Unraveling spatial and temporal heterogeneities of very slow rock-slope deformations with targeted DInSAR analyses

Spaceborne radar interferometry is a powerful tool to characterize landslide activity. However, its application to very slow rock slope deformations (displacement rates < 5 cm/yr) in alpine environments remains challenging due to low signal-to-noise ratio, severe atmospheric and snow cover effects, and heterogeneous deformation patterns related to complex landslide mechanisms in space and time.

In this study we combine available SqueeSAR^TM data (Sentinel 1A/B ascending and descending, 2015-2017), ad hoc multi-temporal baseline DInSAR processing (2016-2019), GPS data (2015 to 2019) and detailed field mapping to unravel the kinematics, internal segmentation and style of activity of the Mt. Mater deep-seated gravitational slope deformation (DSGSD) in Valle Spluga (Italy). The high relief slope (1500-3000 m.a.s.l.) is made of dominant micaschist and paragneiss of the Stella-Timun complex (Suretta nappe) and ranges in inclination between 33° (< 2500 m a.s.l.) and 25° (> 2500 m a.s.l.). At 2900 m a.s.l. the slope is cut by a sharp triangular headscarp with a vertical downthrow of about 40 m, moving downslope, shallower arcuate scarps mark the transition to two nested large landslides, affecting the slope between 2400 m a.s.l. and 1550 m a.s.l; with highly deformed toes.

Through 2DInSAR decomposition, we highlight the global translational kinematics of the DSGSD. However, regional scale processed PSI data result unsuitable to capture the spatial complexity of the phenomenon at the local scale. To obtain a spatially-distributed characterization of the DSGSD displacement patterns, we process several multi-temporal interferograms and retrieve unwrapped phase and displacement maps according to a process-oriented, targeted approach based on variable temporal baselines (from 24-days to 1-year). In this context: a) 1-year interferograms provide a picture of long-term background DSGSD displacement signals; b) seasonal interferograms highlight displacement trends suggesting a complex response of different slope sectors to hydrological input; c) 24 days interferograms outline a triangular shaped active sector extending between 2500 m a.s.l. and the main DSGSD headscarp, corresponding to the movement of extensive debris cover and overlying periglacial features.

Our analyses clearly outline a composite slope instability and a strong spatial heterogeneity with different nested sectors possibly undergoing different evolutionary trends towards failure. The combined analysis of seasonal interferograms and GPS data further confirm a sensitivity of the different slope sectors to hydrological forcing modulated by snowmelt and rainfalls. The herein results outline the potential of a targeted use of DInSAR, carefully constrained by field geological and morpho-structural data, for the detailed investigation of a complex very slow rock slope deformation successfully unravelling its mechanisms, temporal trends of activity and forcing factors. Ground-truthing by means of GPS data further prove that, in the context of very slow rock deformations, PSI data are useful for a first-order characterization of slope activity and kinematics, but often fail to capture local scale spatial segmentation, temporal trends and associated mechanisms.

Our approach prove to be effective in providing key information for the definition of possible evolutive scenarios for risk analysis and mitigation of a widespread, yet challenging class of slope instabilities.