1. Introduction

[2]  Releases of trace gases as a result of human activity have a potential for causing a major change in Earth's climate. There is no doubt that the problem of the global climate change has led to controversy, speculation, and confusion. However, despite the many uncertainties concerning the timing and magnitude of changes that still remain, the consensus of most scientists knowledgeable in these matters is that a global climate change will occur.

[3]  The troposphere and stratosphere are expected to warm and cool, respectively. The ozone concentration is expected to decrease. The consequences for the atmosphere above 60 km are considered during the recent years to be very active, and overall results indicate that the global changes caused by trace gases (e.g., by CO₂ and CH₄ doubling) are not only confined to the lower atmosphere but also extend well into the mesosphere, thermosphere, and even the ionosphere [Golytsin et al., 1996; Roble and Dickinson, 1989]. The expected changes should also lead to some alterations in global circulation, in distributions of temperature and composition, and in the response of the atmospheric system to solar and auroral variability.

[4]  Thus long-term changes in the upper atmosphere present an important and topical subject. Traditionally, the systematic long-term changes are called "trends," meaning the linear correlation of the parameter with time. Of course, it is only a first-order approximation, and actually, the trends often show interruptions or even reversals.

[5]  Contemporary studies of the long-term changes (trends) in the upper atmosphere are mainly based on the assumptions that the global warming in the troposphere is accompanied by a "global cooling" of the middle atmosphere and thermosphere, that is, by a thermal contraction which lowers the height of the ionosphere [Ulich et al., 2003]. The results concerning the ionosphere can be masked by solar variability and geomagnetic variations and significantly depend on the region and season.

[6]  The systematic long-term changes could be manifested in the upper atmosphere transport and circulation. Langematz et al. [2003] recently found that the increase of CO₂ content and the decrease of the ozone concentration are actually able to influence the stratospheric circulation, the winter period of the circulation systematically increasing and the summer period of the circulation decreasing. However, contrary to the stratosphere, using the long series of the upper mesosphere temperature data (optical atmospheric emissions), Offermann et al. [2003, 2004] detected an increase of the summer duration of about 12% in 20 years.

[7]  The aim of our work was to investigate the variability of the dates of the lower thermosphere circulation reversals and to reveal there possible trends in the summer and winter circulation periods. We had at our disposal the database of the regular multiyear lower thermosphere horizontal wind measurements in the Irkutsk/Badary (East Siberia, 52^oN, 102^oE) and Collm (central Europe, 52^oN, 15^oE) observatories. The technique of these measurements (space diversity reception method in the LF range) was numerously described [e.g., Kazimirovsky and Vergasova, 2001]. The statistical analysis of data and comparison of the results made it possible to establish the mean climatic norms for the seasonal variations of the prevailing (mean, background) wind and to reveal the longitudinal (regional) variability of the lower thermosphere wind regime possibly associated to the external forcing from below [Kazimirovsky and Vergasova, 2001; Kazimirovsky et al., 1999, 2003].