From the sky to underground: new Remote Sensing solutions ...

55

n. 135-136 – dicembre 2020

GALLERIE E GRANDI OPERE SOTTERRANEE n. 135-136 – dicembre 2020

From the sky to underground: new Remote Sensing solutions in tunnelling and excavation worksDallo spazio al sottosuolo: soluzioni innovative di telerilevamento nei lavori di scavo e costruzione di gallerie

Paolo MAZZANTI a, b *

Serena MORETTO a

Saverio ROMEO a, b

a NHAZCA S.r.l., Italyb Earth Sciences Department, Sapienza University of Rome, Italy* corresponding author: [email protected]

AbstractIn the last decades there has been a growing development of innovative techniques in the field of observational methods, for the characterization and monitoring of natural hazards and structures/infrastructures. In particular, among the emerging technologies, remote sensing approaches play a key role. SAR (Synthetic Aperture Radar) interferometry is definitely one of the most used, effective, and efficient remote sensing solutions, finding several applications both from satellite and ground-based platforms. In addition to InSAR (Interferometric SAR) techniques, the development of innovative optical imaging techniques is contributing to the spread of photo-monitoring services for the monitoring of ground and structures deformations. After a brief introduction about the observational method and geotechnical and structural monitoring, the paper describes the advantages offered by the principal remote sensing technologies in tunnelling and excavation projects. Finally, to better understand the potential of remote sensing technologies, some examples of recent applications are presented.

SommarioNell’ultimo decennio si è assistito ad un crescente sviluppo di tecniche innovative a supporto del metodo osservazionale per la caratterizzazione e il monitoraggio di rischi naturali e di strutture e infrastrutture. In particolare, tra le tecnologie emergenti, i sistemi di telerilevamento rivestono un ruolo significativo. L’interferometria SAR (Synthetic Aperture Radar) è sicuramente una tra le tecniche di telerilevamento più performanti, che trova applicazione sia da piattaforma satellitare che terrestre. Oltre alle tecniche InSAR (Interferometric SAR), lo sviluppo di algoritmi innovativi di analisi delle immagini ottiche sta aprendo all’implementazione di tecniche di foto-monitoraggio, in grado di acquisire informazioni sulle deformazioni superficiali di terreno e strutture completamente da remoto facendo uso di immagini acquisite da numerose tipologie di sensori.Dopo una breve introduzione riguardante il metodo osservazionale ed il monitoraggio geotecnico e strutturale, il lavoro descrive i benefici offerti dalle principali tecnologie di telerilevamento applicate a progetti in sotterraneo. Infine, per meglio comprendere le prospettive fornite da queste nuove metodologie, vengono presentati alcuni esempi di applicazioni recenti.

Key words: Satellite, InSAR, Photo-monitoring, Remote Sensing, Observational Method.Parole chiave: Satellite, InSAR, foto-monitoraggio, telerilevamento, metodo osservazionale.

1. Introduction

Underground activities are crucial for modern communities, from road/rail transport infrastructures to water supply networks, en-ergy production and pipelines, hazardous waste management fa-cilities, etc.Underground construction areas for infrastructures will continue to be fundamental in sustainable development strategies, balanc-ing the growing demands of urbanized areas and the environ-mental safeguard purposes. The need for underground projects generated, in recent years, a fast-growing market in continuous and increasing evolution.

Tunnelling infrastructures are essential for transportation net-works, representing one of the most useful and complex geotech-nical assets. Given their design, construction and maintenance complexity, the feasibility of a tunnel project is related to several factors, including engineering viability, environmental aspects and economic impacts. To account for the several factors influencing the success of an underground project, risk analysis & manage-ment (RA&M) strategies became an important part of under-ground design and construction activities since the late 1990s [1, 2]. The ‘‘Guidelines for Tunnelling Risk Management’’ pub-lished by the International Tunnelling Association (ITA) in 2004 [1], significantly improved the risk management practice. They

n. 135-136 – dicembre 2020

56

MAZZANTI - MORETTO - ROMEO

describe how RA&M can be implemented in different phases of a tunnelling project, from the early design to the construction phase, representing a milestone in this context.Geotechnical and structural monitoring activities become a stand-ard in the risk assessment and management strategies. In fact, underground works, both in urban and extra-urban environments, have the potential to generate considerable damages during both the construction and operation phases, such as: ground move-ments and settlements, noise and vibrations, long term changes to geology, hydrology and water resources. These damages can be mitigated by detailed modelling and design and prevented moni-toring activities. Over the last 30 years a strong increase of monitoring techniques has been observed [3]. Additionally, scientific and technological advances in the field of monitoring have led to the possibility of observing and controlling several key parameters with different technologies and instruments, thus representing valuable tools in supporting the RA&M strategies and decision processes.While traditional “contact” monitoring methods (i.e. inclinom-eters, piezometers, clinometers, total station, GNSS, etc.) are able to provide accurate measurements at specific points and times, measuring both surface and subsurface parameters, the use of remote sensing techniques greatly increased thanks to their capability to obtain continuous and highly accurate infor-mation about the ground deformation over wide areas without the requirement of in-situ installation of equipment [4]. Innova-tive remote sensing techniques such as Satellite and Terrestrial SAR (Synthetic Aperture Radar) Interferometry (InSAR) and PhotoMonitoring have proven to be powerful monitoring tools in several applications, among which: natural hazards, transpor-tation and tunnelling, energy, oil & gas, mining, glaciology and cultural heritage [5, 6, 7].After a brief introduction about the geotechnical and structural monitoring topic, the paper describes the principal remote sens-ing solutions able to provide value-added information in tunnel-ling and excavation projects, highlighting the potential of SAR monitoring techniques in four case histories.

2. Why monitoring?

The observational method was first proposed by Karl Terzaghi and Ralph Peck to achieve the design challenges, balancing the project security/safety requirements and the economic impact, knowing that each geotechnical project can be characterized by unexpected events [8, 9]. The observational method was proposed to continuously improve and review the project design during the construction phase based on the “observation” and “learning” approach (also known as “learn-as-you-go”), pursu-ing the safety conditions and avoiding an unsustainable impact on economic resources. Peck [9] proposed that construction work should be started using the most probable design; if the monitoring records exceed the predicted behaviour, predefined contingency plans would be triggered. The spread of the observational method has been associated with the development of geotechnical monitoring solutions. In

the 1970s – 1980s geotechnical monitoring become a recognized discipline in engineering projects [3, 4, 10] and several success-ful examples demonstrating the importance of the observational method in geotechnical engineering projects are now available in literature [11, 12, 13].Monitoring activities can be extremely useful in different phase of the life cycle of a tunnel: during the design, the construction and the operational phase.According to Mazzanti [4] and Brunetti [14], monitoring cam-paigns can be generally classified in 3 main categories, based on their aims:– knowledge monitoring, aimed at characterizing the area/pro-

cess under investigation supporting the identification and de-limitation of potential risks. It can be implemented through dif-ferent technologies in the design and operational phases, for standard maintenance purposes and for risk assessment and screening after a paroxysmal event for its characterization;

– control monitoring, aimed at quantitatively follow the evolu-tion of well-known problems and phenomena to be applied during the construction phase, operational and maintenance phases in correspondence of critical segments, for risk man-agement purposes and for the verification and monitoring of high risk areas;

– emergency monitoring, aimed at providing alerts when a well-known risk become unacceptable during construction and op-erational phase in very high-risk areas; it can be considered one of the most powerful non-structural risk mitigation strat-egies, preserving the operational activities and the safety re-quirements.

The most common observations and measurements in tunnelling projects can be summarize as follows:– baseline analyses;– assess the stability of the underground structure;– verify the design assumptions and mathematical models and

improve geological and numerical models;– adjust excavation activities;– identify critical sectors and control their evolution through

time;– control the influence of excavation works on surrounding ar-

eas;– infrastructure safety management.For these purposes, nowadays, several technologies can be used. The measurement of a single parameter is usually not enough; thus complex, multi-parameter, monitoring plans are typically de-veloped. The challenge of a suitable monitoring program lies in identifying the key monitoring parameters and instruments to be used. The choice of the specific technique is performed during the monitoring planning phase according to the site-specific condi-tions and monitoring purposes. Since the early 20s, the advent of remote sensing technologies has allowed implementing extensive monitoring programs, inte-grating accurate and local contact monitoring equipment with continuous ground measurements [12]. The technological ad-vance and the increasing processing capabilities allow obtaining big amount of measurements, also in near real time, contributing to the increasing use of remote sensing services.

57

n. 135-136 – dicembre 2020

FROM THE SKY TO UNDERGROUND: NEW REMOTE SENSING SOLUTIONS IN TUNNELLING AND EXCAVATION WORKS

3. Why Remote Sensing?

Remote sensing techniques represent a powerful tool for ground deformation monitoring during underground excavation works, offering a wide spectrum of monitoring solutions in the design, construction, operational and maintenance phases. They can be successfully applied in several crucial operations, such as: i) route planning; ii) investigating the disturbance caused by tun-nel excavation; iii) detection and control of geohazard-prone areas; iv) structural health monitoring (SHM) of surrounding buildings/structures; v) risk management and risk-informed decision-making. The main advantages in using remote sensing technologies (both satellite and ground-based), are the following:– performing historical analysis to be used as baseline informa-

tion (for satellite-based);– achieving a panoramic view of the area of interest, implement-

ing cost effective big area analysis;– reducing interaction with monitored area;– monitoring remote/inaccessible areas; – increasing the spatial information density of measurement

points;– identifying and understanding unknown processes.Among the most used remote sensing technologies, Satellite and Terrestrial SAR Interferometry represent unique solutions for the monitoring of ground and structures surficial displacements, thanks to the capability to provide information about the ongoing and past deformations with millimeter accuracy [5, 15, 16, 17]. The main benefits of satellite SAR interferometry [18,19,20] are the capability of monitoring large areas with high accuracy and the peculiarity to perform historical analyses thanks to archive satellite SAR images available since 1992. Terrestrial SAR interferometry, on the other hand, thanks to its high sampling frequency of data acquisition (from few minutes to seconds) and its high accuracy in displacement measurement (up to decimal millimeter order), represents a suitable solution for real-time slopes and structures monitoring, being currently considered an effective technology for emergency monitoring purposes [15].Thanks to the active nature of RADAR sensors, the satellite and terrestrial InSAR systems do not require on-site, contact equip-ment, allowing to monitor the deformations completely remotely. However, artificial reflectors can be installed on the ground (i.e., the so-called corner reflectors) to improve the data accuracy and the density of measurement points.

3.1. Satellite SAR Interferometry

Satellite SAR Interferometry (InSAR) is an active remote sensing technique based on a microwave imaging system with cloud-pen-etrating ability and day and night operational capability (all-time, all-weather monitoring system) [21]. The main advantages in using satellite InSAR can be summarized as follows: i) performing historical deformation analysis of the area of interest; ii) performing both local-scale analysis (e.g. single building) and wide area analysis (city, regional and national scale)

reaching millimeter accuracy in displacement monitoring [22, 23, 24, 25, 26, 27, 28]. The first global systematic acquisition of SAR imagery started in the early 1990s, with ERS-1 and -2 satellites (European Space Agency - ESA). Since then, several satellite constellations have been placed in orbit from different space agencies (European, Canadian, Japanese, Italian, German, Spanish, etc.), collecting im-ages at global level with revisit times of the order of few days (for examples the recent ESA Sentinel-1 constellation has a revisit time of 6 days, only over Europe). Ground deformations can be retrieved through the Differential InSAR (DInSAR) approach, based on the phase difference analysis between two SAR images collected in different times, allowing to estimate the displacement occurred between the two acquisitions in the so-called interferogram [18, 19].Several SAR images collected with the same acquisition ge-ometry over the same area (interferometric stack) are used in the Advanced Differential InSAR (A-DInSAR) approaches to perform multi-temporal analysis overcoming the limitations of the DInSAR technique due to the influence of the atmospheric phase screen (APS) and residual topography [18, 19, 29, 30]. The Persistent Scatterers Interferometry (PSI) is one of the most effective and used A-DInSAR techniques, based on the analysis of specific targets on the Earth’s surface (called Per-sistent Scatterers, PSs) characterized by long time-coherent behaviour. The main outputs of a PSI analysis are (Figure 1): i) the defor-mational trend during the investigated time period; ii) the time series of displacement; iii) the height of the target on the ground and iv) cyclic (non-linear) deformation due to several factors (e.g., temperature variations).It is important to highlight that, using InSAR technique, displace-ment is measured along the so-called Line of Sight (LOS), i.e. along the direction between the satellite sensor and the target on the ground.The satellite InSAR analyses presented in this article have been performed using the PSI approach based on the following key processing steps [18]:– selection of the master image;– slave images coregistration;– Ground Control Point (GCP) selection and synthetic DEM

computation;– preliminary parameters estimation (average velocity/non-lin-

ear displacements and residual height) using the Persistent Scatterers Candidates (PSc);

– atmospheric phase screen removal and final estimation over all pixels.

The peculiarities of satellite InSAR make it a suitable tool for knowledge and control monitoring purposes and for the follow-ing applications in tunnelling and excavation works:– Ground deformation baseline analysis to be performed before

the construction activities;– monitoring the tunnel induced ground deformations during

excavation works;– identifying and understanding unknown and ongoing defor-

mational processes (subsidence, settlements, etc.);

n. 135-136 – dicembre 2020

58


– control and verify the interaction between underground works and surface anthropic activities.

3.2. Terrestrial SAR Interferometry

Terrestrial SAR Interferometry (TInSAR, also referred as GBIn-SAR) is a ground-based radar monitoring technique based on the same operational principles of Satellite SAR Interferometry (InSAR). It is an all-time (night and day), all-weather and high-accuracy technique based on an active radar sensor that emits microwaves (Ku band) and receives the return of scattering ob-jects [15, 31]. The TInSAR equipment is usually composed by a radar sensor that moves along a linear rail. Depending on the equipment used, the sampling rate ranges from a few minutes to a few seconds. TInSAR monitoring can be performed by install-ing the equipment on a stable platform with a panoramic view of the scenario to be monitored (up to 4 km away); one of the main advantages is that such a technique does not require the installation of contact sensors or artificial reflectors within the monitored area thus, can be considered as a fully remote sens-ing technique.The phase value of SAR images - collected at different times - is exploited in order to retrieve sub-millimeter accuracy displace-ment measurements. According to the site-specific conditions, the

system’s accuracy can range between some tenths of a millimeter to few millimeters.The typical outputs are represented by a displacement map of the investigated scenario and the related time series of displacement for measurement points characterized by high backscattering fea-tures (Figure 2). As stated for Satellite SAR Interferometry, it is important to highlight that displacement is measured along the so-called Line of Sight (LOS), i.e. along the direction between the ground-based sensor and targets on the ground.The high data sampling frequency and spatial resolution, the con-tinuous monitoring capabilities (both in time and in space) and the high accuracy in terms of displacement monitoring, make TInSAR a suitable technique for all the monitoring purposes (knowledge, con-trol and emergency). In the latest years, TInSAR has demonstrated to be a powerful technique for geotechnical and structural monitoring for ground and structural instability processes [12, 15, 16, 32, 33], thanks to the same features described above, such as: i) the “pano-ramic view” capability; ii) the continuous spatial information; iii) the high information density and long sensing distance.

3.3. Future outlook: PhotoMonitoring

In addition to radar-based technologies, innovative solutions like the PhotoMonitoring referred to different image processing tech-

Figure 1. Example of A-DInSAR results obtained from different satellite sensors (ERS, Envisat and COSMO-SkyMed), in different time periods over the same area. For every measurement point, it is possible to observe the displacement rate in mm/year. Negative values (yellow to red) are points that move away from the satellite, while positive values (light blue to dark blue) are points that move towards the sensor. Green points are stable. In the bottom right corner, the 22 years cumulative time series of the measurement point highlighted with the white star is reported (modified from [26]).

59

n. 135-136 – dicembre 2020


niques are now available for change and displacement monitoring in geotechnical and structural applications [7, 14, 34, 35, 36].PhotoMonitoring, represents an innovative remote monitoring solution that allows to obtain accurate information regarding possible modifications, in terms of changes and displacements, occurring on ground and structures. This is possible thanks to the use of image analysis techniques of multi-temporal scenes, collected at different times over the same area, integrating ad-vanced image-processing algorithms such as: Change Detection (CD), Digital Image Correlation (DIC), Image Classification (IC) and Artificial Intelligence (AI).Thanks to advanced processing methods, it is possible to exploit several images acquired in different times over the same area to perform multi-temporal surficial change and displacement analy-sis, obtaining the time series of displacement of measurement points.PhotoMonitoring technique can be applied to several and different datatype (e.g. optical, thermal, SAR amplitude, etc.), acquired from different platforms (terrestrial, aerial, UAV and satellite) (Figure 3).It has been demonstrated that ground deformations can be detected with an accuracy less than 1/10th of a pixel [35] which can be

translated into millimeter accuracy in the case of high-resolution ac-quisitions, making PhotoMonitoring a valuable technology for differ-ent purposes (knowledge and control monitoring) and applications among which: natural hazards, structural monitoring and mining.

4. Satellite and Terrestrial InSAR in underground engineering projects: case studies

In this chapter, some satellite and terrestrial InSAR monitoring projects in underground/excavation field are presented. In par-ticular, the contribution of satellite SAR interferometry for the ground deformation monitoring during excavation works will be highlighted in two case studies:– the Giustizia – Lolli Tunnel in the urban area of Palermo (Sicily

Region, Southern Italy);– the Santa Lucia Tunnel in the extra-urban area of Valdisotto

Municipality (Lombardy Region, Northern Italy).Moreover, the continuous terrestrial InSAR monitoring activities carried out for alerting purposes in the following projects are presented:

Figure 2. Example of TInSAR displacement map projected over 3D scenario and time series of displacement for selected measurement points related to the 2015 Scillato Landslide in Sicily, IT (modified from [12]). The color of the points indicates the intensity of displacements. Specifically, negative values (light blue to dark blue) are points with deformation direction towards the sensor while green points are considered stable. The location of the radar sensor is approximately at the point of view of the image.

n. 135-136 – dicembre 2020

60


– the San Giovanni Nord Tunnel entrance monitoring during con-struction phase, bordering the Sfalassà Viaduct (A2 Highway, Calabria, Southern Italy);

– the continuous monitoring of a construction site in the urban area of London (UK).

4.1. The Giustizia – Lolli Tunnel satellite InSAR historical analysis in the Metropolitan City of Palermo

In the framework of the railway construction project of the Paler-mo Centrale - Notarbartolo section, the excavation works of the

Giustizia – Lolli Tunnel (known as GN1D - Galleria Naturale Giusti-zia - Lolli) generated subsidence phenomena higher than expected in the design phase, producing damages to surrounding buildings since 2012. In September 2012, 7 buildings were evacuated (from n. 31 to n. 34B in Figure 4) and excavation works have been sus-pended at a distance less than 100 meters from end (Figure 4).Satellite A-DInSAR technique was used after the work suspension to identify the area interested by surficial deformation induced by the tunnelling progression and monitoring its evolution through time. In this paper the results obtained by processing a stack of 114 Sentinel-1 (European Space Agency, ESA) images collected be-tween 2014 and 2017 in the ascending acquisition geometry are presented. The deformational trends have been obtained with millimeter accuracy in the LOS direction (Figure 5). The measure-ment points in correspondence of the evacuated buildings clearly show ongoing settlement processes during the 3 years monitoring period, with cumulative displacements ranging between -7 and -15 mm.Given the significant deviations from the settlements modelled in the executive project, the designers have planned a procedure to manage the aspects related to the safety of citizens (residents, and passers-by), local traffic and construction site operators.Following the technical investigations, the demolition of the criti-cal buildings has been identified as the safest scenario to com-plete the works. The final project provides for the creation of a green area to replace buildings 31bis, 31, 32, 33, 34A.

4.2. The Santa Lucia Tunnel in the extra-urban area of Valdisotto Municipality

On February 18, 2016, due to the occurrence of slope instability phenomena, an ordinance prescribing the evacuation of a private building was issued by the Municipality of Valdisotto (Lombardy Region, Italy).

Figure 3. Example of PhotoMonitoring analyses with a 2D displacement map retrieved by DIC analysis performed on very high-resolution optical imagery. It is reported the displa-cement related to a large earth-slide. The landslide boundary is highlighted in red, while the displacement field is highlighted with white arrows (modified from [14]).

Figure 4. Floor plan of the tunnel route and damages in the surrounding buildings. Buildings from 31 to 34B were evacuated in September 2012. (from http://www.artasicilia.eu/old_site/web/vas/vas_progetti/nodo_pa_tratta_a/dir.php modified).

61

n. 135-136 – dicembre 2020


The evacuated building was located in the right side of the Adda river about 150 meters far from the Santa Lucia tunnel realized be-tween April 2013 and Autumn 2015 on the “SS38 Passo Stelvio”. To characterize the slope instability phenomena, technical surveys, and different monitoring activities (both contact and remote sys-tems) have been activated by the Municipality, among which the A-DInSAR historical analysis of the area. The satellite A-DInSAR analyses were performed with the following purposes:– to characterize the slope gravitational process identifying the

periods of activity (knowledge monitoring [4];– to extend the monitored area to better understand and con-

strain the slope instability phenomenon;– to achieve insights and information about the possible trig-

gers and the eventual destabilizing effects of the anthropic and tunnelling activities. In this scenario, beyond the Santa Lucia Tunnel, the presence of a hydraulic tunnel located 300 meters upstream of the evacuated building has been assessed.

The following interferometric datasets were analysed:– ERS1/ERS2 (from ESA), counting 35 images acquired in the

ascending geometry between April 1992 and October 2000;– Envisat (from ESA), including 42 images acquired in the as-

cending geometry between November 2004 – September 2010;

– COSMO-SkyMed (from ASI, Italian Space Agency), including 34 images acquired in ascending geometry between March 2012 and September 2016;

– Sentinel-1 (from ESA), including 39 images acquired in as-cending geometry between March 2015 and August 2016.

The multi-sensor A-DInSAR results revealed a clear increase in the velocity trends through time in correspondence of the evacuated area (Figure. 6). From the ERS and Envisat analyses the scenario

appear to be stable between 1992 and 2010 (i.e. no displace-ments were recorded). COSMO-SkyMed results shown an average displacement rate of -4 mm/year while the more recent Sentinel-1 dataset showed up to -35 mm/year of velocity trend.The time series of displacement obtained by merging the data processed from the three satellite missions describe the non-lin-ear displacement behaviour of the area of interest, showing a clear trend change occurred between May and July 2013.

4.3. The San Giovanni Tunnel entrance continuous monitoring by TInSAR

Due to the project’s strategic importance and high-risk site-spe-cific conditions related to landslide processes, the slope inter-ested by the San Giovanni excavation activities on the A2 Highway has been widely explored since 2007 and carefully monitored by an integrated monitoring system composed of traditional instru-ments (inclinometers, piezometers, load cells, and topographic measures) a weather station, an automatic camera and a Terrestrial SAR interferometer [16, 17, 36, 37].More in detail, the superficial deformational behaviour of the un-stable slope and the man-made structures has been carried out for 23 months with continuous 24/7 TInSAR monitoring system. The TInSAR monitoring data were interpreted in relation to the working stages, optical image acquired by the automatic camera and the rainfall records during the whole monitoring period. The integrated monitoring approach played a key role in the monitor-ing activity, allowing the correlation between observed deforma-tions and triggering factors [16, 17, 36, 37] (Figure 7). During the monitoring period, a high activity of the slope under investigation has been observed, mainly consisting in: i) shallow slope instability

Figure 5. A-DInSAR results obtained with Sentinel-1 satellite images for the period October 2014 - November 2017. Negative values (yellow to red colours) indicate direction of movement away from the satellite.

n. 135-136 – dicembre 2020

62


events related to the residual effects of the landslide occurred few months before the beginning of the TInSAR monitoring; ii) slid-ing of debris due to excavation activities and iii) deformations of engineering structures (i.e. gabions) associated to intense rainfall events [28].During the monitoring period, the TInSAR resulted to be the only system able to capture the deformational behaviour of the entire slope, thus representing a reference point for emergency monitor-ing purposes ensuring the work safety conditions [28].

4.4. The continuous monitoring of a construction site in the urban area of London (UK)

In the framework of an excavation site related to the realization of new apartments in the city centre of London (UK), involving an area of about 7,000 square meters, the continuous 24/7 TInSAR monitoring has been carried out for 10 months for emergency purposes. The activities aimed at monitoring the surrounding structures and assets, controlling their stability over time, and ensuring the safety of the workers. Monitoring points were defined

by the signal backscatter along the Line of Sight (LOS) between the sensor and the investigated scenario, which included the sur-rounding structures’ walls, the capping beam, the secant piles (as they became exposed during excavation), and the ground surface (Figure 8).The TInSAR monitoring platform was installed on the roof of the cabins located within the working site, in a safe location. The system was finally anchored on the roof of worksite containers in order to avoid possible oscillations which could have affected the measurements (Figure 8).Monitoring data, acquired with a scan interval of 5 minutes and validated twice a day, were daily provided to the contractor by a web-based platform. Although the monitored area was very noisy due to construction works (e.g. crane and vehicles motion), a filtering process, based on the amplitude of reflected radar waves, was implemented in order obtain right and reliable displacement time series. Also in this case, the acquired data were interpreted in relation to the working stages, which were also monitored thanks to an on-field camera. As showed in Figure 8c, thanks to the high values of mean reflectivity, due to the urban context, it was pos-

Figure. 6. Multisensor A-DInSAR analyses of the area of interest. The white arrows show the location of the evacuated building. Time series of displacement retrieved with ERS, Envisat, CSK and S1 satellite sensors in correspondence of the evacuated structures are reported. The Santa Lucia and hydraulic tunnels are reported in red and blue, respectively.

63

n. 135-136 – dicembre 2020


sible to easily identify the main buildings and structures within the Thermal SNR (Signal to Noise Ratio) map. It is important to highlight that the more the SNR is high, the more the accuracy is expected to be good.The use of such an innovative technology, coupled with several contact measurements, allowed to follow an integrated monitoring approach, very useful in this type of working context.

5. Discussions and Conclusions

The potential of interferometric SAR monitoring solutions in un-derground projects has been highlighted in four case histories. We showed how satellite InSAR represents an effective solution to investigate tunnelling induced subsidence and settlements processes in both urban and extra-urban environments. Satel-lite InSAR, especially the multi-temporal A-DInSAR approaches, is now considered a unique opportunity for historical ground motion investigations, to be used for recording the initial site-conditions before the excavation is started (i.e. baseline analy-sis). Thanks to the capability of current SAR satellite constella-

Figure 7. a) TInSAR displacement map along the LOS obtain between July 1, 2008 and January 31, 2009. b) Surficial displacements observed with terrestrial PhotoMonitoring techniques based on Digital Image Correlation between January 17, 2009 and January 18, 2009. c) TInSAR time series of displacement correlated with excavation phases and rainfall events (modified from [36]).

tions in terms of both spatial and temporal coverage, it is now possible to implement long-term monitoring plan based on A-DInSAR techniques, thus monitoring the evolution of ground deformations over time during the different excavation phases, possibly detecting deformation anomalies where to focus the attention, allowing decision makers to plan further investigations with ground-based surveys. The capability of Terrestrial InSAR in terms of high sampling fre-quency and accuracy make it possible to use the technique to im-prove the know-how about a deformational process and to control its evolution over time for emergency purposes. We showed how the technique can be successfully used in geotechnical and struc-tural monitoring activities to continuously monitor both natural phenomena (e.g. geohazard processes interacting with tunnelling activities as for the landslide involving San Giovanni Nord Tunnel entrance) and structures (e.g. the buildings surrounding the ex-cavation site in London).Further, the most innovative PhotoMonitoring technologies have been presented, offering a great opportunity for both long term monitoring and quick analyses from different platforms (terrestri-al, aerial and satellite), taking advantage of the relatively low-cost

n. 135-136 – dicembre 2020

64


sensors. We showed how it is possible to obtain high accuracy dis-placement data just using two optical images (e.g. simple photos) acquired in different times over the same scenario (Figure 7b), representing a suitable monitoring solution to be used in alterna-tive or integration of other techniques. The increasing technological innovation in the monitoring field made this science as useful as complex. Geotechnical and struc-tural monitoring in complex engineering projects (as in excava-tion works) is a multi-purpose activity. The major challenges lies in identifying the key parameters to be monitored (e.g. strain, displacement, pore water pressure, etc.) and to properly design the monitoring plan accounting for site-specific conditions and needs, thus selecting the most appropriate traditional and in-novative instruments to be used. Especially in large engineering projects, the planning and design of geotechnical and structural monitoring is becoming more and more important. There is not a unique option to be implemented and the most suitable ac-tivities must be identified to provide a more efficient and cost-effective monitoring plan.

References

[1] ESKESEN S.D., TENGBORG P., KAMPMANN J., VEICHERTS T.H. (2004) – Guidelines for tunnelling risk management: International Tunnelling

Association, Working Group No. 2. Tunnelling and Underground Space Technology , 217–237.

[2] GRASSO P., SOLDO L. (2017) – Risk analysis-driven design in tunnel-ling: the state-of-the-art, learnt from past experiences, and horizon for future development. Innov. Infrastruct. Solut.

[3] DUNNICLIFF J. (1988) – Geotechnical instrumentaion for monitoring field performance. John Wiley & sons.

[4] MAZZANTI P. (2017) – Toward transportation asset management: what is the role of geotechnical monitoring? J Civil Struct Health Monit.

[5] BOZZANO F., ESPOSITO C., FRANCHIA S., MAZZANTI P., PERISSIN D., ROCCA A. & ROMANO E. (2015) – Understanding the subsidence process of a quaternary plain by combining geological and hydrogeological mod-elling with satellite InSAR data: The Acque Albule Plain case study. Remote Sensing of Environment, 168, 219–238.

[6] URBINI S., BIANCHI-FASANI G., MAZZANTI P., ROCCA A., VITTUARI L., ZA-NUTTA A., GIRELLI V.A., SERAFINI M., ZIRIZZOTTI A., FREZZOTTI M. (2019) – Multi-Temporal Investigation of the Boulder Clay Glacier and North-ern Foothills (Victoria Land, Antarctica) by Integrated Surveying Tech-niques. Remote Sensing 11(12):1501.

[7] MAZZANTI P., CAPOROSSI P., MUZI R. (2020) – Sliding Time Master Digital Image Correlation Analyses of CubeSat Images for landslide Monitoring: The Rattlesnake Hills Landslide. Remote Sens. 12(4), 592.

[8] TERZAGHI K. (1937) – Settlement of structures in Europe and meth-ods of observations. American Society of Civil Engineers. Proceedings 63, pp 1358-1374.

Figure 8. a) TInSAR instrument installed within the working area; b) optical image of the working site, the coloured polygons correspond to the areas highlighted in the Thermal SNR map (c); c) Thermal SNR map (dB): mean reflectivity values over time (how the structures reflect the radar signal).

65

n. 135-136 – dicembre 2020


[9] PECK R.B. (1969) – Advantages and limitations of the observational method in applied soil mechanics. Géotechnique, 19(2):171-187.

[10] DUNNICLIFF J. (1993) – Geotechnical instrumentation for monitoring field performance. John Wiley & Sons.

[11] KOVÁRI K., LUNARDI P. (2000) – On the observational method in tunneling. Proceedings International Conference on Geotechnical & Geological Engineering. Melbourne, Australia, 19-24 November 2000.

[12] MORETTO S., BOZZANO F., BRUNETTI A., DELLA SETA M., MAJETTA S., MAZZANTI P., ROCCA A., VALIANTE M. (2018) – The 2015 Scillato Land-slide (Sicily, Italy): deformational behavior inferred from Satellite & Terrestrial SAR Interferometry. FMGM Rio de Janeiro, 2018.

[13] HWANG C., HUNG W.C., LIU C.H. (2008) – Results of geodetic and geotechnical monitoring of subsidence for Taiwan High Speed Rail op-eration. Nat Hazards (2008) 47:1–16 DOI 10.1007/s11069-007-9211-5

[14] BRUNETTI A., MAZZANTI P. (2019) – Il monitoraggio dinamico delle strutture e delle infrastrutture con Interferometria Radar Terrestre. In-genio web (https://www.ingenio-web.it/).

[15] ANTONIELLI B., CAPOROSSI P., MAZZANTI P., MORETTO S., ROCCA A. (2018) – InSAR & PhotoMonitoring For Dams and Reservoir Slopes Health & Safety Monitoring. Vingt-Sixième Congrès Des Grands Bar-rages Autriche, juillet 2018. doi:10.3217/978-3-85125-620-8-227.

[16] MAZZANTI P., BOZZANO F., BRUNETTI A., ESPOSITO C., MARTINO S., PRESTININZI A., ROCCA A., SCARASCIA MUGNOZZA G. (2015) – Terres-trial SAR Interferometry Monitoring of Natural Slopes and Man-Made Structures. Engineering Geology for Society and Territory – Springer.

[17] MAZZANTI P., BOZZANO F., CIPRIANI I., PRESTININZI A. (2014) – New insights into the temporal prediction of landslides by a terrestrial SAR interferometry monitoring case study. Landslides – Springer.

[18] KAMPES B.M. (2006) – Radar Interferometry Persistent Scatterers Technique. Ed. Dordrecht, The Netherlands - Springer.

[19] FERRETTI A., PRATI C., ROCCA F. (2001) – Permanent Scatterers in SAR Interferometry. IEEE Transactions on Geoscience and Remote Sensing, Vol. 39, No. 1, January 2001.

[20] CROSETTO M., MONSERRAT O., CUEVAS-GONZÁLEZ M., DEVANTHÉRY N., CRIPPA B. (2016) – Persistent Scatterer Interferometry: A review. ISPRS Journal of Photogrammetry and Remote Sensing 115 (2016) 78–89

[21] FERRETTI A., MONTI-GUARNIERI A., PRATI C., ROCCA F., MASSONNET D. (2007) – InSAR Principles: Guidelines for SAR Interferometry Processing and Interpretation. ESA Publications, TM-19. ISBN 92-9092-233-8.

[22] COLESANTI C., FERRETTI A., PRATI C., ROCCA F., (2003) – Monitoring landslides and tectonic motions with the Permanent Scatterers Tech-nique. Engineering Geology, 68, pp. 3–14.

[23] ROCCA A., MAZZANTI P., PERISSIN D., BOZZANO F., (2014) – Detec-tion of past slope activity in a desert area using multi-temporal DIn-SAR with ALOS PALSAR data. Italian Journal of Engineering Geology and the Environment, 14/1, 35-49 [Scopus, I.F. 0,5]. DOI: 10.4408/IJEGE.2014-01.O-03

[24] ROCCA A., MAZZANTI P., BOZZANO F., PERISSIN D., (2015) – Ad-vanced Characterization of a Landslide-Prone Area by Satellite A-DInSAR. G. Lollino et al. eds., Engineering Geology for Society and Territory, Volume 5, Springer International Publishing, Switzerland, 177-181

[25] COSTANTINI M., FERRETTI A., MINATI F., FALCO S., TRILLO F., COLOMBO D., NOVALI F., MALVAROSA F., MAMMONE C., VECCHIOLI F., RUCCI A., FUMA-GALLI A., ALLIEVI J., CIMINELLI M.G., COSTABILE S., (2017) – Analysis of

surface deformations over the whole Italian territory by interfero-metric processing of ERS, Envisat and COSMO-SkyMed radar data. Remote Sensing of Environment. In press. https://doi.org/10.1016/j.rse.2017.07.017

[26] BOZZANO F., ESPOSITO C., MAZZANTI P., PATTI M., SCANCELLA S. (2018) – Imaging Multi-Age Construction Settlement Behaviour by Advanced SAR Interferometry. Remote Sens., 10, 1137; doi:10.3390/rs10071137

[27] URBINI S., BIANCHI-FASANI G., MAZZANTI P., ROCCA A., VITTUARI L., ZANUTTA A., GIRELLI V.A., SERAFINI M., ZIRIZZOTTI A., FREZZOTTI M. (2019) – Multi-Temporal Investigation of the Boulder Clay Glacier and North-ern Foothills (Victoria Land, Antarctica) by Integrated Surveying Tech-niques. Remote Sens. 2019, 11(12), 1501; https://doi.org/10.3390/rs11121501

[28] DI LISA M., HOSSAM ELDIN A.A., MAZZANTI P., MORETTO S.(2020) – Inferring the Creep Settlement Behavior of Rockfill in Reclaimed Lands by Advanced SAR Interferometry and Numerical Modeling: An Example from Arabian Gulf

[29] FERRETTI A., P.C. (2000) – Nonlinear subsidence rate estimation using permanent scatterers in differential SAR interferometry. IEEE Transactions on Geoscience and Remote Sensing, 385, 2202–2212. doi:10.1109/36.868878.

[30] HANSSEN R.F. (2005) – Satellite radar interferometry for deforma-tion monitoring: a priori assessment of feasibility and accuracy. Inter-national Journal of Applied Earth Observation and Geoinformation, 6(3), 253-260.

[31] ANTONELLO G., CASAGLI N., FARINA P., LEVA D., NICO G., SIEBER A.J., TARCHI, D.: Ground-based SAR interferometry for monitoring mass movements, Landslides 1, 21–28, 2004.

[32] BOZZANO F., ESPOSITO C., MAZZANTI P., INNOCCA F., ROMEO S. (2020) – Urban Engineered Slope Collapsed in Rome on February 14th, 2018: Results from Remote Sensing Monitoring. MDPI, Geosciences 2020, 10, 331. doi:10.3390/geosciences10090331.

[33] DI MATTEO L., ROMEO S., KIEFFER D.S. (2017) – Rock fall analysis in Alpine area by using a reliable integrated monitoring system: results from the Ingelsberg slope (Salzburg Land, Austria) – Bull. Eng. Geol. Environ. 2017, 76, 413-420.

[34] BOZZANO F., CAPOROSSI P., ESPOSITO C., MARTINO S., MAZZANTI P., MORETTO S., SCARASCIA MUGNOZZA G., RIZZO A.M. (2017) – Mechanism of the Montescaglioso Landslide (Southern Italy) Inferred by Geologi-cal Survey and Remote Sensing. Advancing Culture of Living with Land-slides – Springer.

[35] CAPOROSSI P., MAZZANTI P., BOZZANO F. (2018) – Digital Image Cor-relation (DIC) Analysis of the 3 December 2013 Montescaglioso Land-slide (Basilicata, Southern Italy): Results from a Multi-Dataset Investiga-tion. International Journal of Geo-Information – MDPI.

[36] MAZZANTI P., CAPOROSSI P., MUZI R. (2020) – Sliding Time Mas-ter Digital Image Correlation Analyses of CubeSat Images for land-slide Monitoring: The Rattlesnake Hills Landslide (USA). Remote Sens. 2020, 12(4), 592; https://doi.org/10.3390/rs12040592

[36] BOZZANO F., CIPRIANI I., MAZZANTI P., PRESTININZI A., 2011. Displace-ment patterns of a landslide affected by human activities: insights from Ground Based InSAR monitoring. Natural Hazards, 59, 3, 1377-1396, doi:10.1007/s11069-011-9840-6, ISSN: 0921-030X.

[37] BOZZANO F., MAZZANTI P., PRESTININZI A. (2008) – A radar platform for continuous monitoring of a landslide interacting with an under-construction infrastructure. Italian Journal of Engineering Geology and Environment, 2, 35-50, ISSN: 1825-6635.

From the sky to underground: new Remote Sensing solutions ...

Documents

Transcript of From the sky to underground: new Remote Sensing solutions ...