Predicting a Volcanic Eruption … by Hooking It Up to the Internet

Predicting a Volcanic Eruption … by Hooking It Up to the Internet
Part of a special series from

Why you should care

Data is allowing the unpredictable to become predictable.

OZY and Predix from GE — the cloud-based development platform built for industry — have partnered to bring you an inside look at the future of digital industries, where people, data and productivity meet.

This past August, volcano explorer Sam Cossman descended 1,200 feet into the fiery mouth of a 150-degree Fahrenheit crater. His mission: to install more than 80 sensors inside Nicaragua’s active Masaya volcano, which spews a waterfall-like flow of orange lava from a cliff. The volcano is just 12 miles from Nicaragua’s capital, Managua, and threatens the lives of all living in its shadow. The connected devices Cossman is installing — which are small enough to fit in the palm of a hand — can precisely measure in real time the volcano’s atmospheric pressure, temperature and gas types, as well as seismic and gravity information, and more. All of this data can then be stored and analyzed on a platform like GE’s Predix. “We are basically bringing the first volcano online,” Cossman says.

Eight hundred million people around the world live within 62 miles of an active volcano.

This is the beginning of the world’s most advanced early-warning system for volcanic eruptions. But the technology can assist with the prediction of other natural disasters, too. Whether within deep fiery craters or under the sea, cloud-connected sensors are delivering a rush of unprecedented insight, ranging from predictions about the severity and epicenter of an earthquake to a tsunami’s speed and direction of travel. However, it’s not without danger: Sensor installers travel to the extremes of the Earth — inside scorching caverns, deep within the sea, on the ground in areas prone to earthquakes.

The opportunity to save lives is unprecedented. With systems like these in place, first responders and the public would have access to crucial information (the data is available to anyone interested in using it to develop monitoring applications). For example, when a tsunami is detected, data might also help forecast when the water is expected to reach a specific location and help pinpoint the closest safe ground. The Industrial Internet is revolutionizing machinery in the skies, too. Global Hawk — the unmanned aircraft of the National Oceanic and Atmospheric Administration and NASA — flies 60,000 feet above cyclones to provide a 3-D view of storm structures. The drone’s sensors supply “large amounts of real-time data on wind speed, moisture, pressure and temperature that help refine forecasts of how storms change in intensity,” says Gary Wick, NOAA project scientist for the Global Hawk experiment.

You might live near one of these treasure troves of sensors without even knowing it. On the west coast of Canada, for example, there are underwater cable observatories in the Pacific Ocean that continually measure key elements like water pressure, temperature and movement. Meanwhile, land-based sensors are measuring similar properties along the coastline. In 2014, these sensors fed data to the web-enabled Awareness Research Network, which detected a 7.3 magnitude earthquake 11,000 kilometers away in Indonesia.

But it’s about even more than Mother Nature and helping out humanity. This cutting-edge concept ushers in a wave of possibilities across all sectors of the digital industry, where business is built upon and transformed by software-defined machines and connected, responsive and predictive solutions. When a volcano is hooked up to the Internet, a digital plan is created — called a digital twin — and sensors continually stream data about its behaviors and track how inside and outside elements relate to one another. This digital simulation can predict behaviors and improve efficiencies on a host of large machinery. Think turbines on wind farms, jet engines and manufacturing plant machines. Disruptions or mechanical failures are anticipated before they happen, production bottlenecks are prevented, data systems and analysts remain connected in real time and obstacles or issues are corrected more quickly than ever before. The big wins: improved uptime and productivity.

Still, when it comes to predicting natural disasters, this technology isn’t quite at the point where it can send accurate warnings early enough to allow people to evacuate. Sensors provide invaluable real-time data, but it’s also important to incorporate historical data into algorithms, which requires skilled professionals who can properly interpret the information. To complicate matters, experts might “not necessarily agree immediately about what the data are showing them,” says Jessica Ball, a volcanologist and Ph.D. fellow with the U.S. Geological Survey. Monitoring lets experts forecast events, but predicting an exact time and location is, Ball says, “impossible, since volcanoes are natural systems and, to some extent, are inherently unpredictable.”

When smart machines are connected to the Industrial Internet, the ability to analyze and predict behaviors and events is dramatically increased. Connecting equipment, whether on the factory floor, in the skies or on the ground — and yes, even inside a volcano — to the cloud allows for unprecedented predictive capabilities born out of digital industry. And the ability to thwart potential catastrophe — environmental or industrial — is good news for everyone.

Predix from GE is enabling the adoption of powerful, secure and scalable solutions built for the industrial app economy. It’s industrial-strength strength, powering the future of industry. Get Connected.


Numbers and factoids --- fodder for your next cocktail party.