A comparison of hydrological models with different level of complexity in Alpine regions in the context of climate change

Abstract

Abstract. This study compares the ability of two degree-day models (Poli-Hydro and a hybrid degree-day implementation of Alpine3D) and one full energy-balance melt model (Alpine3D) to predict the discharge on two partly glacierized Alpine catchments of different size and intensity of exploitation, under present conditions and climate change as projected at the end of the century. For the present climate, the magnitude of snowmelt predicted by Poli-Hydro is sensibly lower than the one predicted by the other melt schemes, and the melting season is delayed by 1 month. This difference can be explained by the combined effect of the reduced complexity of the melting scheme and the reduced computational temporal resolution. The degree-day implementation of Alpine3D reproduces a melt season closer to the one obtained with its full solver; in fact, the onset of the degree-day mode still depends upon the full energy-balance solver, thus not bringing any particular benefit in terms of inputs and computational load, unlike with Poli-Hydro. Under climate change conditions, Alpine3D is more sensitive than Poli-Hydro, reproducing discharge curves and volumes shifted by 1 month earlier as a consequence of the earlier onset of snowmelt. Despite their benefits, the coarser temporal computational resolution and the fixed monthly degree days of simpler melt models like Poli-Hydro make them controversial to use for climate change applications with respect to energy-balance ones. Nevertheless, under strong river regulation, the influence of calibration might even overshadow the benefits of a full energy-balance scheme.

Snow sensitivity to temperature and precipitation change during compound cold–hot and wet–dry seasons in the Pyrenees

“…Our estimates of snow sensitivity to temperature and precipitation change were based on the delta approach, which considers changes in temperature and precipitation based on climate projections for the Pyrenees (Amblar-Francés et al, 2020) but assumes that the meteorological patterns of the reference period will be constant over time. In this work we used a physical-based snow model, since it provides better results for future snow climate change estimations than degree-day models (Carletti et al, 2022). FSM2 is a physicsbased model of intermediate complexity, and the estimates of snow densification are simpler than those from more complex models of snowpack.…”

Turbulence in The Strongly Heterogeneous Near-Surface Boundary Layer over Patchy Snow

“…The Dischma valley is an alpine valley with its valley axis in southeast-northwest direction. It has been comprehensively investigated over the course of multiple observational and modelling studies (Lehning et al 2006;Bavay et al 2009;Brauchli et al 2017;Wever et al 2017;Schlögl et al 2018a,b;Gerber et al 2019;Carletti et al 2022;Reynolds et al 2023) The location of an automatic weather station (AWS) including an eddycovariance (EC) sensor (81000 Ultrasonic Anemometer, R.M. Young Company, Traverse City, Michigan, USA) at 5 m above the surface and two movable EC measurement towers (T1 and T2) is also marked on the orthophoto.…”

Operational snow-hydrological modeling for Switzerland

“…DeBeer and Pomeroy (2017) contend that temperature-index model approaches may yield acceptable results in certain topographic settings and climatic conditions, but the spatial variability in the snowpack energy balance must be considered, especially in cold regions, windy conditions, and increasingly complex terrain. Therefore, distributed energy-and mass-balance snow cover models are increasingly being applied in snow hydrological simulations (Bavay et al, 2013;Warscher et al, 2013;Gallice et al, 2016;Painter et al, 2016;Brauchli et al, 2017;Magnusson et al, 2017;Hedrick et al, 2018;Shakoor et al, 2018;Carletti et al, 2022). In particular, the consideration of temporally and spatially varying surface energy fluxes is crucial for snow-hydrological simulations in complex terrain, extreme weather events, and glacierized catchments, where the presence of local wind systems enhances the importance of turbulent heat exchange for the energy balance of snow and ice surfaces (Shea and Moore, 2010;Sauter and Galos, 2016;Freudiger et al, 2017;Mott et al, 2020).…”

“…To overcome this limitation, downscaling techniques are needed to bridge the scale gap between forcing data and the spatial resolution at which snow models are run (Kruyt et al, 2022). The choice of downscaling methods for different meteorological variables is especially important in modeling mountain snow cover dynamics, as several studies have shown that resolving terrain effects on wind, radiation, and precipitation is crucial in simulating seasonal snow cover dynamics on the mountain slope scale (Schneiderbauer and Prokop, 2011;Musselman et al, 2015;Vionnet et al, 2017;Mott and Lehning, 2010;Reynolds et al, 2020;Carletti et al, 2022). Therefore, downscaling techniques need to be carefully chosen to accurately represent the impact of topography on snow accumulation and ablation, especially in complex mountainous regions.…”