Erebus science – thermal imaging

Erebus science – thermal imaging

Infrared cameras are a great way to take thermal measurements of a volcano from a distance. A thermal camera has been used at Erebus for several years. There, it provides the opportunity to look not just at changes to heat output, but also at the activity of the lava lake.  Nial Peters, one of the Volcanofiles and a PhD student at Cambridge, has been operating the camera for the past three field seasons and looking at the data from it. Nial first went to Erebus as a field assistant for Aaron Curtis, who we interviewed last season, working in the ice caves – so he knows the volcano well. Here’s his email interview from the 2012-13 field season, telling us about his work.

Volcanofiles: What is a thermal camera?

Nial: Pretty much the same as an ordinary camera, except that the sensor records IR radiation rather than visible. Objects that are radiating a lot of heat show up as bright. Note that is not necessarily the same as saying hot objects show up bright – a high temperature silver object may show up as less bright than a cooler black object!

Volcanofiles: Perhaps it’s fairly obvious why you’d want to use a thermal camera on a volcano, then? What sort of things can you look for in the footage from the Erebus lava lake?

Nial: Well, perhaps not so obvious. Of course you can use a thermal camera to do the obvious things like measure heat output from the lava lake and many people have done such studies in the past. The reason I use a thermal camera is because it is capable of imaging the lake through a far thicker volcanic plume than a normal camera. Even on days when the lake is invisible to the naked eye, you can still record clear IR images. I am not so interested in the actual temperature readings from the camera, I am using the data to look at the surface velocity of the lake as it convects. You can also record the Strombolian eruptions of of the lake and measure things like refill time.

The start of a Strombolian eruption

The thermal camera can reach high time resolutions…

…so it gives a good record of explosions

Volcanofiles: You’ve spent a lot of time building things so that you can collect data with the thermal camera. How do you set up the camera in the field? And what have you been working on these past few months?

Nial: The thermal camera we have does not store images on-board like most cameras do. Instead it is designed to stream images over an Ethernet link, using the GenICam interface. This means that it requires a computer to operate. The first year we used the camera, we set up a microwave Ethernet link to the crater rim and ran the camera from a PC in the hut. However, my goal was to have the camera run year-round and this setup was too power hungry and unreliable for this.

The system this year is totally different. The camera is being controlled by a ARM based single board computer (SBC) (it’s a Blue Chip Technologies RE2 board if anyone is interested) running Ubuntu Linux. I have written some custom control software for the camera based on the open-source GenICam project Aravis.

The camera and its computer (Photo: N. Peters)

The software captures images, does a very limited amount of preprocessing, and then compresses the images into PNG files. The images are stored locally on a solid-state hard-disc. The SBC also runs a server program which can send the images and some environmental data (power consumption, temperature, etc.)  in realtime over the microwave ethernet link (when it is operational – in other words, during the field season) so that we can keep an eye on things while we are here. The whole system is designed to run reliably by itself for an extended period, with lots of error checking and correction built into the software, GPS time synchronisation and a watchdog program to restart the whole system should something go badly wrong.

Nial with the microwave antenna used for transmitting data to the hut.

The weak link in the system is the power supply. In total the camera system uses about 11W, which is generated by a solar array and some wind generators situated 0.5km away (conditions on the rim are too harsh for solar panels and wind generators). An inverter is used to boost the voltage to 230V AC and power is then transmitted up a cable to the rim where it is stepped down and rectified to 12V DC. The whole power system has been replaced this year using low-temperature rated components and tougher cable. Hopefully this will mean that we can sustain power to the rim year round, but this is a challenging environment so we will see!

Most of my work for the past months has been developing and testing the new camera system (both software and hardware). It is probably the most complex thing I have ever made and I am really pleased that so far it has worked flawlessly (over 600,000 images captured so far!).

Mounting the camera on its tripod at the crater rim

Volcanofiles: This is your third season running a thermal camera on Erebus, and there is data from older field seasons too. What have you measured with it in the past?  And what else are you hoping to do with it this year?

Nial: The motion tracking is probably the most important thing that I have done with the data so far. This picks up the periodic behaviour of the lake very well, and shows that the lake has been doing the same thing for as long as we have been measuring it! It also shows the recent decrease in the size of the lake (it is now a quarter of the size it was two years ago). This year will be more of the same, but with a year-round dataset hopefully we can see in more detail how the lake is changing.

Volcanofiles: You’ve had four field seasons on Erebus. Does the novelty start wearing off? How have things changed for you since the first time you were up there?

Nial: Certainly it is not as exciting as it was the first time, but then nothing is, once you have some idea of what to expect. It is still an awesome place to come to though, and I am still thrilled that I get the opportunity to work here.

I guess the biggest change for me since my first season has been the transition from working in the caves to working at the crater rim. I’m still really interested in the work that is being done in the caves, but particularly this year I have not had the time to get involved. Season to season there is always a bit of change as different people come and go, but I suppose that it is more similar than it is different.

Volcanofiles: How’s the season going so far?

Nial: Pretty well I suppose. Everything was up and running in record time this year. Of course no field season would be complete without everything breaking and that has started to happen now with one broken gas sensor, a broken spectrometer and no liquid nitrogen left for the FTIR. But these things are to be expected, most of the equipment is being pushed to its limits here and so some downtime is inevitable. As I already said, the new power system is almost complete and the thermal camera system is working well – I am confident that we will get many more months of data after we leave, even if it doesn’t quite make it through the winter.

Nial with the themal camera in the field

Volcanofiles: Thanks, Nial – we look forward to seeing some winter data from the thermal camera!

When everything’s running during the field season, you can see the most recent thermal camera images on the Mount Erebus Volcano Observatory site here.

Leave a Comment