Water shortage in relation with climate change is endured in many regions and is expected to increase in the next decades. To cope with this increasing pressure, the use of a mixture of tree species is nowadays promoted in commercial forestry instead of the standard monoculture widely practiced. The first objective of establishing mixed forest stands is to maintain productivity under the insurance hypothesis; at least one species of the mixture would resist and be resilient to different biotic and abiotic pressures. A postulate often claimed is that mixed forest stands would also help to save water, under the main hypothesis of niche complementarity; the competition for water would be lesser when two or more different species uptake it at different times, or at different locations. However the literature on the effects of mixing species on the forest ecosystem water balance showed contrasted and sometimes contradictory results. Where and under which conditions the complementary effect occurs remain to be specified. This question is addressed in the OPTMix project (Oak Pine Tree Mixture, https://optmix.irstea.fr) designed to test the effect of mixing oak (Quercus petraea) and pine (Pinus sylvestris) on different functions (wood production, biodiversity, resource availability among which, water) of the ecosystem. Thirty three plots divided in three replications were established in the national forest of Orléans, section of Lorris-Les-Bordes, and where mixed stands were compared to even-aged monospecific stands of either oak or pine. Two stand densities were managed, a relatively high one corresponding to traditional silviculture and a low one, assumed to reduce water consumption by trees at the ecosystem level. In each plot, 3 subplots were set up and designed to measure soil water content at 3 depths (-20, -40 and -60 cm from soil surface) as well as the temporary water table depth. Water resource has been measured every 2 hours and recorded every 12 h in dataloggers since the year 2014. In parallel tree water fluxes and water potentials were measured on critical periods of water deficit. Data recording and analyses are still in progress. So that preliminary results should be interpreted with caution. There are as follows. Mixing species had overall little influence on the different measured water variables. This probably results from the shallow soils of the Orleans' forest where both species share a reduced soil volume so that roots take off water in the same soil layers. However oak was a little less constrained by drought when mixed with pine, whereas pine was slightly affected by the presence of oak. These results are in accordance with species functional traits, oak regulates little water spending, whereas pine is much more conservative, i.e. its transpiration is reduced during drought. Consequently a little more water is available for oak when neighbour tree is a pine rather than an oak, decreasing thus competition. Consequences at the ecosystem scale will be discussed as far as the water preservation is concerned.