Scale-dependent Variability in Global Analyses and Prediction Models

In their 1972 paper entitled “Free oscillations of a discrete stratified fluid with application to numerical weather prediction”, R. Dickinson and D. Williamson discussed the normal modes of linearized primitive equations as a useful tool not only for addressing the problem of initialization of numerical weather prediction (NWP) models, but also for the identification of model modes which have significance for climate simulations and for comparison of amplitudes of model modes with those observed in the real atmosphere.

This talk will present recent applications of the normal mode function (NMF) decomposition for the quantification of scale-dependent spatio-temporal variability in global analyses and prediction models. The NMFs facilitate the discussion of the multivariate nature of some of the leading modes of atmospheric variability, especially in the tropics where the role of the large-scale inertio-gravity modes (e.g. the Kelvin waves) in the initial state for NWP, in the growth of forecast uncertainties and in the climate system modelling is not well understood.

The NMF framework has been successfully used to evaluate 1) the information content of observations in data assimilation for NWP, 2) the spectra of analysis and forecast uncertainties, 3) the impact of inertia-gravity modes on the mass-wind coupling near the equator, 4) atmospheric variability and its trends on many scales. In addition, the NMFs provide a novel relationship between the bias and simulated spatial and temporal variance by a climate model in comparison with verifying reanalyses.