Cover crops are known to provide a large number of ecosystem services such as nitrate catch crop and green manure effect. However, the impact of cover crops on water balance is little studied. References reported an increase of evapotranspiration through an increase of cover transpiration that could decrease water drainage. Nevertheless, according to the cover crops management the volume of this modification. There is no consensus on available water for the next cash crop and on the changes in soil water profile which could be explained by the temporal distribution of the rainfall in interaction with cover crop growth rate and date of destruction. We set up an experiment to measure the impacts of cover crops and their management on soil water dynamics during the fallow period between two main cash crops. Ethiopian mustard and crimson clover were sown on July 31 as a mixture cover. A bare soil serves as control. We tested two dates of destruction, in November for a short growing period and in April for a long cover crop growth period. For cover crops destroyed in November, we used two types of residues management, i.e. i) one part of cover crop was destroyed by crushing and left in mulch and ii) the other part was destroyed by ploughing. These four modalities were replicated in four blocks in a spilt-plot design. We measured soil water profiles by gravimetric measurements once a month until 1.2 meter depth by layers of 0.2 meter. Experiments will finish in April 2018. At the present time, we have compared the effect of cover crops versus bare soil on water content in soil profiles. At initial state, both soil water content were equivalent. Therefore, it allows evaluating effect of cover crops on soil water content during the experiment period. Five weeks after sowing, we could see less water in the first part of soil between surface and 0.5 m depth, indicating the water uptake by cover crop, even if the difference was not significant. Three months after sowing we observed a significant lower water content in soil under the cover crop treatment, reduced by ca.50 mm for the whole profile. The measurements also indicated the presence of cover crop roots in deep layers after only 2 months of growth. Two weeks after cover crops destruction, without rainfall, the difference between the cover crop and the bare soil was again ca.50 mm, confirming the important uptake of water. This difference raises questions about the management of cover crops and the impact of the date of destruction on the amount of water available for the next crop. In the French southwestern conditions, this result could be a crucial issue for the emergence and the development of the subsequent cash crop. For the end of the experiment, we expected to measure differences between the three cover crops treatments and the bare soil in order to quantify the effect of cover crop and its mode of destruction on the soil water profile.