Initial results: Spectral Classes O, B and A

One of the benefits of doing astronomy in late December at this latitude (48 N) is that one can begin observing at 5:00 pm, wrap things up by 9:00 pm and still have an early bedtime. the following observations were made on December 20 and 23, 2023.




Obs Date

profile /



10 Lac

5:25 pm



bet Cep

7:11 pm



87 Psc

6:10 pm




6:37 pm



28 And

5:14 pm

Obviously, the best match was for the star Alfirk. I cannot explain the deviations for the other profiles. All profiles were created with the same master flat frame; all used the same instrument response, and all were taken within an hour on the same evening except 10 Lacerta three days earlier.

In addition to trying to improve my workflow, I want to take a look at rectified profiles for these observations (by dividing out the continuum) so I can more readily compare my results with those in Walker’s Atlas. I will save that for another post.

Capturing a Spectrum of NEOWISE

On Sunday evening, July 19 the sky from Whidbey Island was clear all the way down to the horizon, so we had an excellent opportunity to take a few more images of comet NEOWISE. We headed for the parking lot at Greenbank Farm where we had a good view of the northwestern sky. It would have been nice if the parking lot lights had been off, but since the camera was pointing in the opposite direction, it wasn’t too much of a problem.  We had good luck leveling the telescope mount and aligning it with the celestial pole by viewing the North Star through the polar scope. We used Arcturus for a quick one-star alignment. The mount (iOptron CEM25) was tracking very well.

This time I remembered to bring along my intervalometer so I could take exposures longer than 30 seconds. The image below is the result of 13 minutes of total exposure time (5 frames each of 90s, 45s, 15s, 5s and 2s). I had the idea of using the shorter exposures to avoid over-exposing the comet nucleus, but that turned out not to be much of an issue.


We easily see the bright head or ‘coma’ and the fanned out dust tail. Within the coma is the comet nucleus, a small (~ 5km) clump of rock, dust, water ice and frozen gases such as carbon dioxide, methane and ammonia, which can be described as a “dirty snowball” or an “icy dirtball”. The bright coma we see is much larger. From this photo I estimate the coma to be about 230,000 kilometers wide.

If you look carefully, you may see another tail from the coma—straight and angled up slightly relative to the dust tail. This second tail is a result of ionized gas emanating from the nucleus of the comet. Both tails are the result of material coming off of the nucleus. Dust particles drift away and are spread out by pressure from sunlight, producing the prominent dust tail. Various gases are ionized by ultraviolet light from the sun. Being much smaller than dust particles, these ionized atoms and molecules are more strongly affected by sunlight pressure and are propelled along a straighter trajectory.

My estimate of the size of the coma is based on the field of view of my camera and lens. The width of the coma in the image is about 102 pixels, or 2.8% the height of the frame (3648 pixels). The field of view of my telephoto lens is 274 arc-minutes high, so the coma subtends an angle of 2.8% of 274 arc-minutes, or 0.00223 radians. According to NASA, the NEOWISE comet is about 100 million kilometers from Earth. So the width of the coma must be (0.00223)x(100,000,000 km) = 230,000 km, compared to the diameter of the nucleus of only 5 km.

We were in good shape for taking long exposures so we decided to try our luck capturing a spectrum of the comet nucleus. For this, I attached a Star Analyzer 100 diffraction grating onto the front of the telephoto lens.

Star Analyser 100
Star Analyser 100 diffraction grating (Paton Hawksley, UK) attached to a 300mm lens on a Canon 6D DSLR.

The diffraction grating separates the light into its constituent colors. With the Star Analyser in place, the aperture of the lens is reduced to about 10%, so capturing spectra really needs longer exposures, but I settled for a total exposure of 12 minutes.

In the image below, you see what the comet looks like through the diffraction grating. The comet is on the left (its tail streaming upward). The rainbow-colored line to the right is the spectrum of the coma, while the blended colors are from light from the tail. Since the coma is relatively compact, its spectrum is more distinct than for the tail. You can see brightening in the blue, green and yellow parts of the spectrum.

Note the spectra of other stars that happen to be in the field of view.

comet through diffraction grating
Five-minute exposure of comet NEOWISE through Star Analyser 100

As a reference, I also captured a spectrum of Vega which was in a good position, almost directly overhead. It is so bright, it doesn’t require a very long exposure. A 9-second exposure was sufficient. Vega is a good reference because it has prominent hydrogen absorption lines, especially the hydrogen beta line at 4861 Angstroms.

To analyze the spectra, I rotated the images to make them horizontal and cropped them.

Cropped spectrum of NEOWISE (5-min exposure)

Cropped spectrum of Vega (9-second exposure)

The next step was to open the Vega spectrum in the spectral analysis program RSpec and calibrate it by identifying the hydrogen beta line. In this way I could associate wavelengths in Angstroms with pixel values along the x-axis. I got a scale factor of 2.23 Angstroms per pixel.

Vega profile
Light intensity vs wavelength for Vega. Peak at the far left corresponds to the undiffracted (0th order) image of Vega.

Next, by opening the spectrum of NEOWISE and using the previously determined scale in Angstroms, we have a profile for the comet.

NEOWISE profile
Spectrum profile for comet NEOWISE

Note that there is considerably more noise in the profile (random vertical variation) due to the fact that the comet is so much dimmer than Vega.

The camera is only sensitive to wavelengths between about 4000 and 7000 Angstroms, so we cropped the plot outside to within  this range to produce the following.

I am not certain that I have identified the peaks correctly, but wavelengths for C2 and CN were listed in the RSpec software as typical in comets. ‘C2’ refers to neutral carbon diatomic molecules; ‘(CN)2’ refers to cyanogen. The closest I could find for an element with an emission around 5577 Angstroms was the singly-ionized state of the oxygen atom, which I’ve labeled as ‘OI”. Not sure whether that makes any sense, however. I would invite suggestions from those who know more about spectra than I do.