AbstractThe global warming simulations of the general circulation models (GCMs) are generally performed with different ozone prescriptions. We find that the differences in ozone distribution, especially in the upper tropospheric and lower stratospheric (UTLS) region, account for important model discrepancies shown in the ozone‐only historical experiment of the Coupled Model Intercomparison Project Phase 5 (CMIP5). These discrepancies include global high cloud fraction, stratospheric temperature, and stratospheric water vapor. Through a set of experiments conducted by an atmospheric GCM with contrasting UTLS ozone prescriptions, we verify that UTLS ozone not only directly radiatively heats the UTLS region and cools the upper parts of the stratosphere but also strongly influences the high clouds due to its impact on relative humidity and static stability in the UTLS region and the stratospheric water vapor due to its impact on the tropical tropopause temperature. These consequences strongly affect the global mean effective radiative forcing of ozone, as noted in previous studies. Our findings suggest that special attention should be paid to the UTLS ozone when evaluating the climate effects of ozone depletion in the 20th century and recovery in the 21st century. UTLS ozone difference may also be important for understanding the intermodel discrepancy in the climate projections of the CMIP6 GCMs in which either prescribed or interactive ozone is used.