Low-cost volcanic activity detection with Raspberry Pi cameras

A team of researchers from Chile, Australia, the US, and the UK have found a more affordable method to capture volcanic activity by detecting sulphur dioxide (SO2) emissions. While SO2 cameras do already exist, they’re usually expensive, so the team harnessed the more affordable Raspberry Pi cameras to create their solution.

mount etna with sicily in the foreground
Volcanic celebrity: Mount Etna in Sicily

How does SO2 photography work?

Volcanoes emit more sulphur dioxide when they are more active, so monitoring emissions is one way to keep tabs on volcanic activity, helping to predict when an eruption is likely. SO2 strongly absorbs invisible ultraviolet light with a particular range of wavelengths, and also strongly absorbs invisible infrared light with a particular range of wavelengths: all these wavelengths are present in the sunlight that gets through the atmosphere, so if you get a camera that can detect them, and you point it at the sky above a volcanic region, any patches you see in your image that are darker than the rest of the sky are where the sulphur dioxide is. It has absorbed the UV and IR light so not so much of it gets to the camera.

Instrument installation on Lascar volcano, Chile (Photo credit: T. Wilkes). (C, D) Instrument installation on Kīlauea volcano, Hawaii, United States
Figure (A, B) Instrument installation on Lascar volcano, Chile (Photo credit: T. Wilkes). (C, D) Instrument installation on Kīlauea volcano, Hawaii, United States (Photo credit: T. Pering).

Spotty coverage

Unfortunately, because SO2 cameras have typically been very expensive, it’s rare to install them permanently in locations where they would be useful. With the exception of a few bigger volcanoes (Stromboli, Etna, Kīlauea), most regions rely on spotty coverage, which is not good for hazard assessment or for volcanological research. Developing a cheaper method of SO2 photography — one that can be installed permanently — would greatly reduce risk as well as helping researchers who seek to understand volcanic activity.

How does the Raspberry Pi-based solution work?

In Raspberry Pi cameras, as in most digital cameras, there is what’s called a Bayer filter in front of the sensor, and this filters out UV light. (It does other clever colour filtering too, but the important thing here is that it filters out UV.) The volcano researchers here have found that they can remove the Bayer filter by dissolving it away with stuff a lot like nail polish remover; this means they can take photos in which SO2 shows up.

FIGURE 1(A) Layout of permanent SO2 camera. Most parts are mounted to the case using custom 3D prints. The housing here is a Peli 1400 Case; other systems use larger cases for extra mounting space. (B) Front view of instrument optical systems.

The setup also uses the WittyPi HAT, a real-time clock and power management system for Raspberry Pi. It wakes the volcano detector at the appropriate times and allows other components to power down when not in operation, which greatly reduces power consumption. Accompanying solar panels allow for continuous sulphur dioxide measurement during daylight hours.

The full research paper is available to read for free from frontiersin.org. Here’s a list of the researchers and where they’re from:

  • Thomas Wilkes, Department of Geography, University of Sheffield, Sheffield, UK
  • Tom Pering, Millennium Institute on Volcanic Risk Research—Ckelar Volcanoes, Antofagasta, Chile
  • Felipe Aguilera, Departamento de Ciencias Geológicas, Universidad Católica del Norte, Antofagasta, Chile
  • Susana Layana, U.S. Geological Survey, Hawaiian Volcano Observatory, Hilo, HI, United States
  • Patricia Nadeau, U.S. Geological Survey, Cascades Volcano Observatory, Vancouver, WA, United States
  • Christoph Kern, Department of Physics, University of Sydney, Sydney, NSW, Australia
  • Andrew McGonigle, Programa de Magister en Ciencias Mención Geología, Universidad Católica del Norte, Antofagasta, Chile
  • Chengxi Zhu, Cambridge Advanced Imaging Centre, University of Cambridge, Cambridge, UK

1 comment
Jump to the comment form

whheydt avatar

Iceland could probably use whole bunches of these.

Reply to whheydt

Leave a Comment