Abstract This study presents a reanalysis of ozone (O3) exposure experiments performed on deciduous broadleaf oak species in the Mediterranean region and a proposal of critical levels to improve the O3 risk assessment in this area for these widely distributed forest species. Two experiments performed in Spain and Italy were considered, and the following 3 oak species were studied: Quercus pyrenaica, Q. faginea and Q. robur. All the experiments were performed with irrigated potted seedlings growing in Open-Top Chambers exposed to different O3 levels (with charcoal-filtered air as the control treatment) for two consecutive growing seasons. The Phytotoxic Ozone Dose above an instantaneous threshold of 1 nmol O3 m−2 s−1 (POD1) was calculated by applying a Jarvis type model for the estimation of the stomatal conductance (gs), and by adopting a big-leaf resistive scheme to account for the O3 deposition on the vegetation. Two parameterisations were used for the gs multiplicative model: one species-specific based on the “local” gs measurements performed during each experiment, and the other “generic” based on the “Deciduous Mediterranean broadleaf” parameterisation described in the Manual on Methodologies and Criteria for Mapping Critical Loads and Levels and Air Pollution Effects, Risks and Trends of the UN/ECE (CLRTAP, 2015). The two different parameterisations were used to derive dose-response functions and ozone critical levels for the biomass loss of the deciduous oak species. The dose-response functions for roots and total biomass were statistically significant, with both the parameterisations tested (p