-
Paper 71 - Session title: Terrain subsidence and landslides I
16:10 Landslide movement and basal geometry revealed by InSAR: a case study of Cascade landslide complex, WA
Hu, Xie (1); Lu, Zhong (1); Pierson, Thomas (2); Wang, Teng (1); Kim, Jinwoo (1); Cecere, Thomas (3) 1: Southern Methodist University, United States of America; 2: U.S. Geological Survey, Vancouver, WA, United States of America; 3: U.S. Geological Survey, Reston, VA, United States of America
Show abstract
Detection of landslide movement over forested terrains has long been problematic, particularly for the Cascade landslide complex (Washington, USA) located along the Columbia River Gorge. Although parts of the landslide complex have been found reactivated in recent years, the characteristics and the magnitude of motions have not been systematically studied. Here we apply time-series InSAR processing strategies to study the spatial distribution, temporal behavior, and basal geometry of the landslide movement using spaceborne SAR data from L-band ALOS-1 PALSAR-1 (2007-2011) and ALOS-2 PALSAR-2 (2014-2016) and C-band Sentinel-1A (2014-2016). The independent measurements from different sensors confirm the spatial extent of the ~8-km2 reactivated lobe, while other parts of the landslide complex remained generally stable. An integration of measurements from different radar looking geometries constrains the 2-dimentional displacement velocity field over the active lobe. The resolved slope-parallel displacement pattern is different from that of line-of-sight measurement. The largest downslope movement, at a rate of ~0.2 m/yr, turns out to be around the southwest boundary of the active part. InSAR-derived surface displacement further allows us to invert for the landslide thickness and its basal geometry. The temporal oscillations of the landslide movement are correlated with the precipitation, implying that seasonal movement is hydrologically driven. The seasonal motions have a frequency that is similar to that of regional ground oscillations due to mass loading by stored precipitation and subsequent rebound, which were observed at a nearby GPS station. However, the magnitude of the motion at the active slide is much exaggerated, suggesting higher capacity for hydrological loading on the thick and porous landslide body. In addition, time-series radar backscattering captures the incipient motion related to the 2008 Greenleaf Basin rock avalanche, not previously recognized by traditional SAR/InSAR methods. The approach used in this study can be used to identify active landslides in forested terrain, to constrain landslide basal geometry, to track the seasonal movement of landslides, and to identify previously unknown landslide hazards.
Presentation
[Authors] [ Overview programme] [ Keywords]
-
Paper 159 - Session title: Terrain subsidence and landslides I
17:10 Satellite radar interferometry for the early detection of landslides: example from the Moosfluh landslide (Switzerland)
Strozzi, Tazio (1); Wegmüller, Urs (1); Caduff, Rafael (1); Raetzo, Hugo (2); Delaloye, Reynald (3) 1: Gamma Remote Sensing, Gümligen, Switzerland; 2: Federal Office for the Environment, Hazard Prevention Division, Bern, Switzerland; 3: Department of Geosciences - Geography, University of Fribourg, Switzerland
Show abstract
Numerous media in Switzerland (www.rts.ch www.srf.ch, www.lenouvelliste.ch, www.nzz.ch) and around the world (e.g. http://blogs.agu.org) reported at the end of September 2016 that the movement rate of the Moosfluhe landslide above the left flank of the Great Aletsch Glacier in Switzerland has accelerated significantly. The municipality of Riederalp with support from the Canton of Valais communicated that the very large rockslope failure with an estimated area of more than 1 square kilometre and a volume of more than 200 million cubic metres was moving at a rate of more than 20 cm per day (http://gemeinde.bettmeralp.ch/data/Pressemitteilung-Kanton-27_9_2016.pdf). Numerous cracks and rockfalls have been observed and the hiking trails in the sector have been closed. Nevertheless, a rapid collapse of the entire mass was judged to be very unlikely.
According to the open scarp and graben structures, the Moosfluhe mass movement could have been consecutive to post-glacial rebound of the slope by the end of the Younger Dryas, approximately 11,000 years Before Present (BP). The total displacement since the Younger Dryas was roughly 15 to 30 m with a mean deformation rate during the Holocene of a few mm/yr and probable phases of inactivity. Significant recent movements of the Moosfluh landslide were first detected with use of ERS InSAR (Strozzi et al., JGR, 2010). The displacement was visible between 1992 and 1998 in inter-annual interferograms and for a 105 day time interval in the summer of 1999. Subsequent interferograms from ALOS PALSAR, ENVISAT ASAR and TerraSAR-X revealed an acceleration of the landslide from about 4 cm/yr during the 1990’s to more than 20 cm/yr in 2008. The significant acceleration of the quiescent landslide since the 1990’s appeared to be mainly the result of continued debutressing of the valley flanks from the retreating glacier.
In order to endorse the INSAR results and to obtain additional information on the landslide kinematic, airborne photographs taken over Great Aletsch Glacier at the end of the summers of 1976, 1995, and 2006 were photogrammetrically analyzed and a network of differential GNSS points was measured twice each year since 2007. The photogrammetric analysis confirmed no significant movement (i.e., <1 cm/yr) between 1976 and 1995 and the differential GNSS data, complemented with TerraSAR-X interferograms, indicated that the rate of movement at the centre of the landslide decreased in the summer of 2010 and 2011 to about 10 cm/yr and then increased again since 2012 to reach more than 70 cm/yr in 2015. Superimposed to the accelerating trend, velocities were higher during the summer seoson and periods of peak activity occurred after years of high precipitation rate.
In response to the rapid glacier retreat and the increase in velocity of the Moosfluhe landslide, a comprehensive permanent monitoring system has been installed in the last few years at the terminus of the Great Aletsch Glacier by national authorities, research institutions (Glueer et al., Geophysical Research Abstracts, 2015) and the cableway operator (www.gruenenfelder.ch/aktuell/news_moosfluh-rutschung). The system includes GNSS stations, automated total stations, climate sensors, and high precision tilt meters. Movements and accelerations of the Moosfluhe landslide are thus nowadays automated followed in detail.
In our contribution we will discuss the recent evolution of the Moosfluh landslide from satellite InSAR and differential GNSS and highlight the unique role of satellite SAR interferometry in the early recognition of landslides.
[Authors] [ Overview programme] [ Keywords]
-
Paper 162 - Session title: Terrain subsidence and landslides I
16:30 Sentinel-1 Data Help Capture Pre-failure Signatures of Slope Instability – Toward Forecasting of the Temporal Occurrence of Landslides with the Aid of Multi-temporal Interferometry
Wasowski, Janusz (1); Bovenga, Fabio (2); Nutricato, Raffaele (3); Nitti, Davide Oscar (3); Chiaradia, Maria Teresa (4) 1: CNR-IRPI, Italy; 2: CNR-ISSIA, Italy; 3: GAP srl c/o Politecnico/Universita di Bari, Italy; 4: Dept. of Physics, Politecnico/Universita di Bari, Italy
Show abstract
The regularity and higher frequency of acquisitions of Sentinel-1A/B (S-1) with respect to earlier ESA’s satellite C-band sensors (ERS1/2, ENVISAT) represent clear advantages for users of multi-temporal interferometry (MTI) products. The utility of the IW acquisition mode of S-1 for regional scale slope instability detection through MTI has already been demonstrated, e.g., via studies of landslide-prone areas in Italy. In this work, we explore the potential of S-1 data for local (site-specific) scale landslide monitoring and capturing pre-failure signs of slope instability. This is done by using examples of two unstable slopes from different environmental settings and MTI through the Persistent and Distributed Scatterers processing capability of the SPINUA algorithm.
The first case regards a hilltop town in the Apennine Mts., whose stability is threatened by a large (~600 x 300 m2), slow moving deep landslide. We processed over 50 S-1A images acquired since October 2014. The comparison of the MTI results with those based on ERS and ENVISAT imagery shows that a much higher number of radar targets is obtained from S-1A data (e.g., from ~2 to 5 times higher, respectively on the landslide and in the overall area of interest, including also the town and peri-urban areas). With more targets, we can better depict the spatial movement pattern and local velocity gradients, which is important for geotechnical assessment. Furthermore, the lateral boundaries of the landslide can now be delimited in more detail, overcoming the mapping uncertainties typical in cases of very slow, deep landslides affecting urbanized areas. This offers invaluable information for local inhabitants/property owners and for engineering scale hazard assessment. Importantly, the MTI from S-1A data also revealed an accelerating trend with a nearly doubled velocity of the displacements with respect to those in the earlier period covered by ERS-ENVISAT data. The higher frequency of S-1A acquisitions (about 30/year in this case) helped highlighting the non-linearity of surface deformations within the faster displacement phase, whose timing was consistent with the increase in landslide movements detected through subsurface inclinometer monitoring and field observations. The latter demonstrated that this faster movement phase coincided with (or was preceded by) a failure of the landslide toe, which occurred in the inhabited area.
The second case represents an example of a retrospective investigation of a huge (over 2 km long, few tens of m deep) landslide, which occurred in 2016 in an important open-cast coal mine in central Europe. The apparently sudden failure disrupted the mining operations, destroyed in part the mining machinery and resulted in high economic losses. In this case, we exploited over 60 S-1A/B images acquired since November 2015. Despite the presence of spatial gaps in information (due to intensive surface disturbance by mining operations), the MTI results provided a good overview of the ground instability/stability condition within and outside the active mine. Furthermore, it was shown that the 2016 slope failure was preceded by very slow (generally 1-3 cm/yr) creep-like deformations, already detectable in 2014. Although it would not have been simple to issue a short-term warning of the impeding failure based on the displacement time series, the MTI results showed that the slope had been in the critical instability state some months prior to the landslide event. Furthermore, the spatio-temporal mapping of coherence changes in the unstable area indicated a sharp coherence loss in the last few weeks before the slope collapse.
The above examples demonstrate that by securing long-term, regular, high-frequency acquisitions over wide-areas, the Sentinel-1 mission facilitates a more effective use of MTI in slope hazard assessment. We note further improvement thanks to the availability of S-1B data (e.g., more frequent measurements to forewarn potential instability hazards). This has practical impacts on landslide monitoring activities and will aid future research on slope failure forecasting. Thanks to this and free imagery, the site-specific investigations relying on MTI will become even more feasible and cost-effective for non-scientific users (e.g., engineering geology/geotechnical consulting) and commercial services (e.g., Rheticus®).
Acknowledgments
We thank ESA for ERS, ENVISAT and Sentinel-1 & Sentinel-2 data.
Presentation
[Authors] [ Overview programme] [ Keywords]
-
Paper 213 - Session title: Terrain subsidence and landslides I
17:30 Landslide Displacement Monitoring by InSAR Analyses with Persistent and Distributed Scatterers: A Case Study of Danba County, China
Dong, Jie (1); Zhang, Lu (1,2); Liao, Mingsheng (1,2); Gong, Jianya (1,2,3) 1: State Key Laboratory of Information Engineering in Surveying, Mapping, and Remote Sensing, Wuhan University, China; 2: Collaborative Innovation Center for Geospatial Technology, Wuhan University, China; 3: School of Remote Sensing and Information Engineering, Wuhan University, China
Show abstract
Differential SAR interferometry (DInSAR) is able to detect ground deformation with wide coverage and sub-centimeter accuracy, but seriously influenced by inaccurate external DEM, decorrelation and atmospheric phase screen (APS). In consequence, advanced time series InSAR techniques were developed to overcome the limitations of DInSAR. The Persistent Scattterers InSAR (PSI) technique exploits persistent scatterers (PSs) exhibiting high phase stability during the entire image stack. However, in the case of landslide monitoring in rural mountainous areas, the vegetation coverage and complex terrain lead to a poor number of reliable high-quality PS points. Improving the density of reliable points is a key factor to achieve a better understanding of the landslides' extension and dynamics. Although Small BAseline Subset (SBAS) method focuses on the distributed scatterers (DSs) and is more suitable for nonurban environment, the multi-looking operation decreases the spatial resolutions.
In contrast, inspired by the idea of SqueeSAR method, we combine both the PS and DS targets to effectively solve the problem of sparsity of measurements points for landslide investigation in mountainous area. Meanwhile, two different strategies are implemented to replace the ones in SqueeSAR. The first is to employ generalized likelihood ratio test (GLRT) on complex covariance matrix, instead of KS/AD test to identify statistically homogenous pixels (SHPs), which is more robust for a limited number of SAR images. The second is to use the fast phase-linking method to solve the nonlinear minimization when applying the maximum likelihood estimation (MLE) to retrieve optimal phases at DS points. After such preprocessing, they are further processed following the standard PSI procedure together with PS points.
In this study, we focus on the Jiaju Landslide in Danba County in western Sichuan Province in China. This region is east to Qinhai-Tibet Plateau and belongs to Minshan-Qionglai alp, characterized by steep terrain and alpine valleys with its altitude varying from 1,700 meters to 5,520 meters. Because of its plateau monsoon climate along with the strong polymetamorphism and special tectonics, rock fall, landslides and debris flow break out frequently and widely. Jiaju is an ancient landslide, above which the top one of the most beauty villages in China is located. The flowering tourism in the last decade promotes the rapid expanding of the village. And the continuous increase of the buildings load, the uncontrolled slope cutting, the water discharging and the fluvial abrasion induce the reactivation of this ancient landslide. It presents a huge threat to the local village life safety and economic development.
SAR data acquired by both L-band ALOS PALSAR and C-band ENVISAT ASAR from ascending orbits are used in our study. In particular, a total number of 19 fine-beam mode ALOS/PALSAR datasets were collected between December 2006 and January 2011, while only 9 scenes of Stripmap-mode ENVISAT ASAR images were available over a shorter period from August 2007 to June 2008. Furthermore, 20 stationary GPS stations were uniformly distributed on the landslide body, as well as 2 base station installed at surrounding stable area.
The improvements of our proposed method are demonstrated by comparison with regular PSI and SBAS. In addition, the resultant displacement information is assessed by cross-validation between PALSAR and ASAR, as well as in-situ GPS measurements. Lastly, preliminary correlation between time series displacements of measuring point and triggering factors (rainfall, human activity, earthquake, etc.) is analysed to investigate the driving mechanisms for landslide motion.
Presentation
[Authors] [ Overview programme] [ Keywords]
-
Paper 271 - Session title: Terrain subsidence and landslides I
16:50 Detection of damages due to slow landslide through the three-dimensional Finite Element modeling of DInSAR measurements and in situ surveys
Castaldo, Raffaele (1); De Novellis, Vincenzo (1); Lollino, Piernicola (2); Manunta, Michele (1); Tizzani, Pietro (1) 1: Istituto per il Rilevamento elettromagnetico dell'ambiente, Napoli, Italy; 2: Istituto di Ricerca per la Protezione Idrogeologica, Bari, Italy
Show abstract
The analysis of structure damages connected with the kinematics of the landslide phenomena has been analyzed in a large number of scientific studies; the approaches range from the study of two- dimensional slope models, suitable for landslides bodies with sliding surface depth about constant and significantly lower than the landslide length, to more sophisticated three-dimensional Finite Element (FE) models aimed at detecting the different kinematical sectors along the slope area. In this context, the DInSAR technique has been broadly used to detect and monitor surface displacements related to mass movement and slope instabilities, by benefiting from the DInSAR capability to provide dense displacement maps. Among the DInSAR techniques currently available, the Small BAseline Subset (SBAS) approach has well demonstrated its capability to monitor the deformations related to mass movement phenomena with high spatial density of measure points.
In the present work, the Ivancich landslide, which affects a residential area outside the historical center of the Assisi town (Central Italy), has been selected as a representative case study to highlight the capability of advanced 3D FE modeling as complementary tool to perform effective risk analyses of slow landslide processes and accurate urban development planning, also thanks to the big amount of available information. Following this strategy, we exploit the in situ data in order to build up the FE domains and the DInSAR measurements to calibrate the modelled displacement field. In particular, the proposed FE geometry is constrained by using 7 litho-stratigraphic cross-sections and 62 stratigraphic boreholes information. Subsequently, we combine the benefits of a deterministic numerical approach with statistical methods aimed at improving and optimizing the obtained numerical solution in order to analyze and interpret the ground deformations measured in the whole landslide area. The unknowns model parameters, represented by the viscosities of shear band, are searched by benefiting from SBAS-DInSAR results retrieved by processing 39 SAR images collected by the Cosmo-SkyMed (CSK) constellation in the 2009 - 2012 time span. The achieved results allow us to explore the spatial and temporal evolution of the slow-moving phenomenon and, via comparison with the geomorphological data, to derive a synoptic view of the kinematical activity of the urban area affected by the Ivancich landslide.
Finally, a comparative study between the structure damages, revealed along the landslide limits (Figure 1) and the modelled shear rate is proposed (Figure 2). This analysis of the structure damages compared with the model results reveals that ruptures are most severe along the boundary between active and inactive landslide sectors where the shear rate values are high. Hence, the proposed 3D FE modeling tool represents a valuable support for landslide risk analyses and urban development planning within a specific territory area affected by complex slow-moving landslide processes.
Presentation
[Authors] [ Overview programme] [ Keywords]
-
Paper 396 - Session title: Terrain subsidence and landslides I
17:50 Comprehensive Geophysical Model and Criticality Prediction for a Large Deep-seated Gravitational Slope Deformation, Fels Glacier Slide, Alaska
Rabus, Bernhard T; Eppler, Jayson; Pichierri, Manuele Simon Fraser University, Canada
Show abstract
We carry out a comprehensive analysis of historic and recent activity of the Fels Glacier slide, the largest of several deep-seated gravitational slope deformations situated within a few kilometers of the Alaska pipeline corridor near where it crosses the Denali fault. Preliminary studies after the very large M7.9 3 November 2002 Denali earthquake posed questions on pre vs. post-quake dynamics, geo-physical and meteorological drivers, and the criticality of the identified instable slopes whose catastrophic failure would pose severe threats to pipeline integrity directly through the runout, or indirectly through blocked drainage causing subsequent outburst flooding. Most of these questions can now be answered thanks to a large multi-sensor InSAR data set with a favorable combination of look geometries and resolutions spanning the period 1991 to 2016. The data set contains rich information about the temporal evolution of the 3D surface displacement field of the Fels Glacier slide, which exhibits strong seasonal variations with summer and winter maximum downslope velocities up to 100 cm/year and 20cm/year, respectively. Spatially the slope deformation is comprised of several differentially moving nested lobes that produce large displacement gradients and discontinuities. The complexity and variability of the displacement patterns, combined with seasonal snow cover, correspondingly lead to both severe spatial aliasing and discontinuous coherence even in the shortest temporal repeat interferograms, which in the past has prevented meaningful phase unwrapping and quantitative retrieval of the displacement information for the Fels Glacier slide.
Very high resolution TerraSAR-X staring spotlight data acquired for two recent seasonal cycles of the Fels Glacier slide provide unprecedented spatial detail and high coherence, which combined with a novel adaptive iterative demodulation technique, for the first time allow us to unwrap confidently the phase of all preceding coarser, spatially aliased interferograms from ERS, and RADARSAT-1 and RADARSAT-2. As part of this processing all InSAR data is first resampled to a common high-resolution geocoded reference frame where the spatio-temporal surface displacement maps are then derived in form of fall line and emergence vector components using an available high resolution airborne SAR DEM. Besides the time series of displacement maps, detailed field data on surface and structural geology of the sliding slope, as well as glacier down-wasting, and meteorological time series records are inputs to constrain comprehensive geophysical modeling of the Fels Glacier slide with a state-of-the-art 3D discontinuous finite element method.
Results suggest that meteorology (snowmelt and precipitation) is the (obvious) direct driver leading to the large observed seasonality and can quantitatively explain the slide activity. However, glacier down-wasting and seismic activity through altering the “rheological” parameters both crucially influence the long-term response behavior of the Fels Glacier slide including predictions of criticality to “excessive driver value events”. The parametrization with the highest likelihood to explain all model inputs within error bounds suggests the 2002 earthquake has weakened rheological parameters dramatically, leading to a more than factor 3 increased response tothe same meteorological drivers immediately after the quake when compared to the recent period after negative exponential stabilization with a time constant of about 4 years has occurred. Criticality modeling involving the geophysical model and an inverse velocity method points to the possible importance of the late fall timing of the 2002 Denali earthquake preventing a catastrophic failure event at the Fels Glacier slide. Preliminary results place an early summer timing of the earthquake combined with a moderate positive precipitation anomaly near the stability bounds of catastrophic failure.
Presentation
[Authors] [ Overview programme] [ Keywords]
-
Paper 549 - Session title: Terrain subsidence and landslides I
18:10 Round Table Discussion 1/2
All, All ESA, Italy
Show abstract
Terrain subsidence and landslides I
[Authors] [ Overview programme] [ Keywords]