Skip over navigation

Statistical emulation and interpretation of transient climates

Speaker: Elisabeth Moyer, University of Chicago
Series: CEE Departmental Seminars
Location: Bowen Hall Auditorium
Date/Time: Monday, April 22, 2013, 4:30 p.m. - 6:00 p.m.

Abstract:

Elisabeth MoyerClimate projections are made computationally demanding by the thermal inertia of the ocean, which means that any changes in radiative forcing require several thousand years to be fully manifested as climate change. That timescale makes it prohibitive to use state-of-the-art coupled climate models in many impacts studies, including those on optimal mitigation strategies. We demonstrate a technique for rapid "emulation" of the output of large climate models for arbitrary climate forcings given a pre-computed library of climate model output, developed as part of the University of Chicago's Center for Robust Decision-making on Climate and Energy Policy (RDCEP), that allows the information from large climate models to be brought to policy analysis. Library-based statistical emulation is possible because the behavior of transient climates is sufficiently understandable to be represented with simple functional forms. The RDCEP climate library, currently at over 25,000 model years, also provides insight into two puzzling aspects transient climate behavior of current interest: the suppression in precipitation in warming climates, and the nonlinearity of global heat uptake with temperature. These processes imply that long-term precipitation and temperature increases due to rising CO2 are larger than would be inferred from current observations.  Long coupled-model runs demonstrate that precipitation suppression is not a direct radiative effect, as it is typically described, but is a transient effect driven by heat uptake as the ocean warms. Long model runs also suggest that long-term nonlinearity in ocean heat uptake is not a function of nonlinear ocean physics but is the geometric result of locally linear feedbacks in a system with differing rates of warming. The results suggest that committed global warming may in fact be inferred from short-term observations.