Meteor Detection Systems: The Saskatchewan Fireball Network

Camera	Location	Longitude	Latitude	Elevation	Operator
fireball1	Winnipeg, Manitoba	97.080 W	49.901 N	230 m	Larry Gundrum
fireball3	Saskatoon, roof of Physics Building	106.634 W	52.132 N	569 m	Gordon Sarty
fireball4	Lucky Lake, Saskatchewan	107.264 W	51.154 N	724 m	Tenho Tuomi
fireball5	Regina, Saskatchewan	104.587 W	50.251 N	575 m	Martin Beech
fireball6	Yorkton, Saskatchewan	102.476 W	51.216 N	500 m	Jim Huziak
fireball7	Dauphin, Manatoba	100.192 W	50.993 N	420 m	Ron Lupack

Coordinates for fireballs 3, 4 and 7 are from a GPS receiver using WGS 84 coordinates. Coordinates for fireballs 1, 5 and 6 are from Google Earth. Coordinates for fireball5 are approximate until it comes on-line.

Here is a small diary of interesting events caught by the cameras:

Buzzard Coulee Fireball - November 21, 2008

A bright fireball on November 21, 2008 caught lots of media attention, see on-line article from Astronomy magazine. Meteorites were found, see another article with some photos by Rick Huziak. Below are 3 frames from the fireball4 (then in Saskatoon) video and an image of a meteorite I found, with Garry Stone, when I joined Alan Hildebrand's search team on December 4. More pictures from that search can be found by clicking on the meteorite image below. Another search in the spring of 2009 was organized. I have been out looking a second time, for pictures of that trip, look here. See Bob Johnson's blog on his Buzzard Coulee experience (beware - the site is a busy bandwith hog). And I have been out looking a third time and a fourth time to be a BBC extra.

Here's one of my favorite images. Two nearly simultaneous Perseid meteors at 08:04:59 on August 8, 2007 UT from fireball3 (in Saskatoon):

The northern lights are frequent in Saskatoon, but are not normally bright enough to trigger the fireball cameras. When they do the results are spectacular. Below are some northern light videos from fireball3 in Saskatoon:

March 8-9, 2008
|event-20080309-065424-03.m2v |event-20080309-065614-03.m2v |event-20080309-065630-03.m2v |event-20080309-065632-03.m2v |event-20080309-065650-03.m2v |event-20080309-065653-03.m2v |event-20080309-065657-03.m2v |event-20080309-065701-03.m2v |event-20080309-065710-03.m2v |event-20080309-065721-03.m2v |event-20080309-070007-03.m2v |event-20080309-070012-03.m2v |event-20080309-070037-03.m2v |event-20080309-070131-03.m2v |event-20080309-070134-03.m2v |event-20080309-070159-03.m2v |
Get the VLC plug-in if you cannot view these videos.

Camera Construction Pictures

Originally, I used a plexiglass base for the meteor cameras (2 have been made).

But the heat from the heating resistors melted the plexiglass!

So the plexiglass has been replaced with aluminum. For those who are interested in building your own allsky cameras, there are four 500 ohm resistors hooked up in parallel to 120V. So that's 28.8W x 4 = 115W. It's probably overkill, but I've never had any snow, ice or condensation on the domes. Even after the wildest blizzard. There's a thermostat, behind the video camera in this view, that shuts off the resistor heaters at about 15C. So, in Saskatchewan they are on most of the time.

The meteor cameras where constructed in my garage and represent the continuation of my telescope building projects that include an 8 inch f5 Newtonian (under blanket), a 3 inch refractor (behind camera) and an 18 inch f5 Newtonian (biggest object in picture). The second meteor camera may be seen on top of the tool cart. Also visible is an inclinometer for visual fireball sightings and the barrel upon which the 8 and 18 inch mirrors were ground. [The plexglass base camera is awaiting its aluminum upgrade here.]

Here are fireball3 (foreground) and fireball4 (background) on the roof of the Physics building on the University of Saskatchewan campus in Saskatoon. fireball4 has since been moved to near Lucky Lake, Saskatchewan (see below).

fireball4 was moved to Tenho Tuomi's place near Lucky Lake on February 16, 2009.

There it is on the roof of the house.

fireball4 joins the suite of weather and magnetometer equipment in Tenho's basement.

On February 22, 2009, fireball1, the refurbished old Sandia camera from Laird, took up fireball4's old spot for a check out run.

On May 12, 2009, Larry Gundrum, who I met searching for meteorites at Buzzard Coulee took fireball1 to Winnipeg. Here is fireball1 on the roof of Larry's house.

Here's fireball6, installed 2010 on a school roof in Yorkton.

Here's fireball7, installed on Ron Lupack's roof in Dauphin.

fireball3 and fireball4 both caught their first simultaneous meteor on March 1, 2009 as shown in the images above. The data were automatically sent to a central computer named io which then computed an orbit shown below.

fireball1/3/4/6 usage - Rob's mid 2008 software

vidgrab (old version was metgrab) is a program Rob wrote that just records video to disk.
The --sdl means display it in a window and --dry means don't actually write to disk.

When you want to restart the detection, run ~/scripts/startup.sh or (safer) reboot the machine: sudo reboot

For rejecting stationary events, the ~/meteor/etc/evproc.conf file needs to have a "reject spot" line added.

How to kill and restart asgard:

$ killall asgard
$ killall evctrl
$ killall dnotify

evctrl is used to start the daemon when the sun has sufficiently set and to stop it when it rises.

dnotify is a program that will run a script when something in a specified directory changes. Rob uses it on the older ubuntu installs to automatically run the asgard processor program on event files as they are detected. It has been superceded by something called inotify, and the evwait program uses this (but can't on the fireball3/4 systems) - it has the big advantage that it's very easy to reject events if there are multiple triggers in a row (above some set limit) - Rob has a workaround for the dnotify systems to do the same thing, but he's not sure if it's set up on fireball3/4.

Restart with ~/scripts/startup.sh (or reboot)

fireball2 usage

fireball2 is not used any more in favor of asgard, but it and the sentinel box are still around and functional.

To start up the Sentinel software for running the camera

• Logon to fireball2.
• Enter the passphrase for the ssh key (only needs to be done if the system has been rebooted).
• cd sentinel_source2
• sudo make rtlinux
  • this will initialize the rtlinux modules (if they haven't already been loaded) and load the smodule (required for the meteor detection software).
  • only needs to be done if the system has been rebooted.
• cd ..
• ./sentuser2
  • this assumes parallel port is using 0x378 (IRQ 7)
  • ./sentuser accesses the parallel port on 0x278 (IRQ 9)
  • you can determine which parallel port interupt is being used from "sudo more /var/log/messages | grep parport0" (but it shouldn't change from the setup).
• Start remotely using: screen ./sentuser2
  This will allow the remote term to be killed without stopping sentuser2. To get back into the running sentuser2, login again and type screen -r

To begin detecting

• reset
  • sends a reset pulse to the sentinel box
• init
  • loads the control code into the video decoder and encoder
• mask maskf.dat
• threshold 70
  • this isn't necessary if the mask has been created with the correct threshold but 70 works well for fireball2 with maskf.dat
  • see the Video Sentinel Users Manual for instructions on creating a mask
• autoexpose
  • this captures a stacked calibration frame after an event
• timeon 18:30:00 (hh:mm:ss)
  • doesn't work, use if you want to test the threshold and capture video
• timeoff 05:30:00 (hh:mm:ss)
• trigger
  • this enables the detection of above-threshold changes in the video image
  • if the timeon/timeoff commands have been used
• stop
  • only used if timeon/timeoff has been used, otherwise detections should stop at the default time off (6:00am)

Viewing the videos

• ./sdisplay
• space-bar: Replay currrent video
• right-arrow: Load and play next video (replays current video if no next)
• left-arrow: Load and play previous video
• home: Load and play first video in directory
• end: Load and play last video
• del: move current file to Trash and load and play next video
• esc: Exit sdisplay

To test if the Sentinel box is functional

• test
  • 200 frames of pseudo-random data are written to the Sentinel box and read back and verified
  • takes several minutes to complete
  • there should be 0 errors at the end of the test

To capture an image

• capture image.dat
• open an other command window and use "convert -size 640x480+0 -depth 8 gray:image.dat image.bmp"
  • to convert the image to a bitmap
• xv image.bmp
  • to view the image

Archiving of video captures

• Every morning a cron job is set to run scp_script.sh
• For this script to suceed, scp, ssh and public keys must be set up to work without user intervention

Setup of ssh public keys

1. ssh-keygen -t dsa
   • system will ask for a passphrase, do not leave empty or ssh will not be secure
2. scp .ssh/id_dsa.pub fireball@microquasar.usask.ca:key.pub
3. ssh fireball@microquasar
4. cat key.pub >> ./ssh/authorized_keys
   • if there is no authorized_keys file, ssh to the fireball system and it will be created
5. chmod 700 .ssh
6. chmod 600 .ssh/authorized_keys
   • ssh is very picky about the permissions on these files, if these are wrong it will not accept the public key
7. rm key.pub
8. exit
9. ssh fireball@microquasar
   • you should now be asked for a passphrase instead of a password
   • if this worked go on to the next step, otherwise check if permissions are correct
10. install keychain from http://www.gentoo.org/proj/en/keychain/
    • sudo rpm -Uvh keychain-2.4.3.1-1.noarch.rpm
    • add the following to .bash_profile
      • keychain id_dsa
      • . ~/.keychain/$HOSTNAME-sh
    • logoff and log back on and you should be asked for the passphrase (or source ~/.bash_profile)
    • this program will enable you to only have to enter the passphrase once (when the system has been rebooted), instead of for every session.

1. Some possibly useful software: the Japanese UFOCapture suite.
2. NASA Meteoroid Environment Office (MEA)
3. North American All Sky Camera Database
4. NASA WORKSHOP ON METEOR VIDEO OBSERVATIONS AND ANALYSIS | My presentation