The objective of this paper is to deliver the most accurate ozone (O<sub>3</sub>) and carbon monoxide (CO) climatology for the pure troposphere only, i.e. exclusively from the ground to the dynamical tropopause on an individual profile basis. The results (profiles and columns) are derived solely from the <b>M</b>easurements of <b>OZ</b>one and water vapour by in-service <b>Al</b>rbus air<b>C</b>raft programme (MOZAIC) over 15 years (1994–2009). The study, focused on the northern mid-latitudes [24–50° N] and [119° W–140° E], includes more than 40 000 profiles over 11 sites to give a quasi-global zonal picture. Considering all the sites, the pure tropospheric column peak-to-peak seasonal cycle ranges are 23.7–43.2 DU for O<sub>3</sub> and 1.7–6.9 × 10 <sup>18</sup> molecules cm<sup>−2</sup> for CO. The maxima of the seasonal cycles are not in phase, occurring in February–April for CO and May–July for O<sub>3</sub>. The phase shift is related to the photochemistry and OH removal efficiencies. The purely tropospheric seasonal profiles are characterized by a typical autumn–winter/spring–summer O<sub>3</sub> dichotomy (except in Los Angeles, Eastmed – a cluster of Cairo and Tel Aviv – and the regions impacted by the summer monsoon) and a summer–autumn/winter–spring CO dichotomy. We revisit the boundary-layer, mid-tropospheric (MT) and upper-tropospheric (UT) partial columns using a~new monthly varying MT ceiling. Interestingly, the seasonal cycle maximum of the UT partial columns is shifted from summer to spring for O<sub>3</sub> and to very early spring for CO. Conversely, the MT maximum is shifted from spring to summer and is associated with a summer (winter) MT thickening (thinning). Lastly, the pure tropospheric seasonal cycles derived from our analysis are consistent with the cycles derived from spaceborne measurements, the correlation coefficients being <i>r</i>=0.6–0.9 for O<sub>3</sub> and <i>r</i>>0.9 for CO. The cycles observed from space are nevertheless greater than MOZAIC for O<sub>3</sub> (by 9–18 DU) and smaller for CO (up to 1 × 10 <sup>18</sup> molecules cm<sup>−2</sup>). The larger winter O<sub>3</sub> difference between the two data sets suggests probable stratospheric contamination in satellite data due to the tropopause position. The study underlines the importance of rigorously discriminating between the stratospheric and tropospheric reservoirs and avoiding use of a~monthly averaged tropopause position without this strict discrimination in order to assess the pure O<sub>3</sub> and CO tropospheric trends.