Ensemble of global landslide susceptibility

This ensemble of global landslide susceptibility maps at a 36-km spatial resolution on the EASEv2 grid has been created through a combination of blocked random cross validation and predictor variable perturbation. In total, the ensemble has 2500 members, each obtained from a logistic regression equation with 5 predictor variables. Note that these predictor variables vary between ensemble members, as do the logistic regression parameters (study design of data-driven predictor variable selection with cross validation). For details, please see Felsberg, A., Poesen, J., Bechtold, M., Vanmaercke, M., and De Lannoy, G. J. M. (2022): Estimating global landslide susceptibility and its uncertainty through ensemble modelling, Natural Hazards and Earth System Sciences.

Provided are the individual ensemble member maps (FelsbergEtAl2022_LSS_ensMemb_all.nc4c), as well as the ensemble average and standard deviation maps (FelsbergEtAl2022_LSS_ensavg.nc4c, FelsbergEtAl2022_LSS_ensstd.nc4c). More details can be found in the README.