Decision-making under a changing climate

Climate change projections come with a "cascade of uncertainty" due to the chain of models used to obtain them. How can we plan for a changing climate in these conditions ? This lecture will introduce concepts and methods for decision-making when there is no consensus on how to define a planning problem or its uncertainty. This is a situation called deep uncertainty, where :

• multi-objective optimization is used to propose alternatives reflecting competing perspectives, and
• selected alternatives are stress-tested to understand what plausible futures they may be vulnerable to.

The example of run-of-river hydropower design will illustrate the concepts and methods, by demonstrating how their use highlights the limitations of long-held engineering practices in the face of deep uncertainty. For instance, they show that the primary performance metric for design (net present value) can lead to financially vulnerable designs in drying regions.