The following may partly explain this situation. Risk assessments were initially developed by the petroleum sector and were used by the same sector to make

decisions. Through the process linked to the Management plan, other sectors were invited into a settled culture for assessing risk. However, it should be pointed out that the way knowledge gaps are addressed in the preparatory report for the first Management plan [19] matches the approach to refine the impact assessments. The report gives the impression that MEK inhibitor the listed knowledge gaps are possible to fill, that at least some should be filled and that filling knowledge gaps will increase the quality of advice to decision making [9]. The recent and present

efforts to refine impact assessments correspond to several of the listed knowledge gaps. The particular framing of the risk assessment decides which uncertainties are relevant to discuss and which are HTS assay not. In this case, refinements of worst-case scenarios and risk assessments are legitimate while questioning the narrow scope is less so. A worst-case scenario can therefore be understood as a certain way of packaging uncertainty in the policy debate. A consequence of this framing is that representatives from the petroleum sector emphasise the low probability (e.g. in [57]), while environmental NGOs emphasise the great impacts of a

major oil spill (e.g. in [58]). The discussions on uncertainties in this paper suggest that value-laden choices are made (consciously or unconsciously) at three stages: when deciding the scope of risk assessments, when methodological choices are made, and when deciding how results should be presented. It is possible that the policy debate Florfenicol would have been different if other environmental impacts had been emphasised (smaller oil spills, irreducible changes, etc.), if qualitative approaches were more central or if the presentation of risks had been different. Such choices may favour one political action over another. The same issues arise with various handlings of uncertainty. All these choices are value-laden because they have the potential to influence perceptions on what is at risk, how high the risk is, and what ought to be done with regard to the issue. The substantial uncertainties addressed in this paper suggest that the final decision is more a value question than a scientific question. In this case a ranking of risks could be performed to ensure a wider scope on the issue.