Over the last decades, synthetic aperture radar (SAR) images and SAR interferometry (InSAR) have revolutionized Earth observation, allowing for geophysical monitoring of Earth surface processes with centimeter-to-millimeter precision. Accurate measurement of ground displacement is essential for the comprehension of natural hazards, such as earthquakes, and the detection of the smallest ground (transient) displacement is of uttermost importance to better image the dynamics of active faults, especially in tectonic contexts that undergo low deformation rates. However, detecting small deformation signals in raw SAR images remains a significant challenge because of the significant noise level affecting the data (e.g., speckle noise, tropospheric and ionospheric perturbations). Multiple and successful InSAR mass processing methods, including state-of-the-art noise correction methods, have been developed over the last decade, but all rely on intensive computing of massive databases, a tedious procedure that cannot be applied yet at a global scale. Furthermore, because of the low probability of finding earthquakes in intraplate continental settings, automatic detection of such signals in such settings is currently out of the question with InSAR data.

Here, we leverage deep learning to enhance the detection of deformation (e.g., dislocation-like signals) directly from raw SAR images. Our deep-learning-based approach offers the potential to (1) retrieve potential deformation below the noise threshold, thus improving sensitivity, and (2) precisely localize regions of interest from full acquisitions to serve as input for InSAR pipelines, reducing the need to process entire datasets and significantly accelerating computation. Also, deep learning methods can process large-scale images much faster, enabling the creation of dense and extensive detection catalogs for subsequent analysis.

How to cite: Costantino, G. and Jolivet, R.: Detection of dislocation-like signals in raw SAR images with deep learning, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-6358, https://doi.org/10.5194/egusphere-egu25-6358, 2025.

The intensification of the effects of climate change on mountainous geohazards underlines the critical need for advanced tools to monitor geohazards like landslide and permafrost associated hazards. In this study, we present a novel active reflector specifically designed for C-band synthetic aperture radar (SAR) applications, optimized for Sentinel-1 missions. The reflector is engineered to receive vertically polarized signals and reflect them in both vertical and horizontal polarizations, significantly improving the signal-to-noise ratio in DInSAR processing.

To assess its performance, the reflector was deployed on the Clot de la Menera rock glacier in Andorra. DInSAR analysis revealed subtle surface movements, indicative of an active permafrost layer. The reflector's dual-polarization capability enhances measurement accuracy and reliability by providing a brighter and more consistent measurement point.

This study underscores the potential of advanced SAR instrumentation to enhance monitoring in complex terrains where InSAR techniques does not achieve measurements.

How to cite: Monserrat, O., Espin, P., Luzi, G., and Barra, A.:  Enhancing DInSAR Measurements of Mountanious geohazards using a Novel Active Reflector for Sentinel-1 C-Band, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-10568, https://doi.org/10.5194/egusphere-egu25-10568, 2025.

The LuTan-1 (LT-1) mission is China’s first civil L-band differential interferometric SAR (D-InSAR) satellite system, comprising the 01 Group A and B satellites, which were successfully launched in 2022. LT-1 has been extensively utilized for large-scale topographic mapping, geohazard risk identification, and natural resource management. Since June 2023, the LT-1 satellites have entered the strip1 mode to acquire repeat-pass observation data for long-term ground deformation monitoring. However, the initial orbit position vectors lacked sufficient precision, and without external orbit correction data, accurate initial offset estimation between image pairs could not be achieved. This limitation rendered conventional cross-correlation-based region registration algorithms ineffective, posing significant challenges for automated SAR image registration and geocoding. Moreover, long-baseline data introduced registration noise errors, further reducing observation accuracy. To address these challenges, we implemented a neural network-based feature point matching technique to estimate the initial offset between SAR image pairs. Additionally, a block-based registration approach was adopted to suppress registration noise. These methods were applied to the D-InSAR data  processing of the Ji Shishan, Gansu earthquake (Mw 6.2) on December 18, 2023. The results demonstrate that our approach successfully achieved accurate and automated region registration and geocoding while improving interferometric coherence and phase quality.

How to cite: Tan, Y. and Hu, J.: A Neural Network-Based Block Region SAR Image Registration Algorithm, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-7514, https://doi.org/10.5194/egusphere-egu25-7514, 2025.

Multi-temporal synthetic aperture radar interferometry (SAR, MT-InSAR) has been widely recognized as an effective technique for monitoring surface deformation and marking a significant advancement in satellite geodesy to millimeter-level precision. As one of the most representative MT-InSAR methods, permanent scatterer interferometry (PS, PSI) focuses on the most elite pixels over the temporal and spatial scales of SAR images. The selection of PS points is the cornerstone of the excellent performance of PSI, directly influencing the accuracy and density of surface deformation products. Most traditional methods use thresholds to divide PS and non-PS pixels, and their results will no longer be accurate when the surface deformation patterns deviate from the prior model. Benefiting from the development of deep learning, data-driven methods have been widely proposed in recent years and exhibit superior efficiency. However, existing approaches do not fully exploit the contextual relationships between phase, amplitude, time, and spatial dimensions. This will result in the selected PS points showing representative only in certain dimensions.

Therefore, this paper proposed a novel deep learning method for PS selection that leverages the temporo-spatial context features of amplitude images and interferometric phase. Specifically, a pseudo-Siamese temporo-spatial vision transformer (ViT) architecture is employed to process input amplitude and phase time-series stacks simultaneously. In the backbone, the positional information is incorporated into the image tokens via the temporal and spatial embedding layers, and the local features in the context of the time series images are derived by the temporal and spatial encoder. Through a feature fusion module, multi-scale features from amplitude and phase are synergistically integrated. Then, it is output to the decoding head, and the high-quality PS points are predicted pixel by pixel through a multilayer perceptron.

The proposed model was trained on a dataset containing time-series SAR amplitude images and interferometric phase stacks of Barcelona, acquired by the TerraSAR between 2009 and 2011. The dataset includes 8,689 samples for training and 965 samples for validation, with data pre-processing and PS annotation performed using the SUBSIDENCE software from the Universitat Politècnica de Catalunya. To address the class imbalance between PS and non-PS points, the focal loss function was employed. The proposed model was evaluated using metrics like intersection over union (IoU), F1-score, precision, and recall. PS points selected by our method are validated via qualitative and quantitative comparisons against other state-of-the-art methods.

Experimental results indicate that the proposed method markedly improves the density, precision, and phase integrity of PS points. Compared to traditional methods, the proposed model yields more complete and continuous PS point details on buildings and man-made infrastructure, reduces false points, and improves computational efficiency. Additionally, the proposed method performs robustly across diverse land types and is extendable to distributed scatterer (DS) pixel selection. All model codes and training configurations will be available at https://dagshub.com/zhangyfcsu/pssformer.

How to cite: Zhang, Y., Mallorqui, J. J., Wang, W., Qiu, Y., Chen, Y., and Liu, L.: PSSformer: Permanent Scatterers Selection Method for SAR Interferometry based on Temporo-Spatial Vision Transformers, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-4530, https://doi.org/10.5194/egusphere-egu25-4530, 2025.

Fault movements, even when gradual and subtle, can significantly impact the stability of urban infrastructures, posing challenges for construction. This study uses Interferometric Synthetic Aperture Radar (InSAR) to monitor and analyze ground deformation affected by fault activity in several urban areas undergoing constant development: Silver Creek Fault (California, USA), Santa Monica Fault (California, USA), the Para Fault Zone (Adelaide, Australia), and a fault located in the Canary Wharf area, in the city of London (UK).

In urban environments, monitoring surface motion and ground stability is critical due to high population density and complex land use, which increases vulnerability in the area. Traditional in-situ monitoring approaches face a challenge when analyzing large areas and moreover to detect ground displacement movements over a larger area. In this regard, satellite remote sensing techniques offer an advantage to analyze fault-related ground displacement across entire cities or large urban areas due to the large spatial coverage of satellite imagery compared to only-ground instrumentation traditional methods.

Specifically, InSAR offers a reliable, non-intrusive approach for detecting subtle fault-related movements. When utilizing high-resolution sensors, this technique effectively evaluates ground displacement with millimetric precision (1–2 mm) and achieves geolocation accuracy within metrics scales (1–2 m). This allows the analysis of the fault-related ground deformation in detail, even at the scale of a single infrastructure.

The different case studies presented in this work show the effectiveness of InSAR not only to identify faults and their impact on urban areas, but also to quantify ground movements linked to fault areas, this is movements not only caused by fault displacement but affected by them, such as deformation caused by groundwater extraction. The study also shows how fault zones may affect these deformations by either amplifying or physically limiting them.

How to cite: Giralt, A., Reyes, J., Camafort, M., Palanisamy, S., Amherdt, S., Pallarès, J., Urricelqui, C., Garmendia, A., Albiol, D., and Devanthéry, N.: InSAR to monitor fault-related ground movement: An effective approach for urban environments, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-18381, https://doi.org/10.5194/egusphere-egu25-18381, 2025.

On January 24, 2020, the Mw 6.8 Elazığ earthquake occurred on the Eastern Anatolian Fault (EAF) at the indentation zone where the Arabian Plate converges with the Anatolian Plate. It was one of the largest earthquakes on the EAF in the last century before the devastating February 6, 2023, Mw 7.8 and Mw 7.6 doublet earthquakes, separated by ~9 hours. The 2023 Mw 7.8 Kahramanmaraş mainshock propagated along a splay of the EAF, while the 2020 Elazığ earthquake originated near Lake Hazar and propagated southeast to the northern termination of the Pütürge segment. These events suggest the Pütürge segment remained locked during the 2020 and 2023 earthquakes.

This study analyzes the pre-seismic and postseismic deformation of the 2020 Mw 6.8 Elazığ earthquake using Sentinel-1 SAR interferometry to assess the seismic potential of the ~40 km long Pütürge segment and the northeastern EAF zone. We employ small baseline (SBAS) inversion algorithms to analyze time series data from ascending tracks (AT116, AT43) and descending tracks (DT123, DT21), using 402, 96, and 321 interferograms, respectively, for the postseismic phase, and ~1100 interferograms for the pre-seismic phase (2015–2020). We process geocoded unwrapped interferograms, correct errors, reduce tropospheric phase delays using ECMWF ERA5 products and estimate average velocities.

Our results reveal postseismic creep of up to ~25 mm/yr propagating towards the Pütürge segment, while minimal creep was observed in the descending track during the pre-seismic phase of the Mw 6.8 Elazığ earthquake. The faults responsible for the February 6, 2023, Mw 7.8 and Mw 7.6 earthquakes remained locked during this period. This geodetic analysis provides critical insights into the interseismic and postseismic coupling of the Pütürge segment within the EAF zone.

How to cite: Javed, M. T., Barbot, S., and Braitenberg, C.: Seismic Potential and Creep Analysis of the Pütürge Segment (Eastern Anatolian Fault Zone): Insights from SAR Interferometry for the 2020 Mw 6.8 Elazığ and 2023 Mw 7.8 Kahramanmaraş Earthquakes, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-18055, https://doi.org/10.5194/egusphere-egu25-18055, 2025.

The distribution of slow-moving landslides has significance in landscape modification and hazard assessment. Monitoring such landslides is more challenging because of their spatial and temporal variability, particularly in tectonically active regions. In these regions, structural discontinuities exert significant control, requiring risk assessment at both local and regional aspects. The Karakoram-Hindu Kush-Himalaya (KHH), i.e., the orogenic belt in northern Pakistan, is a natural laboratory for studying the geological hazards due to its neotectonism, high seismicity, and diverse precipitation patterns. This region encompasses a complex geological framework, including the Nanga Parbat Syntaxis, Hazara-Kashmir Syntaxis, Hunza fault system, Tirich Mir Suture Zone, Shyok Suture, and the Indus-Kohistan Suture zones. These structures prompt various land sliding activities, yet inventories of slow-moving landslides remain scarce in northern Pakistan. The region is also traversed by the China-Pakistan Economic Corridor (CPEC) through the Karakoram Highway (KKH), an old Silk Road, where landslides severely threaten infrastructure and transportation. Here, we report our recent work regarding the actively slow-moving landslide distribution in this 19350 km² region. We combine the InSAR phase-gradient stacking technique with a deep learning-based YOLOv3 network to detect localized deformation from thousands of wrapped interferograms. Analyzing eight years of Sentinel-1 data (2016-2024), we detected and mapped 1,066 active slow-moving landslides in the Hazara Kashmir region. Further, we extended this analysis to the Khunjerab-Chitral alternate route of the CPEC, detecting 859 active landslides along this corridor. Several large, rapidly moving landslides were also recognized, posing significant risks to underlying villages and the route’s stability. These results are validated using optical imageries and field observations to create the first comprehensive inventory of slow-moving landslides in northern Pakistan. Validation against geomorphological features, published landslides, and field observation confirmed an overall precision of 87%, with detected targets corresponding to landslide features, while 13% were classified as false detections. This study underscores the critical need for monitoring and managing geological hazards in this rapidly uplifting tectonic region.

How to cite: Ahmad, S. M. and Teng, W.: InSAR-Derived Localized Deformation and Slow-Moving Landslide Inventory in Northern Pakistan, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-3486, https://doi.org/10.5194/egusphere-egu25-3486, 2025.

Raniganj, India is a well-known coal mining region characterized by high coal productivity and ongoing land subsidence. Land subsidence can be due to various factors such as mining activities, coal fire, total water storage change, atmospheric loading, oceanic loading, groundwater over-extraction, etc., but mining activities in the region are accredited to be one of the major sources of land subsidence. Despite its complex hydrological environment, where significant contributions arise from surface and subsurface water systems linked to the Ganges River system and proximity to the Bay of Bengal, non-mining factors' role in regional deformation patterns has not been thoroughly investigated. This study attempts to identify the potential sources of the ongoing subsidence in the region using various Earth Observation and Global Positioning Station (GPS) datasets. The deformation pattern of the area was analyzed using ground-based GPS measurements and the interferometric SAR (InSAR) technique with Sentinel-1 Synthetic Aperture Radar data. Seasonal variations in deformation, including pre-monsoon, co-monsoon, and post-monsoon periods, were assessed using total water storage (TWS) changes from the Gravity Recovery and Climate Experiment (GRACE) and Global Land Data Assimilation System (GLDAS) datasets. However, GRACE's coarser resolution and data gaps posed challenges for finer-scale interpretation. To address this, high-resolution datasets such as precipitation, Normalized Difference Vegetation Index (NDVI), and land surface temperature data were utilized in conjunction with Artificial Intelligence (AI) and Machine Learning (ML) techniques to downscale GRACE-derived TWS data, enabling higher-resolution insights into groundwater variability. This comprehensive approach provides a deeper understanding of the causative factors of land deformation in the region, especially the interactions between groundwater changes and other environmental variables. Such insights are crucial for informed land use and planning in this economically and environmentally sensitive region.

How to cite: Ghosh, D., Yadav, M., Yadav, A. K., Yadav, A. K., Kannaujiya, S., and Roy, P. N. S.: Integrated Analysis of Land Subsidence in the Raniganj Coal Mining Region, India Using Multi-Source Earth Observation and AI-Enhanced Techniques, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-12540, https://doi.org/10.5194/egusphere-egu25-12540, 2025.

For being able to accomplish the Sustainable Development Goals is needed to understand the dynamic of the ecosystems where the human activities are developed. For these reasons to understand the baseline and monitoring the environmental variables is fundamental. Earth Observation plays a key role in the management of the anthropic activities.

In the recent years, Lithium become one of the key raw resources to accomplish the Sustainable Development Goals, because is the main component for avoid the use of fossil fuel. Atacama Desert is one of the main places where Lithium is extracted, but also is a fragile ecosystem, due to the presence of endangered fauna (Flamingos) and highly specialized communities of organisms (extremophile microorganisms for example).

It´s been considered the Laguna Tebenquinche for analysing the impact of the land use and land cover change, and its impact on ground deformation (Persistent Scatterer Interferometry) from Sentinel 1. The last one was computed using the CTTC´s processing chain. Vegetation dynamics, water presence and soil moisture has been obtained using NDVI, NDWI and NDMI indexes from Sentinel 2 data (level L2, S2_SR_HARMONIZED) from Google Earth Engine. Both analyses considered the period between 2022 to 2024.

The first analyses were performing the identification of the presence or absence of persistent scatterers and their relationship to the land cover. Then it was conducted an analysis of related to the ground deformation´s mean velocity and the effects of the surface dynamics from Sentinel 2. These results were compared with precipitation rates, temperature of the air, air moisture and underground water levels in the salt flat.

The results show a difference between the northern area of the lagoon, which have a mean ground deformation velocity between -5 to -2 mm/yr, versus the southern part, which has a mean ground deformation velocity between -5 to 2 mm/yr. For the coverage, the northern part of the lagoon has been flooded temporarily and with increase of soil moisture.

For the precipitation, from the end of November 2022 the rain in the salt flat (LZA12-3 station) increase, but it’s now only needed to consider the rain in the salt flat, also the rain in the upper part of the catchment. The Cerro Cosor Station shows a high value of precipitation since January 2024 to April 2024. The water surface and vegetation surface show a relationship with the precipitation pattern, but there is no direct relationship with the soil moisture time series.

Seasonal analysis of surface coverage could help to improve the understanding of the dynamic of the temporal cinematic of the Persistent Scatterers. The integration of Earth Observation helps us to understand and model relationships between climatic events (like ENSO), the hydrological dynamic of the lagoon, connections between the lagoons and the aquifers, evaluation of possible overexploitation of groundwater and saline intrusion, impacts of the climate change on the ecosystem and conditions for the local flora and fauna.

How to cite: Olea-Encina, P., Ramlie, M. C., Crosetto, M., and Monserrat, O.: Relationship between ground deformation time series and coverage dynamics: laguna Tebenquinche case study, Salar de Atacama, Chile, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-19864, https://doi.org/10.5194/egusphere-egu25-19864, 2025.

Portugal's coastal regions face significant challenges due to climate change, with potential GDP reductions of 2% to 5% by 2100, primarily driven by erosion and sea-level rise. The dense occupation of coastal areas increases vulnerability, underscoring the need for detailed studies to model future scenarios and implement mitigation strategies. The Atlantic oceanographic forcing impinges the soft sediment coastline and causes further stress on the erosion-prone coastal stretches.

This project aims to assess coastal erosion along Portugal's western edge, focusing on areas such as Quiaios, Cova Gala, São Pedro de Moel, and the Pedrogão dunes. Coastal retreats are analyzed using the InSAR (Interferometric Synthetic Aperture Radar) technique, complemented by traditional Earth observation methods and topo-bathymetric data to refine this methodology.

The research also employs the PSInSAR (Persistent Scattering Interferometry Synthetic Aperture Radar) technique to study subsidence in rocky coastal areas and evaluate risks. It is also applied to monitor spurs and coastal vegetation, analyzing its relationship with erosion processes.

The PSI technique was chosen for its ability to provide precise measurements of ground displacements, making it effective for beach monitoring and reducing atmospheric noise. By processing InSAR data, it enables millimeter-scale measurements of ground displacement along the satellite’s line of sight, using a point cloud of persistent scattering (PS) elements.

Besides long-term trends, detailed focus will be on determining impacts of coastal storms on the sediment dynamics and resilience capacity of the studies coastal systems. Results will also contribute to the establishment of high-resolution erosion rates which will allow better coastal planning.

This work is supported by the Portuguese Fundação para a Ciência e Tecnologia, FCT, I.P./MCTES, through national funds (PIDDAC): UID/50019/2025 and LA/P/0068/2020 (https://doi.org/10.54499/LA/P/0068/2020).

How to cite: Nunes, A. and Costa P. J. M., P.: Monitoring Coastal Erosion and Subsidence on the western coast of Portugal using PSInSAR and InSAR, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-18798, https://doi.org/10.5194/egusphere-egu25-18798, 2025.