Hydrologic models partition flows into surface and subsurface pathways, but their calibration is typically conducted only against streamflow. Here we argue that unless model outcomes are constrained using flow pathway data, multiple partitioning schemes can lead to the same streamflow. This point becomes critical for biogeochemical modeling as individual flow paths may yield unique chemical signatures. We show how information on flow pathways can be used to constrain hydrologic flow partitioning and how improved partitioning can lead to better water quality predictions. As a case study, an agricultural basin in Ontario is used to demonstrate that using tile discharge data could increase the performance of both the hydrology and the nitrogen transport models. Watershed‐scale tile discharge was estimated based on sparse tile data collected at some tiles using a novel regression‐based approach. Through a series of calibration experiments, we show that utilizing tile flow signatures as calibration criteria improves model performance in the prediction of nitrate loads in both the calibration and validation periods. Predictability of nitrate loads is improved even with no tile flow data and by model calibration only against an approximate understanding of annual tile flow percent. However, despite high values of goodness‐of‐fit metrics in this case, temporal dynamics of predictions are inconsistent with reality. For instance, the model predicts significant tile discharge in summer with no tile flow occurrence in the field. Hence, the proposed tile flow upscaling approach and the partitioning‐constrained model calibration are vital steps toward improving the predictability of biogeochemical models in tiled landscapes.

In this study, we assess the impact of forcing data errors, model structure, and parameter choices on 1‐D snow simulations simultaneously within a global variance‐based sensitivity analysis framework. This approach allows inclusion of interaction effects, drawing a more representative picture of the resulting sensitivities. We utilize all combinations of a multiphysics snowpack model to mirror the influence of model structure. Uncertainty ranges of model parameters and input data are extracted from the literature. We evaluate a suite of 230,000 model realizations at the snow monitoring station Kühtai (Tyrol, Austria, 1,920 m above sea level) against snow water equivalent observations. The results show throughout the course of 25 winter seasons (1991–2015) and different model performance criteria a large influence of forcing data uncertainty and its interactions on the model performance. Mean interannual total sensitivity indices are in the general order of parameter choice < model structure < forcing error, with precipitation, air temperature, and the radiative forcings controlling the variance during the accumulation period and air temperature and longwave irradiance controlling the variance during the ablation period, respectively. Model skill is highly sensitive to the snowpack liquid water transport scheme throughout the whole winter period and to albedo representation during the ablation period. We found a sufficiently long evaluation period (>10 years) is required for robust averaging. A considerable interaction effect was revealed, indicating that an improvement in the knowledge (i.e., reduction of uncertainty) of one factor alone might not necessarily improve model results.