Evaluation and verification of the WetSpa model based on selected rural catchments in Poland The paper presents results of calibration and verification of the WetSpa model, which enables the modelling of rainfall-runoff process based on mass and energy balance in the soil-plantatmosphere system in the catchment. It is a model with distributed parameters, using the structure of raster GIS model to determine the spatial diversity of the catchment environment. This enables simulation of runoff from the catchment, including: precipitation, evapotranspiration, interception of plant surface and soil cover, infiltration and capillary rise in soil and groundwater runoff. Simulated processes depend on the required non-distributed parameters, which were calibrated based on hydrometeorological data from the three rural catchments with different physical and geographical characteristics: Mławka, located in the Wkra basin in Central Poland and the rivers Kamienna and Sidra, which are tributaries of the upper Biebrza in north-eastern part of the country. Distributed catchment parameters were specified on the basis of digital soil maps, land use maps and digital elevation model using GIS techniques. Non-distributed model parameters were calibrated for the three catchments using automatic techniques based on the PEST algorithm. The obtained values of these parameters were scrutinized in order to analyse differences resulting from various characteristics of the study areas. The quality of the model was verified upon dependent and independent data. Appropriate quality measures, including Nash-Sutcliffe efficiency measure, were used to assess model quality. For two catchments (the Sidra and Kamienna) the model showed a satisfactory quality for modelling high flows, it was, however, not satisfactory for low flows. The values for the Mławka catchment justified the assessment of the model quality measurements as very good and good. The factors most affecting the process of river outflow formation were determined using the analysis of model sensitivity to relative changes in parameter values. It was found that the evaluation of the model quality depended largely on the quality of meteorological data and proper parameterization of the soil cover.