The Stratospheric Aerosol and Gas Experiment (SAGE) II has provided unprecedented information of water vapor distribution in the upper troposphere. For the purpose of comparison with output from climate models, the present study assesses the impact of the SAGE II sampling strategy on the tropospheric water vapor climatology in a general circulation model. Since water vapor is sampled only in ''non-cloudy'' regions in the SAGE strategy, the sampled water vapor concentration is smaller than the real climatology. This difference is associated with two factors. One is the water-vapor sampling frequency, the other is the humidity variability inside and outside the clouds. It is shown that maximum difference is at around 300 to 500 mb where it reaches up to 40% in the zonal mean humidity. ¿ American Geophysical Union 1995