Recently, Cooke et al. (Cooke et al. 2022 R. Soc. Open Sci. 9, 211165. (doi:10.1098/rsos.211165)) used a three-dimensional coupled chemistry-climate model (WACCM6) to calculate ozone column depths at varied atmospheric O2 levels. They argued that previous one-dimensional (1-D) photochemical model studies, e.g. Segura et al. (Segura et al. 2003 Astrobiology 3, 689-708. (doi:10.1089/153110703322736024)), may have overestimated the ozone column depth at low pO2, and hence also overestimated the lifetime of methane. We have compared new simulations from an updated version of the Segura et al. model with those from WACCM6, together with some results from a second three-dimensional model. The discrepancy in ozone column depths is probably due to multiple interacting parameters, including H2O in the upper troposphere, lower boundary conditions, vertical and meridional transport rates, and different chemical mechanisms, especially the treatment of O2 photolysis in the Schumann-Runge (SR) bands (175-205 nm). The discrepancy in tropospheric OH concentrations and methane lifetime between WACCM6 and the 1-D model at low pO2 is reduced when absorption from CO2 and H2O in this wavelength region is included in WACCM6. Including scattering in the SR bands may further reduce this difference. Resolving these issues can be accomplished by developing an accurate parametrization for O2 photolysis in the SR bands and then repeating these calculations in the various models.