BSU students study avalanches through sound waves

Credit: Matt Standal / KTVB

Scott Havens is pioneering the research of avalanches through soundwaves at Boise State University.

Print
Email
|

by Matt Standal

Bio | Email | Follow: @KTVBMatt

KTVB.COM

Posted on January 13, 2014 at 4:55 PM

Updated Tuesday, Jan 14 at 4:11 AM

BOISE -- Scott Havens is originally from San Jose, California, but traded the Bay Area for Colorado, Jackson Hole, and Boise nearly a decade ago.

This Ph.D student is studying avalanches, and he's using sound waves to do it.

He's part of the Cryosphere Geophysics and Remote Sensing Group at Boise State University.

The team of about five graduate and Ph.D students has built and installed an array of five infrasound microphones in one of the state's most avalanche prone areas, a 12-mile section of Idaho State Highway 21 known as "Avalanche Alley."

Infrasound is the sound below what humans can hear. It's also the sound avalanches make as they displace air when rocketing down a steep mountain slope.

Havens' team is working with professors at BSU to build and implement the infrasound microphones to track them. The devices are inexpensive and similar to those used in cell phones. They're also used to measure other low-frequency natural disasters, including volcanoes and tornadoes.

Havens says the team's five infrasound microphones are placed about 100 feet apart, just south of the canyon leading to Avalanche Alley. Much like the space between your ears, the space between the microphones allows researchers to discern what direction the sound is coming from. The system also includes a solar panel to power the microphones, an instrument that translates the signals, and small computer to record the data.

Right now, Havens and his team have to tromp through the snow to retrieve the data from the array. When complete, the system will include a satellite modem to enable wireless data transmission and allow researchers to remotely monitor the most common avalanche pathways on about two miles of the highway in real time.

To achieve this technical capability, Havens has enlisted the help of computer science graduate student Gabriel Trisca. Both work to match the frequency signals to known avalanches reported by drivers and road crews in the area.

The process is time consuming, sometimes taking weeks to analyze a specific snow storm.

That's why Trisca is working on a Java-based computer program to recognize avlanche sounds and automate the process. Specifically, he's using artificial intelligence to create something called a neural network that works to enable pattern recognition in computers. 

"Now, since we have so much data, it's a big problem if Scott had to look through all of it and label it," Trisca told KTVB. "So, what we're trying to do is teach a computer to do as well as Scott does in identifying these different things."

The goal is to create a real-time avalanche monitoring system that can be used by the Idaho Transportation Department on State Highway 21. The system would allow forecasters to respond to avalanches faser and keep drivers safe. Havens says he could envision the system growing to allow detection of avalanches on a much larger scale too.

"Making them low cost enough that you can deploy multiple of these microphones across an entire mountain range and detect all avalanches that are occuring," Havens told KTVB.

Both Havens and Trisca hope to automate the infrasound avalanche detection system by the end of the year.

According to reports, Boise State is the only University in North America studying avalanches through sound.

Print
Email
|