Upward saline groundwater seepage is leading to surface water salinization of deep lying polders in the Netherlands. Identifying measures to reduce the salt content requires a thorough understanding and quantification of the dominant sources of water and salt on a daily basis. However, as in most balance studies, there are large uncertainties in the contribution from groundwater seepage. Taking these into account, we applied a probabilistic (GLUE) end-member mixing approach to simulate two years of daily to weekly observations of discharge, salt loads and salt concentration of water pumped out of an artificially drained polder catchment area. We were then able to assess the contribution from the different sources to the water and salt balance of the polder and uncertainties in their quantification. Our modelling approach demonstrates the need to distinguish preferential from diffuse seepage. Preferential seepage via boils contributes, on average, 66 % to the total salt load and only about 15 % to the total water flux into the polder and therefore forms the main salinization pathway. With the model we were able to calculate the effect of future changes on surface water salinity and to assess the uncertainty in our predictions. Furthermore, we analyzed the parameter sensitivity and uncertainty to determine for which parameter the quality of field measurements should be improved to reduce model input and output uncertainty. High frequency measurements of polder water discharge and weighted concentration at the outlet of the catchment area appear to be essential for obtaining reliable simulations of water and salt fluxes and for allotting these to the different sources.