Convolution neural network (CNN) is an effective and popular deep learning method which automatically learns complicated non-linear mapping from original inputs to given labels or ground truth ...through a series of convolutional layers. This study focuses on detecting landslides from high-resolution optical satellite images using CNN-based methods, providing opportunities for recognizing latent landslides and updating large-scale landslide inventory with high accuracy and time efficiency. Considering the variety of landslides and complicated backgrounds, attention mechanisms originated from the human visual system are developed for boosting the CNN to extract more distinctive feature representations of landslides from backgrounds. As deep learning needs a large number of labeled data to train a learning model, we manually prepared a landslide dataset which is located in the Bijie city, China. In the dataset, 770 landslides, including rock falls, rock slides, and a few debris slides, were interpreted by geologists from the satellite images and digital elevation model (DEM) data and further checked by fieldwork. The landslide data was separated into a training set that trains the attention boosted CNN model and a testing set that evaluates the performance of the model with a ratio of 2:1. The experimental results showed that the best F
1
-score of landslide detection reached 96.62%. The results also proved that the performance of our spatial-channel attention mechanism was fairly over other recent attention mechanisms. Additionally, the effectiveness of predicting new potential landslides with high efficiency based on our dataset is demonstrated.
The Ms 8.0 Wenchuan earthquake on May 12, 2008 triggered tens of thousands of landslides and produced large amounts of loose material. The loose material accumulated in gullies or on slopes provides ...abundant sources for the consequent debris flows, which will endanger resettled residents and destroy urban reconstruction. During the 5years following the Wenchuan earthquake event, heavy rainfalls have already induced a great number of debris flows in the earthquake-damaged area, resulting in serious casualties and property losses. The co-seismic landslides and the debris flows induced by rainfalls in the Mianyuan River basin are analyzed using multi-temporal remote sensing images. More than 2000 landslides were triggered and about 4.0×108m3 of loose material was generated in the Mianyuan River basin, and the volume of erodible material is 6.0×107–1.6×108m3. There were 1.27×107m3 of bed load sediments and 6.3×105m3 suspended load sediments transported into the river system in the past 5years. The active mass movements will last for a long period in the Mianyuan River basin.
•Characteristics of post-seismic landslides following the 2008 Wenchuan earthquake•2259 landslides mapped in the Mianyuan River basin, with a volume of 4.0×108m3•The volume of erodible material is 6.0×107–1.6×108m3 in the Mianyuan Riverbasin.•Only 1.27×107m3 of bed load has been transported into the Mianyuan River.•Active mass movements will last for a long period in the Mianyuan River basin.
Landslides are catastrophic natural hazards that often lead to loss of life, property damage, and economic disruption. Image-based landslide investigations are crucial for determining landslide ...susceptibility and risk. In practice, satellite images have been widely utilized for such investigations; however, they still require significant labor and time resources. In this study, we propose an image-based two-phase data-driven framework for detecting and segmenting landslide regions using satellite images. In phase I, an object detection algorithm, Faster-RCNN, is trained to detect the landslide location within the large-scale satellite images. The bounding boxes of each landslide location are proposed and visualized. In phase II, we crop the satellite images into small images using the location information of the bounding boxes. Next, we use a boundary detection algorithm to identify the boundary information of each detected loess landslide to strengthen the segmentation performance. Finally, we improve the architecture of the segmentation U-Net by integrating additional inception blocks with dilation to enhance the landslide segmentation performance. A total of 150 local loess landslide occurrences in northern China are selected as our case study to validate the effectiveness, efficiency, and universality of the proposed two-phase framework. Segmentation of loess landslides is considered a challenging task due to the intrinsic nature of vague boundary information. The proposed framework is compared with the conventional U-Net and other recent benchmarking landslide segmentation algorithms. Computational results indicate that the proposed framework produces more accurate segmentation of loess landslides compared with the other tested benchmarking algorithms.
Two successive landslides within a month started in October 11, 2018, and dammed twice the Jinsha River at the border between Sichuan Province and Tibet in China. Both events had potential to cause ...catastrophic flooding that would have disrupted lives of millions and induced significant economic losses. Fortunately, prompt action by local authorities supported by the deployment of a real-time landslide early warning system allowed for quick and safe construction of a spillway to drain the dammed lake. It averted the worst scenario without loss of life and property at least one order of magnitude less to what would have been observed without quick intervention. Particularly, the early warning system was able to predict the second large-scale slope failure 24 h in advance, along with minor rock falls during the spillway construction, avoiding false alerts. This paper presents the main characteristics of both slope collapses and damming processes, and introduces the successful landslide early warning system. Furthermore, we found that the slope endured cumulative creeping displacements of > 40 m in the past decade before the first event. Twenty-five meter displacement occurred in the year immediately before. The deformation was measured by the visual interpretation of multitemporal satellite images, which agrees with the interferometry synthetic aperture radar (InSAR) measurement. If these had been done before the emergency, economic losses could have been reduced further. Therefore, our findings strengthen the case for the deployment of systematic monitoring of potential landslide sites by integrating earth observation methods (i.e., multitemporal satellite or UAV images) and in situ monitoring system as a way to reduce risk. It is expected that this success story can be replicated worldwide, contributing to make our society more resilient to landslide events.
On 18 November 2017, a magnitude Ms. 6.9 (Mw 6.4) earthquake struck Nyingchi, Tibet Autonomous Region, China, which is located in the famous Grand Canyon region of the Yarlung Zangbo River in the ...eastern Himalayas, Tibet. The Nyingchi event was a thrusting event, with a focal depth of 12 km at 29.87° N and 95.02° E. According to emergency investigations and remote sensing, the Nyingchi event triggered at least 1820 co-seismic landslides. The landslides mainly occurred in the Grand Canyon region within an area of 527 km
2
(the inner area of Namcha Barwa tectonic node). The landslide distribution characterises an obvious hanging wall effect and is classified as “small concentration region and large landslide distribution area”. The failure patterns mainly consist of rock falls, rock avalanches, and deposit failures. One co-seismic landslide partially blocked the Yarlung Zangbo River, and the barrier lake remained. Different influencing factors, such as the seismic fault, river, slope aspect, slope angle, rocks, and elevation, have different influences on landslide occurrences, and the co-seismic landslides in the hanging wall area and footwall area present obviously different characteristics. Additionally, the post-earthquake effect impacted the recent Sedongpu landslide.
Zhouqu County in Gansu Province, Northwest China, is typically highly prone to landslides. On July 12, 2018, a landslide blocked the Bailong River near Zhouqu County, posing a serious threat to the ...life and property of local residents and the safety of infrastructure. Small baseline subset interferometry synthetic aperture radar technology (SBAS-InSAR) was adopted to identify the potential active landslides in the surrounding area of Zhouqu County, using ascending and descending orbit Sentinel-1 satellite images taken from October 2017 to December 2018. The surface deformation areas detected by SBAS-InSAR were verified by optical remote sensing image interpretation and field investigation, and a total of 23 active landslides were identified finally. The deformation characteristics of four typical landslides are analysed in detail using deformation velocity and rainfall data. It is found that the deformation velocity of landslides in this area is mainly affected by rainfall and there is a lag effect. The results can provide a reference for the prevention and control of landslide risk in Zhouqu County.
At 5:38 am on the 24th June, 2017, a catastrophic rock avalanche destroyed the whole village of Xinmo, in Maoxian County, Sichuan Province, China. About 4.3 million m
3
of rock detached from the ...crest of the mountain, gained momentum along a steep hillslope, entrained a large amount of pre-existing deposits, and hit the village at a velocity of 250 km/h. The impact produced a seismic shaking of ML = 2.3 magnitude. The sliding mass dammed the Songping gully with an accumulation body of 13 million m
3
. The avalanche buried 64 houses; 10 people were killed and 73 were reported missing. The event raised great concerns both in China and worldwide. Extensive field investigation, satellite remote sensing, UAV aerial photography, and seismic analysis allowed to identify the main kinematic features, the dynamic process, and the triggering mechanism of the event. With the aid of ground-based synthetic aperture radar monitoring, the hazard deriving from potential further instabilities in the source area has been assessed. The preliminary results suggest that the landslide was triggered by the failure of a rock mass, which had been already weakened by the M
s
7.5 Diexi earthquake in 1933. Several major earthquakes since then, and the long-term effect of gravity and rainfall, contributed to the mass failure. The high elevation, slope angle, and vegetation cover in the source area hinder geological field investigation and make hazard assessment difficult. Nonetheless, monitoring and prevention of similar collapses in mountainous areas must be carried out to protect human lives and infrastructures. To this aim, the integrated use of modern high-precision observation technologies is strongly encouraged.
Landslide susceptibility mapping is an effective approach for landslide risk prevention and assessments. The occurrence of slope instability is highly correlated with intrinsic variables that ...contribute to the occurrence of landslides, such as geology, geomorphology, climate, hydrology, etc. However, feature selection of those conditioning factors to constitute datasets with optimal predictive capability effectively and accurately is still an open question. The present study aims to examine further the integration of the selected landslide conditioning factors with Q-statistic in Geo-detector for determining stratification and selection of landslide conditioning factors in landslide risk analysis as to ultimately optimize landslide susceptibility model prediction. The location chosen for the study was Atsuma Town, which suffered from landslides following the Eastern Iburi Earthquake in 2018 in Hokkaido, Japan. A total of 13 conditioning factors were obtained from different sources belonging to six categories: geology, geomorphology, seismology, hydrology, land cover/use and human activity; these were selected to generate the datasets for landslide susceptibility mapping. The original datasets of landslide conditioning factors were analyzed with Q-statistic in Geo-detector to examine their explanatory powers regarding the occurrence of landslides. A Random Forest (RF) model was adopted for landslide susceptibility mapping. Subsequently, four subsets, including the Manually delineated landslide Points with 9 features Dataset (MPD9), the Randomly delineated landslide Points with 9 features Dataset (RPD9), the Manually delineated landslide Points with 13 features Dataset (MPD13), and the Randomly delineated landslide Points with 13 features Dataset (RPD13), were selected by an analysis of Q-statistic for training and validating the Geo-detector-RF- integrated model. Overall, using dataset MPD9, the Geo-detector-RF-integrated model yielded the highest prediction accuracy (89.90%), followed by using dataset MPD13 (89.53%), dataset RPD13 (88.63%) and dataset RPD9 (87.07%), which implied that optimized conditioning factors can effectively improve the prediction accuracy of landslide susceptibility mapping.
Landslide disasters occur frequently in the mountainous areas in southwest China, which pose serious threats to the local residents. Interferometry Synthetic Aperture Radar (InSAR) provides us the ...ability to identify active slopes as potential landslides in vast mountainous areas, to help prevent and mitigate the disasters. Quickly and accurately identifying potential landslides based on massive SAR data is of great significance. Taking the national highway near Wenchuan County, China, as study area, this paper used a Stacking-InSAR method to quickly and qualitatively identify potential landslides based on a total of 40 Sentinel SAR images acquired from November 2017 to March 2019. As a result, 72 active slopes were successfully detected as potential landslides. By comparing the results from Stacking-InSAR with the results from the traditional SBAS-InSAR (Small Baselines Subset) time series method, it was found that the two methods had a high consistency, with 81.7% potential landslides identified by both of the two methods. A detailed comparison on the detection differences was performed, revealing that Stacking-InSAR, compared to SBAS-InSAR may miss a few active slopes with small spatial scales, small displacement levels and the ones affected by the atmosphere, while it has good performance on poor-coherence regions, with the advantages of low technical requirements and low computation labor. The Stacking-InSAR method would be a fast and powerful method to qualitatively and effectively identify potential landslides in vast mountainous areas, with a comprehensive understanding of its specialty and limitations.
On 14 August 2021, a Mw 7.2 earthquake struck the Tiburon Peninsula, Haiti, with an epicenter at 18.434° N, 73.482° W and a focal depth of approximately 10 km. Combining multiple high-resolution ...satellite images and data of topographic, geological and seismological factors, this study evaluates the spatial and size distributions of the coseismic landslides triggered by this event and their corresponding controlling factors. The results show that the 2021 Mw 7.2 event, whose seismogenic fault is the Enriquillo–Plantain Garden Fault, triggered at least 8444 landslides over an area of ~2700 km2. The total area of those triggered landslides was 45.6 km2 and they were concentrated in the western section of the Tiburon Peninsula, especially within Pic Macaya National Park (6100 landslides occurred in or near this park, 72.2 % of the total), and 89.4 % of the landslides were distributed in the hanging wall area. High landslide concentrations (≥landslide frequency of 25/km2) are more prevalent in higher-elevation areas (≥1000 m). In areas at elevations ≥1000 m, more landslides are concentrated on the middle-lower mountain slopes; the landslide concentration is inversely proportional to the elevation and positively correlated with slope and local relief. The seismogenic fault area is typified by a high landslide concentration. Limestone is the dominant rock in the study area, as is the case for the 2010 earthquake, and rainfall has a positive relationship with the landslide concentration. For landslides whose areas ≥1000 m2, the total number and area of coseismic landslides from the 2021 event are both larger than those from the 2010 event.
•The 2021 Haiti earthquake triggered at least 8444 landslides covering a total area of 45.8 km2.•Landslides are concentrated in the hanging wall and areas of high-relief, most notably in Pic Macaya National Park.•In high elevation area, high landslide concentration is distributed in middle-low slopes.•Limestone is dominant rock, rainfall and PGA have positive relationship with concentration.•For landslide area ≥ 1000 m2, total number and area of landslides from 2021 event are larger.