*2.3.2 Terrestrial radar interferometry*

In the last two decades, TRI revealed to be a valuable tool to monitor glaciers [9, 34–42]. TRI concerns the analysis of the phase difference between two radar acquisitions, which is directly related to the target displacement component parallel to the LOS. Typical radar apparatuses can provide spatially distributed displacement data in an area of several square kilometres with an operative range of a few kilometres. Radars are active sensors, as such, TRI can be applied during the night and severe meteorological conditions. Moreover, TRI measurements have sub-millimetre sensibility in an optimal context. However, the processing is not trivial and it requires high computational costs. Particularly complicated is the phase wrapping solution, which depends on the phase 2 periodicity and which is

related to the sensor-to-target range. Moreover, TRI is quite sensitive to possible morphological change of the scattering surface and that causes signal decorrelation and extreme atmospheric conditions can heavily affect the measurements [43, 44]. In glaciological contexts, long distances, morphological surface changes and severe meteorology are common and TRI processing must be handled carefully.

#### **2.4 Glacier internal structure**

GPR has been widely used as a geophysical method for the study of internal glacier properties. A variation in electrical permittivity creates dielectric interfaces and subsequent reflections that can be analysed. GPR can be used for the definition of firn-ice transition, the detection of subglacial cavities and the ice thickness [45]. GPR systems include a transmitter and a receiver antenna. Typical operating frequencies vary between 10 and 15 MHz, for the investigation of glaciers having depths of hundreds of meters, to 400–600 MHz, for shallow investigations. Different factors can limit the effectiveness of the technique, such as debris cover of the ice surface or highly crevassed areas that can create scattering or absorption phenomena that reduce the possibility of investigation of the glacier sub-surface. Processing of radar data normally implies many steps, which include (i) low-frequency filtering, caused mainly by surface reflection; (ii) selection of a time gain to correct for the amplitude divergence; (iii) temporal and spatial filtering for improving the signal-to-noise ratio; (iV) deconvolution and (v) migration [46].

GPR apparatuses are usually lightweight and compact and they can be easily transported by walking or snowmobile, which allows at acquiring a large number of 2D radar profiles. However, helicopter-borne surveys provide the most versatile platform and they have been used for detecting glacier thickness [47, 48], intraglacial features [49] and snow accumulation [50].

#### **2.5 Data integration**

Spatially distributed deformation data provide wide information on the investigated process. Nevertheless, common remote sensing apparatuses only provide specific displacement components or punctual measurements and the integration of different sensors is necessary to obtain spatially distributed 3D data.

Dematteis et al. [6] proposed an innovative solution to obtain 3D displacement using DIC and TRI data integration. DIC and TRI provide different and complementary displacement components that can be coupled to obtain a three-dimensional representation of the surface kinematics. The necessary conditions to couple the different data are that their maps must have the same spatial resolution in the same coordinate system (CS). Therefore, a geometric transformation is required to represent both data in the same CS, which is usually associated with a georeferenced DEM.

A different approach of data integration entails the merging of DIC and RTS data. RTS provides 3D displacement in specific points, while DIC can measure spatially distributed data. Therefore, their integration allows obtaining the displacement direction and versus using RTS data, while the DIC results give the spatial distribution.

## **3. Case study: Planpincieux-Grandes Jorasses glaciers**

The Planpincieux and Grandes Jorasses glaciers form a unique polythermal glacial complex located on the Italian side of the Grandes Jorasses peak (Mont

**129**

kinematics.

*Close-Range Sensing of Alpine Glaciers DOI: http://dx.doi.org/10.5772/intechopen.92841*

ages for the absence of snow.

Giordan et al. [7].

**3.1 Monitoring campaigns**

105 m3

Blanc massif), in the Ferret valley (**Figure 1**). The glaciers have approximately a South-East aspect and the elevation ranges from 2600 m asl to 4200 m asl. The accumulation area of the Grandes Jorasses Glacier is formed of two 45° steep cirques, which merge in an icefall at 3500 m asl. In the left cirque is located the Whymper Serac, whose front is at an elevation of 3800 m asl (**Figure 2b**). According to Pralong and Funk [51], this portion is classified as an unbalanced hanging glacier. As such, the serac progressively increases its volume and when its shape reaches unstable geometry, the serac collapses. This cycle follows an irregular periodicity and the time between the break-offs ranges from a few years to more than a decade. Usually, the unstable ice chunk has a volume of the order of

, which can collapse at once or in several pieces. The instability dynamics is

driven only by the geometry and it is not linked to temperature or water percolation. Therefore, the fracture can also occur during the cold season, when the collapse might easily trigger a large snow avalanche that would seriously threaten the underlying buildings and the road at the valley bottom. The last events happened in August 1993, June 1998 [11] and September 2014 [12]. The first one caused the fatality of eight mountaineers, but the ice avalanche did not cause further dam-

The Planpincieux Glacier topography presents three distinguished zones: the accumulation area, 3000–3500 m asl, is formed of two steep cirques that merge in a wide plateau at 2900–3000 m asl, and two lobes constitute the ablation area. The right lower lobe (**Figure 2a**) is 32° steep on average and it is quite crevassed. Its terminus ends in correspondence of a bedrock cliff that causes frequent calving. In the past, several collapses occurred and, in a few cases, they endangered the bridge of the Montitaz stream that originates from the glacier snout. Further information on the Planpincieux Glacier can be found in

In the last decades, the Planpincieux-Grandes Jorasses Glacier complex has become an open-air laboratory where innovative remote sensing techniques have

The Planpincieux Glacier is observed by two monoscopic time-lapse cameras placed in the opposite side of the Ferret valley, at a distance of 3800 m from the glacier. The monitoring station is equipped with two solar panels and an electric cell for power supply. It is remotely controlled by a Raspberry Pi 3 connected to the server of the Geohazard Monitoring Group (GMG) of the Research Institute for Geo-Hydrological Protection (IRPI), in Torino, Italy. A robotised webcam has been installed in 2018 to survey the station functioning. The system is active since August 2013 and it acquires images at hourly frequency. In the period August 2013–December 2019, it collected more than 35,000 images and it is probably the longest continuous series of hourly images in the European Alps. The images are processed with the DIC technique to estimate the surface glacier

The Grandes Jorasses Glacier is being monitored since 2010 by an RTS installed in the Planpincieux hamlet at a distance of 4800 m. The RTS measures every 2 h the position of the prisms installed onto and in the vicinity of the Whymper Serac (**Figure 2b**). Due to the extreme meteorological conditions and the exceptional sensor-to-target range, the prisms are not always visible and gaps in the measurement series are frequent, especially during the cold season. Snowfalls and strong wind occasionally cause the loss of some prisms and the intervention of Alpine

been developed to monitor the glacier activity [6–12].

### *Close-Range Sensing of Alpine Glaciers DOI: http://dx.doi.org/10.5772/intechopen.92841*

*Glaciers and the Polar Environment*

**2.4 Glacier internal structure**

related to the sensor-to-target range. Moreover, TRI is quite sensitive to possible morphological change of the scattering surface and that causes signal decorrelation and extreme atmospheric conditions can heavily affect the measurements [43, 44]. In glaciological contexts, long distances, morphological surface changes and severe

GPR has been widely used as a geophysical method for the study of internal glacier properties. A variation in electrical permittivity creates dielectric interfaces and subsequent reflections that can be analysed. GPR can be used for the definition of firn-ice transition, the detection of subglacial cavities and the ice thickness [45]. GPR systems include a transmitter and a receiver antenna. Typical operating frequencies vary between 10 and 15 MHz, for the investigation of glaciers having depths of hundreds of meters, to 400–600 MHz, for shallow investigations. Different factors can limit the effectiveness of the technique, such as debris cover of the ice surface or highly crevassed areas that can create scattering or absorption phenomena that reduce the possibility of investigation of the glacier sub-surface. Processing of radar data normally implies many steps, which include (i) low-frequency filtering, caused mainly by surface reflection; (ii) selection of a time gain to correct for the amplitude divergence; (iii) temporal and spatial filtering for improving the

GPR apparatuses are usually lightweight and compact and they can be easily transported by walking or snowmobile, which allows at acquiring a large number of 2D radar profiles. However, helicopter-borne surveys provide the most versatile platform and they have been used for detecting glacier thickness [47, 48], intragla-

Spatially distributed deformation data provide wide information on the investigated process. Nevertheless, common remote sensing apparatuses only provide specific displacement components or punctual measurements and the integration

Dematteis et al. [6] proposed an innovative solution to obtain 3D displacement using DIC and TRI data integration. DIC and TRI provide different and complementary displacement components that can be coupled to obtain a three-dimensional representation of the surface kinematics. The necessary conditions to couple the different data are that their maps must have the same spatial resolution in the same coordinate system (CS). Therefore, a geometric transformation is required to represent both data in the same CS, which is usually associated with a georeferenced DEM. A different approach of data integration entails the merging of DIC and RTS data. RTS provides 3D displacement in specific points, while DIC can measure spatially distributed data. Therefore, their integration allows obtaining the displacement direction and versus using RTS data, while the DIC results give the spatial

The Planpincieux and Grandes Jorasses glaciers form a unique polythermal glacial complex located on the Italian side of the Grandes Jorasses peak (Mont

of different sensors is necessary to obtain spatially distributed 3D data.

**3. Case study: Planpincieux-Grandes Jorasses glaciers**

meteorology are common and TRI processing must be handled carefully.

signal-to-noise ratio; (iV) deconvolution and (v) migration [46].

cial features [49] and snow accumulation [50].

**2.5 Data integration**

**128**

distribution.

Blanc massif), in the Ferret valley (**Figure 1**). The glaciers have approximately a South-East aspect and the elevation ranges from 2600 m asl to 4200 m asl. The accumulation area of the Grandes Jorasses Glacier is formed of two 45° steep cirques, which merge in an icefall at 3500 m asl. In the left cirque is located the Whymper Serac, whose front is at an elevation of 3800 m asl (**Figure 2b**). According to Pralong and Funk [51], this portion is classified as an unbalanced hanging glacier. As such, the serac progressively increases its volume and when its shape reaches unstable geometry, the serac collapses. This cycle follows an irregular periodicity and the time between the break-offs ranges from a few years to more than a decade. Usually, the unstable ice chunk has a volume of the order of 105 m3 , which can collapse at once or in several pieces. The instability dynamics is driven only by the geometry and it is not linked to temperature or water percolation. Therefore, the fracture can also occur during the cold season, when the collapse might easily trigger a large snow avalanche that would seriously threaten the underlying buildings and the road at the valley bottom. The last events happened in August 1993, June 1998 [11] and September 2014 [12]. The first one caused the fatality of eight mountaineers, but the ice avalanche did not cause further damages for the absence of snow.

The Planpincieux Glacier topography presents three distinguished zones: the accumulation area, 3000–3500 m asl, is formed of two steep cirques that merge in a wide plateau at 2900–3000 m asl, and two lobes constitute the ablation area. The right lower lobe (**Figure 2a**) is 32° steep on average and it is quite crevassed. Its terminus ends in correspondence of a bedrock cliff that causes frequent calving. In the past, several collapses occurred and, in a few cases, they endangered the bridge of the Montitaz stream that originates from the glacier snout. Further information on the Planpincieux Glacier can be found in Giordan et al. [7].

## **3.1 Monitoring campaigns**

In the last decades, the Planpincieux-Grandes Jorasses Glacier complex has become an open-air laboratory where innovative remote sensing techniques have been developed to monitor the glacier activity [6–12].

The Planpincieux Glacier is observed by two monoscopic time-lapse cameras placed in the opposite side of the Ferret valley, at a distance of 3800 m from the glacier. The monitoring station is equipped with two solar panels and an electric cell for power supply. It is remotely controlled by a Raspberry Pi 3 connected to the server of the Geohazard Monitoring Group (GMG) of the Research Institute for Geo-Hydrological Protection (IRPI), in Torino, Italy. A robotised webcam has been installed in 2018 to survey the station functioning. The system is active since August 2013 and it acquires images at hourly frequency. In the period August 2013–December 2019, it collected more than 35,000 images and it is probably the longest continuous series of hourly images in the European Alps. The images are processed with the DIC technique to estimate the surface glacier kinematics.

The Grandes Jorasses Glacier is being monitored since 2010 by an RTS installed in the Planpincieux hamlet at a distance of 4800 m. The RTS measures every 2 h the position of the prisms installed onto and in the vicinity of the Whymper Serac (**Figure 2b**). Due to the extreme meteorological conditions and the exceptional sensor-to-target range, the prisms are not always visible and gaps in the measurement series are frequent, especially during the cold season. Snowfalls and strong wind occasionally cause the loss of some prisms and the intervention of Alpine

guides it is necessary for the installation of new targets. Moreover, the Whymper Serac is continuously monitored by a 4800-m-far monoscopic camera. This survey is active since 2010 and the serac surface displacement is estimated with feature tracking of the hourly photographs.

Besides these continuous monitoring systems, in the past, several measurement campaigns have been conducted to increase the glacier understanding and to develop new monitoring techniques of glaciological close-range remote sensing. In **Table 1**, the complete list of the surveys conducted since 2010 is presented and the related references are reported when available.
