Journal Article

Statistically-based regional landslide susceptibility assessment in the UNESCO global geopark Caminhos dos Cânions do Sul (Brazil)

Martins, Tiago Damas, Bonini, José Eduardo, da Silva Pereira, Susana, Sugiyama, Marina Tamaki de Oliveira, Gomes, Maria Carolina Villaça, Vieira, Bianca Carvalho

2025

Overview

The scenic value of landscapes within Geoparks is often attributed to the geomorphological processes that have shaped them in the past or up to the present day, including landslides. However, these processes also pose significant threats to the integrity of geosites and the safety of visitors, highlighting the need for risk prevention and mitigation plans for geohazards. The Caminhos dos Cânions do Sul Geopark (southern Brazil) lacks landslide inventories and susceptibility maps, essential for conducting practical geohazard risk analyses. This study addresses this gap by compiling a landslide inventory of the major events over the past 30 years, using a rule-based Object-Based Image Analysis (OBIA) approach, and assessing the susceptibility for four modeling domains within the Geopark using the Information Value method. Seven independent variables (aspect, slope, topographic wetness index, terrain ruggedness index, geomorphons, elevation, and curvature) were selected, resulting in 120 combinations for each modeling domain. For each predisposing factor combination, model performance was assessed using the area under the Receiver Operating Characteristics (ROC) curve, and the conditional independence of variables was evaluated. The best models for each domain were selected based on the criteria of conditional independence, goodness of fit, and number of variables. The final landslide susceptibility map was produced by merging the best three models’ results. The resulting susceptibility classification indicates that many geosites are located in areas with moderate to very high susceptibility or within zones likely to experience material transport or deposition.

• Publisher: Springer Nature
• Subject: Global Geoparks Network / Information value method / Serra Geral Escarpment