Propagating information from snow observations with CrocO ensemble data assimilation system: a 10-years case study over a snow depth observation network

Abstract

Abstract. The mountainous snow cover is highly variable at all temporal and spatial scales. Snowpack models only imperfectly represent this variability, because of uncertain meteorological inputs, physical parameterizations, and unresolved terrain features. In situ observations of the height of snow (HS), despite their limited representativeness, could help constrain intermediate and large-scale modeling errors by means of data assimilation. In this work, we assimilate HS observations from an in situ network of 295 stations covering the French Alps, Pyrenees, and Andorra, over the period 2009–2019. In view of assimilating such observations into a spatialized snow cover modeling framework, we investigate whether such observations can be used to correct neighboring snowpack simulations. We use CrocO, an ensemble data assimilation framework of snow cover modeling, based on a particle filter suited to the propagation of information from observed to unobserved areas. This ensemble system already benefits from meteorological observations, assimilated within SAFRAN analysis scheme. CrocO also proposes various localization strategies to assimilate snow observations. These approaches are evaluated in a leave-one-out setup against the operational deterministic model and its ensemble open-loop counterpart, both running without HS assimilation. Results show that an intermediate localization radius of 35–50 km yields a slightly lower root mean square error (RMSE), and a better spread–skill than the strategy of assimilating all the observations from a whole mountain range. Significant continuous ranked probability score (CRPS) improvements of about 13 % are obtained in the areas where the open-loop modeling errors are the largest, e.g., the Haute-Ariège, Andorra, and the extreme southern Alps. Over these areas, weather station observations are generally sparser, resulting in more uncertain meteorological analyses and, therefore, snow simulations. In situ HS observations thus show an interesting complementarity with meteorological observations to better constrain snow cover simulations over large areas.

Spatio-temporal information propagation using sparse observations in hyper-resolution ensemble-based snow data assimilation

“…The spatio-temporal DA techniques that we explored herein also have wider implications. An immediate possible operational application could be to integrate information obtained from the typically sparse national snow-monitoring networks into high-resolution distributed physically based snow simulations, building on the work of Magnusson et al (2014); Cluzet et al (2022), as an alternative to approaches based purely on statistical interpolation (Fassnacht et al, 2003;Collados-Lara et al, 2020). In a similar vein, as an extension of the work of Odry et al (2022), the spatio-temporal snow DA approach presented here allows for the fusion of sparse manual snow surveys into high-resolution distributed snowpack simulations.…”

“…A particularly promising potential application is the assimilation of snow depth acquisitions from the ICESat-2 laser altimeter (Enderlin et al, 2022;Deschamps-Berger et al, 2022), which records data along linear tracks that exhibit a discontinuous pattern in space. A straightforward extension of the experiments herein could involve joint DA using (nearly) spatially continuous satellite retrievals together with sparser retrievals or ground-based measurements.…”

“…If the objective is to reconstruct the SWE distribution in the absence of direct SWE observations, snow depth is likely the most useful variable since it explains most of the SWE variability in many regions (Sturm et al, 2010). Assimilating snow depth also enables one to benefit from the wealth of in situ snow depth measurements from automatic stations or manual sampling (Smyth et al, 2019) or from emerging satellite products (Marti et al, 2016;Lievens et al, 2022;Deschamps-Berger et al, 2022). Unfortunately, the assimilation of in situ or remotely sensed snow depth poses several challenges.…”

“…Cluzet et al (2021) suggested alternative promising approaches to spatiotemporal snow DA using a PF in synthetic experiments with a semi-distributed geometry by combining a form of domain localization with observation error inflation. Cluzet et al (2022) extended these approaches using data from a real snow depth observation network over the French Alps and Pyrenees, showing marked improvements compared to both the open-loop and the current operational approach of Météo-France.…”

Radar based high resolution ensemble precipitation analyses over the French Alps

“…Methods based on ensemble data assimilation algorithms have been developed to process these observations, (Magnusson et al, 2017;Cluzet et al, 2021). According to Cluzet et al (2022) and Deschamps-Berger et al (2022), these methods primarily use snow observations to compensate for errors in the precipitation forcing of the snow cover model. Quantifying the uncertainties in the precipitation fields is therefore essential to fully benefit from the assimilation of snow observations.…”