Robert J. Lempert
Michelle E. Miro
The effects of climate-related natural hazards pose a significant threat to sustainable development in Latin America and the Caribbean (LAC) region (Barandiarán, Esquivel et al. 2018) and in particular its transportation sector. Risk Management provides an appropriate framework for assessing and mitigating the impacts of climate change and other climate-related natural hazards on transportation and other systems and choosing actions to enhance their resilience (Jones, Patwardhan et al. 2014; Lempert, Arnold et al. 2018). Risk Management also forms the foundation of the Inter-American Development Bank’s (IDB) Disaster Risk Management Policy (IDB 2007), the Bahamas commitment (IDB 2016) and the Disaster and Climate Change Risk Assessment Methodology for IDB Projects (Barandiarán, Esquivel et al. 2018).
However, analysts and policymakers involved in transportation planning, policy, and investment face significant challenges in managing the risks triggered by the effects of climate change. Climate change impacts the lifespan of roads, airports, and railroads as they have time horizons that surpass 40 years, thus making it harder (if not impossible) to forecast with confidence all relevant future events that will affect such infrastructure. In addition, the climate has already changed, so the return frequency of storms, for example, and other extreme events may now be different than suggested by the historical record in ways that are not always currently well understood. Implementing Risk Management under conditions of such uncertainty can prove difficult (Lempert, Arnold et al. 2018).
Past climate is no longer a reliable predictor of future climate and there is a high level of uncertainty about how climate has and will change in the future. However, waiting for these uncertainties to be resolved does not offer a path forward for transportation planners, who still need to consider future climate and other conditions when developing long-term infrastructure plans. To support long-term planning, climate models, while far from perfect, can offer useful insights into future climate and are helpful when they are used appropriately. Considerations of future climate should also weigh multiple objectives (e.g., reliability, cost-effectiveness, and equity) and other socio-economic or policy conditions, as many decisions will prove effective or provide benefits under multiple future conditions.
In developing plans, weighing benefits, and considering future conditions, planners should not mistake well-characterized risk for conditions of Deep Uncertainty. Well-characterized risk exists when planners and engineers can confidently use single joint probability distributions (i.e., predictions) to describe hazard, exposure, and vulnerability that contribute to risk. In contrast, we define Deep Uncertainty (Lempert, Popper et al. 2003) as:
Deep Uncertainty occurs when the parties to a decision do not know or do not agree on the likelihood of alternative futures or how decisions or actions are related to consequences.
As described below, Deep Uncertainty occurs when the parties to a decision do not know or do not agree on the likelihood of alternative futures or how decision or actions are related to consequences. DMDU enables Risk Management under such conditions. Decision Making Under Deep Uncertainty (DMDU) enables Risk Management under conditions of Deep Uncertainty, that is when risks cannot confidently be quantified.
This guidebook was prepared for and funded by the IDB and is intended to help IDB team leaders, technical experts, planning and executing agencies, and consultants in conducting an analysis of DMDU, which is one approach to the thinking process of evaluating and making decisions under a Risk Management context. This approach and document are therefore aligned with the Disaster and Climate Change Risk Assessment Methodology for IDB projects (IDB 2018) as an approach that applies to system or portfolio analyses.
Specifically, this guidebook introduces and provides guidance on applying methods for DMDU to transportation planning and reviews several such methods, including scenario planning, Adaptive Pathways, and robust decision making (RDM). This review is geared towards supporting the incorporation of DMDU methods into IDB’s transportation sector funding and planning processes. In the risk calculation, instead of an “agreement on assumptions,” DMDU methods pursue an “agreement on potential actions.” That is, the DMDU methods refrain from making explicit predictions about which future will occur in the risk calculation, and instead focus on evaluating potential feasible actions for associated risks and benefits. The focus of such an approach addresses uncertainty not by an explicit numerical quantification, but by selecting robust actions that will maximize benefits across the likely range of potential future conditions.
Section 2 provides a brief summary of risk and iterative Risk Management, its current application to transportation, how the Disaster and Climate Change Risk Assessment Methodology for IDB Projects (Barandiarán, Esquivel et al. 2018) implements these ideas, and how this guidebook supports this IDB methodology for the transportation sector. Section 3 discusses the new challenges generated by climate change, and summarizes information about current and future climate change and climate impacts on transportation in the LAC region. Section 4 introduces decision making under Deep Uncertainty (DMDU) as applied to transportation and reviews several such methods, including scenario planning, Adaptive Pathways, and robust decision making (RDM). The final section offers implications and recommendations for IDB.