Principal Investigator C Schlosser
Co-investigators John Reilly , Chris E Forest , Ronald Prinn , Michael Follows , Henry Jacoby
Project Website http://globalchange.mit.edu/focus-areas/earth-system-response
A challenging aspect of Earth system research is model evaluation, given that historical data is relatively limited in time and space, subject to considerable measurement error, and that the record captures only one realization of an Earth system subject to natural variability. The Joint Program is applying a variety of statistical approaches to evaluate model performance and update parameters so that models best replicate observations. One of these approaches draws from detection and attribution literature to determine key parameters and their uncertainty -- including climate sensitivity, the rate of heat uptake by the ocean, and the response of the system to aerosol and other forcings. Another direction of research draws on inverse methods to improve estimation of flux models, including those related to anthropogenic emissions. Ultimately this work could lead to real-time data assimilation to improve model performance and prediction, and to evaluate and monitor mitigation efforts.
A key contribution in this area has been joint likelihood estimates of climate sensitivity, ocean heat uptake, and aerosol forcing as well as the role of natural forcings (Quantifying uncertainties in climate system properties with the use of recent climate observations; and Estimated PDFs of climate system properties including natural and anthropogenic forcings). This work indicates that ocean heat uptake may be slower than is often predicted by mainstream ocean models, and hence warming the atmosphere in response to greenhouse gases would occur faster than many models project. These estimates are important inputs into our forward projections of climate system uncertainty.
Inverse methods have been used to constrain estimates of various anthropogenic and natural sources of methane (Atmospheric modeling of high- and low-frequency methane observations: Importance of interannually varying transport). This research can be contrasted with bottom-up measures of sources of emissions that use experimental methods to estimate emissions from representative sources (for example, a rice field) and then extrapolate them to the world using increasingly complex models to extrapolate. In the end these are highly complementary approaches and the goal of the Program is to fuse inverse methods with modeling to update in real-time model parameters to provide a best fit of data and then apply those models in a prognostic mode.