Aircraft data from the Stratosphere-Troposphere Exchange Project (STEP) and the Equatorial Mesoscale Experiment (EMEX) flights conducted on February 2, 1987, off northern Australia are used in cumulus cloud and photochemical models to determine the effects of convection on upper tropospheric O3 production. Ozone production is calculated as the amount integrated over cloud outflow layers for the first 24 hours after convection. Ozone production with convection is compared to ozone formation in undisturbed conditions. Model simulations of the EMEX 9 convective system indicate lower tropospheric air relatively rich in CO and low in NOx exiting in cloud outflow, slightly depressing the rate of O3 formation in the middle and upper troposphere. Other convective complexes, 800--900 km upstream, caused even greater perturbations to measured profiles of CO, NOx, O3, and H2O and implied a 15--20% reduction in the rate of O3 production from 14.5 to 17 km. The greatest factor affecting O3 formation in the upper troposphere in the STEP/EMEX flight might have been lightning-produced NOx. We estimate that O3 production from 12 to 17 km is 2--3 times more rapid than it would be with no lightning. This STEP/EMEX event adds to a climatology of half a dozen cases we have analyzed to determine the effects of convection on free tropospheric O3 production. The study region represents the ''maritime continent'' in contrast to continental regions studied previously. Relatively small quantities of species from the lower troposphere were transported to the upper troposphere because of the relatively weak vertical velocities in the storm and because chemical species gradients had been minimized by frequent convection prior to the February 2 event. Earlier in the convective season, the chemical consequences of a single episode might have been more substantial. ¿ American Geophysical Union 1993