To investigate the real linkage and effectiveness of using the AOT40 index and ozone stomatal flux (FO 3 ) in the assessment of physiological alteration/leaf injury on clover clones sensitive to ozone and Quercus ilex plants, two statistical techniques – Partial Least Squares (PLS) and Neural Net Analysis (NNA) – were applied. Different results were obtained in relation to the statistical method chosen. Linear methodologies applied to clover highlighted the role of temperature (TEMP) and O 3 concentration (O 3 Mean) in affecting photosynthesis (PHOTO), leaf injury, and stomatal conductance (COND). In Quercus plants, COND was linearly correlated to two environmental variables, TEMP and Vapour Pressure Deficit (VPD), and to two physiological variables, PHOTO and Leaf Transpiration (TRASP), whereas PHOTO was correlated with TEMP, sO 3 , COND and sub-stomatal CO 2 /external CO 2 ratio (Ci/Ca). These linear relationships were, in part, modified by NNA. In fact, non-linear relationships between environmental variables, and morphological and physiological variables were evident, suggesting caution when risk assessments are made on ozone concentration-based critical levels. Both plant types showed a relationship with FO 3 that negatively affected leaf injury and PHOTO in clover and Quercus plants, respectively, suggesting that ozone flux-based critical levels were more effective in linking with leaf injuries or reduction in carbon metabolism.