Variability of beta-plane zonal jets

The dynamics of the mid-latitude atmospheric jet is an important component of internal variability in real and modelled climate. The variability may also potentially affect the response to deterministic external forcing, with implications for seasonal prediction including the ‘signal-to-noise’ paradox. Recent research has used ad hoc probabilistic approaches to investigate the paradox but has given little dynamical insight into the behaviour observed in models. This motivates further dynamical study of the factors determining variability and response to forcing. We use a simple stochastically forced barotropic model containing the essential mechanisms for beta-plane jet variability to conduct a range of numerical experiments. We consider first the dependence of the behaviour on the damping time scale and on the amplitude and latitudinal width of stochastic forcing that is statistically homogeneous in longitude. We consider leading empirical orthogonal functions of the zonal mean wind velocity, use these as quantifiers of jet behaviour, and analyse the amplitude, latitudinal structure, and autocorrelation time scale of the simulated variability. We move on to examine cases where there is imposed longitudinal variation. An appropriate decorrelation time scale of the zonal jet stream could be displayed by the model, partly depending on the damping time scale. We conduct experiments with applied forcing to determine whether the basic prediction of the fluctuation-dissipation theorem, that response to forcing is proportional to the autocorrelation time scale for natural variability, holds in this system. This simple dynamical model has good implications for a physical understanding of jet persistence and the signal-to-noise paradox.