College of Engineering | Apply to CSU | Disclaimer | Equal Opportunity Statement | Privacy | Search CSU

Wave-mean flow positive feedbacks associated with sudden stratospheric warmings

You must be on the CSU network—either physically or using VPN—to watch this or any of the videos on this site.

September 16, 2014
Jeremiah Sjoberg
Hosted by Thomas Birner (advisor), Dave Thompson, Wayne Schubert, Rolando Garcia (Affiliate Faculty), Richard Eykholt (Physics)


Sudden stratospheric warmings — most often characterized by zonal mean zonal wind easterlies at 60N, 10 hPa — represent the largest dynamical event of the wintertime polar stratosphere. Despite this, the predictability remains low, in part because the forcing of these warming events involves a nonlinear positive feedback between planetary scale waves and the zonal wind of the polar stratosphere. In the wave-mean flow positive feedback, wave forcing decelerates the mean flow, allowing enhanced upward wave propagation, which then further decelerates the mean flow, etc., until the mean flow no longer supports wave propagation. This positive feedback process is crucial for the initiation of such events. Because the associated low predictability stems from poorly resolving initiation, this dissertation focuses on increasing mechanistic understanding of the wave-mean flow positive feedback associated with sudden stratospheric warmings.

A simple model of wave-mean flow interaction is the first tool utilized here. In the original form of the model, constant bottom boundary wave forcing, set by geopotential height perturbations, results in a zonal wind state that oscillates between positive values (westerlies) and negative values (easterlies). We present a reformulation of the bottom boundary condition which allows for specification of the upward wave activity flux. Unlike with the original bottom boundary condition, we may precisely set the wave amplitudes propagating into the model domain. With this reformulated model, steady incoming wave fluxes lead to a steady zonal wind response. The oscillating state from the original model is found to rely on a representation of the positive feedback which is too strong. Transient forcing experiments in the reformulated simple model support previous results that there is a preferential wave forcing time scale on the order of 10 days for sudden stratospheric warmings. Forcing the model near this preferential time scale most efficiently drives the positive feedback.

Lower stratospheric wave fields in reanalysis data show supporting evidence for these preferential wave forcing time scales prior to sudden stratospheric warmings. Pulses of wave activity flux are also analyzed in reanalysis data, and a set of pulses which are a novel proxy for strong wave-mean flow positive feedback are found. The zonal wind near these pulses display the expected characteristics of the positive feedback: strong precedent zonal winds and strong subsequent wind decelerations. This proxy is thus a useful diagnostic for the wave-mean flow positive feedback.

A general circulation model forced by idealized planetary scale topography is employed to perform high order experiments. By stepwise increasing the height of the topography, we find that the frequency of sudden stratospheric warmings within the model increases nonlinearly to a maximum at moderate topographic heights and then strongly jumps down to a lower, steady value for still higher topography. Analyzing the proxy for positive feedback here reveals that the positive feedback is strongest in the range of topographic heights associated with the largest occurrence of sudden warmings, and also that preferential wave forcing time scales on the order of 10 days are upheld.