Search Sonardyne

Follow us

Search Sonardyne

Follow us

Case study

How to optimise carbon storage monitoring with marine robotics

Partners: Fugro, National Oceanographic Centre (NOC), British Geological Survey (BGS) and Plymouth Marine Laboratory (PML)

June 24, 2021

For a long time, while carbon capture and storage (CCS) in offshore underground reservoirs has been widely regarded as a major way we can help reduce carbon emissions, it’s failed to attract the up-front investment needed to make it work.

In today’s far more climate conscious world, sentiment and interest in CCS has very much changed. Significant projects are now being planned, not least in Europe, where projects are moving forward in Norway, Netherlands and the UK. Carbon storage licenses are being awarded and wells being drilled specifically for carbon capture and storage.

There is some experience. Since 1996, CO2 from natural gas production on the Norwegian shelf has been captured and reinjected into formations beneath the seabed. But these projects are relatively small – and the potential is vast. On the UK Continental Shelf alone there’s at least 78 gigatonnes of CO2 potential storage capacity – some 200 times the UK’s 2016 emissions*.

It’s not going to be a small task to achieve the visions that operators from Equinor in Norway to BP in the UK are promoting, such as the Northern Lights project to the Northern Endurance Partnership.

From capture to transport by pipeline and injection into a suitable geological formation offshore, there’s a lot to process. But the challenges do not end there. What happens to the CO2 once injected and will we know if it finds a leak path to the surface?

Detecting CO2 in the marine environment

These are concerns that led, back in 2014, to a three-year funded research programme focused on developing the capability to reliably detect CO2 in the marine environment – a whole store capability. Funded by the Energy Technologies Institute (ETI), the project was delivered by a consortium of experienced companies including Fugro, National Oceanographic Centre (NOC), British Geological Survey (BGS) and Plymouth Marine Laboratory (PML) and Sonardyne.

It became apparent early on that different storage sites pose different types of risks, so different monitoring regimes would be required throughout each store’s lifecycle; such as pre-injection survey, ongoing monitoring during injection and post-closure monitoring. Our concept was shaped around building a ‘system of systems’ to enable CO2 storage operators to be able to conduct a risk-based plan for environmental monitoring, using different techniques at different times.

Four key technology elements for carbon storage monitoring

In broad terms, this systems approach identified four key elements. The first is a low-power and hence long-endurance autonomous underwater vehicle (AUV) to provide cost-effective wide-area coverage survey during baseline and repeat environmental surveys using a combination of our Solstice side scan sonar and chemical sensing. The second and third elements were seabed landers capable of detecting and monitoring any leakage at high-risk locations. These comprise of two different landers, one using an active sonar and the second combining passive sonar and chemical sensing.

The active sonar lander, based on our Sentry integrity monitoring system (IMS), gives sensitive and reliable automated leak detection capability across a wide area. For instance, around an injection well, Sentry can monitor an area of over 2.3 million square metres, to help visualise that its equivalent to around 325 football pitches. The passive sonar and chemical lander, uses the smarts from our underwater acoustics capabilities and is capable of both detection of leaks, but offers improved verification the potential to estimate leak rates at shorter ranges.

Featured in this story

Sentinel sonar head



Want to find out more about this project?

We can tell you more about this story

The fourth element is a surface gateway to enable communication between a shore- based monitoring office and the underwater systems. Such a gateway can be deployed from a fixed platform, from a moored buoy or from an uncrewed surface vessel (USV), many variants of which are now readily available in the market for over-the-horizon data harvesting missions. In fact, we now have a range of payloads suited specifically and now commonly used on operator’s USVs for their requirements, we also offer our own end-to-end data-harvesting service, when you just want the data and not the worry about the interfaces involved in getting it.

Acoustic data harvesting, from seabed to shore

On the ETI project, we used our wideband acoustic communications between the underwater landers and a buoy on the surface with all data forwarded on via satellite communications to a server. This is the type of set-up we deploy on long-term monitoring projects such as tsunami monitoring systems. It’s well-proven and used globally. Display and interpretation of the monitoring data can be simply integrated into a third party system; in this case a modified version of Fugro’s Metis software package. Built on GIS technology, this allowed non-expert users to access a web portal in which they could see the data visualised and run off reports.

The leak target was deployed in the North Sea, east of Bridlington, close to the location of what had been the proposed White Rose storage complex – a project which lost out under the closure of the CCS competition in 2015, but is now part of the previously mentioned Endurance project (named after a specific geological structure targeted for CO2 storage).

The National Oceanographic Centre’s (NOC’s) Autosub Long Range (ALR) was then set to work, deployed from the small port at Bridlington and towed a short distance off the coast. After the ALR performed a series of tests to demonstrate safe navigation, the leak – a small CO2 leak – was turned ‘on’ with a flow rate of between 16 and 20 litres per minute of gas at depth, depending on the state of the tide. This is equivalent to a leak of just less than 125 kilogrammes of CO2 per day, or around 45 tonnes per year. Although it sounds large, a leak of this size is actually only capable of changing the pH of the surrounding seawater by a few parts in a thousand. Such changes occur on an hourly basis due to entirely natural processes related to tidal mixing and respiration of all types of life in the ocean.

CO2 leak hunting with big data

With the leak “on”, ALR performed a series of different wide-area and fine-area search patterns over five days to search for the leak. The sensor hub on the vehicle processed in real-time a complex set of Solstice sonar, physical and chemical sensor data into useful information. The sonar data alone amounted to around 10 megabytes of raw data, every second, corrected for vehicle motion.

Automatic target recognition algorithms were used to identify any leaks or regions of interest. The system would then score these regions of interest and save a small “snippet” of the sonar image data. At regular intervals throughout the survey, ALR would surface and send back data via satellite, including navigation data, chemical and physical sensor data and details of snippets of sonar data from detected leaks – an example of which can be seen below. The operator could then request the highest scoring images of the leaks to be transferred, allowing a high confidence of the detection of a leak whilst the survey was still in progress.

All of the uploaded data was simultaneously transferred to an internet server which allowed presentation and interpretation using Fugro’s Metis software – an intuitive data delivery platform that allows metocean, vehicle navigation, chemical and sonar snippet data to be combined and displayed. This allowed data sharing across a wide team and supported operational decision making.

During the five days, the ALR travelled a total of 270 km and could have surveyed 54 sq km of seabed in normal operation. However, for the purposes of the demonstration, a total of 16.1 sq km was actually surveyed, because we wanted to work near to the simulated leak and test multiple approaches to sensing the leak. Throughout its mission, the ALR was remotely controlled from the shore, mostly from the NOC’s control room in Southampton.

The project consortium demonstrated a functional “system of systems” which can provide operators of offshore CO2 storage sites with a high confidence in their safe operation and to assist in the provision of regulatory compliance. A survey was carried out over a realistic area and leaks were detected during different demonstration phases using chemical sensors, passive acoustic and active acoustic methods. Back then, we proved that it was possible to conduct shore-to-field-to-shore environmental survey operations using a long-endurance AUV operating well in excess of normal AUV deployments was also a significant result.

This method of working showed that you only need a small local deployment team, supported by remote shore-based operations and a data interpretation team – something that we’re more used to doing on a regular basis these days using USVs!

The mission also included a high element of automatic data reduction to allow communication of critical information back to shore in near real time to enable human-in-the-loop mission planning such as changing the mode of operation. It is also entirely possible, and has been demonstrated elsewhere, that further reduction of human decision making can be achieved to reduce operator intervention.

The project demonstrated that it is possible to build highly cost-effective and autonomous sensing systems with on-board intelligence, which are both simple to deploy and operate and which could be very cost competitive with vessel-based or vessel supported AUV survey operations. At the same time, the project members also developed two flexible seabed lander packages which are capable of extended duration deployment, typically six months to a year. These can provide localised and still also wide-area monitoring, automated processing of data subsea and communication of that information to surface.

Looking beyond carbon capture, the potential applications of such integrated marine robotic and intelligent remote sensing technologies are many and varied across ocean science, renewables, security and naval domains.

Get in touch with you would like to learn how we can support your CCS project.


Have you got a similar project?

Challenge us to solve your problem

Challenge us to solve your problem