Practical Estimation of Landslide Kinematics Using PSI Data

Kinematics is a key component of a landslide hazard because landslides moving at similar rates can affect structures or collapse differently depending on their mechanisms. While a complete definition of landslide kinematics requires integrating surface and subsurface site investigation data, its practical estimate is usually based on 2D profiles of surface slope displacements. These can be now measured accurately using Persistent Scatterer InSAR (PSI), which exploits open access satellite imagery. Although 2D profiles of kinematic quantities are easy to retrieve, the efficacy of possible descriptors and extraction strategies has not been systematically compared, especially for complex landslides. Large, slow rock slope deformations, characterized by low displacement rates (<50 mm/year) and spatial and temporal heterogeneities, are an excellent testing ground to explore the best approaches to exploit PSI data from Sentinel-1 for kinematic characterization. For three case studies, we extract profiles of different kinematic quantities using different strategies and evaluate them against field data and simplified numerical modelling. We suggest that C-band PSI data allow for an effective appraisal of complex landslide kinematics, provided that the interpretation is (a) based on decomposed velocity vector descriptors, (b) extracted along critical profiles using interpolation techniques respectful of landslide heterogeneity, and (c) constrained by suitable model-based templates and field data.

Automatic classification of landslide kinematics using acoustic emission measurements and machine learning

Founded on understanding of a slope’s likely failure mechanism, an early warning system for instability should alert users of accelerating slope deformation behaviour to enable safety-critical decisions to be made. Acoustic emission (AE) monitoring of active waveguides (i.e. a steel tube with granular internal/external backfill installed through a slope) is becoming an accepted monitoring technology for soil slope stability applications; however, challenges still exist to develop widely applicable AE interpretation strategies. The objective of this study was to develop and demonstrate the use of machine learning (ML) approaches to automatically classify landslide kinematics using AE measurements, based on the standard landslide velocity scale. Datasets from large-scale slope failure simulation experiments were used to train and test the ML models. In addition, an example field application using data from a reactivated landslide at Hollin Hill, North Yorkshire, UK, is presented. The results show that ML can automatically classify landslide kinematics using AE measurements with the accuracy of more than 90%. The combination of two AE features, AE rate and AE rate gradient, enable both velocity and acceleration classifications. A conceptual framework is presented for how this automatic approach would be used for landslide early warning in the field, with considerations given to potentially limited site-specific training data.

Recharge response and kinematics of an unusual earthflow in Liechtenstein

Understanding landslide behavior over medium and long timescales is crucial for predicting landslide hazard and constructing accurate landscape evolution models. The behavior of landslides in soil that undergo periodic displacements, termed earthflows or compound soil slides, is especially difficult to forecast at these timescales. This is because velocities can increase by orders of magnitude over annual to decadal timescales, due to processes such as changing recharge conditions, erosion of the landslide toe, and retrogression of the landslide head. In this paper, we provide a detailed analysis of the Schlucher landslide, an unusual earthflow that is perched above the village of Malbun, Liechtenstein. This landslide had been displacing by 10 to 20 cm/year until 2015, when displacements on the order of 2 m/year occurred from 2016 to 2018. These large displacements damaged landslide mitigation measures, caused numerous surface deformation features, and threatened the local population downstream of the earthflow. This landslide has an unusually long monitoring record, with accurate displacement and climatic data available since 1983. We analyze this nearly 40-year monitoring time series to estimate recharge from snowmelt and rainfall, and its correlation with displacement. We also analyze recently collected, high-resolution surface and subsurface data in order to understand landslide response to recharge, landslide kinematics through time, and catastrophic failure potential. We find that interannual displacements can be explained with variations in recharge; however, periodic surges with recurrence times of tens of years must be explained by other mechanisms. In particular, recharge into the landslide during the recent acceleration (2016 to 2018) was not anomalously high. Instead, we argue that loss of internal strength is responsible for this recent acceleration period, and that this mechanism should be considered when forecasting the surge potential for certain earthflows and soil slides.

When image correlation is needed: combining very dense radar-amplitude and optical times series for unravelling the complex dynamics of a not so slow slow-moving landslide in the tropics

&lt;p&gt;Slow-moving landslides exhibit persistent but non-uniform motion at low rates which makes them exceptional natural laboratories to study the mechanisms that control the dynamics of unstable hillslopes. Here we leverage 4.5+ years of satellite-based radar and optical remote sensing data to quantify the kinematics of a slow-moving landslide in the tropical rural environment of the Kivu Rift, with unprecedented high spatial and temporal resolution. We measure landslide motion using sub-pixel image correlation methods and invert these data into dense time series that capture weekly to multi-year changes in landslide kinematics. We cross-validate and compare our satellite-based results with very-high-resolution Unoccupied Aircraft System topographic datasets, and explore how rainfall, simulated pore-water pressure, and nearby earthquakes control the overall landslide behaviour. The landslide exhibited seasonal and multi-year velocity variations that varied across the landslide kinematic units. While rainfall-induced changes in pore-water pressure exerts a primary control on the landslide motion, these alone cannot explain the observed variability in landslide behaviour. We suggest instead that the observed landslide kinematics result from internal landslide dynamics, such as extension, compression, material redistribution, and interactions within and between kinematic units. Our study provides, a rare, detailed overview of the deformation pattern of a landslide located in a tropical environment. In addition, our work highlights the viability of sub-pixel image correlation with long time series of radar-amplitude satellite data to quantify surface deformation in tropical environments where optical data is limited by persistent cloud cover and emphasize the importance of exploiting synergies between multiple types of data to capture the complex kinematic pattern of landslides.&lt;/p&gt;

Precursory and post-failure analysis of landslide deformation in Danba County, China using optical remote sensing and Multi-temporal InSAR (MTI) methods with corner reflectors

&lt;p&gt;On 17 June 2020, a large debris flow triggered by continuous heavy precipitation hit the Danba County in southwest China, blocked the river and a barrier lake was formed. Meanwhile, on the other side of the river, a large-scale landslide was triggered due to the reactivation of the ancient landslide body. Then an evacuation of more than 20000 people leaving their home town was urgently conducted.&lt;br&gt;This study exploits multi-sensor remote sensing techniques to assess landslide deformation, precursory deformation and post-failure motion of Danba landslide. We start with optical remote sensing images using the cross correlation method to investigate the overall information about this collapse, such as magnitude and moving direction of the sliding. Two high-resolution remote sensing optical images from Planet are processed right before and after the failure.&lt;br&gt;Moreover, we apply the advanced Multi-temporal InSAR (MTI) techniques such as Persistent Scatterer Interferometry (PSI) and Small Baseline Subsets (SBAS) to analyze the precursors of the landslide over the long term. Based on the results of optical remote sensing, the descending Sentinel-1 data in 2014-2020 are extensively exploited with a better geometry of satellite observation. The long-term and transient of the deformation are analyzed against variations of precipitation, and then the related early warning systems are further explored.&lt;br&gt;The last stage of the work is the monitoring of current movements in the collapse region after the failure. It is explored by using multiple SAR datasets including C-band Sentinel-1 and X-band TerraSAR-X (TSX) high-resolution SAR images. With the help of the field works by our collaborators, stable artificial corner reflectors (CR) are deployed on selected sites to evaluate their performance in deriving landslide kinematics. Different from the traditional Triangle CR (TCR), the new design of dihedral CR (DCR) are introduced and exploited on the scene. The performance of this new design towards MTI processing and sub-pixel offset-tracking processing is examed and tested in this study. Results are presented and further discussed for a better assessment of Danba landslide.&lt;br&gt;The results of this paper can provide new strategies for developing an early warning system in this landslide using remote sensing technologies. Besides, the post-failure results are compared with the pre-event analysis, which could give an associated and comprehensive understanding of the whole landslide kinematics.&lt;/p&gt;

KINEMATIC MODEL OF THE SLOW-MOVING KOSTANJEK LANDSLIDE IN ZAGREB, CROATIA

The interpretation of landslide kinematics provides important information for those responsible for the management of landslide risk. This paper presents an interpretation of the kinematics of the slow-moving Kostanjek landslide, located in the urbanized area of the city of Zagreb, Croatia. The sliding material (very weak to weak marls, often covered with clayey topsoil) exhibits plastic, rather than rigid behavior. Due to this reason, and low landslide velocities, landslide features, such as main scarps or lateral flanks, are barely noticeable or do not exist in most of the landslide area. The data used for the kinematic interpretation were obtained from 15 GNSS sensors, for the period of 2013-2019. The monitoring data revealed a different spatial and temporal distribution of landslide velocities, resulting as a consequence of geomorphological conditions and forces that govern the landslide movements. Temporally, eight periods of faster movements and seven periods of slower movements were determined. Spatially, velocities measured in the central part of the landslide were higher than on its boundaries. The interpretation of the surface (horizontal and vertical) displacements and the direction of movement reveal a new insight into the engineering geological model and provide important information for the management of the Kostanjek landslide risk.

Investigation of active movement prior to artificial slope landslides from multi-temporal InSAR: a case study of Northern Taiwan

&lt;p&gt;Detection of slope instability using Interferometric Synthetic Aperture Radar (InSAR) can aid the understanding of landslide kinematics and prevent the related geological hazards. However, conventional InSAR techniques often fail in the retrieval of deformation measurements in mountainous areas with dense vegetation and complex terrain, thus resulting in diminished information of slope movement. In this study, we propose a new multi-temporal InSAR method to improve the spatial coverage of measurement points by jointly exploiting persistent scatterers (PS) and distributed scatterers (DS). Particularly, topographic errors and tropospheric delays are well-considered according to their spatial and temporal characteristics. We applied this method to retrieve the historic displacements prior to the collapse of an artificial slope in Northern Taiwan using 15 ALOS/PALSAR images. The derived results suggest a pre-landslide movement with a rate of approximately -30 mm/year in the radar line-of-sight (LOS) direction. Meanwhile, the time series displacements reveal that the temporal behaviors of downslope movement are correlated with local rainfall and seismic activities. The study helps to analyze the slope instability in Northern Taiwan.&lt;/p&gt;