Machine learning approaches for predicting land subsidence in Ca Mau: XGBoost, Random Forest, and MAF

The damage to pipeline infrastructures caused by reactive soils has been a critical challenge for asset owners. Sustainable backfilling materials have recently gained interest to stabilize highly reactive zones as a pre-emptive approach towards sustainability. In this study, two adjacent sections of a sewer pipeline trench in Melbourne, Australia were backfilled with two blends of 100% recycled aggregates. The sites were monitored for ground deformations during October 2020–February 2022 (17 months) using surveying techniques. Interferometric synthetic aperture radar (InSAR) techniques and algorithms were also employed to estimate the ground movements of the sites and surrounding regions. The cross-validation of deformation results achieved from both techniques enabled an in-depth analysis of the effectiveness of the recycled aggregates to address reactive soil issues in urban developments. Observational deformation data and their spatiotemporal variation in the field were satisfactorily captured by the InSAR techniques: differential InSAR (DInSAR), persistent scatterer interferometry (PSI), and small baseline subset (SBAS). The SBAS estimations were found to be the closest to field measurements, primarily due to the analysis of zones without well-defined geometries. This study’s contribution to existing knowledge defines the spatiotemporal influence of sustainable backfill in areas with reactive soil through field data and satellite imaging. The relationship between InSAR techniques and actual field behavior of sustainable backfill can be a baseline for the growing construction that may be challenging to perform field monitoring due to resource constraints.

Persistent scatterer interferometry (PSI) has been proved to be an advanced Interferometric Synthetic Aperture Radar (InSAR) technique used to measure and monitor terrain deformation. Two of the critical problems with InSAR have been the effect of the refractive atmosphere and decorrelation on the interferometric phases due to long spatial-temporal baseline. The low density of persistent scatterers (PS) in non-urban areas affected by spatial-temporal decoherence more seriously has inspired the development of alternative approaches. Sentinel-1 (S1) has improved the data acquisition throughout, and compared to previous sensors, increased considerably the differential interferometric SAR (DInSAR) and PSI deformation monitoring potential. This paper describes an innovative methodology to process S1 SAR data. Different with PSI, its most original part includes two key processing stages: high and low frequency splitting from wrapped phases, prior to atmospheric filtering, and final direct integration to generate the complete deformation with time series containing linear and nonlinear components. The proposed method has two fundamental advantages compared with traditional PSI approach: the final monitoring results with excellent coverage of coherent points and the generation of active maps even for the areas with serious deformation in short term to break through the inherent limitation of PSI. The effectiveness of the proposed tools is illustrated using a case study located in Catalonia (Spain). This methodology has supposed a definitive step towards the implementation of DInSAR based techniques to support decision makers against geohazards. In this work, the deformation procedures happened in three different areas of the Catalonia (Spain) are presented and analysed. The maximum accumulated subsidence of over – 60 cm induced by mining activity can be detected by proposed methodology with nice coverage from January 2017 to January 2019. These reported cases illustrate how DInSAR based techniques can provide detailed terrain deformation for geohazard activity with complex topographical conditions. The active deformation areas map can be generated in fast aimed at geohazard risk early warning and management.

This Special Issue hosts papers related to deformation monitoring in urban areas based on two main techniques: Persistent Scatterer Interferometry (PSI) and Synthetic Aperture Radar (SAR) Tomography (TomoSAR). Several contributions highlight the capabilities of Interferometric SAR (InSAR) and PSI techniques for urban deformation monitoring. In this Special Issue, a wide range of InSAR and PSI applications are addressed. Some contributions show the advantages of TomoSAR in un-mixing multiple scatterers for urban mapping and monitoring. This issue includes a contribution that compares PSI and TomoSAR and another one that uses polarimetric data for TomoSAR.

Landslides are the most catastrophic geological hazard in hilly areas. The present work intends to identify landslide susceptibility along Karakorum Highway (KKH) in Northern Pakistan, using landslide susceptibility mapping (LSM). To compare and predict the connection between causative factors and landslides, the random forest (RF), extreme gradient boosting (XGBoost), k nearest neighbor (KNN) and naive Bayes (NB) models were used in this research. Interferometric synthetic aperture radar persistent scatterer interferometry (PS-InSAR) technology was used to explore the displacement movement of retrieved models. Initially, 332 landslide areas alongside the Karakorum Highway were found to generate the landslide inventory map using various data. The landslides were categorized into two sections for validation and training, of 30% and 70%. For susceptibility mapping, thirteen landslide-condition factors were created. The area under curve (AUC) of the receiver operating characteristic (ROC) curve technique was utilized for accuracy comparison, yielding 83.08, 82.15, 80.31, and 72.92% accuracy for RF, XGBoost, KNN, and NB, respectively. The PS-InSAR technique demonstrated a high deformation velocity along the line of sight (LOS) in model-sensitive areas. The PS-InSAR technique was used to evaluate the slope deformation velocity, which can be used to improve the LSM for the research region. The RF technique yielded superior findings, integrating with the PS-InSAR outcomes to provide the region with a new landslide susceptibility map. The enhanced model will help mitigate landslide catastrophes, and the outcomes may help ensure the roadway’s safe functioning in the study region.

Landslides cause significant socioeconomic impacts on people and national infrastructures like railways and roads and are considered one of the common geohazards that demand more attention. In Sweden, many areas are prone to landslides due to the presence of underlying quick-clay sediments, which may lead to minor to large slides. Ground deformation monitoring in such hazardous areas is important for a better understanding of the landslide processes and mitigation of hazards. Over the last decade, Interferometric Synthetic Aperture Radar (InSAR) time-series techniques, such as Persistent Scatterer Interferometry (PSI) and the Small Baseline Subset (SBAS), have become a crucial tool for ground surface deformation analysis. SBAS and PSI use SAR data to retrieve the time-series cumulative phase of the Persistent Scatters (PS). The main objective of this study was to demonstrate the advantage of using advanced InSAR time series analysis for a better understanding of surface deformation before a landslide event. We focused on the recent landslide on the E6 Sweden-Norway highway near Stenungsund in Southern Sweden, which occurred on 23 September 2023. Sentinel-1 SAR data was collected between 2018 and September 2023, with ascending flight direction to measure the pre-event deformation in the landslide zone. We used Alaska Facility (ASF) on-demand product processes based on Hybrid Pluggable Processing Pipeline (Hyp3) to search, process, and download time series Sentinel-1 data. We also used Miami INsar Time-series software in PYthon (Mintpy) to perform cloud-based SBAS processing using unwrapped interferograms stack derived from Sentinel-1 time series data. In addition, we employed Basic PSI products (ground motion in the Line-of-Sight (LOS) direction) provided by the European Ground Motion Service (EGMS). The initial SBAS results and EGMS Basic products for the same ascending orbit showed continuous deformation on the highway segment in the landslide zone over the last EGMS update period, 2018 to 2022 for the PSI results and 2018 to 2023  for the SBAS results. The first five-year period of the EGMS Basic and Ortho products, i.e., 2015-2021, was also checked and showed the same results over the longer period between 2015 and 2021. Both sets of PSI and SBAS results agree on the annual cm-level (10-15 mm/year) subsidence rate of the highway before the landslide, with SBAS analysis yielding more measurement points, especially in the vegetated and unbuilt areas. The initial results showed that the SBAS technique could provide more information within the hazardous zone; nevertheless, due to Sentinel-1 C-band data, the quality of the results can be degraded by coherence variations in the vegetated areas.  The comparison of preliminary results of InSAR data processing and available EGMS products provides insights into ground movements, facilitating a comprehensive understanding of evolving conditions before the landslide. Nevertheless, the results emphasize the importance of incorporating advanced time series InSAR techniques for continuously monitoring infrastructures such as railroads and highways to support sustainable development and natural hazard assessments.

Persistent Scatterer Interferometry: A review

Correct use of multi-temporal Interferometric Synthetic Aperture Radar (InSAR) datasets to complement geodetic surveying for geo-hazard applications requires rigorous assessment of their precision and accuracy. Published inter-comparisons are mostly limited to ground displacement estimates obtained from different algorithms belonging to the same family of InSAR approaches, either Persistent Scatterer Interferometry (PSI) or Small BAseline Subset (SBAS); and accuracy assessments are mainly focused on vertical displacements or based on few Global Navigation Satellite System (GNSS) or geodetic leveling points. To fill this demonstration gap, two years of Sentinel-1 SAR ascending and descending mode data are processed with both PSI and SBAS consolidated algorithms to extract vertical and horizontal displacement velocity datasets, whose accuracy is then assessed against a wealth of contextual geodetic data. These include permanent GNSS records, static GNSS benchmark repositioning, and geodetic leveling monitoring data that the National Institute of Statistics, Geography, and Informatics (INEGI) of Mexico collected in 2014−2016 in the Aguascalientes Valley, where structurally-controlled land subsidence exhibits fast vertical rates (up to −150 mm/year) and a non-negligible east-west component (up to ±30 mm/year). Despite the temporal constraint of the data selected, the PSI-SBAS inter-comparison reveals standard deviation of 6 mm/year and 4 mm/year for the vertical and east-west rate differences, respectively, thus reassuring about the similarity between the two types of InSAR outputs. Accuracy assessment shows that the standard deviations in vertical velocity differences are 9−10 mm/year against GNSS benchmarks, and 8 mm/year against leveling data. Relative errors are below 20% for any locations subsiding faster than −15 mm/year. Differences in east-west velocity estimates against GNSS are on average −0.1 mm/year for PSI and +0.2 mm/year for SBAS, with standard deviations of 8 mm/year. When discrepancies are found between InSAR and geodetic data, these mostly occur at benchmarks located in proximity to the main normal faults, thus falling within the same SBAS ground pixel or closer to the same PSI target, regardless of whether they are in the footwall or hanging wall of the fault. Establishing new benchmarks at higher distances from the fault traces or exploiting higher resolution SAR scenes and/or InSAR datasets may improve the detection of the benchmarks and thus consolidate the statistics of the InSAR accuracy assessments.

Landslide classification and identification along Karakorum Highway (KKH) is still challenging due to constraints of proposed approaches, harsh environment, detail analysis, complicated natural landslide process due to tectonic activities, and data availability problems. A comprehensive landslide inventory and a landslide susceptibility mapping (LSM) along the Karakorum Highway were created in recent research. The extreme gradient boosting (XGBoost) and random forest (RF) models were used to compare and forecast the association between causative parameters and landslides. These advanced machine learning (ML) models can measure environmental issues and risks for any area on a regional scale. Initially, 74 landslide locations were determined along the KKH to prepare the landslide inventory map using different data. The landslides were randomly divided into two sets for training and validation at a proportion of 7/3. Fifteen landslide conditioning variables were produced for susceptibility mapping. The interferometric synthetic aperture radar persistent scatterer interferometry (PS-InSAR) technique investigated the deformation movement of extracted models in the susceptible zones. It revealed a high line of sight (LOS) deformation velocity in both models’ sensitive zones. For accuracy comparison, the area under the curve (AUC) of the receiver operating characteristic (ROC) curve approach was used, which showed 93.44% and 92.22% accuracy for XGBoost and RF, respectively. The XGBoost method produced superior results, combined with PS-InSAR results to create a new LSM for the area. This improved susceptibility model will aid in mitigating the landslide disaster, and the results may assist in the safe operation of the highway in the research area.

The time-series interferometric synthetic aperture radar (InSAR) techniques has the potential of estimating the land subsidence at a high accuracy of nearly in millimeter scale. The major limitation of the present InSAR methods like persistent scatterer interferometry (PSI) is that they measure land deformation velocities only in the direction of the radar line of sight (LOS). Hence, the real prospects of the deformation velocity, i.e. the horizontal and vertical components of velocity remain uncertain. This study is focusing on retrieving PSI based multi-dimensional land deformation velocity along horizontal and vertical directions in an urban region by using ascending and descending pass Sentinel-1 datasets. The area for this study is Lucknow city in Uttar Pradesh state of India. The estimated LOS velocity vectors from ascending and descending pass time-series datasets are utilized for retrieving multidimensional subsidence velocity. Further, the retrieved time-series land subsidence is correlated to the groundwater level variation during the considered time span.

<p>Cultural heritage is a key element of history as the ancient monuments and archaeological sites enrich today’s societies and help connect us to our cultural origins. The project entitled ''SpaCeborne SAR Interferometry as a Nonivasive tool to assess the vulnerability over Cultural hEritage sites (SCIENCE)'' has as ultimate objective to predict the vulnerability of the archaeological sites to ground deformation in time and space and protect them against natural/man-made damage. The SCIENCE project aims to develop, demonstrate, and validate, in terms of geotechnical local conditions and monuments’ structural health, SAR interferometric techniques to monitor potential ground deformation affecting the archaeological sites and monuments of great importance. </p><p>During the last few years, spaceborne Synthetic Aperture Radar (SAR) interferometry has proven to be a powerful remote sensing tool for detecting and measuring ground deformation and studying the deformation’s impact on man-made structures. It provides centimeter to millimeter resolution and even single buildings/monuments can be mapped from space. Considering the limitations of conventional MT-InSAR techniques, such as Persistent Scatterers Interferometry (PSI), in this project a two-step Tomography-based Persistent Scatterers (PS) Interferometry (Tomo-PSInSAR) approach is proposed for monitoring ground deformation and structural instabilities over the Ancient City Walls (Ming Dynasty) in Nanjing city, China and in the Great Wall in Zhangjiakou, China. The Tomo-PSInSAR is capable of separating overlaid PS in the same location, minimizing the unfavorable layover effects of slant-range imaging in SAR data. Moreover, the demonstrations are performed on well-known test sites in China and in Greece, such as: a) Ming Dynasty City Walls in Nanjing, b) Great Wall in Zhangjiakou, c) Acropolis complex of Athens and d) Heraklion walls (Crete Island), respectively.</p><p>In particular, in the framework of SCIENCE project are processed several radar datasets such as Sentinel 1 A & B data of Copernicus program and the high resolution TerraSAR-X data. The products of Persistent Scatterers Interferometry (PSI) are exported in various formats for the identification of the persistent scatterers using high resolution optical images, aerial photographs and fusing with high accuracy Digital Surface Models (DSM). In addition, the validation of the results is taking place through in-situ measurements (geological, geothechnical e.t.c) and data for the cultural heritage sites conditions.</p><p>SCIENCE project’s final goal is the risk assessment analysis of the cultural heritage monuments and their surrounding areas aiming to benefit institutions, organizations, stakeholders and private agencies in the cultural heritage domain through the creation of a validated pre-operation non-invasive system and service based on earth observation data supporting end-user needs by the provision knowledge about cultural heritage protection. In conclusion, SCIENCE project is composed by a bilateral consortium of the Greek delegation of Harokopio University of Athens, National Technical University of Athens, Terra Spatium S.A, Ephorate of Antiquities of Heraklion (Crete), Acropolis Restoration Service (Athens) of Ministry of Culture and Sports and by the Chinese delegation of Science Academy of China (Institute of Remote Sensing and Digital Earth) and  International Centre on Space Technologies for Natural and Cultural Heritage (HIST) under the auspices of UNESCO (HIST-UNESCO).</p>

Landslides are a major natural hazard which can cause significant damage, economic loss, and loss of life. Between the years of 2004 and 2016, 55,997 fatalities caused by landslides were reported worldwide. Up-to-date, reliable, and comprehensive landslide inventories are mandatory for optimized disaster risk reduction (DRR). Various stakeholders recognize the potential of Earth observation techniques for an optimized DRR, and one example of this is the Sendai Framework for DRR, 2015–2030. Some of the major benefits of spaceborne interferometric Synthetic Aperture Radar (SAR) techniques, compared to terrestrial techniques, are the large spatial coverage, high temporal resolution, and cost effectiveness. Nevertheless, SAR data availability is a precondition for its operational use. From this perspective, Copernicus Sentinel-1 is a game changer, ensuring SAR data availability for almost the entire world, at least until 2030. This paper focuses on a Sentinel-1-based Persistent Scatterer Interferometry (PSI) post-processing workflow to classify landslide activity on a regional scale, to update existing landslide inventories a priori. Before classification, a Line-of-Sight (LOS) velocity conversion to slope velocity and a cluster analysis was performed. Afterwards, the classification was achieved by applying a fixed velocity threshold. The results are verified through the Global Positioning System (GPS) survey and a landslide hazard indication map.

Urban infrastructures have become imperative to human life. Any damage to these infrastructures as a result of detrimental activities would accrue huge economical costs and severe casualties. War in particular is a major anthropogenic calamity with immense collateral effects on the social and economic fabric of human nations. Therefore, damaged buildings assessment plays a prominent role in post-war resettlement and reconstruction of urban infrastructures. The data-analysis process of this assessment is essential to any post-disaster program and can be carried out via different formats. Synthetic Aperture Radar (SAR) data and Interferometric SAR (InSAR) techniques help us to establish a reliable and fast monitoring system for detecting post-war damages in urban areas. Along this thread, the present study aims to investigate the feasibility and mode of implementation of Sentinel-1 SAR data and InSAR techniques to estimate post-war damage in war-affected areas as opposed to using commercial high-resolution optical images. The study is presented in the form of a survey to identify urban areas damaged or destroyed by war (Islamic State of Iraq and the Levant, ISIL, or ISIS occupation) in the city of Mosul, Iraq, using Sentinel-1 (S1) data over the 2014–2017 period. Small BAseline Subset (SBAS), Persistent Scatterer Interferometry (PSI) and coherent-intensity-based analysis were also used to identify war-damaged buildings. Accuracy assessments for the proposed SAR-based mapping approach were conducted by comparing the destruction map to the available post-war destruction map of United Nations Institute for Training and Research (UNITAR); previously developed using optical very high-resolution images, drone imagery, and field visits. As the findings suggest, 40% of the entire city, the western sectors, especially the Old City, were affected most by ISIS war. The findings are also indicative of the efficiency of incorporating Sentinel-1 SAR data and InSAR technique to map post-war urban damages in Mosul. The proposed method could be widely used as a tool in damage assessment procedures in any post-war reconstruction programs.

Climate change represents an important cause of subsidence, especially in coastal cities affected by changes in surface water level and water table. This paper presents a complementary study of Interferometric Synthetic Aperture Radar (InSAR) and Ground Penetrating Radar (GPR) for the early detection of subsidence and sinkhole phenomena. The methodology was applied to a coastal urban area in Galicia, northwest Spain (humid region), showing apparent signs of subsidence and building settlement during the last two years. Two different InSAR methods are compared for the period from June 2021 to March 2022: PSI (Persistent Scatterer Interferometry) and SBAS (Small Baseline Subsets), and the average deformation velocities obtained resulted in −3.0 mm/yr and −4.1 mm/yr, respectively. Additional GPR data were collected in January 2022 to validate the InSAR results, which detected subsidence in agreement with the persistent scatters obtained from the PSI method. This is crucial information to plan preventive maintenance.

Interferometric synthetic aperture radar (InSAR) techniques, such as persistent scatterer interferometry (PSI) or SAR tomography (TomoSAR), deliver three-dimensional (3-D) point clouds of the scatterers' positions together with their motion information relative to a reference point. Due to the SAR side-looking geometry, minimum of two point clouds from cross-heading orbits, i.e., ascending and descending, are required to achieve a complete monitoring over an urban area. However, these two point clouds are usually not coregistered due to their different reference points with unknown 3-D positions. In general, no exact identical points from the same physical object can be found in such two point clouds. This article describes a robust algorithm for fusing such two point clouds of urban areas. The contribution of this paper is finding the theoretically exact point correspondence, which is the end positions of façades, where the two point clouds close. We explicitly define this algorithm as “L-shape detection and matching,” in this paper, because the façades commonly appear as L-shapes in InSAR point cloud. This algorithm introduces a few important features for a reliable result, including point density estimation using adaptive directional window for better façade points detection and L-shape extraction using weighed Hough transform. The algorithm is fully automatic. Its accuracy is evaluated using simulated data. Furthermore, the proposed method is applied on two TomoSAR point clouds over Berlin with ascending and descending geometry. The result is compared with the first PSI point cloud fusion method (S. Gernhardt and R. Bamler, “Deformation monitoring of single buildings using meter-resolution SAR data in PSI,” ISPRS J. Photogramm. Remote Sens., vol. 73, pp. 68-79, 2012.) for urban area. Submeter consistency is achieved.

Persistent scatter interferometry (PSI) based algorithms are the conventional tools used to aid in the detection of constant ground motion in a set of radar images within a concrete time span. Despite the fact that slight variations in the ground motion trend exist, the conventional linear PSI technique, used on a regular basis worldwide, is able to retrieve these variations incorporating its contribution to the general linear trend of the time series (TS). However, when the ground motion does not have linear dependence on time — for instance, in the case of different trends and even strong variations, such as the alternation of periods of heave and subsidence — the linear deformation model to differentiate/isolate the atmosphere is not applicable. This is because the linear method approach of decorrelation of atmosphere over time and its spatial correlation fails to comply. A new approach is therefore needed to be able to retrieve accurate measurements, irrespective of the motion being linear or nonlinear. With this new approach, a valuable, cost-effective technique to monitor motion either in civil engineering projects or in naturally occurring events with a nonlinear deformation pattern is provided. The objective of this paper is to present a new approach to the conventional methodology in order to tackle the nonlinearity problem, not only for the cases where it is not possible to apply the conventional PSI, but also for those that may benefit from the application of this new nonlinear methodology accuracy of the TS, despite the longer processing time. The methodology used to achieve this objective involves the use of conventional PSI algorithm techniques, combined with a nonlinear module which is based on the advanced differential Interferometric synthetic aperture radar (InSAR) technique with the aid of lineal-dependent models, such as error residual height (ERH) dependent on the spatial baseline of the satellite orbits and thermal compensation dependent on temperatures. In this paper, a comparison between both techniques in the same case study and with the same radar acquisitions is presented. The area chosen as a case study is a zone of the City of London, an area where nonlinear motion, triggered by tunnelling works, has been clearly discerned. Lineal processing was shown to possibly underestimate or misestimate the ground motion detected in the TS in a nonlinear scenario, while the new methodology was able to reflect when exactly the area under study was stable and when the period of motion started.