Is seeking certainty in climate sensitivity measures counterproductive in the context of climate emergency? The case for scenario planning
本文探讨了使用贝叶斯方法缩小气候敏感度测量范围可能适得其反，因为这会忽略已知未知和意外情况，并可能违反预防原则，因此建议采用情景规划决策框架。
Climate emergency is fast becoming the overriding problem of our times and rapid reductions in carbon emissions a primary policy focus that is liable to affect all aspects of society and economy. A key component in climate science is the “climate sensitivity” measure and there has been a recent attempt using Bayesian updating to narrow this measure in the interests of “firming up the science”. We explore a two stage argument in this regard. First, despite good intentions, use of Bayes sits awkwardly with uncertainty in the form of known unknowns and surprise. Second, narrowing the range may have counterproductive consequences, since the problem is anthropogenic climate change, and there are asymmetric effects from under-response in the face of irreversible and ampliative effects. As such, narrowing the range using Bayes may inadvertently violate the precautionary principle. We take from this that there is a case to be made for scenario focused decision frameworks.