Patagonian Andes Landslides Inventory: The Deep Learning’s Way to Their Automatic Detection

Abstract

Landslide inventories are crucial to studying the dynamics, associated risks, and effects of these geomorphological processes on the evolution of mountainous landscapes. The production of landslide maps is mainly based on manual visual interpretation methods of aerial and satellite images combined with field surveys. In recent times, advances in machine learning methods have made it possible to explore new semi-automated landslide detection methodologies using remotely detected images. In this sense, developing new artificial intelligence models based on Deep Learning (DL) opens up an excellent opportunity to automate this arduous process. Although the Andes mountain range is one of the most geomorphologically active areas on the planet, the few investigations that use DL mainly focus on mountain ranges in Europe and Asia. One of the main reasons is the low density of landslide data available in the Andean areas, making it difficult to experiment with DL models requiring large data volumes. In this work, we seek to narrow the existing gap in the availability of landslide inventories in the area of the Patagonian Andes. In addition, the feasibility and efficiency of DL techniques are studied to develop landslide detection models in the Andes from the generated datasets. To achieve this goal, we generated in a manual process a datasets of 10,000 landslides for northern Chilean Patagonia (42–45°S), being the largest freely accessible landslide datasets in this region. We implement a machine learning model, through DL, to detect landslides in optical images of the Sentinel-2 constellation using a model based on the DeepLabv3+ architecture, a state-of-the-art deep learning network for semantic segmentation. Our results indicate that the algorithm detects landslides with an accuracy of 0.75 at the object level. For its part, the segmentation reaches a precision of 0.86, a recall of 0.74, and an F1-score of 0.79. The correlation of the segmentation measured through the Matthews correlation coefficient shows a value of 0.59, and the geometric similarity of the correctly detected landslides measured through the Jaccard score reaches 0.70. Although the model shows a good response in the testing area, errors are generated that can be explained by geometric and spectral relationships, which should be solved through new training approaches and data sets.

The Yinshan Mountains Record over 10,000 Landslides

“…Liesbet Jacobs et al [38] created the first landslide inventory based on archival information in the Rwenzori Mountains, detailing 48 landslide events and providing an important reference for landslide research in the region. Bastian Morales et al [39] created a landslide dataset for the Patagonian Andes (42-45 • S), containing 10,000 landslides. He pointed out that the lack of geohazard research in the region is mainly due to the low density of available landslide data, which makes it difficult to perform deep learning studies.…”

Developing a more complete understanding of tropical montane forest disturbance ecology through landslide research

“…However, high-resolution satellite data paired with computer vision for object and change detection (Zhong et al, 2020;Amatya et al, 2021;Lu et al, 2022) unlocks the ability to identify past landslide activity and monitor landslides in near real-time at regional and even continental scales (Yang et al, 2022). For example, this approach has recently been used to detect and map landslides in Nepal (Prakash et al, 2021;Meena et al, 2022), Taiwan, China, Japan (Ghorbanzadeh et al, 2021(Ghorbanzadeh et al, , 2022, and the Patagonian Andes (Morales et al, 2022). In the Andes, Morales et al (2022) applied a convolutional neural network to Sentinel-2 images to develop a 10,000-landslide inventory covering approximately 20,000 km 2 .…”

“…For example, this approach has recently been used to detect and map landslides in Nepal (Prakash et al, 2021;Meena et al, 2022), Taiwan, China, Japan (Ghorbanzadeh et al, 2021(Ghorbanzadeh et al, , 2022, and the Patagonian Andes (Morales et al, 2022). In the Andes, Morales et al (2022) applied a convolutional neural network to Sentinel-2 images to develop a 10,000-landslide inventory covering approximately 20,000 km 2 . Their model, the first automated landslide detection model in the region, was most accurate in areas with vegetation cover (Morales et al, 2022), suggesting this approach will work well across forested regions of the Andes.…”

A research on the relationship between landslide area changes and environmental factors in the southern Tibetan plateau

“…With the robust development of machine learning, researchers have been able to explore different landslide classification or zoning methods using algorithms such as logistic regression, support vector machines, decision trees, and random forests (Mohan et al, 2021). In recent years, numerous studies have utilized high-resolution imagery and deep learning models based on Convolutional Neural Networks (CNNs) to extract landslides, achieving promising results (Ghorbanzadeh et al, 2019;Sameen and Pradhan, 2019;Shinde et al, 2019;Morales et al, 2022). Although the aforementioned optical remote sensing image analyses for landslide extraction are typically limited to smaller spatial scales, they have provided valuable insights into the distribution and characteristics of landslides.…”