Six campaigns to measure vertical profiles of water vapor and ozone in the eastern tropical Pacific have taken place at San Cristóbal, Galapagos between 1998 and 2000 as part of SOWER/Pacific, as well as one campaign in Watukosek, In-donesia in December 2001. In all campaigns water vapor was measured by the NOAA/CMDL frost-point hygrometer. Previous campaigns with the NOAA/CMDL hygrometer took place in the western Pacific region in 1993, in north eastern Brazil in 1997, thus providing a uniform data set of tropical upper tropospheric and lower stratospheric water vapor. These observations provide important data elucidating the longitudinal and seasonal variation of water vapor in the tropical upper troposphere and lower stratosphere and of the dehydration processes taking place as air crosses the tropical tropopause region.

The campaigns at San Cristóbal (0.9ºS, 89.9ºW) took place in March/April 1998 and March 1999, in September 1998 and September 1999, and in November/December 2000. The different meteorological conditions between these campaigns provide interesting results, even for campaigns in the same season. The first campaign in March/April of 1998 took place at the end of the very strong El Niño of 1997/98, while the others took place under the La Niña conditions. Furthermore, the campaigns cover nearly one full phase of the quasi-biannual oscillation (QBO).

Saturation is an indicator and one of the important prerequisites for the dehydration of air in the upper troposphere. All observations show a high relative humidity with respect to ice (RHI) in the uppermost troposphere. Typical values for RHI are between 35% and 80%, with values in excess of 100% during most spring campaigns. Extremely low water vapor mixing ratios at the tropopause have only been observed in the western Pacific region, but not in the eastern Pacific and South American region. Extreme deep convection, reaching the tropical tropopause is one of the import processes, which can dehydrate air. This process was observed in the western Pacific region, but not in the eastern Pacific region, where deep convective events never reached the tropopause. Nevertheless, the data indicate ongoing dehydration even in the eastern Pacific region during the northern winter and spring months through slow ascending motion. However, data from Brazil indicate that dehydration through this slow ascending motion is not ubiquitous and probably limited to regions with cold enough tropopause temperatures. Furthermore there is strong evidence, that Kelvin wave activity at the tropical tro-popause contributes to the dehydration in the tropical upper troposphere. The data from outside the western Pacific region provide strong evidence both in ozone and water vapor for stratospheric contributions to the air below the tropopause, pro-viding additional support for the transitional region below the tropical tropopause.