Geochemical insights into the internal dynamics of debris avalanches. A case study: The Socompa avalanche, Chile

Authors

  • Régis Doucelance,

    Corresponding author
    1. Laboratoire Magmas et Volcans, Université Blaise Pascal, Clermont-Ferrand, France
    2. CNRS, UMR 6524, LMV, Clermont-Ferrand, France
    3. IRD, R 163, LMV, Clermont-Ferrand, France
    Search for more papers by this author
  • Karim Kelfoun,

    1. Laboratoire Magmas et Volcans, Université Blaise Pascal, Clermont-Ferrand, France
    2. CNRS, UMR 6524, LMV, Clermont-Ferrand, France
    3. IRD, R 163, LMV, Clermont-Ferrand, France
    Search for more papers by this author
  • Philippe Labazuy,

    1. Laboratoire Magmas et Volcans, Université Blaise Pascal, Clermont-Ferrand, France
    2. CNRS, UMR 6524, LMV, Clermont-Ferrand, France
    3. IRD, R 163, LMV, Clermont-Ferrand, France
    Search for more papers by this author
  • Chantal Bosq

    1. Laboratoire Magmas et Volcans, Université Blaise Pascal, Clermont-Ferrand, France
    2. CNRS, UMR 6524, LMV, Clermont-Ferrand, France
    3. IRD, R 163, LMV, Clermont-Ferrand, France
    Search for more papers by this author

Abstract

One way to infer the internal dynamics of debris avalanches is to characterize the heterogeneity of their deposits. Here we present high-precision Sr-Nd isotope compositions, plus major and trace element concentrations, of matrix samples and rock fragments from the Socompa debris-avalanche deposit (Chile). The Socompa blocks are easily identifiable in the field, but distinguishing substrate debris from disaggregated material formed at the volcano is difficult to do with only field criteria. Combining isotope data with field observations can help with this. Measured Sr and Nd isotope ratios show significant variations, defining a binary mixing array where matrix and rock deposits overlap. This testifies to the mixing of crushed rocks during collapse and/or movement. Assimilation of Socompa basement appears to be variable; overall, it is far lower than was previously proposed. Comparison between matrix and block samples in contact, over the whole surface area of the deposit, shows that the isotopic heterogeneity increases from source to front. Close to the Socompa, matrices are resulting from simple crushing of adjacent rocks. At the front, rock samples with distinct compositions are found in a close relationship with matrices that result either from mixing of these (or some of these) rocks or from crushing of basement material. Between the source and the front, the efficient mixing of Socompa rocks (and basement rocks) generates matrices with isotopic compositions distinct from those of the blocks they are in contact with. We interpret these results as being due to more efficient vertical mixing during the avalanche emplacement.