Previous: Ext. Abst. Next: Data Up: Ext. Abst.

Introduction

The study of the variability of stratospheric temperatures would ideally require homogeneous, long-recorded, high vertical resolution data. Existing observations differ in type of measurements, length of time period and time-space sampling. Data from ground-based instruments such as lidar, radiosonde, rocketsonde cover rather long periods but are not uniformly distributed around the globe; conversely, satellite data provide global coverage and uniform distribution permitting a time-space analysis of pattern variability. On the other hand the vertical resolution is much coarser.

While there is an overall agreement on  a negative stratospheric temperature trend over the last 2-3 decades, the amplitude of the trend depends on what is used for its inference (WMO report, 1998). Besides, the estimation of the trend requires a precise knowledge of the influence of the solar cycle, volcanic eruptions, QBO, and ENSO which all play an important role in the stratospheric temperature variability.

The purpose of this work is to attempt to discriminate the effect of the previously-discussed factors on the variability of the stratospheric temperatures, so that a trend can be estimated. In a first step, in order to quantify the uncertainties in our dataset, comparisons between our record and the data set of the Stratospheric Group of the Free University of Berlin (FUB) obtained in a independent way are presented. Then, the response to the QBO, ENSO and solar forcings are isolated from the long-term linear trend using a multi-parameter least squares fit analysis.

Previous: Ext. Abst. Next: Data Up: Ext. Abst.