The December 2022 issue of IEEE Spectrum is here!

Close bar

New Deep-Sea Pressure Sensor Could Monitor Dangerous Undersea Faults

A new self-calibrating pressure sensor is helping scientists monitor underwater seismic activity

4 min read
A close-up photo of a modified pressure sensor sitting on a metallic table in a research lab.
Two quartz crystals rest within the olive green housings at the base of this modified pressure sensor in a lab at the University of Washington.
Photo: Dana Manalang

For humans, almost everything is harder underwater. This also applies to seismic monitoring. On land, seismologists can festoon fault lines with wireless instruments, but offshore their options are much more limited. This is a problem because 90 percent of naturally-occurring earthquakes happen underwater.

“Our sampling of what’s beneath us is very, very weak and very low resolution,” says Andrew Lamb, a geophysicist for the U.S. Geological Survey. “There’s a lot of uncertainty to it.”

One way to monitor underwater seismic activity is to place pressure sensors on the seafloor. The quartz crystals within these sensors are piezoelectric, meaning they produce a small electric charge when placed under pressure. The sensors can therefore measure the pressure, or weight, of the water above them as a way to detect the very slight rise and fall of the seabed below.

Such instruments can detect up-and-down movement of the seafloor on a sub-millimeter scale. Measuring the rise and fall of different sections along a fault can help geologists determine where tectonic plates are “locked up” or have gotten stuck as they try to move past each other. In these spots, pressure builds up between the plates, producing earthquakes when they break free.

But pressure sensors are prone to drift, which means their readings lose accuracy over time. Sensor drift can be as severe as 10 centimeters a year, while the seabed itself may only shift by a few millimeters. This makes the readings useless to scientists, since the drift is greater than the actual movements they wish to measure.

To correct that error, deep sea pressure sensors are calibrated frequently, usually with the help of a remotely operated vehicle. That’s an expensive proposition, though, so few pressure sensors rest on the seabed today.

A marine geophysicist and electronic engineer from the University of Washington are now testing a new self-calibrating pressure sensor that could be deployed on the seafloor as a low-cost, long-term way to monitor seismic activity. Instead of waiting for an ROV, the sensor periodically calibrates itself against the air pressure contained within its own titanium housing.

It was built by Paroscientific Inc. in Bellevue, Wash. and modified by Dana Manalang in her UW lab. The original instrument included two pressure sensors, a barometer, and an accelerometer. She added a valve and actuator to enable the pressure sensors to switch from measuring external water pressure to internal air pressure, and back again.

In June, the team added the device to a collection of instruments sprouting from an undersea cable off the coast of California. There, the seabed is very stable, which means any change in the readings would be evidence of sensor drift. The team is now collecting acceleration and pressure data 40 times a second from two separate sensors contained within the device, which operate independently of each other, and should produce the same results if they are calibrated correctly.

“We’ve only been collecting data for about 100 days, but so far it appears to be working,” says William Wilcock, a marine geophysicist at UW. “It’s clear that it’s going to remove most of the drift. Whether it’s removing 90 percent of it, or 99 percent of it, we don’t know. So far, we can’t see that it’s not removing 100 percent of it.”

An illustration of the Cascadia Subduction Zone off the coast of Washington and Oregon. Image: Alicia.iverson via, Wikimedia Commons

The group will present their results in December at the annual meeting of the American Geophysical Union. If it works, Wilcock says he could imagine installing 100 sensors for long-term monitoring of the Cascadia Subduction Zone, a 1,000-kilometer fault line off the coast of Washington and Oregon with a roughly 15 percent chance of producing a massive earthquake with a magnitude of 8.0 or higher in the next 50 years.

Data from a deep sea sensor network could provide help scientists understand what’s happening along that fault line, Wilcock says. “In the Cascadia Subduction Zone, the part of the plate that’s locked and could produce one of these magnitude-8 or-9 earthquakes is all offshore. Some people think it’s locked all offshore, some people think it’s locked only in the subduction zone. There’s evidence in Oregon that it’s not fully locked but slowly creeping. Just using land-based data, you can’t tell any of this,” he says. “You really need to make observations on the seafloor.”

Manalang estimates that, once installed, a sensor network could remain in place for at least 10 years. This is assuming the sensors could be connected to an undersea cable for power and communications. She adds that they’re also working on a low-powered version that could run on a battery.

There are a handful of other instruments available for underwater seismic monitoring, but they are either too expensive to be deployed en masse, or require frequent upkeep. In one scheme, scientists place acoustic GPS transmitters on the ocean floor that constantly ping a buoy or seaglider above with their position, as they rise and fall and slide with tectonic plates. 

Tom Brocher, a research scientist for the U.S. Geological Survey, says scientists’ broader understanding of earthquakes has been hindered by a lack of accurate instrumentation on the seafloor. “This would be a big step upward,” he says of the UW sensor, should it prove to eliminate sensor drift. Other major subduction zones near Alaska and Japan would be good candidates for a sensor network of their own.

The Conversation (0)

Why Functional Programming Should Be the Future of Software Development

It’s hard to learn, but your code will produce fewer nasty surprises

11 min read
A plate of spaghetti made from code
Shira Inbar

You’d expectthe longest and most costly phase in the lifecycle of a software product to be the initial development of the system, when all those great features are first imagined and then created. In fact, the hardest part comes later, during the maintenance phase. That’s when programmers pay the price for the shortcuts they took during development.

So why did they take shortcuts? Maybe they didn’t realize that they were cutting any corners. Only when their code was deployed and exercised by a lot of users did its hidden flaws come to light. And maybe the developers were rushed. Time-to-market pressures would almost guarantee that their software will contain more bugs than it would otherwise.

Keep Reading ↓Show less