International Snow Science Workshop Proceedings - Montana State University Library

Item: Assimilation of modis observations of snowpack surface properties into one year of spatialized ensemble snowpack simulations at a field site in the French Alps

• • Title: Assimilation of modis observations of snowpack surface properties into one year of spatialized ensemble snowpack simulations at a field site in the French Alps

    Proceedings: International Snow Science Workshop Proceedings 2018, Innsbruck, Austria

    Authors:

    • Bertrand Cluzet  [ Univ. Grenoble Alpes, Universit e´ de Toulouse, Me´ te´o-France, CNRS, Centre d’E´ tudes de la Neige, Grenoble, France ]
    • Jesus Revuelto  [ Univ. Grenoble Alpes, Universit e´ de Toulouse, Me´ te´o-France, CNRS, Centre d’E´ tudes de la Neige, Grenoble, France ]
    • Matthieu Lafaysse  [ Univ. Grenoble Alpes, Universit e´ de Toulouse, Me´ te´o-France, CNRS, Centre d’E´ tudes de la Neige, Grenoble, France ]
    • Marie Dumont  [ Univ. Grenoble Alpes, Universit e´ de Toulouse, Me´ te´o-France, CNRS, Centre d’E´ tudes de la Neige, Grenoble, France ]
    • Emmanuel Cosme  [ UGA-CNRS-IRD, IGE, Grenoble, France ]
    • Francois Tuzet  [ Univ. Grenoble Alpes, Universit e´ de Toulouse, Me´ te´o-France, CNRS, Centre d’E´ tudes de la Neige, Grenoble, France ]

    Date: 2018-10-07

    Abstract: Detailed snowpack modelling is crucial for avalanche hazard forecasting, glaciological modelling and hydrological studies, but its use is currently limited by its level of uncertainty. Ensemble forecasting approaches are commonly used to quantify the associated uncertainties. Combined with satellite data assimilation, they can help reduce the modelling errors. In this study, an ensemble simulation chain accounting for both meteorological and modelling uncertainties was used to simulate snowpack conditions in the mountain range ("massif") of Grandes-Rousses covering an area of about 500km2 for various elevations, aspects and slopes during the 2013-2014 winter. This modelling chain involves perturbed meteorological forcings from ARPEGE-SAFRAN analysis system and multi-physics ensemble version of snowpack model Crocus called Ensemble System Crocus (ESCROC). In addition, visible and near infrared satellite data from MODIS sensor were retrieved in the same area and study period. Such data convey precious information on the snowpack surface impurities content, snow microstructure properties and snowpack extent. A comparison with ensemble outputs is presented to assess their potential for data assimilation with a particle filter. Results show that there is a high potential of assimilation of MODIS observations into ensemble semi-distributed simulations of snowpack if transformed variables are used to tackle observation systematic biases. This could lead to a significant improvement in snowpack modelling accuracy at the massif scale in a near future.

    Object ID: ISSW2018_O04.10.pdf

    Language of Article: English

    Presenter(s):

    Keywords: snow modelling, ensemble, remote sensing, MODIS, data assimilation, particle filter.

    Page Number(s): 338-343

  • Get Full Article

Back to Home page