P1.12
Non-Gaussian probability distributions: What are their implications for predictability?
Philip Sura, NOAA-CIRES Climate Diagnostics Center, Boulder, CO; and M. Newman, C. Penland, and P. Sardeshmukh
Atmospheric circulation statistics are not strictly Gaussian. Small bumps and other deviations from Gaussianity are often interpreted as implying the existence of distinct and persistent nonlinear circulation regimes associated with relatively high predictability. Similar deviations, however, also occur in linear systems perturbed by multiplicative noise, and can be associated with much lower predictability. Multiplicative noise is generally identified with state-dependent variations of stochastic feedbacks from unresolved system components, and may in some instances be treated as stochastic perturbations of system parameters. It is shown that introducing such perturbations in the damping coefficient of large-scale linear Rossby waves can produce deviations from Gaussianity very similar to those observed in the joint probability distribution of the first two Principal Components (PCs) of weekly-averaged 750 mb streamfunction data of the past 52 winters. A closer examination of the Fokker-Planck probability budget shows that the observed deviations cannot be understood without accounting for multiplicative noise, and are unlikely the result of deterministic nonlinear interactions between the two PCs. Thus the observed non-Gaussian probability distributions do not necessarily imply the existence of persistent nonlinear circulation regimes, but are rather more consistent with the expected distributions for a linear system perturbed by multiplicative noise.
Poster Session 1, Lorenz Symposium Posters
Thursday, 13 January 2005, 9:45 AM-9:45 AM
Previous paper Next paper