Argania spinosa (the argan tree) is a slowgrowing tree endemic of Morocco, growing on semiarid areas where no other tree species can live. With the aim of predicting temporal changes in A. spinosa woodlands under a probable increase in aridity, we set off to investigate these questions: how do A. spinosa physiological attributes respond to variations in climatic conditions and seasonality, and which is the set of attributes that most affects tree response to environmental conditions? In three study sites, Beni Snassen (North), High-Atlas (Mountain) and Admine Forest in Agadir (Coastal), gas exchange measurements, photochemical efficiency, leaf water potential and different leaf attributes were monitored in February, July and November of 2006. The Mountain site presents the most continental climate. Trees in this site were the most stressed in summer, having the lowest midday leaf water potential values, photochemical efficiency and assimilation rates. We found a Ψ md threshold around -4 MPa, below which stomatal conductance responds linearly to Ψ md . Plants from the North area never reached this threshold during the study period. Although leaf pigments presented a clear seasonal pattern, leaves from Coastal trees exhibit the highest content for each season. The three study sites were separated by two discriminate functions obtained by canonical discriminant analysis. In summer, the Mountain population is separated from the other sites mainly by assimilation rate and F v /F m , while in winter transpiration rates and chlorophyll content are the main discriminant variables. Our study shows that A. spinosa trees adjust their physiological status and leaf attributes to environmental conditions allowing plants to thrive under a dry climate. Under a scenario of global change, the distribution of the argan tree likely shifts to milder areas.