Leaf or needle ozone uptake was estimated for young trees at seven experimental sites across Europe using a stomatal conductance simulation model. Dose–response relationships based on cumulative leaf uptake of ozone (CUO) were calculated using different hourly ozone flux thresholds and these were compared to dose–response relationships based on daylight AOT40, which is currently used within the UNECE Convention on Long-Range Transboundary Air Pollution (CLRTAP). Regression analysis showed that the CUO–biomass response relationships were highly significant for both coniferous and broadleaf trees, and independent of which ozone flux threshold was applied. On the basis of this regressions analysis, an hourly flux threshold of 1.6 nmol m−2 s−1 (CUO>1.6) is proposed as the most appropriate for all species categories in deriving dose–response relationships. The analysis indicated that the current critical level for ozone impacts on European forests of AOT40 10 ppm h may not protect the most sensitive receptors and that critical levels for AOT40 and CUO>1.6 of 5 ppm h and 4 mmol m−2, respectively, are more appropriate. The research identified weaker dose–response relationships for the CUO exposure index compared with AOT40. Distinguishing between sensitive and less sensitive species substantially improved the CUO–biomass response relationships although, still, to a lesser extent than when exposure was expressed as AOT40.