The stability of probabilistic methods of evaluating expected strong ground motion levels is examined as a function of probability level. A case history for Vancouver, British Columbia, is used to show that when input parameters are screened on the basis of compatibility with low probability calculations, robust (±25 per cent) results are obtained for probabilities in the range of 10−2 to 10−4 per annum. There is no inherent loss of stability with decreasing probability.
Probabilistic approaches are not in conflict with deterministic approaches, since deterministic data can be incorporated into the analysis. The usefulness of the probabilistic approach lies in providing a framework for evaluating uncertainties.