• Welcome to Orpington Astronomical Society.


New version SMF 2.1.4 installed. You may need to clear cookies and login again...

Main Menu

Meteor Network

Started by Rick, Sep 08, 2021, 17:28:01

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.


About a month ago, while the Perseids were putting on their annual show, I placed a bunch of orders for bits of kit to put together as a meteor camera. The image collection device is a Sony IMX291 camera with a 4mm lens, and the processing device is a Pi 4. It took a while for the bits to arrive, and a while longer for me to put them all together and get them working. I finally attached the camera to the side of my house last weekend, and the system is now calibrated and ready to run.

Most of the information needed to set the thing up is on the Gloobal Meteor Network wiki ( https://globalmeteornetwork.org/wiki/index.php?title=Main_Page ). There are also pointers on the UK Meteor Network website ( https://ukmeteornetwork.co.uk/ ).

I've a clear view out towards the north, and this is the coverage map for my camera.

The largest area is for meteors at about 100 kms altitude. The other two areas are for meteors at around 70 kms and 25 kms altitude. Most meteors won't get much lower than 70 kms.

On Monday night the system spotted 122 events, and last night it spotted 195, so there's quite a lot dropping out of the sky at the moment. Here's my pick of the night:

Of course, the forecast for most of the rest of the week is not so great, and I suspect my camera weatherproofing may get a bit of testing, but the main point of the system is that it mostly gets on with the observing by itself, without needing too much intervention, and if the sky clears in the early hours the camera will be there to look out for meteors.


That is really cool.  Check this guy out in near Minsk:  https://www.astrobin.com/users/Astronominsk/
I wonder if he is part of a wider European network.

See for example this one: 



Couldn't find a corresponding station this GMN list ( http://istrastream.com/rms-gmn/ ), but there are multiple meteor monitoring networks, so he may belong to a different one, or his camera may not have reported in for a while.

(Heute Nacht ist es bew├Âlkt.)


It looks like a great project Rick and really interesting.  It's the kind of thing I really should attach to my observatory.



Not a lot of work involved, but a few minor gotchas to be avoided, especially with the camera hardware. I think my total came in at under 200 quid,  but I did re-purpose one or two things. Some folk have put the entire set-up into a camera housing that just needs a PoE cable, but the Pi 4 gets quite warm...


A great project, Rick.  Look forward to some updates soon!

Thanks also for the 'Minsk' link Roberto ~ some cracking Moon pictures amongst his portfolio.




Quote from: Rick on Sep 08, 2021, 23:34:39I think my total came in at under 200 quid, but I did re-purpose one or two things.

Not quite under 200, but plenty of change from 250, partly because I ended up getting a 4GB Pi 4 and a fast 128GB SD card rather than the 2GB Pi 4 and 64GB SD card that's the present minimum recommendation.


Some great results.



I've been impressed by the way the system can grab data through holes in the clouds. Each morning it produces a stack of the frames on which it thinks there are meteors it has managed to calibrate against the starry background. Last night was the first night it failed to see any. I'm not totally convinced by the meteors it spotted on some nights, but it has certainly picked out one or two through holes in the clouds.


Finally got the system hooked up to a couple of archive sites:

The main GMN archive site at http://istrastream.com/rms-gmn/
The GMN site has timelapse video and a whole lot else if you look at an individual station's page.

The UKMON archive at https://archive.ukmeteors.co.uk/latest/index.html
The UKMON "Live" site ( https://ukmeteors.co.uk/live/#/ ) will also be getting images from it.


...and it's interesting to see what comes out of the data the system uploads. The UKMon system comes up with orbits for meteors for which it has sufficient matches. One particularly bright meteor my camera caught just over a week ago ended up being analysed thus:


Turns out it was a sporadic (not associated with any particular shower).


Three more from the night of 1st/2nd October. The first was fairly faint, a sporadic with its aphelion in the asteroid belt and perihelion near the orbitof Venus:


The second was rather more spectacular, and was a sporadic with its aphelion out beyond Jupiter and perihelion between the orbits of Venus and Mercury:


One rather different one was this (though it wasn't as spectacular), which came in from outside the solar system from well above the ecliptic and met the Earth rather than diving underneath and then swinging back up and out:



Amazing info Rick.  The third one came all the way from outside the SS to end its days here  :o



The interesting information comes out because there are enough cameras monitoring the sky that some meteors get seen by several cameras in different locations, and that allows trajectories to be calculated with some accuracy. That in turn allows the interesting oddballs to be spotted. Without the triangulation it may be pretty, but not much else.

One of the things to come out if this kind of data collection is the discovery or refining of known meteor showers. In the last few days there's been talk of one in the southern hemisphere (far enough south that it's not one we'd see any meteors from) associated with Comet 15P/Finlay. The radiant's in Ara, so they're dubbed the "Arids"... (See here.)


Incidentally, there's an article about meteor monitoring starting on page 36 of the October issue of Astronomy Now (which came out last month).


Thanks Rick

I'll look the article up.



I didn't catch this one on my camera as it was too cloudy here, but a couple of cameras under clearer skies did:


Note the eccentricity of the orbit. It's another sporadic inbound from interstellar space.


Quote from: Rick on Oct 28, 2021, 19:15:51These automatic cameras are collecting a lot of data which allows meteoroid orbits to be calculated, and that will, among other things, help identify new showers.

The Global Meteor Network announced one such discovery yesterday evening:

On October 24, the GMN cameras in Europe recorded an outburst of an unknown meteor shower. The peak was very sharp, most meteors were observed within a period of 30 minutes.

You can read more about it here:

The meteors were on a very interesting orbit - it looks like they were pieces of a sunskirting comet which probably disintegrated a long time ago, and they only now reached Earth after returning from the Oort cloud.

Thanks to everyone for their hard work and contributions, no other smaller regional network could have observed this shower.


Quote from: Rick on Oct 29, 2021, 09:52:58The meteors were on a very interesting orbit

It wasn't a particularly good night for UK meteor observers, but a couple of cameras did catch one of the meteors in the newly-identified cluster, and you can see the UKMON orbital analysis here:



One of the weak points in the Raspberry Pi setup is its use of an SD card for main storage. Provided you get a good one this is not usually too much of a problem, but when things do go wrong the troubleshooting can be interesting...

A couple of nights ago my camera seems to have had more than the usual number of detections*, and this caused the SD card to hit the stops. At some point the Pi froze, and then got re-booted. It then entered a rather strange state, where the SD card was in read-only mode, but it was running in an overlay filesystem mode. This makes it look as if file changes are taking effect, but when re-booted it's back to square one. I went through that cycle a couple of times before realising what had happened. Unfortunately, the SD card itself seems to be stuck in read-only mode. This is a known failure mode, and is at least slightly better than being stuck in write-only mode, as at least this way you can get the data off...

Mwanwhile, it is clearly time to get a new SD card, and re-build the software. I think I will also get an SSD and use that instead of the SD card for all the data directories.

* Friday 5th November... I wonder why... (Err... No I don't**). :boom:
** A pity, as the four previous nights were cracking nights for meteor detections, with six or more active showers and over 70 meteors each night.


The necessary bits arrived, and I've re-built the system today. It's busy looking at the clouds at the moment. ClearTonight.com doesn't give it any chance of spotting any meteors, but maybe a small hole in the clouds will drift past at the right time...

I now have the software and system files on a new rather smaller SD card, and the data is going to a USB3 SSD. At least that way, if the data disk fills up then it won't kill the system. I've not yet worked out thether the original SD card is permanently read-only, or whether it can be revitalised, but until I'm happy the new system is behaving I'll keep the read-only version of the old one around.


Hi Rick

Re your previous link below from 29/10:~

"You can read more about it here:

I don't understand any of the technical stiff but do really like your input and the reports.  In the link above, images 9 and 13, also showed like a series of white blobs elsewhere in the sky - would these have been satellites which the recorder was catching?

- Hugh


Hi Hugh,

Quote from: Hugh on Nov 09, 2021, 10:40:55Re your previous link below from 29/10:~

"You can read more about it here:

I don't understand any of the technical stiff but do really like your input and the reports.
The "heat map" display is using "Sun-centred geocentric ecliptic co-ordinates", which is a bit of a mouthful, but means that the ecliptic (the plane of Earth's orbit) is represented by the horizontal line through the middle, and the top and bottom points are perpendicular to that plane. It's a projection useful for giving an idea of objects' orbits. The various blobs indicate the number of meteors coming from that particular direction, so a very bright spot indicates a direction from which a lot of meteors were coming. That map is for one 24 hour period between (about) noon on 24th October and noon on 25th October. The brightest spot is the Orionids, and other known showers are identified by little rings. (Here's a table of abbreviations) The yellow arrow points at a cluster of meteors without a ring round them, which is the suspected new shower.

Quote from: Hugh on Nov 09, 2021, 10:40:55In the link above, images 9 and 13, also showed like a series of white blobs elsewhere in the sky - would these have been satellites which the recorder was catching?

In the photographs lower down, the meteors are usually slightly tapered trails with a more or less noticeable head. Satellites, if they vary in brightness, usually do so in a symmetric manner. Aircraft confuse the issue further by having flashing lights, so you end up with bright dots on the trails. All the trails are interrupted at intervals by the camera's electronic shutter. Fast-moving objects have longer dashes and fewer breaks. Satellites and aircraft are usually much slower than meteors, and the system knows that and usually ignores them, but there's often more than one trail on an image, so parts of aircraft and satellite trails often appear on images with meteors. At this time of year aircraft are the main photo-bombers. Figure 9 has a very obvious aircraft trail at the top, Figure 4 has one close to the meteor, and Figure 13 has one at the bottom. The last is from a different type of camera, and there's also evidence of interlacing on the image.

At other times of year, common photo-bombers are moths, bats and birds.

Possibly the most common source of artefacts is internal reflections from bright lights (including the Moon). My camera gets interference from various neighbours' rottweiler lights... :chase:


Yesterday's BAA Webinar (available on YouTube at https://youtu.be/eAGyNELqqIM ) was about the recovery of the Winchcombe meteorite earlier this year. It gives quite a bit of detail about various meteor-monitoring networks.


...and last Friday's BAA Radio Astronomy Group webinar was also about the recovery and analysis of the Winchcombe meteorite. You can watch it here: https://youtu.be/zgGRL15RgEw