Climate Attribution through Enhanced Fingerprinting

DESCRIPTION: CLDERA will enable climate attribution in currently unachievable scenarios with a novel approach to uncover and employ "pathways”, defined as the chain of physical processes between source and impacts and their spatio-temporal evolution. Novel computational modeling techniques will fuse with large-scale observational data from the 1991 Mt. Pinatubo eruption to elucidate the dominant pathways between source and impacts. These pathways will be crucial constraints, helping to cull the possibilities in attribution. CLDERA's success will not only be a transformation in climate statistical approaches for attribution but also for the national security and policy considerations of decision-makers.

The aim of climate attribution in CLDERA is to determine the dominant source for a given impact-it complementarily reframes the forward simulated pathways problem to identify and rank all possible sources that could have led to that impact. Dominant drivers from the simulated and observed pathways will be used in this thrust to constrain the problem of finding the predominant source for an observed impact. Two classes of methods will be explored: optimization and statistical. Fingerprinting is the state of-the-art statistical approach for climate attribution that detects the significance of a source contributing to an impact, but it can result in physically meaningless source-impact correlations because it is often unbounded from etiological relationships. We will expand fingerprinting into a multivariate analysis with the inclusion of pathway information. This is anticipated to strengthen the signal-to-noise ratio and enable demonstration of fingerprints for single emission, short-term Mt. Pinatubo effect while also ensuring physically meaningful correlations.

Dr. Kate Marvel's research group shall support the attribution thrust by assisting in the development of enhanced fingerprinting. She is actively working on developing fingerprints at multiple spatiotemporal scales with the goal of increasing the signal-to-noise ratio through better use of physical information and quantifying uncertainties arising from observations, models, and gaps in understanding.