To cope with increasing water scarcity, breeding programs targeting for more efficient crop require comprehensive knowledge of carbon- and water use-related processes, including photosynthesis, the regulation of transpiration (by stomatal closure) and water use efficiency (the ratio of carbon gain to water use). These questions have been scarcely studied in fruit trees, yet are of major economic importance. A main limitation results from the lack of high-throughput techniques to characterize the physiological variability in these processes for hundreds of trees in the field as required to performing genetic analyses. We screened the European apple diversity for photosynthesis, by unique combination of high-throughput and in planta measurements applied on two core-collections in Montpellier (France) and Bologna (Italy), respectively. A progressive soil water deficit was ensured and continuously monitored in both orchards, through summer irrigation withholding, in order to evaluate the genotypic variability in response to drought. We tested the relevance and applicability of the semi-empirical IPL index, computed from chlorophyll fluorescence measurements, as a high-throughput indicator of leaf photosynthesis. The validity of this proxy was assessed through fine measurements of photosynthesis on 6 cultivars common to both collections. We validated the robustness of IPL through model calibration across sites, days and periods of measurements. We then undertook IPL measurements in the whole collections (>250 cultivars). A large variability of IPL was observed with highly significant effects of the genotype and watering scenario (well-watered vs. water deficit). These results pave the way to genome wide association studies (GWAS) to identify the genomic regions controlling the variations detected.