As of 25 July 2021, my meteor camera is live and operational, fixed to the south wall of my garage and taking video of the sky every night.
To be clear, this involved next-to-no actual complex study on my part, just the purchase of a few items off the internet, downloading some software that someone else wrote and following the assembly instructions provided by people who had already done it before. It was easy, just big-kid-lego, really.
In fact, I was following a recipe for a meteor cam created by a group of enthusiasts called the Global Meteor Network (GMN). Damir Segon, Pete Gural, Denis Vida, Dario Zubovic, Mike Mazur and Patrik Kukic (with assistance from others also) have all had a hand developing the equipment configuration and software over a number of years, culminating in the Raspberry Pi Meteor Station (RMS) design. The GMN is the collection of installed RMSs and the centralised server at the University of Western Ontario they connect to.
There have been, and are, other interconnected meteor monitoring efforts, notably Cameras for All-Sky Meteor Surveillance (CAMS) but GMN has an advantage in the whole project being open-source, so the analysis methods are reviewable by anyone, the captured data is shared using an Open Data CC-BY-4.0 license so the data is accessible to any researcher, the ability to compute orbits is not limited to a single individual, and the station equipment cost is significantly less.
I discovered this project one day when it was mentioned by Hamish Barker, an astronomer who is the Convenor of the region’s astronomy club, the Nelson Science Society’s Astronomy Section. He had established the first GMN RMS in Aotearoa New Zealand, and wondered if I would be interested in setting up another camera with overlapping coverage of the sky so it would be possible to triangulate the trajectories of any observed meteors. It’s a global citizen science project, it is astronomy that I can do every night with zero effort, and if we are lucky we might be able to track down a fallen meteor — uh, yeah, I’m interested. In fact, it was precisely the sort of project I think is really cool, because it involves a very modest set-up cost, an afternoon’s fiddling with electronic components, an evening or two of installation, setup and configuration, and after that it is continuously in action, collecting Real Usable Data whenever it can peak through the clouds.
If you want to give it a go too, here is where you start.
While it might be possible to free-wheel a meteor camera setup of your own design, GMN recommends a specific set of hardware to enable the global collaboration and integration and to ease support issues as all involved are volunteers. So, for instance, the ‘brains’ in the RMS is a Raspberry Pi, and if you do not have one, you need to buy one. I know. I know, you might have an old laptop lying around that you could use instead, BUT a Raspberry Pi is incredibly inexpensive and you’ll save both yourself and others time and wasted effort if you just follow the tried and tested recipe.
Here are the main components that were required as at the time I was assembling my RMS:
- A Raspberry Pi 4 with 2GB RAM (a Raspberry Pi 3B+ with 1GB is actually sufficient, but the trend is toward the added capabilities of the Model 4), with 5V/3A power supply, 128GB micro SD card, and a specific recommended case/housing
- A Real Time Clock accessory for the Raspberry Pi
- A specific recommended security camera housing which includes a bracket needed to mount the camera
- A specific bare-board IP security camera sporting a Sony IMX291 sensor with a supplied Power over Ethernet power-and-communication cable
- One of four recommended lenses; for dark sky sites like mine a 4mm f/0.95 lens providing a 88°x45° field of view is recommended (in light polluted urban areas, an 8mm f/0.9 lens with a 40°x20° field of view is recommended)
- A Power over Ethernet Injector to supply power to the camera using a network cable into the camera’s PoE cable
Note that the official and up-to-date, full and complete recommended parts list is found on the GMN website, here. Don’t rely on my list above (which is also why I am not linking to specific items from this page – use the official parts list if you do this yourself).
The GMN provides a Raspberry Pi operating system image to put on your micro SD card; it is based on Raspberry Pi OS, configured for RMS operation and connection to the GMN, with all necessary software pre-installed. There are deployment instructions that need to be stepped through but even if you don’t understand what is going on they are trivial to do and before long you will be able to capture meteors just like me.
Because the cameras are continuously recalibrating, they can do quite detailed photometry measurements, and from those measurements some quite interesting estimates can be made, such as the initial mass of the meteoroid and it’s bulk density.
Why bother? Because right now we don’t even properly understand the scale of this area of study. We believe 43+/- 14 t of meteoroids enter earth’s atmosphere every day – although some estimates have been made of over 250t/day. Getting better data, with high-precision measurements means it is possible to produce accurate models that can be used for very important things such as predicting meteoroid impact risk on spacecraft.
Back in 2013 the Chelyabinsk fireball produced a bright flash, an atmospheric explosion with an air blast and shock wave, causing many initially to think it was a nuclear attack. It caused widespread damage, especially with blown out windows, and something like 1200 were injured as a result. It could have been a lot worse – the energy released has been estimated to be equivalent to between 400,000,000 and 600,000,000 kg of TNT (which is about 20x to 30x Hiroshima explosions). This was produced by a meteoroid that was possibly just 19m in size, and maybe 12,000,000 kg in mass. The meteorite that was recovered ended up being about 600kg, so 99.995% of the meteor vapourised in the atmosphere. We actually only know of about 1% of all Chelyabinsk-sized meteoroids. This sort of meteoroid is right on the limits of our Near Earth Object detection capability, and in this case it came at us from the sunlit side of the earth, so it was as if it was coming at us while we were blinded by sun-strike on the windscreen.
Sky-survey projects are scanning the skies for large and potentially hazardous objects like this and bigger. But there is also a lot to learn from the smaller objects that are harder to see until they interact with our atmosphere.
Some meteor showers are unexpected, and because there hasn’t been any real global monitoring effort until now, some events are missed entirely – eg there are no recorded observations of the 2012 draconid outburst.
How can we improve our understanding and reduce uncertainties? Well, that is why we now have a citizen science project to monitor meteor activity in the atmosphere.
With multiple cameras observing a meteor, the trajectory can be calculated to determine its initial position and direction of travel and initial velocity. From there a meteor’s orbit can be calculated. When this is done for lots of meteors an overall picture can be constructed of where the meteoroids are and perhaps what the parent body was can be inferred from that data. The GMN RMS network can provide the high-precision data required in order to produce accurate models that can be used to predict meteoroid impact risk on spacecraft.
Right now there are hundreds of contributors to the project in at least 20 countries – but mostly in Europe, Canada, the UK and the USA. We need a lot more GMN RMS stations in all longitudes, but we especially need more stations in southern hemisphere latitudes. If you would like to be part of the project, head over to globalmeteornetwork.org and check out the wiki which has information on how to get started and build your own RMS meteor camera, and sign up for the newsgroup forum on groups.io.