Vivien He researched, designed, built, and tested the device over the summer. The device was roughly $100. (Photo from Vivien He)

 

When an earthquake occurs, having even just a few seconds of warning can make a big difference. A heads-up gives people a chance to protect themselves. This is what ShakeAlert, an earthquake early warning system, aims to do. But the system is expensive and very consumer-friendly. Recently, a high school junior built a cheaper solution using parts that totaled less than $100.

 

Vivien He, a student at Palos Verdes Peninsula High School in Rolling Hills Estates, California, designed a low-cost earthquake detection system that can easily be adopted in homes. He was influenced by the lack of seismic noise, which quieted down by a median average of 50% worldwide due to COVID. Wondering if it was possible to measure seismic noise herself, He spent a summer researching, designing, building, and testing the device.

 

The device is made up of a geophone, a component that converts ground motion into electrical signals, an analog-to-digital converter to digitize the signals, and a Raspberry Pi computer that handles additional processing. The analysis is done via Python, which then activates an onboard alarm when it detects an earthquake. It will even send text alerts to friends and family.

 

As fate would have it, He finished writing code right before an earthquake hit Southern California. When she woke up, she found the device successfully captured the earthquake and that the waveform perfectly matched data from a U.S. Geological Survey station 4 kilometers from her home. Since then, her device has recorded all earthquakes in the area.

 

So, what’s the next step? He has already redesigned the device. The latest version features all the components stacked in a last-cut acrylic case roughly the size of a Rubik’s Cube. Next, He hopes to extend her monitoring system to include several devices that can communicate with each other and offer alerts over a larger area. She also wants to include learning machine algorithms to make sure the system can chart true earthquakes.

 

Have a story tip? Message me at: http://twitter.com/Cabe_Atwell