The evaluation of avalanche release depth distributions represents a major challenge for hazard management in mountaineous regions. This depth constitutes an important input ingredient of hazard mapping procedures. This PhD thesis presents a rigorous formalism in which these distributions are expressed through a coupling of mechanical and meteorological factors. The stability criterion of a layered snowpack is investigated using a finite-element analysis accounting for the spatial heterogeneity of weak-layer mechanical properties. Considering that an avalanche can occur only if the snowfall depth exceeds a critical value corresponding to a stability criterion, release depth distributions obtained from the mechanical model are coupled with the distribution of 3-day extreme snowfalls. We show that this coupled model is able to reproduce field data from 369 natural slab avalanches in La Plagne (France). Not only the power-law tail of the distribution, corresponding to large slab depths, but also the core of the distribution for shallow slab depths, are well represented. Small to medium-sized avalanches appear to be controlled mainly by mechanics, whereas large avalanches and the associated power-law exponent, are influenced by a strong mechanical-meteorological coupling. Finally, we demonstrate that the obtained distribution is strongly space dependent, and, using max-stables processes allowing a rigorous spatial interpolation, our coupled model is used to obtain release depth maps for given return periods in the whole French Alps.
