Strategy to reduce subjectivity in landslide susceptibility zonation by GIS in complex mountainous environments

In the last decade, statistical modelling of landslide susceptibility by GIS has become a major topic of research. Despite some advancements, particularly through the bayesian approach based on the concept of "weight of evidence", some unsolved questions may be asked: (i) what is the influence of the input data set on the quality of the simulations and (ii) how can we resolve the effect of violating the conditional independence assumption. This paper presents a strategy to collect relevant geographical information for reducing subjectivity in landslide susceptibility zonation in complex mountainous environments and stresses the different ways to reduce information redundancy.