How fiber optics are changing our understanding of the subsurface

In a matter of months, fiber optic cables on Johan Sverdrup will have generated more data than Equinor’s almost 50-year history put together. Here’s how we made it happen - and what it means.??

About 140 kilometers west of Stavanger, Norway, lies the Johan Sverdrup field and offshore platform. Here, we’ve placed fiber optic cables, the very same type of cable that delivers the Internet to your home, along our wells. While there’s little need for Netflix in the middle of a reservoir, there is a need for data and that’s why we’re using these fiber optic cables. With a slight modification we can use them as microphones and temperature sensors:

“There’s nothing special about the fiber cables themselves, but what’s special is the unit we’ve put on top. It’s called an interrogator and sends laser pulses down the cable, records what’s returned and converts that into either sound or temperature data,” project leader H?kon Sunde Bakka explains. The result is known as Distributed Acoustic Sensing (DAS) and Distributed Temperature Sensing (DTS). Sensors record these data at every meter along the cable, which might not sound like that much. But a single cable in the wells on Johan Sverdrup can reach lengths of up to 5,000 meters.

That means we have 5,000 microphones from just one cable and one single microphone records data at 10,000 Hz – giving us 50 million samples per well each second. In more human terms, the amount of data is the equivalent of 10,000 Netflix movies per second.?Quite impossible to watch or binge on its own, but thankfully we have the right people to sort that out. “We move all this data into a high-performance cluster that’s been developed specifically to compress these 10,000 “movies” into a single ‘movie’ that our engineers can watch. It's all thanks to taking the process capabilities of the subsurface and translating them into IT,” H?kon explains.

Traditionally, the process of viewing this data took 2-3 weeks, if everything went according to plan, and involved flying hard drives to shore via helicopter. Now, it takes 2-3 seconds and is livestreamed. “Live streaming opens the door to automation. Now, we can use software and machine learning to take us from a world where we need to constantly watch the data, to one where we are told exactly when we need to watch it. All enabled through and scaled by the tight integration between our Omnia and Azure platforms,” H?kon says.

Heatmaps to the people

These insights are created in several ways, and applications is one. One of these apps is Fo.tone, a webapp that handles the live streamed data and presents it together with other wellbore data – to enable a whole new world of information. “Reservoir- and production engineers having access to this kind of data means they can optimize production in the wells. When every well produces for tens of millions Norwegian kroner every day, one can just imagine the potential impact,” Fo.tone team lead Kjetil Tonstad explains.

The data is presented as a heatmap, a visualization that might remind you more of a 70’s wallpaper than data itself. An inconsistency in temperature, for example, will show as a different color and is more easily spotted with this heatmap.

Fo.tone heatmap showing the acoustic energy in a demo wellbore

Applications like Fo.tone integrate production data, and engineers might be looking at different types or variations of data from the same source. Saying it’s a very data-dense app is no understatement, which made for a challenge for the developers to handle.

“Finding a way to display all these different types of data in a way that made sense while also being responsive is a real challenge. We wanted to make it responsive so that you could freely look up the data you need and personalize the app,” Fo.tone front-end tech lead Frits Talbot says. If you’re curious and want to read more about what software development challenges the team faced, check out the Fo.tone story on Loop.

How technology increases safety

So, what are some of the areas we could use this technology for? Safety is always our top priority and an important place to start. Fiber optic sensor data can help us improve leak detection, for example through spotting a discrepancy in a heatmap like Fo.tone’s.

“A sudden temperature drop displayed would indicate that there is a leak. Early detection of leaks means the potential damage is less severe and remediation could be cheaper,” H?kon explains. These data also allow us to see where water comes into the well, which tells us a lot about both the well and reservoir’s behavior during the production or injection phase. For wells that lack this technology, we would get the information months after water production started in a well, significantly reducing our understanding of the underlying processes. ?

“All these lessons are quite different from what we would get if we went into a well months after it started producing water. Now, we can see how the well acts as it is happening. This can provide us with more insight when both steering and planning production or wells in the future,” H?kon says.

For Johan Sverdrup, there's work underway for a gas lift monitoring system that tells us when gas lift valves are operating in an abnormal state. “It tells us when we’re doing something to the valves that forces them to not behave as they should. Our assumption is that when that happens, the valves are experiencing increased wear, which over time can lead to a break,” H?kon explains.

“By knowing what a safe operating window is and what is not, we can start to challenge the limit set by our vendors. These limits tell us what we can do with the valves, meaning we can extend that operating window based on our own observations. This lets us reduce gas lifts and by doing so use less power, which means overall lower emissions,” he adds.

Future of fiber optics

This entire endeavor might seem very specific to wellbore monitoring, but the data platform is entirely agnostic. It only targets DAS and DTS fiber optic data as a datatype and doesn’t just target wells.

“Essentially, we can take this kind of data from anywhere, copy-paste it into the same platform and reuse the work that’s already been done. A lot of our assets are looking at fiber optics so there is a massive potential in this technology,” H?kon explains.

Martin Linge has the same cables installed as Johan Sverdrup, while Gudrun currently has temperature data (DTS) available and is ready to use acoustic data (DAS) when needed. For our subsea assets, there are already many fiber optic cables installed and ready to be used. Technology isn’t the only driving force behind this work, however – it's collaboration.

“Collaboration truly is key in projects like this. When we started, we were just passionate people with different competencies from a variety of business areas, but now we’ve been able to change the way we work with fiber optic data,” H?kon says. “We love what we do and think it’s super interesting and that’s how we get the good ideas. Keeping people interested and having the right people on the right tasks,” he adds.