Weather and climate have substantial impacts on human activity and the economy, among them damage to life and property from extreme weather. To manage these impacts and mitigate their harms -- to do "adaptation", in other words -- it is helpful to assess the risks ahead of time, by estimating the probabilities of impactful events and how costly they might be. Such risk assessment is practiced in multiple sectors and fields including the insurance industry, infrastructure planning, agriculture, public health, and so on. Yet the methods, data, tools and models differ across sectors, particularly in the extent to which and ways in which they account for possible changes in risk driven by changes in climate. Historical data by itself does not adequately capture such changes; climate models can, in principle, do better, but they bring in additional uncertainties and possible errors. Researchers and actors in different sectors make different choices about how to handle the trade-offs involved in using model simulations versus historical observations, in part because of differences in the impacts that matter to them. But it is not always clear how they decide what trade-offs to make, or if the trade-offs they make are the best possible choices. This project seeks to develop a general framework for assessing the value added by using climate model output to inform decision-making for weather and climate risk. The work applies a theoretical framework from economics in which the va