New process descriptions

The study by Bronstert et al. (2023) focuses on description of infiltration excess (Hortonian surface runoff) in catchment models. In particular, they investigate the importance of micro- and macropores in the infiltration process. The study is based on good datasets from well instrumented infiltration/infiltration-excess experiments and observations at three spatial scales: point, field and catchment (115 km2). Two modelling hypotheses were then tested against these field data, namely an approach without macropores, which is traditionally used in catchment models, and an approach based on double-porosity soil enabling a combined modelling of high infiltration rates in macropores and dampened soil moisture distribution after termination of infiltration. The results from tests at point and field scale suggest that both modelling approaches are capable of reproducing soil moisture dynamics, but that the inclusion of macropores results in more realistic soil hydraulic parameters. The results from catchment scale show that the macropore based approach is more robust in reproducing flood hydrographs for different rainfall intensities and generally outperforms the modelling approach without macropores. Altogether Bronstert et al. (2023) conclude that macropores are of high relevance for infiltration and soil moisture dynamics during periods of high intensity rainfall and therefore should be considered in catchment modelling focusing on simulation of flood events.
While it is well known that stream discharge in vegetated catchments during dry periods can exhibit natural fluctuations of up to 10% daily, the capability of catchment models to reproduce such behavior and explain the underlying processes has so far rarely been tested. Le Cecilia et al. (2022) use the CATHY physically-based integrated surface-subsurface hydrological model to study the complex processes generating diel streamflow fluctuations in a 2.67 km2agricultural catchment in Switzerland. After demonstrating that the model is capable of satisfactorily simulating the diel streamflow fluctuations, including the timing of short-living streamflow peaks attributed to irrigation, the model was subsequently used to test alternative hypotheses for which processes may contribute to these fluctuations. The results show that evapotranspiration is the dominant process generating diel fluctuations, while changes in saturated hydraulic conductivity due to diel soil temperature fluctuations caused an amplitude of the diel streamflow signal 10 times smaller than evapotranspiration.
The study by Riazzi et al. (2022) uses a travel time tracking method to simulate stream electrical conductivity (EC) using high frequency (hourly) monitoring data from the 369 km2 Duck River catchment in Tasmania, Australia. Two modelling approaches are tested. The first approach assumes that evapotranspiration is the only process driving the changes in EC, while the second assumes that the water salinity in catchment storages is a function of water age in these storages. The results show that the two hypotheses are equally successful in simulating EC concentrations and tracking its event and seasonal dynamics, and hence it is not possible to differentiate which of the two underlying hypothesis are better supported by the available observational data. Given that EC data from operational observation networks is much more widely available than other tracer data, Riazzi et al. (2022) conclude that using EC data to calibrate travel time models is a promising approach.
In contrary to other papers studying individual hydrological processes or the impact of process descriptions on discharge or solute fluxes at the catchment output, Gaur et al. (2022) evaluate how well a spatially distributed hydrological model is able to reproduce observed spatial patterns within the catchment. They use the MIKE SHE with 5 km x 5 km resolution to simulate the hydrological response of the 19,276 km2 Subarnarekha catchment in Eastern India. The study compares model simulations with remote sensing derived patterns for evapotranspiration and soil moisture. The comparison is made using three spatial performance metrics, i.e. joint empirical orthogonal functions (EOF), fractional skill scores (FSS) and spatial efficiency (SPAEF). The results demonstrate the potential and value of hydrological model calibration to observed spatial patterns across time.