CAREER: Diagnosis of forced versus intrinsic low-frequency variability in high-resolution coupled climate models using geostrophic turbulence techniques

Regents Of The University Of Michigan - Ann Arbor, Ann Arbor MI

Investigators

Abstract

Overview: A long-standing question in climate dynamics is the extent to which low-frequency climate variability is intrinsic versus forced. The climate system exhibits variability over a vast range of time scales. Recent findings show that eddy-resolving ocean models exhibit substantial inter-annual variability even when such variability is absent in the atmospheric forcing. This result suggests that intrinsic oceanic nonlinearities are nearly as important as atmospheric forcing in maintaining low-frequency oceanic variability. Earlier work shows that nonlinearities also drive some of the low-frequency variability in atmospheric models. Intellectual Merit: The project addresses an important question of climate science - whether low-frequency variability is free or forced. The application of new tools will provide a useful complement to other approaches used to answer this question. The proposed work builds upon recent research by the principal investigator, in which frequency - and frequency-wavenumber domain spectra, spectral transfers, and spectral fluxes have been diagnosed from eddy-resolving ocean general circulation models in realistic domains, from gridded satellite altimeter data, and from idealized two-layer QG turbulence simulations driven by an imposed baroclinically unstable mean flow. As with spectral transfers and fluxes in wavenumber space, which have long been diagnosed in geostrophic turbulence studies, the spectral transfers and fluxes in frequency space quantify the relative contributions of forcing, nonlinearity, and other processes to the budgets of energy and energy flux. In the idealized two-layer QG turbulence simulations, nonlinearities are the largest terms in the maintenance of low-frequency variance, with forcing and friction playing important but secondary roles. The proposed work will extend this analysis to the oceanic and atmospheric components of coupled climate models, run in both "stand-alone" and fully coupled modes. Broader Impacts: This project will contribute to the quantification and understanding of low-frequency variability, and increase understanding of eddy-resolving coupled climate models, which will soon become widely used tools in climate prediction studies. The project provides funding for a postdoctoral scientist to perform the realistic-domain results, in collaboration with scientists at a leading national climate modeling lab (NOAA/GFDL). The idealized model will be run and analyzed by a graduate student, who has obtained support from an NSF graduate student fellowship. Undergraduates will be integrated into the research, as the principal investigator (PI) has been doing since 2006. Collaboration with the University of Ghana will help to develop earth science capacity in a continent where this capacity is severely lacking. The PI and three members of his group--a postdoc of Ghanaian descent, and two U.S. graduate students--will visit the oceanography department of the University of Ghana for two weeks each summer. At University of Ghana, the PI will give lectures on physical oceanographic topics, and the PI's postdoc and students will help Ghana oceanography students develop skills such as using Matlab, using satellite altimeter products, and attaining familiarity with ocean models. The project builds upon the PI's continuing interest in the development of African science, engendered during his experience as a Peace Corps volunteer teacher in Ghana, and is consistent with the substantial investment of the PI's institution in Africa.