#O03 Evaluation of small-scale, nonlinear physical processes in climate simulations
Abstract
Secular changes in Earth’s climate system are driven by large-scale, long-term sequestration of heat within the deep ocean. Small scale processes, such as mixing occurring due to ocean eddies, interacts via nonlinear feedbacks to exert a dominant impact on the transport of mass and heat in the ocean. These small scale processes impact the climate in two ways: vertically, heat is exchanged between the deep ocean and atmosphere via vertical mixing within the oceanic surface boundary layer; Horizontally and internally within fluid layers, heat transport occurs laterally along ocean density layers that periodically outcrop to the surface or deep ocean, providing a mechanism for deep-ocean entrainment of heat.
These vertical and horizontal processes that mix the fluid occur at subgrid resolutions for most climate simulations. The regional refinement capability of MPAS-O has allowed for the direct simulation of important small scale processes in selected regions and their impact on the climate. At such fine resolutions, post-processing diagnostics of mixing impacts is impractical. MPAS-O has in-situ analysis capabilities (e.g., via Lagrangian In-situ Global High-performance particle Tracking (LIGHT) as well as the Eliassen-Palm Flux Tensor) that allow for diagnosis and evaluation of the mixing impact of small scale eddies on the large scale flow. This diagnosis can be used to develop improved parameterizations for models that cannot directly simulate these important phenomena.
Even with this novel regional refinement capability, some processes such as vertical mixing, cannot be resolved explicitly in MPAS-O. Direct simulation of boundary layer mixing in large scale models will not be feasible for many decades, if ever. Thus, Large Eddy Simulations that resolve small scale vertical mixing processes are subsequently used to evaluate the most ubiquitous oceanic vertical mixing parameterization, the Kappa-Profile Parameterization. The influence of this parameterization on important climate phenomena (such as El Nino / Southern Oscillation) is explored. Our testing also demonstrated that improved vertical mixing schemes are needed to improve scalar transport from the surface to the deep ocean and therefore model fidelity.
Ultimately, fidelity of climate simulation depends on the mixing parameterizations used in the ocean model. Only by understanding the complexities of unresolved horizontal and vertical mixing will the correct carbon and heat uptake into the global ocean be computed. These fluxes in the coupled ocean-atmosphere system are essential to describe secular climate changes to the Earth System.