No abstract

A series of experiments are described that examine the sensitivity of the northem-hemisphere winter evolution to the equatorial quasi-biennial oscillation (QBO). The prime tool for the experiments is a stratospheremesosphere model. The model is integrated over many years with the modelled equatorial winds relaxed towards observed values in order to simulate a realistic QBO. In experiment A the equatorial winds are relaxed towards Singapore radiosonde observations in the height region 16-32 km. In contrast to previous modelling studies, the Holton-Tan relationship ( w a d c o l d winters associated with easterly/westerly QBO winds in the lower stratosphere) is absent. However, in a second experiment (run B) in which the equatorial winds are relaxed towards rocketsonde data over the extended height range 16-58 km, a realistic Holton-Tan relationship is reproduced. A series of further studies are described that explore in more detail the sensitivity to various equatorial height regions and to the bottom-boundary forcing. The experiments suggest that the evolution of the northern-hemisphere winter circulation is sensitive to equatorial winds throughout the whole depth of the stratosphere and not just to the lowerstratospheric wind direction as previously assumed.

Equatorial winds in the stratosphere are known to influence the frequency of stratospheric midwinter sudden warmings. Sudden warmings, in turn, influence the Earth's climate both through their direct influence on polar temperatures and through the temperature dependence of ozone depletion in the lower stratosphere. The conventional (Holton-Tan) explanation for the equatorial influence on sudden warmings is in terms of the equatorial winds in the lower stratosphere (-20-30 km) acting as a waveguide for midlatitude planetarywave propagation. This study employs stratospheric-temperature analyses and equatorial rocketsonde wind data extending to 58 km to diagnose the relationship between the northern-hemisphere polar temperatures and equatorial zonal winds at all height levels in the stratosphere. In addition to the recognized Holton-Tan relationship linking the polar temperatures to the quasi-biennial oscillation in equatorial winds in the lower stratosphere, a strong correlation of polar temperatures with equatorial winds in the upper stratosphere is found. We suggest that this may be associated with the strength and vertical extent of the westerly phase of the semi-annual oscillation in the upper stratosphere, although the observations alone cannot provide a conclusive, causal relationship. The main diagnostic tools employed are correlation studies and composite analysis. The results underline the need for continued high quality, equatorial wind measurements at all stratospheric levels.

No abstract