Why and how to use a probabilistic software to assess rockfall hazard?
RocPro3D
January 26th, 2024
Rockfall hazard
To assess rockfall hazard, it is necessary to have a DTM, the distribution of soils and their properties, and the position, shape and mass of blocks likely to fall. However, field observations generally show a degree of natural variability of these parameters, and it’s generally difficult to know the exact variability of these parameters at the site scale.
Two paths can be considered: a deterministic approach or a probabilistic approach.
Given all the variabilities involved, a probabilistic approach is preferred to a deterministic approach, the later requiring perfect knowledge of the values of each parameter (soils mapping and properties, initial conditions of start, blocks properties).
Probabilistic approach
Probabilistic or stochastic (also known as Monte-Carlo, in analogy to casino gambling) simulation software is used to account for this variability by incorporating uncertainty in each parameter.
An analogy may be drawn from the Galton’s statistical experiment, which illustrates the convergence of a binomial distribution to a normal distribution as the number of released balls increases.
In a probabilistic approach, it is therefore essential to carry out a large number of trials to explore the effect of the variability in the probabilistic parameters under consideration on the result.
Probabilistic approach in rockfall simulation
Such an approach is applicable to rockfall hazard assessment provided, among other things, that the probabilistic parameters considered vary within a realistic range and that a large number of tests are performed to cover the whole range of possibilities.
RocPro3D
In RocPro3D, a trial is equivalent to a launched block, corresponding to the release of a ball in the Galton’s board experiment.
It is therefore necessary to throw a rather large number of blocks in order to reach statistical convergence and to allow evaluating the rockfall hazard or dimensioning a protection.
In particular, RocPro3D takes into account the variability of:
- the rockfall conditions (position, velocity, fall height), their size and shape,
- the topography described by the DTM,
- the parameters associated with the physical models of impact (restitution coefficients) and frictional rolling (dynamic Coulomb friction) of the blocks on the DTM.
Would you like to try a stochastic rockfall simulation on one of your cases?
Contact us for a trial version or a demo session
