S. Gulf of Mexico (offshore) and California. These estimates are not actual measurements but are calculated usually from an inventory of equipment multiplied by an estimated VOC emission rate for each piece of equipment. Producers have made estimates for the Mexican Gulf of Mexico and the producing basins of the northeast Atlantic (North Sea, Norwegian Sea, etc.), but documentation of their methods could not be found. In all cases, there is no information on the uncertainty in their estimates or speciation (percent occurrence of each compound) of the VOC, an important factor when considering how much of the VOC might make it back to the sea or what its toxic effect might be. Still, the situation in the above regions is better then the other offshore producing areas where no estimates were found.
Oil and Gas Producers (1994) describes four methods for estimating VOC (and other emissions) from offshore operations. The simplest, Tier 1, estimate is based on total production volume and tends to be conservative. Developing the higher tier estimates requires details concerning platform-specific oil types, fuel consumption, equipment, etc. Since there are thousands of offshore facilities and no central databases, it would be a daunting task to apply these higher tier methods on a worldwide basis or even for North America. MMS has required the Gulf of Mexico operators to provide a Tier 3 estimate by the summer of 2001 but not in time for this report.
For the reasons cited above, the Tier 1 method was used in this report. It requires an estimate of the production volume and a VOC rate per unit produced. E&P Forum (1994) provides estimates based on the information available at the time but a review of the literature revealed more recent and detailed information.
Table D-6 shows the VOC rate for the four regions where VOC estimates have been made. The source of these numbers is given in later discussions. The average of the rates is given in the last row. Mexico (Pemex, 2000) has reported the lowest rate, which was roughly five times lower then in the northeast Atlantic and nearly three times lower then in the U.S. Gulf of Mexico. The smallness of the Mexican number was surprising especially when compared to the U.S. Gulf numbers where oil types were similar and production methods are likely to be at least as clean. Unfortunately no insight was offered into how the Mexican number was calculated so it was difficult to determine the source of the difference. Consequently we took the Mexican rate at face value.
It is of some interest to compare the E&P Forum (1994) VOC rates to those in Tables 2-2 through 2-6. The E&P rate for the N. E. Atlantic of 1.1 × 10−3 (based on an average of the E&P values for Norway and the U. K.) compares well to the 1.17 × 10−3 developed in this study. The only other common region is the Gulf of Mexico. Here the E&P estimate of 2.2 × 10−3 is about four times larger then developed in this study. The source of the discrepancy is fairly obvious: E&P based their estimates on an EPA estimate that was not as recent as the MMS estimates used in this study. E&P also provides estimates for Canada but these are based on onshore fields in Alberta, not a particularly good basis.
The lowest VOC rate in Tables 2-2 through 2-6 was used to estimate the lower bound estimate of VOC for each region, and the highest VOC rate was used to estimate the upper bound estimate. For regions where no VOC were published, the best estimate of VOC emissions was calculated by multiplying the average in Tables 2-2 through 2-6 by the