Ground deformation monitoring with satellite radar sensors

September 10th, 2014, Published in Articles: PositionIT, Featured: PositionIT

 

In the last decade, Synthetic Aperture Radar (SAR) data have been widely used to detect and measure ground deformation phenomena. Either ground-based, airborne or satellite-borne, radar sensors paved the way to a variety of applications ranging from landslide analyses to subsidence and sinkhole monitoring.

While ground-based radar became one of the main tools for implementing early warning systems in mining areas, satellite radar techniques proved extremely effective for wide area mapping over hundreds or even thousands of square kilometres. Indeed, this new “view from space” has been supporting decision makers, providing unique, high quality, displacement data for different applications. Satellite SAR data, in particular, can be used for fault characterisation and calibration of geo-mechanical models in the oil
and gas sector, for monitoring landslides, volcanoes and seismic faults, areas prone to sinkholes, terrain compaction phenomena induced by tunneling and mining activities, and even for monitoring the stability of individual buildings.

Fig. 1: Available SAR satellite constellations.

Fig. 1: Available SAR satellite constellations.

In this article, after a brief introduction on the techniques and the sensors used today, we report some real-life examples related to engineering and mining projects, focusing on what can (or cannot) be obtained today using this new, fascinating, information source.

Satellite radar sensors 

Common to every monitoring technique, new data sources create new opportunities and make possible new applications. 2014 was a very important year for satellite radar sensors. The European Space Agency (ESA) launched the first satellite of the Sentinel constellation, while the Japanese Space Agency (JAXA) launched the new ALOS-2 satellite.

Fig. 2: SqueeSAR applied to a low-resolution data source.

Fig. 2: SqueeSAR applied to a low-resolution data source.

Sentinel-1A is the first step towards a brand new system for earth observation, where both radar and optical sensors will continuously provide up-to-date data about our planet. The C-band radar sensor (wavelength of about 6 cm) mounted on board Sentinel-1A will be able to image the entire globe every twelve days, and after the launch of the twin satellite Sentinel-1B next year, the Earth will be monitored with unprecedented accuracy every six days.

ALOS-2 JAXA’s new SAR satellite is a follow-on mission from the successful ALOS (“DAICHI”) mission. The high-resolution L-band sensor (wavelength of about 24 cm, allowing better penetration in vegetated areas) will provide an updated image of the Earth every 14 days.

Fig. 3: SqueeSAR applied to a high-resolution data source.

Fig. 3: SqueeSAR applied to a high-resolution data source.

These new sensors will provide new radar data for monitoring projects, adding two valuable options to the list of operational satellites. Different radar sensors usually operate at different frequencies and feature different spatial resolutions and revisiting times (Fig. 1).

Over the past few years, there has been increasing attention paid to assessing the impact of human activities on climate and other natural phenomena, leading to further investments in the space segment. Satellite data, however, can be used not only for scientific studies, but also for monitoring specific assets in commercial projects. As a matter of fact, the number of radar images acquired every day over both land and oceans has increased steadily over the years, and this is one of the reasons behind the increasing number of applications of radar data.

InSAR techniques

The acronym InSAR stands for Interferometric SAR and refers to a family of algorithms, rather than to a specific technique. Using interferometric techniques and a set of radar images acquired at different times from the same acquisition geometry, it is possible to measure displacements of just 1 mm on the ground from satellites orbiting the Earth hundreds of kilometres above us: this is the magic of InSAR.

Fig. 4: Change in the planned track of the new Venice-Trieste railway line.

Fig. 4: Change in the planned track of the new Venice-Trieste railway line.

SqueeSAR is the most advanced InSAR processing techniques, and has been widely used to detect and monitor surface deformation phenomena. Depending on the SAR sensor selected as the data source, it can provide high quality displacement data for a variety of applications.

In general, InSAR data can be updated regularly, and can provide new ways to design early warning systems covering entire nations, where satellite information can highlight areas where in situ sensors and continuous monitoring tools should be installed. Rather than replacing conventional technologies, satellite data will work more and more in synergy with airborne sensors and ground-based instruments. Satellite SAR measurements are becoming a companion data source for total stations and GPS-GNSS measurements. The availability of new high-resolution constellations allows one to reach unprecedented density of measurement points (see Figs. 2 and 3).

Fig. 5: Tunnelling activity and induced subsidence (in blue the tunnel track).

Fig. 5: Tunnelling activity and induced subsidence (in blue the tunnel track).

Engineering projects

When planning new infrastructure, be it a pipeline, a railway or a highway, any engineer or geo-technician would consider potential problems related to surface deformation phenomena. In fact, landslides, sinkholes and subsidence phenomena can all threaten new infrastructure and any monitoring technology capable of managing risk related to ground movement can have an immediate impact on risk assessment.

InSAR analysis is a strategic instrument that can also be used for route planning. Knowing the stability of an area and the embankments surrounding a planned road, railway or pipeline route allows construction to be avoided in unstable areas and facilitates lower risk solutions.

Fig. 6: Settlement vs. tunnelling activity.

Fig. 6: Settlement vs. tunnelling activity.

This was the case for the new Venice-Trieste railway line, in Italy, where the original track would have crossed the coastal slopes and the Trieste Flysch formations. No active landslide was known to be present along the track. A SqueeSAR analysis carried out on SAR data already acquired could highlight unstable areas along the route. While geological studies are still in progress to assess the causes of the phenomena, the railway authority decided to change the track to skip such motion prone areas.

Satellite monitoring of new infrastructures

A historical archive of radar data allows the cause-effect relationship between construction (of car parks, underground lines, etc.) and damage introduced to surrounding buildings. InSAR is an effective technique for retrospectively resolving disputes regarding damages resulting from human operations (e.g. excavation) and natural causes (e.g. terrain compaction). Results of the investigations are of particular interest to insurance companies. The availability of historical archives of satellite SAR acquisitions that date back to the nineties, allows new claim assessment procedures that could not be envisaged just a few years ago.

Fig. 7: Time series of displacement. Red – InSAR. Blue – levelling.

Fig. 7: Time series of displacement. Red – InSAR. Blue – levelling.

In the framework of the new high-speed Italian railway connecting Milan to Naples, a new underground station has been completed in Bologna. The tunneling operation created a double-track tunnel with an excavation area of approximately 130 m2, crossing urban tissue at shallow depths (approx. 10 m). A very high density of commercial and residential housing is present in the area. Given this scenario, special attention has been devoted to monitoring the impact of tunneling works on surface structures. Total stations and GPS antenna have been used for monitoring specific locations, while InSAR (SqueeSAR) data provided a synoptic view over a much wider area than could have been monitored with in situ instruments.
Fig. 5 shows the results of the first analysis carried out in 2011 with a clear correlation between tunneling and subsidence visible. Moreover, when looking into the time series of displacement of the subsiding point, a temporal correlation between settlement values and various phases of the tunneling process can be highlighted.

Fig. 8: Radar in the pit.

Fig. 8: Radar in the pit.

The first settlement shown in Fig. 6 corresponds to the excavation of ten explorative micro-tunnels, while the latter is an effect of tunneling works. By comparing satellite data with in situ observations, it was possible to integrate different data sources and to carry out a precision assessment of satellite measurements (Fig. 7).

Mining applications

Fig. 9: Radar on the rim of the pit.

Fig. 9: Radar on the rim of the pit.

The use of InSAR data for monitoring open pit mines is easy to grasp. Satellite measurements can provide a synoptic view of the displacement field affecting the pit and its neighbourhood, which cannot be provided by in situ observations. Satellite data can increase mine safety and make early-warning procedures related to slope failure more effective, as well as highlight areas where ground-based, continuous monitoring tools should be located. InSAR can measure ground movement caused by underground mineral extraction too, and is already used to map the extent and rate of surface deformation occurring over mining sites, as well as abandoned mines. Ground displacement maps are used to identify areas of instability that could indicate potential hazards that may require remedial actions. Satellite data can provide an overview of displacements occurring on waste piles too, mapping areas of instability that could lead to structural collapse of the waste pile.

Fig. 10: Satellite radar flying over the pit.

Fig. 10: Satellite radar flying over the pit.

The mining industry has been working with geodetic instruments since the earliest stages. This is particularly true for open pit mining operations. When working on very steep slopes, the risk of a failure can be extremely high. Geodetic measurements have been used to deal with such a risk, feeding early warning systems of increasing complexity and sophistication. Ground-based radar systems, in particular, are widely used both as a monitoring tool and as a key data source for early warning systems.

Satellite radar measurements rely on the same technical approach, but the sensor is operated from a different acquisition geometry. Figs. 8, 9 and 10 show different line of sights for each radar configuration. Although not the right tool to create an early warning system (since information can be updated only after a new satellite acquisition over the area of interest, typically after a few days from the latest take), they are very sensitive to vertical, rather than horizontal, surface displacements and provide data over very large areas. Last but not least, since all satellite radar sensors operate in the microwave domain (and since wavelengths are a few centimetres long), they can all measure extremely small (millimetre) range variations.

Fig. 11: Cumulative displacement over the whole mining area.

Fig. 11: Cumulative displacement over the whole mining area.

Providing information over the whole mining area

The satellite perspective allows one to monitor a large mining site with a single analysis. The following example shows a mine spread for nearly 20 000 hectares, and it is well monitored with several in-situ sensors.
Fig. 11 shows the displacement field estimated from satellite data over a four-month period. No in situ technique can provide such a picture at reasonable cost. It should be noted that this map was obtained without installing anything on the ground. These data-sets can be easily ingested in modern geographic information systems (GIS), allowing for an effective integration with other data sources. It can also be updated on a regular basis, depending on the application and the sensor to be used.

In pit monitoring

Starting from the synoptic view, a simple close-up allows one to identify and measure displacement occurring in the pit, highlighting portions of the slope affected by significant displacement.

Since these maps are not based on an a priori network of measurement points, an easy visualisation of the displacement is achievable (Fig. 12). In order to complement the displacement information, for every measurement point identified, a proper time series of displacement is provided (Fig. 13).

Fig. 12: Zoomed view from in the pit, displaying displacement in millimetres.

Fig. 12: Zoomed view from in the pit, displaying displacement in millimetres.

Water reclamation facilities and tailings dams monitoring

Following the same approach, a dam that is in a water reclamation facility is an easy target for InSAR monitoring. In this particular case, the analysis of a cross section shows the area where the dam structure is settling faster than any other part and may require closer inspection. Among the other advantages, the no ground sensor monitoring that is served through satellites delivers frequent updates (quarterly or even monthly), without additional cost or manpower. There is also no instrumentation that would impede on normal dam building and maintenance activities.

Fig. 14: Water reclamation tailings operations. Cross section of displacement over a dam.

Fig. 14: Water reclamation tailings operations. Cross section of displacement over a dam.

Conclusions

Modern SAR satellites provide better temporal coverage at a much higher resolution than their predecessors. The shorter timeframe between observations and the increased density of sample points has brought InSAR technology to the forefront of mine site and infrastructure monitoring. Processing algorithms have become more robust and optimised to meet the higher data volumes and shorter timelines. The SqueeSAR examples in this article illustrate the results obtained from an industry leading InSAR processing environment. This means all surface assets of a mine site, manufacturing facilities or transportation network can be continuously monitored from a single source, which provides monthly, quarterly, semi-annual or annual reports to meet an operator’s monitoring and maintenance activities/requirements.

Contact Ian du Toit, Optron, Tel 021 421-0555, idutoit@optron.com

Related Articles

  • Rugged solutions provider appoints technical support manager
  • Economic street level imagery surveys for smaller municipalities
  • Information is nothing without ethical action
  • Setting up a real-time inventory management system
  • GIS software achieves top user ranking