Detecting fingerprints of landslide drivers: A MaxEnt model

Abstract

[1] Landslides are important geomorphic events that sculpt river basins by eroding hillslopes and providing sediments to coastal areas. However, landslides are also hazardous events for socio-ecological systems in river basins causing enormous biodiversity, economic, and social impacts. We propose a probabilistic spatially explicit model for the prediction of landslide patterns based on a maximum entropy principle model (MAXENT). The model inputs are the centers of mass of historical landslides and environmental variables at the basin scale. The model has only three parameters requiring calibration: the threshold for the network extraction, the trade-off factor between model complexity and accuracy, and the threshold of landslide susceptibility. The calibration on a subset of observations detects the environmental drivers and their relative importance for landslides. We employ the model in the Arno basin, Italy, selected because of its widespread landslide dynamics and the large availability of landslide observations. The model reproduces the size distribution and location of over 27,500 historical landslides for the Arno basin with an accuracy of 86% obtained from the variable-landslide inference on about 37% of observed landslides. Future landslide patterns are predicted for 17 A1B and A2 rainfall scenarios and for a multimodel ensemble from 2000 to 2100. We show that potential landslide hazard is strongly correlated with variation in the 12 and 48 h rainfall with a return time of 10 years. As the climate gets wetter, the average probability of landslides gets higher which is shown by the landslide size distribution. Hence, the landslide size distribution is a fingerprint of the geomorphic effectiveness of rainfall as a function of climate change. MAXENT is proposed as a parsimonious model for the prediction of landslide patterns with respect to more complex models. The need for very accurately sampled and delineated landslides is lower than for other prediction models. Moreover, the model informs about the drivers of landslides and their relative importance without assumptions on the main triggering factors. This is important to inform monitoring of environmental variables. Our modeling approach can enhance the planning of socio-ecological systems in river basins by improving the accuracy of landslide prediction in space and time.

Exploiting Maximum Entropy method and ASTER data for assessing debris flow and debris slide susceptibility for the Giampilieri catchment (north‐eastern Sicily, Italy)

“…Among these, Philips et al (2004Philips et al ( , 2006 presented an integrated approach of maximum entropy and GIS technologies for species distribution modelling where the application of this algorithm maximises the entropy in a geographic space. In the present contribution, we exploit the MaxEnt approach to predict the spatial distribution of landslides, with a similar assumption to that described in Convertino et al (2013). Merow et al (2013) illustrate the MaxEnt model architecture as requiring presence only (PO) data and a set of predictors distributed across a regularly gridded space.…”

“…Common practises involve the use of stochastic and/or data mining methods relying on presence/absence techniques (e.g., Eker et al, 2014;Ermini et al, 2005;Pourghasemi et al, 2013;Lombardo et al, 2014) for calibrating the predictive model. In this research we decided to pursue a presence-only approach, which has recently been introduced within the landslide scientific community (Convertino et al, 2013;Davis et Sims, 2013;Park, 2014) by applying the Maximum Entropy (MaxEnt) algorithm (Elith et al, 2011;Phillips et al, 2004 and2006;Phillips and Dudík, 2008), which does not rely on the contribution of negative (no-landslide or absence) cases for calibration.…”

A Hybrid Physical and Maximum-Entropy Landslide Susceptibility Model

“…Each source has characteristic artifacts--LiDAR noise and stepped contour interpolation effects--that were mitigated using 3 × 3 low-pass filters: one for slope and two in succession for curvature. Spatial variation in precipitation intensity as was used in Convertino et al [26], was not considered for our study due to the relatively small size of our study area with very few rain gauges to derive a suitable spatial input. Categorical variables included vegetation, geology, and Boolean 50-m trail and stream buffers.…”

“…A maximum entropy modeling approach was used by Convertino et al [26] for the 9130 km 2 Arno River basin in the Tuscany region of Italy. Felicísimo [39] compared logistic regression, the maximum entropy (MaxEnt) application of Phillips et al [40], multiple adaptive regression splines (MARS), and classification and regression trees (CART) for modeling landslide susceptibility in a region of northern Spain; CART and MaxEnt performed best based upon area under the receiver operator characteristic curve (AUC).…”

“…Maximum entropy models make no statistical assumptions about the variables used as inputs, and as a Bayesian approach focuses on maximizing probabilities, in this case that observations are similar based upon inputs in terms of maximizing entropy in information space which may include environmental space [25]. In the process, the model is parsimonious, with variables incorporated on the basis of their being necessary and sufficient in maximizing prediction accuracy [26].…”