Atmospheric Chemistry and Physics www.atmos-chem-phys.net 1680-7316 1680-7324 4 11/12 2004 10.5194/acp-4-2449-2004 http://www.atmos-chem-phys.net/4/2449/2004/ http://www.atmos-chem-phys.net/4/2449/2004/acp-4-2449-2004.html http://www.atmos-chem-phys.net/4/2449/2004/acp-4-2449-2004.pdf 2449 2464 2004-12-08 Noctilucent clouds and the mesospheric water vapour: the past decade U. von Zahn G. Baumgarten U. Berger J. Fiedler P. Hartogh Leibniz-Institute of Atmospheric Physics, Kühlungsborn, Germany Max-Planck-Institute for Solar System Research, Katlenburg-Lindau, Germany The topic of this paper is the sensitivity of the brightness of noctilucent clouds (NLC) on the ambient water vapour mixing ratio f(H<sub>2</sub>O). Firstly, we use state-of-the-art models of NLC layer formation to predict NLC brightness changes in response to changes in the 80km mixing ratio f(H<sub>2</sub>O) for the two cases of ground-based 532nm lidar observations at 69&deg; N and for hemispheric satellite SBUV observations at 252nm wavelength. In this study, we include a re-evaluation of the sensitivity of NLC brightness to changes in solar Lyman &alpha; flux. Secondly, we review observations of episodic changes in f(H<sub>2</sub>O) and those in NLC brightness, the former being available since 1992, the latter since 1979. To this review, we add a new series of observations of f(H<sub>2</sub>O), performed in the Arctic summer at the ALOMAR observatory. The episodic change exhibited by the Arctic summer means of f(H<sub>2</sub>O) turns out to be quite different from all those derived from annual means of f(H<sub>2</sub>O). The latter indicate that since 1996 a significant reduction of annually averaged upper mesospheric water vapour has occurred at low, mid, and high latitudes. These decreases of f(H<sub>2</sub>O) have been observed over the same time period in which a slow increase of SBUV NLC albedo has occurred. From this scenario and additional arguments we conclude that the cause for the observed long-term increase in NLC albedo remains to be identified. We close with comments on the very different character of decadal variations in NLC brightness and occurrence rate.