Evaluation of SWAT models performance to simulate streamflow spatial origin. The case of a small forested watershed

•Enhanced SWAT hydrological simulation in a very small forested watershed.•Field data, uncommonly used in model calibration, included in the process.•Satisfactory simulation of outlet-discharge and sub-watersheds contribution achieved.•Evaluation of SWAT's capacity to accurately simulate surface/base flow contribution.•The model-field combined approach allowed detecting spatial-temporal uncertainties.

SummaryThe Soil and Water Assessment Tool (SWAT) model has been applied widely in many types of environment with different goals. The purpose of this paper is to assess the ability of SWAT to simulate hydrological processes in the Aixola watershed. Electrical conductivity (EC) was used to estimate water contribution from the two main sub-watersheds. Streamflow contribution from the sub-watersheds varies throughout the year; the larger of the two contributes greater flow in wetter seasons, while the smaller one has more regulation capacity and contributes more in summer. The data obtained from EC were used to calibrate the model, simulating this variability satisfactorily and even more-so when the model was forced during calibration. Additionally, EC measured at the outlet of the watershed was used to make a decomposition of the hydrograph (surface runoff - base flow), comparing the data obtained with those simulated by SWAT. The results showed that the model performed well and identified the source of uncertainties in modelling this watershed. When additional data is included in the calibration, this made it possible to obtain a more realistic hydrological simulation of the Aixola watershed.