This paper investigates the spatiotemporal characteristics and life-cycle of movements within the Joshimath landslide-prone slope over the period from 2015 to 2024, utilizing multi-sensor interferometric data from Sentinel‑1, ALOS‑2, and TerraSAR‑X satellites. Multi-temporal InSAR analysis before the 2023 slope destabilization crisis, when the region experienced significant ground deformation acceleration, revealed two distinct deformation clusters within the eastern and middle parts of the slope. These active deformation regions have been creeping up to −200 mm/yr. Slope deformation analysis indicates that the entire Joshimath landslide-prone slope can be categorized kinematically as either Extremely-Slow (ES) or Very-Slow (VS) moving slope, with the eastern cluster mainly exhibiting ES movements, while the middle cluster showing VS movements. Two episodes of significant acceleration occurred on August 21, 2019 and November 2, 2021, with the rate of slope deformation increasing by 20% (from −50 to −60 mm/yr) and around threefold (from −60 to −249 mm/yr), respectively. Following the 2023 destabilization crisis, the rate of ground deformation notably increased across all datasets for both clusters, except for the Sentinel‑1 ascending data in the eastern cluster. Pre-crisis, horizontal deformation was dominant both in the eastern and middle clusters. Horizontal deformation remained dominant and increased significantly in the eastern cluster post-crisis phase, whereas vertical deformation became predominant in the middle cluster. Wavelet analysis reveals a strong correlation between two acceleration episodes and extreme precipitation in 2019 and 2021, but no similar correlation was detected in other years. This indicates that while extreme rainfall significantly influenced the dynamics of slope movements during these episodes, less strong precipitation had a minimal impact on slope movements during other periods.

The time series of normalized difference vegetation index (NDVI) and interferometric coherence extracted from optical and Synthetic Aperture Radar (SAR) images, respectively, have strong responses to sudden landslide failures in vegetated regions, which is expressed by a sudden increase or decrease in the values of NDVI and coherence. Compared with optical sensors, SAR sensors are not affected by cloud and daylight conditions and can detect the occurrence time of failure in near real-time. The purpose of this paper is to automatically determine the time of failure occurrence using time series coherence values. We propose, based on some existing anomaly detection algorithms, a deep neural network-based anomaly detection strategy that combines supervised and unsupervised learning without a priori knowledge about failure time. Our experiment using July 21, 2020 Shaziba landslide in China shows that in comparison to widely used unsupervised methodology, the use of our algorithm leads to a more accurate detection of the timing of the landslide failure.

Surface deformation induced by the extraction of natural resources constitutes a non-stationary spatiotemporal process. Modeling surface deformation time series obtained through Interferometric Synthetic Aperture Radar (InSAR) technology using deep learning methods is crucial for disaster prevention and mitigation. However, the complexity of model hyperparameter configuration and the lack of interpretability in the resulting predictions constrain its engineering applications. To enhance the reliability of model outputs and their decision-making value for engineering applications, this study presents a workflow that combines a Tree-structured Parzen Estimator (TPE)-based Bayesian optimization approach with ensemble inference. Using the Rhineland coalfield in Germany as a case study, we systematically evaluated six deep learning architectures in conjunction with various spatiotemporal coding strategies. Pairwise comparisons were conducted using a Welch t-test to evaluate the performance differences across each architecture under two parameter-tuning approaches. The Benjamini–Hochberg method was applied to control the false discovery rate (FDR) at 0.05 for multiple comparisons. The results indicate that TPE-optimized models demonstrate significantly improved performance compared to their manually tuned counterparts, with the ResNet+Transformer architecture yielding the most favorable outcomes. A comprehensive analysis of the spatial residuals further revealed that TPE optimization not only enhances average accuracy, but also mitigates the model’s prediction bias in fault zones and mineralize areas by improving the spatial distribution structure of errors. Based on this optimal architecture, we combined the ten highest-performing models from the optimization stage to generate a quantile-based susceptibility map, using the ensemble median as the central predictor. Uncertainty was quantified from three complementary perspectives: ensemble spread, class ambiguity, and classification confidence. Our analysis revealed spatial collinearity between physical uncertainty and absolute residuals. This suggests that uncertainty is more closely related to the physical complexity of geological discontinuities and human-disturbed zones, rather than statistical noise. In the analysis of super-threshold probability, the threshold sensitivity exhibited by the mining area reflects the widespread yet moderate impact of mining activities. By contrast, the fault zone continues to exhibit distinct high-probability zones, even under extreme thresholds. It suggests that fault-controlled deformation is more physically intense and poses a greater risk of disaster than mining activities. Finally, we propose an engineering decision strategy that combines uncertainty and residual spatial patterns. This approach transforms statistical diagnostics into actionable, tiered control measures, thereby increasing the practical value of susceptibility mapping in the planning of natural resource extraction.

Landslides are a major type of natural hazard that cause significant human and economic losses in mountainous regions worldwide. Optical and synthetic aperture radar (SAR) satellite data are increasingly being used to support landslide investigation due to their multi-spectral and textural characteristics, multi-temporal revisit rates, and large area coverage. Understanding landslide occurrence, kinematics and correlation to external triggering factors is essential for landslide hazard assessment. Landslides are usually triggered by rainfall and thus, are often covered by clouds, which limits the use of optical images only. Exploiting SAR data, and their cloud penetration and all weather measurement capability, provides more precise temporal characterization of landslide kinematics and its occurrence. However, except for a few research studies, the full potential of SAR data for operational landslide analysis are not fully exploited yet. This is a very demanding task, considering the availability of a vast amount of Sentinel-1 data that have been globally available since October 2014.In this presentation we summarise all the achievements that were made within the framework of MultiSat4SLOWS project (Multi-Satellite imaging for Space-based Landslide Occurrence and Warning Service), financed within the Helmholtz Imaging 2020 call. The project aims on developing a multi-sensor approach for detection and analysis of the landslide occurrence time and its spatial extent using freely available SAR data from Sentinel-1. Within this project, we generated a reference database based on Sentinel-1 and -2 data for training, testing and validation of deep learning algorithms. The reference database contains various landslide examples that occurred worldwide and include pre- and post-event polarimetric, coherence and backscatter features. Also, we investigated the applicability of SAR/InSAR time-series data for landslide time detection. Finally, we introduce a prototype of a Visual Analytics platform for rapid landslide analysis of spatial and temporal ground deformation patterns and correlation with external triggering factors.

The early detection of slope instability and the monitoring of frequent hazard processes in mountainous regions is of paramount importance due to their sudden occurrence, and the risk of causing numerous fatalities and significant economic damage. The recent collapse of the Brienz/Brinzauls rockslide on June 15, 2023, in an active, deep-seated mountain slope deformation complex in Switzerland, provides a unique opportunity to investigate the evolution of precursors leading up to the collapse. Early identification of accelerating rockmass enabled us to set up a network of five broadband seismometers, strategically deployed to systematically record seismic signals in close proximity, reducing information loss due to attenuation of seismic waves. The internal rock damage dynamics in the displacing rock mass were interacting with external seasonal forcings, such as snow melt and rainfall, for years preceding the collapse at approximately 21:38:00 UTC on June 15, 2023. Seismic events of various types have been detected in the entire landslide complex, characterised by the recurrence of identical seismic events that aggregate prominently within the most rapid compartment, referred to as the "Insel," positioned directly above the village of Brienz. This study aims to investigate the influence of seasonal forcings on accelerating the rate of displacements and to understand how the nature of detected precursors changes over time. We systematically examine the feedback loop between seasonal triggers and gravity-driven internal rock damage under changing stress conditions during fluctuations in compartment velocity. Initially, events exhibit accelerations following periods of precipitation, but subsequently, a runaway acceleration in seismic events was noted even during dry periods. The locations detected reveal communication between the upper and lower parts of the “Insel” mass in the build-up to the main collapse. From June 1 onward, there is a consistent and gradual increase in the mean spectral power of the recurring seismic events, with a rapid escalation observed in the three days leading up to the collapse. Interestingly, on the final day preceding the main collapse, a significant decrease in the mean spectral power was identified. To complement seismic observations, the spatial and temporal changes in pre-failure slope instability for the period 05.2014-06.2023 were also analyzed using Sentinel-1 synthetic aperture radar (SAR) data using a multi-temporal interferometric (MTI) approach. MTI analysis indicates several patches of instability and surface deformation on the slope, along with signs of significant surface displacement of a few centimetres per year, also manifesting in the village of Brienz. To facilitate automatic detection and classification, we apply data science methods to various statistical seismic attributes of the identified precursors. This study contributes to advancing our understanding of the mechanisms leading to rockslide collapses, with the potential to significantly enhance warning system effectiveness.

Satellite Remote Sensing Investigation of 21 July 2020 Shaziba Landslide, China Wandi Wang1,2, Mahdi Motagh1,2, Sara Mirzaee3, Sigrid Roessner1 and Tao Li4Section 1.4 - Remote Sensing and Geoinformatics, GFZ German Research Center for Geosciences, Potsdam, Germany Institute of Photogrammetry and Geoinformation, Leibniz University Hannover, Hannover, Germany Department of Marine Geology and Geophysics, University of Miami, United States GNSS Research Center, Wuhan University, China  We present the results of remote sensing analysis of deformation related to the 21 July 2020 Shaziba landslide in China. The landslide, which occurred following the heavy precipitation from June to August 2020, is located at the Qingjiang River, approx. 30 km from Enshi City in Hubei Province of China.   It destroyed over 60 houses, and by blocking the course of the river, formed a landslide dam, which threatened the safety of people and infrastructure downstream. Although Shaziba landslide occurred in form of reactivation of an old landslide, the landslide prone slope was not instrumented prior to this most recent failure. Therefore, high-resolution remote sensing imagery was used as a very effective source of information for a detailed investigation of the evolution of this slope failure.  We collected the satellite remote sensing data covering a time period from June 2016 to July 2021 and comprise optical and radar data. Firstly, cross-correlation analysis using satellite optical imagery from Planet and Sentinel-2 systems was used to retrieve the lateral direction and magnitude of landslide movements. Next, multi-temporal interferometry (MTI) analysis based on Sentinel-1 and TerraSAR-X SAR data was exploited to obtain pre- and post-failure deformation. Results from different MTI techniques including Persistent Scatterer (PS), Small Baseline Subsets (SBAS), and Eigendecomposition based Maximum-likelihood-estimator of Interferometric phase (EMI) were compared to evaluate the most suitable method for InSAR time-series analysis of deformation related to the evolution of Shaziba landslide. Finally, several high-resolution DEMs derived from TanDEM-X (TDX) data were analyzed using repeat-pass interferometry and stacked together to compensate for the errors related to DEM alignment in order to precisely estimate the landslide volume. The results highlight how the integration of various remote sensing sensors helps to gain a better understanding of landslide evolution process and characterization.