Eyes in the sky: Spotting millimetre-scale movements from space

Dr Sakthy Selvakumaran

Department of Engineering, University of Cambridge

Email: ss683@cam.ac.uk





Dr Sakthy Selvakumaran studied Engineering at the University of Cambridge (MA, MEng). She then lived and worked internationally in design, construction and R&D roles within industry and international development, becoming a Chartered Civil Engineer (CEng, MICE). Her roles in industry ranged from designing new bridges to working on sites to reconstruct housing destroyed by earthquakes. She returned to Cambridge for her PhD, spending part of her doctoral training as a Visiting Researcher with the German Aerospace Center (DLR) and received the Institute of Engineering Technology (IET) Leslie H Paddle Award for her work. Her achievements include being named on the Forbes 30 Under 30 Europe List (2016), and the Financial Times Top 100 Most Influential Women in UK Engineering list (2020), and being appointed to the Young Professionals Panel of the National Infrastructure Commission in the UK (2018). Dr Selvakumaran is currently the Isaac Newton Trust / Newnham College Fellow in Engineering.

Research Interests

Dr Sakthy Selvakumaran’s research interests focus on contributing toward more sustainable and resilient cities. She has particular expertise in using remote sensing (satellite monitoring) to understand its relevance, utilisation, and limitations to urban environments and civil engineering applications. Other research interests include topics which support overseas organisations she works with such as microhydro schemes in the Andes and sustainable wastewater treatment.


Road, tunnels, bridges and other infrastructure form the backbone of our cities. With increasing challenges like climate change and deteriorating structures, the buildings in our cities and increasingly at risk. What if we could use images from satellites to predict when (and what) might be about to fail? Satellites can provide us with millimetre-scale measurements that can be used to spot signs of problems. Combining these kinds of datasets with sensors and measurements on the ground provide new and exciting opportunities from construction of new structures through to maintenance of the existing built environment.

The following are notes from the on-line lecture. Even though I could stop the video and go back over things there are likely to be mistakes because I haven’t heard things correctly or not understood them. I hope Dr Selvakumaran, and my readers will forgive any mistakes and let me know what I got wrong.

The talk was about using satellites to look at very small-scale things on Earth and how we can use them to monitor really small-scale movements of structures.

Dr Selvakumaran works as an engineer and is currently the Isaac Newton trust fellow in engineering at the university of Cambridge. She is also the co-founder of a small company that’s set up to do lots of different types of monitoring

She began her career as a civil engineer and studied engineering at Cambridge.


A civil engineer is a person who practices civil engineering – the application of planning, designing, constructing, maintaining, and operating infrastructures while protecting the public and environmental health, as well as improving existing infrastructures that have been neglected.



Dr Selvakumaran was fascinated by how civil engineers could shape the built environment such as roads, tunnels, railways, water treatment centres etc. and how all of these things have a really high impact on society.

As a creative person she was excited by the idea that she could sketch something on a piece of paper and work out all the different forces that act on the design including the effect of its own weight. In the case of a pedestrian bridge this included working out the effect of people standing and walking on it, seeing how will it be affected by the materials used in building it and how the weather, such as the wind, would affect it.

The process also involved figuring out how all the building materials would work together, how much twist/rotation the structure could withstand and how heavy things needed to be to be able to resist the forces acting on the structure.

Just as exciting was seeing the sketch being built in real life.


The above image shows a 90-tonne piece of steel work being moved into place. It was about 34 metres long and about 6 to 10 metres wide.

It was one single piece that arrived in the back of a truck at about five a.m one morning


Dr Selvakumaran is one of the people in the above image. How small they look indicates how large the steel structure was.

It’s amazing how a sketch of a structure with forces calculated gets built and transformed in real life.

This particular structure became a bridge as part of Crossrail, which is a new train line going east-west across London, and it will be used by pedestrians, cyclists, people in wheelchairs moving from one side to the other. It was a really exciting project to work on.


The finished bridge on Crossrail



Crossrail is a railway construction project under way mainly in central London. Its aim is to provide a high-frequency suburban passenger service crossing London from west to east, to be branded Elizabeth line, by connecting two major railway lines terminating in London, the Great Western Main Line and the Great Eastern Main Line. The project was approved in 2007, and construction began in 2009 on the central section and connections to existing lines that will become part of the route.

The main feature of the project is the construction of a new railway line that will run underground from near Paddington Station via central London and Liverpool Street Station to Stratford. Another almost entirely new line will branch off the main line at Whitechapel in east London. It will run to Canary Wharf, cross the Thames and connect with the North Kent Line at Abbey Wood in south east London.


As a civil engineer, other projects Dr Selvakumaran has worked on include tunnelling. For this she had to learn about different soil types. With some soils it’s a bit like tunnelling through toothpaste which provides some challenges. Other soils are pretty stiff. You need to consider the soil as well as forces when you dig underground

Dr Selvakumaran has also worked on a number of different types of new bridges and structures as well as existing structures to try and keep them safe and increase the number of people that are able to use them.

Her work has also taken her overseas.


The above images show some of the projects she worked on in Peru which involved some pretty challenging conditions.


Peru is a country in western South America.



The photo above was a micro hydra scheme in the Andes. This region was very remote and mountainous meaning not only was it a difficult place to get to but it was really hard to get equipment, e.g., lifting equipment, there. Coupled with the fact that any normal structure built there would be very hard to maintain.


The Andes, Andes Mountains or Andean Mountains are the longest continental mountain range in the world, forming a continuous highland along the western edge of South America.

It extends from north to south through seven South American countries: Venezuela, Colombia, Ecuador, Peru, Bolivia, Chile, and Argentina.


Working in difficult conditions in different places throws up different unique challenges.


The above image is a landslide mitigation scheme in a region of Peru where there were a number of slums that had been built up over time. The problem being that the homes were flimsy and were at risk of being destroyed by landslides. The walls in the scheme were built to dissipate landslides as they came down the valley and take away some of their kinetic energy (energy that they have due to their motion), as they moved down. So, by the time they hit the village. it wouldn’t be wiped away.


Landslide mitigation refers to several man-made activities on slopes with the goal of lessening the effect of landslides. Landslides can be triggered by many, sometimes concomitant causes. In addition to shallow erosion or reduction of shear strength caused by seasonal rainfall, landslides may be triggered by anthropic activities, such as adding excessive weight above the slope, digging at mid-slope or at the foot of the slope. Often, individual phenomena join together to generate instability over time, which often does not allow a reconstruction of the evolution of a particular landslide. Therefore, landslide hazard mitigation measures are not generally classified according to the phenomenon that might cause a landslide. Instead, they are classified by the sort of slope stabilisation method used:

Geometric methods, in which the geometry of the hillside is changed (in general the slope);

Hydrogeological methods, in which an attempt is made to lower the groundwater level or to reduce the water content of the material

Chemical and mechanical methods, in which attempts are made to increase the shear strength of the unstable mass or to introduce active external forces (e.g. anchors, rock or ground nailing) or passive (e.g. structural wells, piles or reinforced ground) to counteract the destabilizing forces.

Each of these methods varies somewhat with the type of material that makes up the slope.


The above images show the creation of earthquake resistant housing.

During her studies Dr Selvakumaran learnt about the effects of earthquakes. Peru is in a region where earthquakes are common.


An earthquake is the shaking of the surface of the Earth resulting from a sudden release of energy in the Earth’s lithosphere that creates seismic waves.

At the Earth’s surface, earthquakes manifest themselves by shaking and displacing or disrupting the ground.


Earthquake epicentres occur mostly along tectonic plate boundaries, and especially on the Pacific Ring of Fire.

An ongoing project in Peru is to build earthquake resistant housing. Normal building materials such as certain types of masonry and concrete will crack and fall when the ground shakes but the Peruvian houses were built with a timber frame and interwoven cane walls. So as the houses shake, they have some flexibility and they’re able to withstand some of the earthquake force. The idea was to build into the structures a resistance and resilience to natural hazards.

Of course, it isn’t just Peru that has problems.

The UK, Ireland, US, Europe and the whole World in general have problems that would concern civil engineers.


The Broadmeadow viaduct, in Ireland, carries the main Dublin to Belfast railway across the Broadmeadow Estuary, about 13 kilometres north of Dublin, just north of Malahide.

The image below is of the Broadmeadow viaduct that failed in 2009 as a train was moving over it.


On Friday 21 August 2009, at 6:30pm, a 20-metre section of the viaduct collapsed. Some reports state that the collapse started while a passenger train was passing over it, others say shortly after. The driver of the Balbriggan to Dublin Pearse service was passing over the viaduct and noticed the section crumbling away. He coasted the train (not increasing the engine revs in case the vibrations made it collapse while the DMU was on it) across it. The train was not derailed and no one was hurt: the driver raised the alarm when he arrived at Malahide.

The train was full of passengers but no one was hurt. Thanks to the driver a tragedy was averted.

This particular bridge failed due to scour.


Bridge scour is the removal of sediment such as sand and gravel from around bridge abutments or piers. Scour, caused by swiftly moving water, can scoop out scour holes, compromising the integrity of a structure.


Diagram of how scour holes are generated

It’s a bit like the effect of sandpaper on wood.


Sandpaper and glasspaper are names used for a type of coated abrasive that consists of sheets of paper or cloth with abrasive material glued to one face.

Despite the use of the names neither sand nor glass are now used in the manufacture of these products as they have been replaced by other abrasives such as aluminium oxide or silicon carbide. Sandpaper is produced in a range of grit sizes and is used to remove material from surfaces


The bridge scour effect occurs when the river suddenly changes path or it floods and it becomes heavier. The way the river moves through and down makes changes to the river bed, which gets “sand papered away”. The piers of the bridge resting on the riverbed become unstable as the riverbed is being worn away because eventually there comes a point when there’s a big hole under the bridge and it collapses.

In the UK there are quite a lot of these types of failures. They are terrifying and cause a risk to life and damage to local economies. They’re also expensive to fix.




Pooley Bridge in Ullswater, Cumbria and the Keswick Railway Path Bridge collapsed following heavy flooding.


An inquest into the death of a police officer who was killed when a bridge collapsed during the Cumbria floods was told today more lives could have been lost had passers-by not intervened.

The hearing was told PC Bill Barker had been directing traffic off the Northside bridge in Workington when it gave way beneath him early on 20 November.


There are many other ways that structures can fail.

The image below shows the I-35 west Mississippi river bridge in the US


The I-35W Mississippi River bridge (officially known as Bridge 9340) was an eight-lane, steel truss arch bridge that carried Interstate 35W across the Saint Anthony Falls of the Mississippi River in Minneapolis, Minnesota, United States. The bridge opened in 1967 and was Minnesota’s third busiest, carrying 140,000 vehicles daily. It had a catastrophic failure during the evening rush hour on August 1, 2007, killing 13 people and injuring 145. The NTSB cited a design flaw as the likely cause of the collapse, noting that a too-thin gusset plate ripped along a line of rivets, and additional weight on the bridge at the time contributed to the catastrophic failure.


The National Transportation Safety Board immediately began a comprehensive forensic engineering investigation that was expected to take up to eighteen months.

A forensic engineering investigation is done in the same way as one done after any crime in order to find out why the collapse happened. There are two reasons for this. Is a person responsible for the event (are they guilty of a crime) and can civil engineers try and learn from any mistakes?

In this particular case the investigators were treating it as a crime scene but there was no indication there was any foul play involved, [but] it’s a crime scene until they can determine what was the cause of the collapse

On November 13, 2008, the NTSB released the findings of its investigation. The primary cause of the collapse was the undersized gusset plates, at 13 mm thick. Contributing to that design or construction error was the fact that 51 mm of concrete had been added to the road surface over the years, increasing the static load by 20%. Another factor was the extraordinary weight of construction equipment and material resting on the bridge just above its weakest point at the time of the collapse. That load was estimated at 262 tonnes, consisting of sand, water and vehicles. The NTSB determined that corrosion was not a significant contributor, but that inspectors did not routinely check that safety features were functional.

In this event the scary thing was that the inspection reports had photos of the bridge and the investigators could actually see that part of the bridge had started to bow out a little bit. There were also signs that the bridge was starting to move in an unusual way.


The bridge had been regularly inspected since 1993 and reports had thrown up problems and in 2005 it was rated as “structurally deficient” and in possible need of replacement. On a separate measure, the I-35W bridge was rated “structurally deficient”, but was deemed to have met “minimum tolerable limits to be left in place as it is.

Visual inspections of the bridge missed things. So there needs to be methods to help identify possible problems.

Another famous bridge collapse that happened more recently was the Morandi bridge


Ponte Morandi (English: Morandi Bridge), officially Viadotto Polcevera (English: Polcevera Viaduct), was a road viaduct in Genoa (Italy), constructed between 1963 and 1967 along Italy’s A10 motorway over the river Polcevera, from which it derived its official name. The bridge is widely called “Ponte Morandi” after its structural designer, the noted engineer Riccardo Morandi.

The bridge was an engineering and architectural landmark since its construction. It connected Genoa’s Sampierdarena and Cornigliano districts across the Polcevera valley. It also provided a critical artery of European route E80, linking Italy and France.

When a 210 metres section of the viaduct collapsed during a rainstorm on 14 August 2018, 43 people died — leading to a yearlong state of emergency in the Liguria region, extensive analysis of the structural failure and widely varying assignment of responsibility.

A huge loss of life, homes and a real disruption to the economy. The sad thing was that inspections had thrown up problems but nobody realised how bad the problems were.



The collapsed part of the bridge is shown in red

There must be better ways to spot very serious problems in structures. Visible inspections can’t show up everything.

Inspection, monitoring and maintenance of infrastructure


Visible bridge inspections are the most common methods of inspection. They can be helpful but there are lots of new technologies that can help keep the bridges and other structures safe,

Visible inspections have been done historically and are common practice for looking after structural assets. When Dr Selvakumaran worked as an engineer in industry one of her jobs was doing visual bridge inspections. This often entailed her sitting in a cherry picker.


An aerial work platform (AWP), also known as an aerial device, elevating work platform (EWP), bucket truck or mobile elevating work platform (MEWP) is a mechanical device used to provide temporary access for people or equipment to inaccessible areas, usually at height. There are distinct types of mechanized access platforms and the individual types may also be known as a “cherry picker” or “scissor lift.”



It turns out that most bridges are only inspected every few years, and usually at a distance. Major road bridges are scrutinised more than other road bridges. For major roads and railway bridges visible inspection means getting up close, often in a cherry picker, taking careful measurements and photographs. However, in bad weather, when it’s raining, snowing and very cold, inspections can take place rather quickly. The engineer might not be focused enough and fail to spot something. Or the flaw might just not be visible to the naked eye.

Dr Selvakumaran was driven to understand what kind of technologies could be used to support engineers like her to keep structures safe for the people around us. To this end she joined a group in Cambridge called the centre for smart infrastructure and construction.



This group does a lot of work around sensing so why not put sensors on the structures in order to give a more data-driven approach to how the structures can be monitored.

Dr Selvakumaran worked with a particular type of sensor which will be discussed later.


So, structures like the Hammersmith flyover age and deteriorate over time.


The Hammersmith flyover is an elevated roadway in West London which carries the A4 arterial road over and to one side of the central Hammersmith gyratory system, and it links together the Cromwell Road extension (Talgarth Road) with the start of the Great West Road. It is one of the first examples of an elevated road using reinforced concrete.

Not only do we have aging, deteriorating structures but we have climate change to contend with.

A lot of bridge stock in the UK was built not long after the end of the second world war and are getting pretty old and decrepit.

Like people, as structures get older and older, they tend to deteriorate and have structural health problems. Couple that with the effects of climate change and lots more flooding and even greater problems arise.

Some people think snow is fun but when you get snow you get more ice. Icy roads inevitably means putting salt on the them. Salt actually accelerates the process of corrosion and so structures are further and further under attack.


Britain is also getting a lot more floods than it used to, which is another effect of climate change and floods cause a lot of damage. This means that structures should be monitored more regularly than every six years or so

Dr Selvakumaran believes that the monitoring of structures should be connected into a more data-centric, data-driven future and we should be collecting as much data as we can to better understand how our existing structures behave as well as coming up with better methods of maintaining them against hazards, climate change and general age and deterioration.

The focus of her work involved satellites and she spent time working with an organization called the German aerospace centre (Germany’s space agency) to see how satellites could be used to support the data-driven vision for protecting our built environment.



The German Aerospace Centre, abbreviated DLR, is the national centre for aerospace, energy and transportation research of Germany. Its headquarters are located in Cologne and it has multiple other locations throughout Germany. The DLR is engaged in a wide range of research and development projects in national and international partnerships. In addition to conducting its own research projects, DLR also acts as the German space agency. As such, it is responsible for planning and implementing the German space programme on behalf of the German federal government. As a project management agency, DLR also coordinates and answers the technical and organisational implementation of projects funded by a number of German federal ministries.


The above image shows TanDEM-X and TerraSAR-X flying in formation. Credit: DLR


The new German Earth observation satellite TerraSAR-X was launched in June 2007. The objective of this five-year mission is to provide radar remote sensing data to scientific and commercial users. The satellite’s design is based on the technology and expertise developed in the X-SAR and SRTM SAR missions (Synthetic Aperture Radar). The sensor has a number of different modes of operation, with a maximum resolution of one meter, and is capable of generating elevation profiles.

TerraSAR-X is the first satellite that was jointly paid for by government and industry. DLR contributed about 80 percent of the total expenses, with the remainder being covered by EADS Astrium. The satellite’s core component is a radar sensor operating in the X band and capable of recording the Earth’s surface using a range of different modes of operation, capturing an area of 10 to 100 kilometres in size with a resolution of 1 to 16 meters.

Everybody is familiar with satellites and what they can do. To some extent we use them every day for, example, our weather forecasts.


The picture above right is an optical image. A satellite is basically taking photos of the Earth. It uses the sunlight being reflected off the surface of the Earth to affect a form of digital camera on the satellite and recorded as a photo.

Satellites can also be used to track progress over time. Below are some images taken by one of Dr Selvakumaran’s colleagues at the German aerospace centre and it shows that we can track development over time.


The images show how things are changing and how things are becoming urbanized.

It isn’t just optical sensors that are used for the monitoring. There are multiple different sensors so the above right image used a sensor that measured height differences.

The different sensors allow classification and show how the urban environment is changing over time.

The satellites are unsurprisingly getting better and better, going from something that could only give a kind of coarse low pixel, low spatial, low time resolution images to finer and higher resolution images. Not only that but more of them are being put in orbit (although not everyone is happy with more satellites orbiting the Earth e.g., astronomers).


Lots of satellites are put in orbit in series, one after the other. This is so a site can be revisited more frequently in order to track development.

One of the satellites involved in the monitoring uses radar. The principle behind it is the same as that used by policeman for checking the speed of cars (although these are being phased out).


Radar is a detection system that uses radio waves to determine the range, angle, or velocity of objects. It can be used to detect aircraft, ships, spacecraft, guided missiles, motor vehicles, weather formations, and terrain. A radar system consists of a transmitter producing electromagnetic waves in the radio or microwaves domain, a transmitting antenna, a receiving antenna (often the same antenna is used for transmitting and receiving) and a receiver and processor to determine properties of the object(s). Radio waves (pulsed or continuous) from the transmitter reflect off the object and return to the receiver, giving information about the object’s location and speed.


A radar speed gun (also radar gun and speed gun) is a device used to measure the speed of moving objects. It is used in law-enforcement to measure the speed of moving vehicles and is often used in professional spectator sport, for things such as the measurement of bowling speeds in cricket, speed of pitched baseballs, and speed of tennis serves.

A radar speed gun is a Doppler radar unit that may be hand-held, vehicle-mounted or static. It measures the speed of the objects at which it is pointed by detecting a change in frequency of the returned radar signal caused by the Doppler effect, whereby the frequency of the returned signal is increased in proportion to the object’s speed of approach if the object is approaching, and lowered if the object is receding.





The Doppler effect (or the Doppler shift) is the change in frequency of a wave in relation to an observer who is moving relative to the wave source. It is named after the Austrian physicist Christian Doppler, who described the phenomenon in 1842.


Change of wavelength caused by motion of the source.


An animation illustrating how the Doppler effect causes a car engine or siren to sound higher in pitch when it is approaching than when it is receding. The red circles represent sound waves.



Christian Andreas Doppler (29 November 1803 – 17 March 1853) was an Austrian mathematician and physicist. He is celebrated for his principle – known as the Doppler effect – that the observed frequency of a wave depends on the relative speed of the source and the observer.

Radar satellites emit signals and receive the reflected signals back. All the signals are processed using interferometric synthetic aperture radar.


Interferometric synthetic aperture radar, abbreviated InSAR (or deprecated IfSAR), is a radar technique used in geodesy and remote sensing. This geodetic method uses two or more synthetic aperture radar (SAR) images to generate maps of surface deformation or digital elevation, using differences in the phase of the waves returning to the satellite or aircraft. The technique can potentially measure millimetre-scale changes in deformation over spans of days to years. It has applications for geophysical monitoring of natural hazards, for example earthquakes, volcanoes and landslides, and in structural engineering, in particular monitoring of subsidence and structural stability.

Synthetic aperture radar (SAR) is a form of radar in which sophisticated processing of radar data is used to produce a very narrow effective beam. It can be used to form images of relatively immobile targets; moving targets can be blurred or displaced in the formed images. SAR is a form of active remote sensing – the antenna transmits radiation that is reflected from the image area, as opposed to passive sensing, where the reflection is detected from ambient illumination. SAR image acquisition is therefore independent of natural illumination and images can be taken at night. Radar uses electromagnetic radiation at microwave frequencies; the atmospheric absorption at typical radar wavelengths is very low, meaning observations are not prevented by cloud cover.


The top right image is a typical SAR image (it is in black and white for a very good reason) and looks almost like a photo. The area it is showing is the south west of England, around the city of Bristol. The Bristol and English Channels can be seen along with a bit of South Wales. The different tones of the image are due to the different amplitudes of the reflected waves.

Objects that show up as very bright are the things that are reflecting radar back to the satellites really well. Things like the mountains, rocks and cities. Urban structures full of concrete and metal will reflect radar really well and show up as bright images too.

Forests and trees are things that won’t show up as well because they will tend to disperse the radar in different directions and not necessarily reflect the radar directly back to the satellite. They will come up a bit darker in the image.

Water comes up very dark indeed because the radar signal hits the water and then just gets reflected all over the place and not back up to the satellite. Therefore, it is only part of the processed image by its absence.

The satellites are continually orbiting the Earth taking images of the same spot over and over again. These images are collected and processed using interferometric synthetic aperture radar (InSAR)


Interferometric Synthetic Aperture Radar (InSAR), or SAR Interferometry, is the measurement of signal phase change between two images acquired over the same area, at different times. When a point on the ground moves, the distance between the sensor and the point changes and so the phase value recorded by the sensor will be affected too.


The change in signal phase (Δφ) is expressed by the equation below:


Where l is the wavelength, ΔR is the displacement in the Line Of Sight (LOS) and α is a phase shift due to different atmospheric conditions at the time of the two radar acquisitions.


An interferogram is the difference of the phase values corresponding to a certain area, i.e., it is a digital representation of a change in surface displacement. It corresponds to a matrix of numerical values ranging from -π to +π (phase variations) and it can be converted into a map – the easiest way to observe whether or not motion has occurred over a certain area.

In simple terms imagine that the satellite is flying overhead and it’s sending a sinusoidal radar wave down to the Earth. It reflects off an object and it goes back up to the satellite. The number of LOS wavelengths returned is noted. The next time the satellite comes around again it sends another radar wave down and it comes back up again. Again, the number of LOS wavelengths returned is noted. If the two values are different, usually a fraction of a wavelength difference, then it shows the object has moved. Because radar uses 3cm microwaves this method can measure millimetres of movement.

It’s amazing to think that a satellite 500 kilometres above the surface of the Earth can measure millimetres of movement

Processing all of the data together, making sure that the images are stacked one on top of each other at the same time, considering atmospheric effects, thinking about the satellite motion and using the various different physics principles produces an image like the one below.


As mentioned earlier some things reflect really well and some don’t so what engineers want to do is pick the points that are reliably reflecting.

Water doesn’t reflect very well so it’s not going to have that many stable reliable reflective points but things like buildings do.


All of the dots on the above image indicate points that reflect really, really well. They are mainly buildings.

A particular relatively reliable point can be chosen and any movement measurements recorded.


Essentially the measurements are one-dimensional line of sight because the radar signal is going down to Earth and then coming back up again.

However, the measurements are being made from various different satellites and in each of these cases they’re measuring in a one-dimensional line of sight,

If an east-west direction measurement is required only one component is needed.


Pythagoras theorem is applied on a large scale. We know what angle the satellite is taking its image at and we know that it’s looking sideways.

If the process is using an angle to the vertical, we can work out what component of that measurement is caused by the movement in a particular direction

Obviously, there’s a number of limitations in doing this including the fact that a one-dimensional line-of-sight measurement can’t be decomposed into three dimensions. However, we can measure some component of it and because different satellite measurements are used, we can combine some of them together and get a closer idea of 3d motion.

Another limitation is caused by the satellite moving north-south over time. As it orbits the Earth it’s taking an image to it’s left or right (an east-west direction) meaning it’s not able to capture an image in the north-south direction as well.

So, on the one hand the satellite can measure millimetres of movement but on the other it can’t capture all directions.

So, it can’t be used for everything but it can still do some pretty exciting stuff

image image


Map of ground movement in the Bank region of London using TerraSAR X data

The above image is called a velocity map. Each of the points are stable reflections. The satellite has taken many measurements over time and collected a stack of measurements.

It is called a velocity map because it is showing how the movements change over time.

The scale shows that the blue regions are sinking and the red regions are rising. Anything that’s a kind of yellow-green is fairly stable.





Point here looks like the ground is sinking. The arrow is referring to the graph.


The plot of each of the measurements is on a millimetre scale and the graph shows that the maximum amount of sinking is only about 20mm (the diameter of 2p coin). Yes, the buildings in that area are moving but in this case it’s not something to be worried about. It’s actually expected because this is the site of the Bank tunnelling work. This is being done to connect a couple of existing stations.

When tunnels are dug some of the earth above the tunnel sinks a little bit. This is to be expected. All engineers need to do is make sure it doesn’t sink too much as that might cause damage to the buildings above.

This imaging was a really exciting way to monitor the little bit of tunnelling happening in a very localized area and show Transport for London that there was only a couple of centimetres of movement.


Transport for London (TfL) is a local government body responsible for the transport system in Greater London, England.


Tadcaster Bridge or Wharfe Bridge spans the River Wharfe in Tadcaster, North Yorkshire, England. The road bridge is believed to date from around 1700. It is the main route connecting the two sides of the town and one of two road crossings in the town, the other being the bridge for the A64 bypass. Tadcaster Bridge partially collapsed on 29 December 2015 after flooding that followed Storm Eva, and reopened on 3 February 2017.


Another application of this imaging is investigating scour failure, which was mentioned earlier.

There was an extended period of flooding in Tadcaster before part of the bridge collapsed and so for this bridge.


This area in red on the map shows a very localized area of collapse of the bridge


The embankment underneath the bridge (pier 5) was worn away until the bridge was no longer supported and that part of the bridge collapsed.

Dr Selvakumaran decided to do a study of the bridge using satellites.

image image

So, she picked the reflective points on the bridge in the satellite imagery and one of them was actually in the middle of the bridge right at the point where it collapsed.

So, she studied satellite images taken before the collapse to see how that point moved over time. She processed the data of the selected point, that related to the bridge, and plotted the small movements between each acquisition taken over time.

Each acquisition was recording the number of waves received by the satellite. Any difference in the number of waves received during each subsequent acquisition would indicate movement had occurred. So, Dr Selvakumaran took acquisition after acquisition and measured fractions of a sine wave difference, which corresponded to tiny movements and plotted those tiny movements over time.

Dr Selvakumaran initially looked at data from 2014 to 2015.


Each of the above points is a satellite data acquisition

There is a little bit of movement and some of the measurement is background noise, not an actual movement, but generally the bridge is staying stable over time.

She plotted a bit more data for 2015 and noticed something strange happening on the 15th of November. The bridge started moving a lot more. The next acquisition showed even more movement compared to what is classified as normal and this was a little alarming and the bridge actually collapsed on the 29th of December.


The results showed there was a sort of warning almost a month and a half before the bridge collapsed



The measurements indicated the point where the bridge was starting to sink, indicating there might be a problem. The pier was already moving down so there might be a hole underneath it.

It would be very useful to the owner of a structure to have prior knowledge of a problem. With the bridge there is no way that a scour hole beneath a pier can be seen, especially during flooding where the water is full of mud and particulates, and it would be too dangerous to send in a diver

The ability of satellites to be able to monitor these very, very tiny movements and give asset owners a warning is an incredibly powerful tool for engineers so not only can failures be spotted but the structures can be monitored for day-to-day behaviour. This kind of movement is completely normal.

Bridges have problems in their day-to-day life. They are designed to expand and contract as the temperature changes or as traffic rolls over them and engineers facilitate that kind of movement by including expansion joints or bridge bearings.



Expansion joint on a bridge

An expansion joint or movement joint is an assembly designed to hold parts together while safely absorbing temperature-induced expansion and contraction of building materials, and vibration, or to allow movement due to ground settlement or seismic activity. They are commonly found between sections of buildings, bridges, sidewalks, railway tracks, piping systems, ships, and other structures.

Building faces, concrete slabs, and pipelines expand and contract due to warming and cooling from seasonal variation, or due to other heat sources. Before expansion joint gaps were built into these structures, they would crack under the stress induced.



An expansion bearing on the Queen Elizabeth II Metro Bridge.

A bridge bearing is a component of a bridge which typically provides a resting surface between bridge piers and the bridge deck. The purpose of a bearing is to allow controlled movement and thereby reduce the stresses involved. Possible causes of movement are thermal expansion and contraction, creep, shrinkage, or fatigue due to the properties of the material used for the bearing. External sources of movement include the settlement of the ground below, thermal expansion, and seismic activity. There are several different types of bridge bearings which are used depending on a number of different factors including the bridge span, loading conditions, and performance specifications

However sometimes bridge bearings get stuck or expansion joints develop problems and the bridge is just not moving in the expected way. This would normally be difficult to spot.

Some bridges have a normal movement of a few millimetres and some of them can move tens of centimetres it all depends on the arrangement of the materials that make up the bridges and the temperature. There are other factors to but the bridges all have some kind of signature and if engineers can start monitoring bridges more regularly and understand what their signatures are, they should be able to spot any unusual movements.

Unfortunately, not all bridges have reflective points and therefore won’t be able to reflect satellite radar very well.

One such bridge is Waterloo bridge in London. So, when bridge data is processed the engineers don’t get a lot of measurements


Waterloo Bridge is a road and foot traffic bridge crossing the River Thames in London, between Blackfriars Bridge and Hungerford Bridge.




Dr Selvakumaran and her team tried to see how they could augment the bridge to reflect better. This involved placing reflectors at certain points along the bridge.

The reflectors work in a similar way to bicycle reflectors.



Bicycle reflectors are made of a large number of small tetrahedrons set together so that there are dips in between in the shape of a corner of a right-angled cube. The two diagrams above show that whatever angle the light beam hits the corners it will reflect back along its original path.

The reflectors on the bridge are the little triangle shapes.



Above left: Annotated SAR amplitude image with bridge marked and corner reflector installation appearing as bright points. Above right: Aluminium corner reflector (circled) installed on bridge pier alongside ATS (Automated Total Stations) prism targets.



The above image shows ATS target prism locations on the bridge (base optical image provided by Google Earth).

The satellite reflectors are just basically three pieces of metal put together so that the satellite signal can hit it and reflect it back to the satellite.


The above image indicates one of the bridge’s piers


The team installed these reflectors at critical points on the bridge so for bridge movement they’re really interested in what each of the piers are doing and whether the expansion joints are working properly.

The bridge was augmented with these reflectors and measurements were recorded using satellites. Traditional surveying techniques were also used.


Prisms, as seen above, are also found on the bridge and form part of the traditional surveying equipment.



Often people can be seen surveying or taking measurements from the ground in three dimensions to see how structures are behaving

Dr Selvakumaran and her team were able to carry out a comparison between traditional surveying and satellite measurements.


In the zoomed in region of the above image there are very bright spots which are the locations of the reflectors. The team were able to track the bridge’s behaviour over time.


The above image is a plot of relative ATS and relative SAR movements between Piers 2 and 6 on the east side of Waterloo bridge.

The blue part of the above graph shows the results from traditional surveying. This can be done more frequently, every few minutes rather than every few days.

The pink part of the above graph shows the satellite’s measurements.

The orange part of the graph shows how the temperature varied over the time period, December 2017 to December 2018. This enabled the team to see how the temperature affected the bridge

The bridge was affected by the changing temperatures. The graph shows that even in the winter it heated up and cooled down on a daily basis causing an expansion and contraction. There was also expansion and contraction during the summer.

The data from the two measuring techniques don’t match perfectly but what the graph does show is that through the satellite measurements the team were able to capture the behaviour of expansion and contraction over the summer and winter.

If the satellite shows the bridge isn’t expanding and contracting like it normally should it doesn’t fix the problem but at least it highlights to the bridge’s owner that there might be a problem and maybe engineers should go and have a look to check its health and diagnose any problems before it becomes serious.

The reality is that once the problem is visible it’s often too late to do something about it. A visible crack would be harder to repair than a crack that isn’t.

The earlier problems are spotted the quicker some preventative measures can be put in place to prevent the problems getting worse.

The monitoring processes can be used to monitor all sorts of different phenomena.


The image above shows another velocity map where the blue regions shows areas that are sinking.

There are some good reasons why parts of central London are sinking but we shouldn’t be worried.

At the moment a new super sewer is being built.


The Thames Tideway Tunnel will be a 25 km tunnel running mostly under the tidal section of the River Thames through central London to capture, store and convey almost all the raw sewage and rainwater that currently overflows into the river.


Proposed route of the super sewer.

London’s sewage system was built in the Victorian era and it was designed for a smaller population.

When it’s a really stormy day and there’s lots of rain the sewage system can’t cope and the sewage overflows directly into the Thames. This is very unpleasant and harms the wildlife that relies on the Thames.

The new tunnel will transport the sewage to a treatment centre rather than dumping it into the Thames.

Some shafts are being dug in central London and when they are put in the ground where the water table is really high there is a limit to how far down, they can go. Too far down and there is water everywhere so what engineers do is they take the water table level down, Taking water out of the ground causes the entire area to sink. This is seen in the Battersea area where everything seems to be sinking. Eventually the water will be released and the ground will go back up again. This whole process of planned sinking and rising can be tracked. It is completely normal. However, in some areas water companies can show their customers the effect of taking too much water out of the ground.



Another big construction project in central London is Crossrail. The project has likely resulted in the slight drop in levels, although extensive compensation grouting has aimed to minimise the changes.


Crossrail is a railway construction project under way mainly in central London. Its aim is to provide a high-frequency suburban passenger service crossing London from west to east, to be branded Elizabeth line, by connecting two major railway lines terminating in London, the Great Western Main Line and the Great Eastern Main Line. The project was approved in 2007, and construction began in 2009 on the central section and connections to existing lines that will become part of the route.

The monitoring techniques can also show up the effects of mining and tunnelling and indicate whether there is a danger of subsidence.

The beauty of the satellite method is that it can be done remotely anywhere in the world and spot things when they’re actually happening.

It can sometimes even spot illegal activities where people are doing things, they shouldn’t be doing, by the very small movements being made.


Satellites can actually measure many things:

Optical imaging and optical measurements;

Monitoring air pollution;

Tracing forest fire scars.

Dr Selvakumaran finds the built environment as exciting as building new stuff and she really does value the opportunity to design and see things come to life

For her it’s all about understanding how we can connect our built environment into the into the digital era and use a whole array of tools including satellites to help support and keep things safe, resilient and ready for the future.

Questions and answers

1) Don’t you have to worry about the atmosphere interfering with the radio waves?

What’s exciting about the satellite radar is that unlike optical satellites you can use it day and night. You don’t have to wait for sunlight or anything like that. Radar passes through clouds so you can use it at any time it’s not affected.

The atmosphere does have an impact but there are ways and means around it. The larger the number of images taken the easier it is to account for random noise as well as atmospheric effects. This gives a more reliable result. Additionally, understanding the area being processed means atmospheric effects can be taken into consideration.

Satellite images can show an entire city such as Cambridge using high resolution satellites.

If the satellite is looking at a couple of things in a very small area the atmospheric difference between them is unlikely to be great so the atmosphere won’t have much of an effect and can be taken into account. The calculations involved are not easy and could probably be the subject for a lecture on their own.

2) I was wondering why concrete and buildings are very good at reflecting light?

Just to correct, it’s not light but radar. If you imagine, for example, things that are metal and you shine a light on them (although we’re not actually talking about light) they will reflect differently from light shining on a tree that is moving or water. Light coming off the metal tends to travel in set directions whereas a tree will scatter light in lots of different directions.

It’s not just the material itself but a combination of material and geometry that can create sharp reflections where the wave bounces back along the path that it took from the source.

Sometimes an object will reflect quite well and sometimes an object needs to be augmented if it reflects badly.

Water or for example a farmer’s field are poor reflectors of radar.

Imagine sending a radar signal into a field of crops and you will get a certain type of signal back. Then the farmer harvests the crops and a radar signal will not be reflecting off the same object as before. There won’t be a coherent, stable reflection anymore so the process is not useful. There are ways and means around this. You could just put corner reflectors into fields and that would help and there’s a couple of other different processing techniques that can be used to take into account those kinds of things but yes, the reflections are a product of material and geometry mostly.

3) The reflectors on the bridge didn’t look very sturdy how do you know the reflector isn’t just blowing in the wind?

That’s the fun of being a structural engineer you do experiments. Dr Selvakumaran designed the brackets. They’re pieces of aluminium and are actually thicker than they look. They look quite small and flimsy relative to the bridge but they are fairly sturdy. They’re basically attached to an arm bracket that has a ball on one side and a rotating arm on the other so it’s almost like an arm joint. They are locked it into place and then everything is bolted tightly together. They were fixed on to Waterloo bridge with the permission of the asset owner of course. They’re very stably stuck there because they’ve got to withstand wind. The aluminium sheeting was designed with holes to facilitate drainage. You’ve also got to think of pigeon loading because they perch on things and make a mess.

4) How long would a satellite be able to measure an object when passing over once?

As the satellite passes over, it images an area. Different satellites do this differently but some of them will just image straight over as they move. The images are taken over very short periods of time so it’s almost like a snapshot. One image is taken at one angle and another is taken at a different angle.

For measuring movement one image is not interesting. You really need more than one image. With two images you can a spot any differences. This is useful for many applications. If you had a sudden flood you could compare an image of the region with an image of the region before the flood and see where the standing bodies of water. You can do an instant flood mapping.

Another example of satellite use involved working with an organisation which was helping in the aftermath of the blast in Beirut.

On 4 August 2020, a large amount of ammonium nitrate stored at the port of the city of Beirut, the capital of Lebanon, exploded, causing at least 204 deaths, 6,500 injuries, and US$15 billion in property damage, and leaving an estimated 300,000 people homeless.


Dr Selvakumaran did some work on damage assessment and some of the organisations working in the remote sensing and satellite community used a couple of SAR images before and after the explosion to see how the area had changed and which areas were most damaged.

You just need a couple of images to do something useful. However, to get over any atmospheric affects you need lots and lots of images (at least 20 to 30).

5) Does it matter whether the surface of a reflector is shiny or is more to do with the material used and does it matter if some gunk ended up on the reflector as well?

I think with pigeons involved we can imagine what sort of gunk we might be discussing here.

Dr Selvakumaran had carried out an experiment and had to keep cleaning the apparatus as the reflectivity was affected by some gunk. You can have reflectors that are painted with a white coating to make them kind of corrosion resistant and they still reflect so it’s not just down to the how shiny the metal

You can do some experiments yourself as there is a service by the European Space Agency called Sentinel where you can download fire radar for free



The European Space Agency is an intergovernmental organisation of 22 member states dedicated to the exploration of space. Established in 1975 and headquartered in Paris, ESA has a worldwide staff of about 2,200 in 2018 and an annual budget of about €6.68 billion (~US$7.43 billion) in 2020.





You can download any area over Europe, like your home, and see what’s reflecting well.

If the reflector gets covered by a pigeon sitting on it it’s not going to be effective. If it gets covered in too much material such as snow it might have a problem. Snow is an issue for certain parts of the world but you can actually design reflectors with different covers and various different configurations

The reflectors on Waterloo bridge were designed for that bridge for time period for the experiment.

The team are actually doing a couple of other trials of different configurations and types but the type of reflector also depends on where you’re installing it, how long it’s being used and what you want to do with it. Each reflector is bespoke and is designed for the particular requirements of the scenario that you’re looking at.

The function is mostly down to the type of radar. SAR radar comes in different bands.



The bands all have different wavelengths

C band is a wavelength that requires a much larger reflector than the X band and this is to do with the physics.

There’s actually a neat little formula for the radar cross sectional area that you need in terms of how well the reflector will reflect.


Radar cross-section (RCS) is a measure of how detectable an object is by radar. Therefore, it is called electromagnetic signature of the object. A larger RCS indicates that an object is more easily detected.

6) Do you have to get someone to clean the reflectors?

I think the audience is worried about the less glamorous side of scientific research.

I think it’s going to take a lot of bird mess before it seriously impacts on the reflector’s reflectivity.

7) I was wondering whether you found anything surprising, that you weren’t expecting? Whether you had seen something completely new? If possible, could you talk about something interesting you had found that was entirely unexpected.

In the UK we tend we tend to declare when we’re doing things like tunnelling. It’s pretty hard to tunnel something without someone realizing it’s happening. We tend to notice things like that.

There was something surprising once. We were doing something on a site and the area around the site was sinking and we didn’t know why. Everybody was blaming everybody else. “What have you done that’s causing the ground to sink?” and we’re saying nothing and then you look at the satellite imagery and you can see that actually someone else further away had been dewatering and they just didn’t understand how far that zone of influence of dewatering had affected and it had actually come as far as our site. The interesting thing about satellites is you can see what is going on in a wide area.


Dewatering is the removal of water from solid material or soil by wet classification, centrifugation, filtration, or similar solid-liquid separation processes, such as removal of residual liquid from a filter cake by a filter press as part of various industrial processes.

If one image is the size of a couple of cities you can measure the scale of motion across the wide area which is different from when you’re doing measurements on the ground

8) Are reflectors only put in place temporarily, just to measure the bridge, say, every five years or are these permanently attached?

The Waterloo bridge project was temporary but “temporary” on a bridge is actually quite long.

There’s are a number of different corner reflectors that have been installed all around the world on a more permanent basis. For example, in Norway as they have a lot of landslide problems. There’s a whole network of them that have been employed to regularly monitor and track and give a landslide warning system in place.

Corner reflectors are also used for calibration purposes and there’s a number that are installed around the world on a permanent basis. You might just pass some of them and think, okay, maybe they’re just a satellite reflector dish but they’re actually calibration systems for radar systems.

9) You mentioned dewatering a couple of times. That’s not something I’m familiar with. I mean I’ve got birch trees in the allotments around the corner for me which suck up quite a bit of water apparently but imagining something more industrial is difficult. So, what are people actually doing when they’re de-watering these sites?


The water table is an underground boundary between the soil surface and the area where groundwater saturates spaces between sediments and cracks in rock. Water pressure and atmospheric pressure are equal at this boundary.

If you can imagine there is water in the ground. Some of you have a water table that’s quite high. If there are people working in the ground, they need to take out the water before they can dig and work in the hole. It’s done in different ways for different circumstances but a simplified explanation of what you can do is to put some sheets in the ground. cut off an area that you’re working in and just basically suck the water out of the ground. Then put it back in when you’re done. You’re just lowering the water table down a bit and then bringing it back up again. It sounds quite simple an explanation but it’s a bit more complicated when actually doing it.

10) Is there any reason why buildings would rise, why things would lift up?

There’s an effect called ground heave.


Ground heave is the upward movement of soils which can occur at any stage of a building or development process. Ground heave is usually associated with clay soil which swells when wet. This then rises upwards – the effect is almost exactly the opposite of subsidence.

If you can imagine that the weight of soil several kilometres down is quite heavy. There’s quite a lot of force downwards. Now if you imagine a case where you suddenly dig a hole in the ground such as digging a basement or you’re cutting into a big hill or you’re developing a new road you suddenly take away a lot of soil. The ground below this soil has a certain amount of moisture and if you take all that weight of soil from it the ground can rise up. Engineers call it heave. The ground rises and heats up and it kind of rebounds against the pressure you’ve just taken off of it and so that’s where you can sometimes get buildings rising.

11) It’s been fascinating to hear about civil engineering as well. It’s one of those things that is around us all the time but you don’t actually hear about what goes into designing these things or the sort of people that are working on them. It’s a really fascinating insight into something that we might not be too familiar with.

It’s a really interesting time for people to go into engineering. When I started there was this focus on working in one sector but now with the rise of data-driven engineering problems are more easily solved.

I’m a member of the institution of civil engineers but there is also engineering technology and electrical engineering. There is also a Society for Imaging Science and Technology.

I do structural engineering and I also do lots of physics-based things.

Engineering involves people working in different disciplines. You can connect all these different pieces together. Yes, I think it’s a really exciting time to become an engineer.

Certainly, there is a typical sort of background for civil engineers and the sort of people you work with

The other exciting thing about civil engineering is that nowadays there’s a lot more ways to do it so you can either go to get a degree at a university or there are apprenticeships available. There are companies that will sponsor you. You work part-time in the company and do your degree part-time,

But in all cases, there’s a lot of scholarships available. There’s a lot of people willing to pay for you to do an engineering degree but you need to study physics and maths. Without those you get stuck.

12) If you lowered the water level how large of an area would be affected? Basically, how localised is the water table, if that makes sense?

That’s really site specific and it depends on what you’re doing and how you’re doing it. It’s like asking how long a piece of string is.

I’m really surprised by so many civil engineering questions. I thought that you all would be intrigued by this idea that something 500km up can measure a distance of two millimetres but apparently, you’re finding the water table stuff more interesting.

13) The students never thought you could do such cool stuff with civil engineering. You’ve certainly been a good ambassador today. Is there anything you’d like to add?

No, I think I’ve covered everything and hopefully the students will realize that actually in both remote sensing and a civil engineer there are some really exciting careers. I really got into the idea of remote sensing to solve problems and keep things safe, but also with the idea of monitoring sustainable development goals. Satellites can play a huge role in that area. It’s really an up-and-coming field. It’s definitely something for the students to look into if they’re interested.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s