Managing Uncertainties in Risk-based Adaptation to Climate Change

Abstract eng:
The changing climate may impose significant threats to infrastructures, human health, agriculture and other economic sectors, and as such climate change may potentially result in setbacks and losses for business and society. It is therefore paramount that adaptation measures are implemented timely and adequately to safeguard assets and resources for present and future generations. This requires rational decision making under a vast amount of uncertainties resulting from the necessary idealizations made in modeling physical processes, orography, etc.; the statistical analysis of climate projections; the uncertainties from modeling impacts to assets (the losses) that the changing climate may cause. A major challenge is to handle this vast amount of uncertainty that grows exponentially in length of the lead time (forecast horizon). Within climate science this exponential growth of uncertainties is called 'cascade of uncertainties'. To assure transparency in handling all uncertainties as well as comparability between different adaptation studies, DNV is developing a Recommended Practice (RP) for adaptation to climate change with the objective to provide a rational framework for adapting structural systems. The framework makes use of a quantitative risk-based approach, where risk is measured as the expected monetary loss. The calculated monetary loss (the risk) may not represent a physically observable monetary loss. Rather, it is a decision variable that embraces all the above uncertainties. Decision strategies are needed for the timing of investments and allocation of resources to safeguard the future in a sustainable manner. The RP provides the basis for formulating such decision strategies. The purpose of this paper is to illustrate the RP and how uncertainties described by the large set of random variables (typically 50 to 100 of which only a few contribute significantly to the total uncertainty) entering the risk model can be managed. Part of the uncertainties can be reduced through (costly and time consuming) information gathering. Because of this, the decision theoretic Value of Information is illustrated as a mean to identify what variables that most effectively may reduce the total uncertainty.

• Export as: BibTeX | MARC | MARCXML | DC | EndNote | NLM | RefWorks
• Add to your basket