Abstract. Land cover and soil properties largely determine how climatic and hydrological regimes interact and produce hydrological stress in aquatic ecosystems. This study aims to clarify the influence of forests, as well as other majoritarian land cover types, on hydrological regime through an experimental design without the main limitations associated with traditional paired-watershed studies. With this aim, we use more catchments and an additional forest descriptor: forest maturity. We focus on flood and drought regimes, as they constitute the extremes of hydrological variability. Specific objectives were to isolate the relative contribution of precipitation and land cover composition to such flow extremes and to contrast the effectiveness of forests (surface and maturity) and other land cover types to predict them. The study was developed in a heterogeneous region located in the Cantabrian Mountains (NW Spain) with different vegetation types and a long history of human disturbance and land use change that allowed a robust experimental design. Regression and partial correlation analyses were developed using hydrological and meteorological data combined through hydrological modelling using IHACRES. Land cover characteristics showed ability to predict both flood regimes and low flows, although low flows were explained mainly by precipitation regimes. Forests showed a stabilization effect on flow regime (lower floods and greater base flows), but the effect was more evident with forest maturity than with surface. Other land cover types showed different effects. Evaluating the role of land cover on hydrological stability requires the use of comprehensive information involving different descriptors and their temporal changes, not only the current surface occupied by each land cover type.