Global change is increasing the pressure on forest ecosystems in the critical zone that supplies valuable ecosystem services and critically affects stream water quantity and quality. Tackling environmental management challenges in the critical zone requires holistic comprehension of the interactions and feedbacks between vegetation and water in forest ecosystems. One critical element in this respect is the need for a thorough understanding of space-time patterns of tree water uptake and the associated first order controls. Much progress was achieved to determine the sources (i.e. soil water, groundwater, stream water) of tree water uptake using stable isotopes of hydrogen and oxygen. Today, it has been shown that access to deeper water sources in the regolith can sustain transpiration during dry periods, thus climate and landscape position and related access to water sources seem to play a critical role on tree water uptake sources. However, much is still unsolved on the role of seasonality and hillslope catena process on tree water uptake under different climatic and geological settings. Within this proposal the hypothesis that ecohydrological processes in hillslopes control sources of tree water uptake and that climate and geology are first order controls will be tested. This will be done by a hillslope comparison study in Luxembourg and Italy, under different climatic conditions and in different geological settings. The three investigated hillslopes are dominated by beech trees (Fagus Sylvatica). For this, stable isotopes of 2H and 18O will be used to determine tree water sources by observing xylem water, soil water, groundwater, and throughfall fortnightly. A Bayesian mixing model will be applied for determining the sources. The proposal contributes to an increased scientific understanding of ecohydrological processes in the critical zone, which is critical for a better management of natural resources under global change and deteriorating forest health in Luxembourg and beyond.