charged to the Great Lakes would be biochemically reduced, or would attach to solids and settle out during the extended residence time in the lakes and would therefore not make it to the ocean. Likewise, closed inland basins such as the Great Salt Lake would not discharge to the sea. (Contact NRC staff to obtain information describing how specific metropolitan areas were classified as contributing to major river basins.)
For the majority of the inland river basins, no usable oil and grease data were available in STORET. In addition, the number of observations for the Hudson, James, Neuse, Sacramento, and Susquehanna rivers was very small (2, 1, 7, 4 and 2, respectively). It was therefore decided to use an alternative procedure based on the unit loads of oil and grease per urban land area and per capita calculated from Steps 1-4 to estimate the contributions of oil and grease from these other river basins. The procedure was as follows:
The unit loads of oil and grease per urban land area calculated from Steps 1-4 were used for the other river basins with the following assumptions:
The Hudson and James rivers were assumed to have unit loads of oil and grease per urban land area of 12.22 g m−2 yr−1, the values calculated from 99 observations in the 1990s on the Delaware River. The high unit loadings on the Delaware River are likely due to the highly industrialized nature of the waterway, and the Hudson and James rivers are also very industrialized.
It was assumed that Alaskan rivers (i.e., Copper, Susitna, and Yukon rivers) did not contribute significant loads of oil and grease to the ocean.
All other rivers for which measured data were not adequate or were unavailable were assumed to have unit loads of oil and grease per urban land area of 1.25 g m−2 yr−1. This value was based on the average annual loading for 1990s data from the Mississippi and Delaware rivers together divided by the urban areas in both basins. Rivers for which this value applied included the Alabama-Tombigbee, Altamaha, Apalachicola, Brazos, Colorado (Texas), Columbia, Neuse, Potomac, Rio Grande, Roanoke, Sabine, Sacramento, St. Lawrence, Santee, San Joaquin, Saskatchewan, Savannah, Susquehanna, and Trinity rivers.
Using data obtained from the U.S. Bureau of the Census (1998) and Statistics Canada (2000), the annual loads per unit land area (Lai) were calculated as follows:
where lai was the unit load for river i as described in Step 5.a. The urban land area, Aui, was calculated in the same manner as described in Step 4 for metropolitan areas in the United States. For metropolitan areas in Canada, Aui was calculated using data from Statistics Canada (2000).
For the United States, metropolitan areas in U.S. Bureau of the Census (1998) were classified as contributing to coastal basins if they fell within one of the 451 coastal counties defined by Culliton et al. (1990). The individual coastal basin metropolitan areas were then aggregated into the appropriate coastal zones in Figure 1-7. The data for 1997 urban land area for metropolitan areas as of June 30, 1996 (U.S. Bureau of the Census, 1998) were then compiled for each coastal zone. Similarly, data from Statistics Canada (2000) for Canadian metropolitan areas along the coast were grouped into the appropriate coastal zones.
The annual load Lai was calculated for urban areas in each coastal zone i in the United States and Canada using Equation I-3. The unit load per urban land area for coastal zone i, lai, was 12.22 g m−2 yr−1 for coastal zone D, and 1.25 g m−2 yr−1 for all other coastal zones. The unit loads were set at higher values for Coastal Zone D because that is the coastal zone to which the Delaware River discharges. (Contact NRC staff to obtain information describing how specific metropolitan areas were classified as contributing to various coastal zones.)
Because almost one-fourth of the crude oil distillation capacity of the United States is located along the Gulf coast (Radler, 1999), the petroleum refining industry discharges a substantial amount of additional oil and grease to coastal waters in that area. To estimate this contribution, data for oil refineries in Louisiana and Texas (from Radler, 1999) were used to estimate the operating capacity of coastal refineries in these states (Table I-4). The petroleum hydrocarbon discharge was determined by multiplying the operating capacity by an assumed rate of hydrocarbon loss that corresponded to effluent guidelines for these discharges (American Petroleum Institute, National Ocean Industries Association, and Offshore Operators Committee, 2001):
6.0 lbs per 1000 barrels of crude produced
3.2 lbs per 1000 barrels of crude produced
Calculations using each of these guidelines were made, and the average of the two calculations was used as a best estimate of the loadings. This discharge was added to the coastal discharge for coastal zone G.