Meteor Spectroscopy and the Amateur

Written by Ed Majden

First a bit of history.

In the second half of the 1800’s attempts were made to observe meteor spectra visually using prisms.  Because of the short duration of meteor phenomenon this was difficult but it was established that meteors produce discrete line spectra.  The bright lines neutral sodium and neutral magnesium where correctly identified visually by experienced observers.

The first photographic meteor spectrum was secured by chance during a routine stellar spectroscopy program by Pickering in 1897 at Harvard. S. N. Blazhko in Russia set up the first successful photographic program in 1904 – 1907.  This pioneer program yielded the spectra of three bright meteors. Up until 1931 only 11 meteor spectra had been secured, mostly by chance, except for the tree obtained by Blazhko.  Canadian astronomer Peter M. Millman while getting his Ph.D. at Harvard was asked to look at the meteor spectra secured thus far.  This resulted in two papers, the first can be down loaded at: and the second at: Peter Millman made meteor spectroscopy a life long interest and was considered a World Authority in this field.

From 1897 to 1958 the total number of known meteor spectra secured was only 318.  This was partly due to the few people engaged in this field and also because it was only possible to obtain spectra of meteors brighter than -2.0 magnitude and brighter with conventional cameras and films available at this time.  Most were obtained using objective prisms but in the latter part of the 1950’s transmission diffraction gratings were introduced to obtain spectra.  This was a great improvement as gratings produce near linear dispersion spectra unlike prisms which have good dispersion at the blue end but crowded dispersion at the red end making line identifications more difficult.

Dr. Millman encourage amateurs to get involved in this field and he published a paper promoting this, Amateur Telescope Making – Book Two, Scientific American included this paper, Meteor Photography.  Few took up this challenge because of the difficulty in securing a meteor spectrum.  The technique is simple but one cannot predict where a meteor bright enough to produce a spectrum crosses the camera field of view in the correct direction so some got discouraged and did not continue trying.  During this period conventional cameras fitted with a dispersive element, a grating or prism were used.  At the end of WW11 good quality aero cameras hit the surplus markets at very reasonable cost and these were adopted for use as meteor spectrographs.

Millman, and later with Ian Halliday and others established the Meanook/Newbrook Meteor Observatories in Alberta and later the Spring Hill Meteor Observatory near Ottawa.  The latter also used radar detection systems to study meteors in conjunction with photographic and visual observations.  Sadly these programs were shut down as a result of budget cuts by the federal government.  Ondrejov Observatory in the now Czech Republic became the centre to carry on this work.  Others have since taken up the challenge mostly in the USA, England, Spain, and Russia.

As mentioned before it was only possible to obtain the spectra of meteors brighter than -2.0 magnitude.  Hi speed cameras like the large aperture Maksutov Cameras were introduced by Gale A. Harvey NASA/LRC, in the late 1960’s and and 1970’s.  These cameras were capable of producing spectra as faint as +1.0 magnitude or so.  This effort produced 746 photographic meteor spectra during the four years they were in operation.  The results of this was published in various journals including Sky & Telescope magazine and IAU Symposium publications.

Also during this time, TV systems were being experimented with increasing the faint magnitude capability down to around +3 visual magnitude.  A paper on this, Spectroscopy of Perseid Meteors with an Image Orthicon by Peter M. Millman, A.F. Cook and C.L. Hemenway was published, refr. NRCC No. 11822 and I believe also in Sky & Telescope.

Since this time military night vision devices using image intensifiers became available and have been adopted by both professionals and some amateurs to obtain and record faint meteors and also meteor spectroscopy. Sirko Molau from the IMO runs a direct meteor recording program using image intensifiers systems and also faint lux security type cameras with fast lenses for his programs.  He can be contacted at if your interested in doing this.

I personally use second generation image intensifier systems to record faint meteor spectra.  As a Canadian I was lucky enough to buy a surplus 2nd Generation 25mm Image Intensifier, type MX9944/UV, before the U.S. government put export restrictions on these devices after 9/11.  Now they are difficult to obtain from U.S. sellers as you have to apply for an export permit.  Sometimes they are available outside of the U.S.A. so I scan eBay looking for them.  New ones are probably too expensive for an amateur’s budget so one hopes that a surplus one still has some life left in it.  You just take your chances buying a surplus intensifier and hope it will work.  Non U.S. made intensifies are also made by other countries, Russia, China, etc.  so all is not lost.  I recently bought a XX1335/Q image intensifier from a British surplus dealer which is nice for meteor spectra as it has a 50mm diameter input screen which will accommodate higher dispersion spectra.  I would love to get a 3rd Generation 25 mm ITT Intensifier as these have a longer life but alas the U.S. export restriction is in effect.

I once asked Canadian meteor astronomer Ian Halliday if this was still considered worth doing.  He said yes, but noted that the utility of running a meteor spectroscopy program for an individual can be difficult.  Conventional photographic meteor cameras require about 100 hours of exposure time to secure one spectrum. That’s a lot of film! One can of course concentrate your efforts during major meteor showers like the Perseids or Geminids to increase your chance of success.  One must obtain a very good spectrum from these showers to get a professional interested in measuring your spectrum.  One can of course attempt to measure your spectra yourself as there are computer programs available that are made for this purpose.  Unlike stellar spectra most meteor spectra have relatively low dispersion so identifying a line can be difficult.  In some cases you must have the experience in knowing the most probable line that should be present in that region.  I once tried a program and had Jiri Borovicka at Ondrejov measure the same spectrum and nearly 1/2 the lines I had identified were not the correct ones even though the computer program said they were correct.

There are other issues to deal with also such as distortions produced by the lenses you use.  It is very desirable to get a high dispersion spectrum but this requires a long focus lens a large grating and large format film, at least 4X5, 8X10 even better but the cost goes up exponentially.  I would love to find a large grating for a Kodak F-2.5 – 12 inch focal length lens and use 8X10 film but alas, this costs money.  I was lucky enough to find a surplus large reflection grating that should work using the method employed by the BAA member Mr. Aires.  A reference for his BAA Journal paper can be found elsewhere on this web site.  I still need to build the camera and find a source of inexpensive  Tri-X or Ilford HP-5 – 8X10 film.  No luck so far.

When doing spectroscopy one should try and work with another person situated 50 or so km away so heights of the start and end point can be arrived at.  One can then study the height of where certain spectral lines become visible or fade out.  One should also use  chopping shutter to arrive at the velocity and also the spectrum of the meteor train between the shutter breaks.  This also allows longer exposures as it takes longer for sky fog to build up on the film as it is exposed to the sky for 1/2 the time.  Our fireball camera network is very useful as it can provide height and velocity of your meteor spectrum if your doing this on your own as I’m doing.  That is, until I  get others in this network interested in doing meteor spectroscopy.

We could even accomplish a first, getting the spectrum of a meteor dropping fireball and recovering the meteorite.  This would answer many questions about the presents and formation of spectral lines by comparing it to the analysis of an actual recovered meteorite.  One can always dream!  😉

Too bad on can’t get a large format ccd detector for meteor spectra but robbing a bank to pay for one is probably not a good idea!  A Polish fireball group did have a nice success using a Canon 20D digital camera using crossed thin film gratings, a first by the way, using crossed gratings.  If you do this, the direction of the meteor flight path is less important.  Attached is their digital camera spectrum.  They deserve congratulations!  One feature is incorrectly identified as Cr at 427.4 nm.  Jiri Borovicka says this is probably an Fe iron line.  This spectrum is also unique as it’s in colour.  B&W is preferred as this simplifies photometry intensity scans as this is established for B&W films.  Jiri says this can be done with colour film also but is more complex to do.

I will try and answer any questions on meteor spectroscopy and if I can’t I will ask my friend Jiri Borovicka for an answer.  Hope some of you take up the challenge!

Ed Majden



Leave a Reply

Your email address will not be published. Required fields are marked *