Alaska represents a unique challenge to understanding petroleum hydrocarbons in the marine environment. Anthropogenic sources of petroleum hydrocarbons are dominated by spills occurring during the extraction or transportation of petroleum. Unlike regions that have undergone a greater degree of urbanization, low-level, chronic releases of petroleum are not significant in Alaska, thus the erosion of petroleum-bearing rock can be significant at a local scale. Conversely, the ecosystems in the coastal and offshore areas of Alaska have had no need to adapt to elevated concentrations of petroleum hydrocarbon and may, therefore, be even more sensitive to future exposures. Continued efforts by the extraction and transportation industries to minimize the potential for spills will be a key component to ensure the health of natural resources in this area.

In addition to the general conclusions discussed above, several specific actions have to be taken to address the issues raised in this report (see Chapters 3, 4, and 5 for greater detail). This study was largely funded by U.S. entities, especially federal agencies. Many of the following recommendations are therefore focused on actions that can be taken by the United States. This does not mean, however, that they are not broadly applicable to the international community. On the contrary, both individual nations and groups of nations should consider how these recommendations could be implemented worldwide.

Diffuse sources (natural seeps and runoff from land-based sources) are responsible for the majority of petroleum hydrocarbon inputs into North American waters, with contributions of 61 percent and 21 percent, respectively. In contrast, discharges from extraction and marine transportation of petroleum are responsible for less than 3 percent of the hydrocarbon inputs. Natural seeps represent the largest single petroleum hydrocarbon input but there is a great range in the uncertainty estimation. Federal agencies especially the U.S. Geological Survey (USGS), the Minerals Management Service (MMS), and the National Oceanic and Atmospheric Administration (NOAA) should work to develop more accurate techniques for estimating inputs from natural seeps, especially those adjacent to sensitive habitats. Likely techniques will include remote sensing and ground truthing. This will aid in distinguishing the effects of natural processes from those of anthropogenic activities. Urban runoff and recreational boating require attention because the spills are chronic and often occur in sensitive ecosystems. For example, the range of uncertainty in estimates of land-based petroleum hydrocarbons is four orders