Mesospheric ozone response to changes in water vapor

[1] Observations made by the Upper Atmosphere Research Satellite (UARS) now extend in time for almost a decade and allow for an improved understanding of seasonal and interannual variations of middle atmospheric minor constituents. In particular, measurements of mesospheric ozone and water vapor made by the UARS Halogen Occultation Experiment (HALOE) reveal significant seasonal and decadal variability. At midlatitudes, the observed variation in water vapor is primarily an annual cycle that peaks in summertime driven by changes in the residual circulation. In contrast to earlier satellite observations, HALOE ozone sunset observations also have an annual cycle that is anticorrelated with the water vapor cycle. The anticorrelation is a consequence of water vapor photolysis producing hydrogen species that destroy ozone. However, HALOE sunrise observations of ozone show no seasonal cycle. Analysis of monthly anomalies between 1991 and 2001 show a long-term water vapor increase of approximately 1%/year. The observed response of ozone depends on altitude and time of day and is strongly negative (up to 4%/year) for sunset observations around 80 km. Again, the response of ozone at sunrise is minimal. Sensitivity tests using a three-dimensional global chemical transport model show good agreement with the observations and confirm that the observed decrease in sunset ozone is a response to increasing water vapor.