Testing parameterizations resolution and spectra测试参数的分辨率与光谱.pptVIP

Testing parameterizations of submesoscale ocean variability:Resolutions and power spectra Alexey Kaplan, Enrique Curchitser, Huei-Ping Huang Lamont-Doherty Earth Observatory of Columbia University William G. Large National Center for Atmospheric Research * * AESOP Planning Workshop Applied Physics Laboratory of the University of Washington Seattle, March 8-9, 2005 Goal: to develop and apply a framework for evaluating the performance of subgrid scale parameterizations in ocean models where the mesoscale is explicitly resolved. Means: A nesting system of models using the Regional Ocean Modeling System (ROMS) with resolutions varying from 40km to 250m for modeling ocean variability in Monterey Bay and the surrounding regions of Pacific Ocean. Variability and spectra: Spatial and temporal variability of all variables of interest will be studied. Spectral wavenumber distributions of ocean variability and energy transfer between different scales will be computed and analyzed, as a way to evaluate the fidelity of model simulations, to investigate the scalability of small-scale variability in the ocean models, observations, and parameterizations, and to infer the desirable features of the parameterization schemes. 0 Approach to testing parameterizations: via separation of impacts of model resolution on the mean model dynamics and on the parameterization of subgrid scale variability. Framework: implementation of a system of spatial pre-filtering for model fields entering the parameterization schemes. Such a pre-filtering system makes it possible to change from the highest (H) to the reduced (R) effective resolution of inputs to a subgrid parameterization scheme, while keeping the resolution of the prognostic dynamical scheme unchanged. Diagnostic parameteriz. Prognostic parameteriz. Dynamical scheme Types of model runs R R R How much is lost? R R H Parameterizations at R affect dynamics? R H H Ho