​

Day 2 Workshop:

​

Climate Change Impacts On Flood Risks and Decision Making

​

Technical Overview

​

Photo credit: Reuters

https://www.reuters.com/article/us-texas-floods-forthood-idUSKCN0YR04U

Purpose:

• To gather modeling approaches and case studies in the burgeoning field of compound flooding.
• To discuss considerations and methods for capturing uncertainty, cognitive, societal and cultural factors for decision-making in response to flood risks.
• To explore opportunities for collaborations and potential uses for specific decision-makers, including DoD installations.
• To summarize the workshop findings.

The workshop will explore opportunities and approaches for multidisciplinary assessment and response to climate change impacts on flood risks by integrating physical modeling, human factors, and decision science.

Goals:

Motivations

• Alters the frequency, intensity, and location of storms, thereby increase vulnerability and risks
• Reshapes DoD strategic, operational, and tactical environments
• Presents an urgent need to understand how climate changes propagate into flood impacts on DoD installations and operational viability

Climate Change as a National Security Threat

Motivations

• Various uncertainty sources are often uncaptured
• Heavy computational needs prevents high-fidelity simulations, full uncertainty quantification and data assimilation, and thus threatens informed decision-making in real-time
• Lack of methodologies that integrate physics models with observation data, and can be rapidly queried in real-time to make evidence-based decisions

Limitations of Current Flood Risk Assessment Tools

Motivations

• Need both forward UQ (propagate uncertainty) and inverse UQ (reduce uncertainty from data)
• Effective decision-making must allow interactions between the human decision-maker and the computational results
• Lack of a systematic framework to support decision-making with end-to-end UQ and human factors responding to flood risks

Need for End-to-End UQ and human factors in decision-making

A Possible Approach

Modeling and Prediction of Climate, Precipitation, and Flood

• Analyze historical and future flood-producing storms and generate regional simulations at scales required to adequately capture organized convection and extreme precipitation

• Produce high-fidelity flood simulations capturing meso-scale topography (𝒪(10-10³) m) and micro-scale features (channel, floodplain, buildings) as well as the dynamically changing rainfall losses due to infiltration and stormwater management infrastructure (𝒪(1-10) m)

Incorporating Factors of Psychology, Stakeholders, and Society

• Cultivate methods to capture human decision-making process of regional stakeholders, focusing in particular on their tradeoffs, barriers and facilitators

• Engage with local and regional stakeholders to address the role of local, regional, and institutional cultures on decision-making that supports resilience for DoD facilities

Expert elicitation and surveying

Quantify and Communicate Uncertainty Towards Decision-Making

• Develop end-to-end UQ “backend” computation capabilities, including dimension reduction, uncertainty propagation, Bayesian inference, and optimal experimental design

• Create UQ “frontend” tools (e.g., visualizations) to communicate uncertainty in ways that improves the quality of decisions

• Find optimal decision rules by maximizing an expected utility under uncertainty