Robustness-based evaluation of hydropower infrastructure design under climate change

Abstract

The conventional tools of decision-making in water resources infrastructure planning have been developed for problems with well-characterized uncertainties and are ill-suited for problems involving climate nonstationarity. In the past 20years, a predict-then-act-based approach to the incorporation of climate nonstationarity has been widely adopted in which the outputs of bias-corrected climate model projections are used to evaluate planning options. However, the ambiguous nature of results has often proved unsatisfying to decision makers. This paper presents the use of a bottom-up, decision scaling framework for the evaluation of water resources infrastructure design alternatives regarding their robustness to climate change and expected value of performance. The analysis begins with an assessment of the vulnerability of the alternative designs under a wide domain of systematically-generated plausible future climates and utilizes downscaled climate projections ex post to inform likelihoods within a risk-based evaluation. The outcomes under different project designs are compared by way of a set of decision criteria, including the performance under the most likely future, expected value of performance across all evaluated futures and robustness. The method is demonstrated for the design of a hydropower system in sub-Saharan Africa and is compared to the results that would be found using a GCM-based, scenario-led analysis. The results indicate that recommendations from the decision scaling analysis can be substantially different from the scenario-led approach, alleviate common shortcomings related to the use of climate projections in water resources planning, and produce recommendations that are more robust to future climate uncertainty.