Cover crops reduce drainage but not always soil water content due to interactions between rainfall distribution and management

•We analysed the impact of different cover crop management on water balance.•Our approach combined two-field experiments with crop-model simulations.•The reduction of drainage was mainly dependant on cover crop sowing date.•Cover crops could reduce soil water content for the next cash crop in dry conditions.•The effect depends on the rainfall distribution and cover crop management interaction. Cover crops are a potential component of agroecological cropping systems, since they may render crop rotations more sustainable. They simultaneously provide multiple ecosystem services, such as decreasing nitrate leaching, decreasing erosion, and increasing soil organic matter. However, cover crops increase evapotranspiration and reduce drainage, which results in a potential disservice for groundwater recharge. Little attention has focused on management of cover crop residues after destruction or their influence on water flux dynamics, particularly in dry and temperate climates. The objective of our study was to analyze and quantify the impact of cover crop management on soil water content and water flux dynamics to understand the main mechanisms of system functioning. We combined a two-year field experiment with crop-model simulations. We performed the field experiment in southwestern France that compared three cover crop treatments, with bare soil as the control. The treatments included (1) living cover crops lasting ca. 9 months from August-April, (2) crushing cover crops in November and leaving them as mulch on the soil, and (3) plowing up cover crops in November to promote residue decomposition and the green manure effect. The STICS soil-crop model was used to predict water fluxes that were not measured and to perform a 20-year independent simulation study based on recent climate series for the experimental site. Our main results indicated that cover crops (1) always reduce water drainage by 20-60 mm compared to that under bare soil; and (2) could significantly reduce soil water content (0-120 cm deep) for the next cash crop by a mean of 20-50 mm, and up to 80 mm in dry spring conditions, but early destruction could decrease this negative impact. The simulations clearly showed that the interaction between climate variability, i.e., rainfall distribution during the fallow period, and cover crop management should be considered to explain the impact of inter-annual variability on the water balance. Thus, destroying cover crops mechanically in late autumn and retain