VV Cep spectroscopy reference star ?

VV Cep 2017-2019 Campaign
Moderator for this forum: Ernst Pollmann
Olivier Thizy
Posts: 370
Joined: Sat Sep 24, 2011 10:52 am
Location: in the french Alps...
Contact:

VV Cep spectroscopy reference star ?

Post by Olivier Thizy »

Hello,


Regarding spectroscopy, I am looking at which reference star to use - while 20 Cep is the reference star for photometry, it shouldn't be used for spectroscopy as too late type.

I have been using hd203307 (close to delta Cyg) which is also in MILES database but ~6° far away.

In François excel, we have 7 Cep (hd204770) which is a closer (~4°) mag 5.4 magV star of type B8V with E(B-V)=0.03.

Bottom line, which reference star do you usually take and which one should we recommend to take so we all use the same reference star?


Cordialement,
Olivier Thizy
http://observatoire-belle-etoile.blogspot.fr/
Olivier Thizy
Posts: 370
Joined: Sat Sep 24, 2011 10:52 am
Location: in the french Alps...
Contact:

Re: VV Cep spectroscopy reference star ?

Post by Olivier Thizy »

From Steve Shore:
I'd suggest using one of the KPNO spectrophotometric standards,

HD 217086 22:54:49.0 +62:27:35 O5 7.49 +0.63

http://adsabs.harvard.edu/cgi-bin/nph-b ... 58a4905502

and use the NOT visibility tool (there's also one from ESO) to check the timings of your standards:

http://www.not.iac.es/observing/forms/visibility/

you can put in a list of stars and also the location of your observatory. I've included screen shots. The star is ideally suited for comparisons (using, for instance, the date 15 Sept. and Grenoble as the site)

Steve
Screen Shot 2015-08-24 at 2.56.06 AM.png
Screen Shot 2015-08-24 at 2.56.06 AM.png (195.26 KiB) Viewed 9364 times
Screen Shot 2015-08-24 at 2.56.40 AM.png
Screen Shot 2015-08-24 at 2.56.40 AM.png (111.25 KiB) Viewed 9364 times
Olivier Thizy
Posts: 370
Joined: Sat Sep 24, 2011 10:52 am
Location: in the french Alps...
Contact:

Re: VV Cep spectroscopy reference star ?

Post by Olivier Thizy »

From Phil Bennett:
Hi Steve, Olivier, Ernst, and others,

I guess I don't see the need to observe a check star for spectroscopy, unless you really want spectrophotometry, in which case you shouldn't use a slit (because the flux trhough the slit will depend in a color-dependent way on the the fraction of the seeing disk transmitted through the slit).

For medium or high-resolution spectroscopy you need to use a slit to optimize the spectral resolution, and this basically rules out the option of obtaining spectrophotometry (flux-calibrated spectra).

Personally, I think we need higher resolution more than we need flux-calibrated spectra -- it's the spectral lines that contain most of the information.

So I wouldn't bother with observing check stars, *especially* ones (i.e., HD 217086) that are 2.5 mags fainter that the program star (VV Cep). If you do that you'll be spending 90% of your time observing the check star, which is a*huge* waste of time.

The other problem with (slitless) spectroscopic of a late-type star like VV Cep is its red color. All the late-type giants and supergiants are somewhat photometrically variable, so you won't find a good spectrophotometric standard of similar color. Therefore, one is forced to use a earlier (hotter) spectral type as the standard, but again, the seeing disk of the star will be refracted by the atmosphere in a manner that depends on the (different) colors of the stars, and you have to be careful to account for this during your reduction.

If you want a spectrum of a spectral standard (not a bad idea -- but keep it *bright*), I recommend mu Cephei. It's about a magnitude brighter so at least you will be spending most of your exposure time on the program star(s). Then you can also do a few spectrophotometric standards, e.g., Vega (for VV Cep) sometime during the night. You should also do some twilight exposures (in deep twilight) to observe a flat field in order to calibrate out the CCD response across the chip.

Personally I would observe VV Cep at medium or hi-res using a fairly narrow slit, and just rectify the continuum, or normalize to Vega. It won't be true spectrophotometry but it won't be that bad either, and you will have better spectral resolution. Just spend most of your time on the program star, and be careful not to saturate H-alpha in the VV Cep spectrum.

Phil
Olivier Thizy
Posts: 370
Joined: Sat Sep 24, 2011 10:52 am
Location: in the french Alps...
Contact:

Re: VV Cep spectroscopy reference star ?

Post by Olivier Thizy »

Phil,


many thanks for your input. Just to clarify, reference stars have different meanings for the amateurs.

A reference star is often used to calculate the instrumental response of our equipment. This includes both the 'true' instrumental response as well as atmosphere response - this is why we are looking for reference stars close to the target, within 5 degrees typically.

Now, there are some methods (which I personally do not apply... yet!) to separate the true instrumentale response (which is fairly stable over time in a permanent observatory/equipment situation) and the atmospheric extinction which can be modeled. This is well described there (in french but can be 'google' translated):
http://www.astrosurf.com/buil/isis/guid ... method.htm

In the mean time, I will switch to 7 Cep as reference star to calculate my overall instrumental response at an airmass close to the target.


Otherwise, we can also use spectrophotometric standards and use them to calibrate our spectra in absolute flux. This can't be done with eShel fiber fed spectrograph, but is possible wih Alpy 600, LISA or Lhires III spectrograph when used with photometric slits. Those slits have two opening; a narrow one to take the spectrum and a large one to capture all the flux and use this to calibrate the spectrum. But you are right, this is time consuming.

Another way to do it is to have B/V photometry done at the same time on the target. Spectra can then be rescaled for absolute flux. This method has been done for exemple by David Boyd to mesure Halpha emission on EE Cep during last eclipse. He presented his results during last OHP spectroscopy workshop and we was able to show that while Halpha emission was varying compared to continuum, the emission was actually constant over time during the eclipse. See his paper on JBAA:
http://arxiv.org/abs/1412.5127v1
David's talk was recorded during OHP and is available here:
https://vimeo.com/136381653


But I undertand that you recommend to focus on high resolution spectroscopy, if possible toward the UV, rather than trying to do absolute flux calibrated spectroscopy on VV Cep.


Also a quick note on flat. Most of us use slit spectrographs that have internal tungsten lamp to make spectroscopic flats. Twilight flats will of course not work on slit spectrographs.


Cordialement,
Olivier Thizy
Vincent Bouttard

Re: VV Cep spectroscopy reference star ?

Post by Vincent Bouttard »

Olivier,

I have used HD208682 at OHP workshop. This star is really close to VV Cep. OK, it's a Be star but at an early state (B2Ve). I don't know how it really impacts the instrumental response.

Vincent
Olivier Thizy
Posts: 370
Joined: Sat Sep 24, 2011 10:52 am
Location: in the french Alps...
Contact:

Re: VV Cep spectroscopy reference star ?

Post by Olivier Thizy »

Vincent,

Some Be star (for exemple delta Sco) have some free-free emission which create some continuum over the atmospheric continuum. I read somewhere that delta Sco true continuum was 30% (or was it 70%?) below what we observe... anyway, in such case, we shouldn't use such star as reference star for the instrumental response.

But I do not know if this kind of continuum emission is often or not in Be stars but just to be sure, I tend to avoid those type of stars as references.


Cordialement,
Olivier Thizy
Thilo Bauer
Posts: 37
Joined: Sun May 31, 2015 4:45 pm
Contact:

Re: VV Cep spectroscopy reference star ?

Post by Thilo Bauer »

Olivier,

good question. Because nobody could provide an answer at OHP, I tried to find a acceptable answer by myself during observation at the OHP.

My personal opinion: Reference stars to calibrate the instrument response should be bright enough. There are two reasons for choosing stars around 3rd magnitude and above: Signal and signal. That means two things: One should require a good quality signal, but not too long exposures which will mean wasted time. Another aspect might be an almost similar air mass for the calibration standard and the target to observe. If this condition is not fulfilled and stars are too far away from each other with different airmass, it will be a mess to calibrate the whole observation site over years.

Calibration of the observing site is, what astronomers did and continue to do with spectro-photometry and integral photometry. Calibration of the site is a very time consuming task by itself. I don't know if one ever tried this for any typical suburban site, like those amateurs are using. I guess weather conditions vary much more on a suburban site, than at a typical, carefully selected location of an astronomical observatory. For me there will also be different areas of sky brightness and directions where large cities may increase seeing and air pollution. Usually, I try to avoid these directions of sight for astronomical observation.

So, my conclusion is: Having no calibrated site, air mass of reference and target should be similar. The reference standard should not be too far away. Ideally the reference should have similar declination to the target. In this case one might try to do alternating observation of target and reference.

To have indicators about magnitude and selection of references stars:
Steve Shore suggested HD 217086 for reference, a young stellar object of magnitude V=7.66. For my 8" Cassegrain, a star of that magnitude looks too faint as reference. For 30 sec exposure of a Canon DSLR and 8" Cassegrain with Alpy, the signal of HD 217086 will be as low as CH Cyg. Means, you don't even find it on the screen of the camera from a single exposure, but there will be enough signal for stacking 10-50 spectra.

Finally, and more intuitively, I selected the only two bright stars listed on "ESO RA ordered list of spectrophotometric standards", that might be good:
  • HR 7001, alpha Lyr
  • HR 153, zeta Cas
Zeta Cas should be good reference for other spectroscopic targets in this region as well, like gamma Cas. Unfortunately, there are not many bright stars listed with declination similar to VV Cep found on ESO list. Alpha Lyr is a bit far away from VV Cep. But, for my site it will be around zenith these months, good for instrumental calibration with air mass close to one.

Summary: I would suggest these additional criteria for spectrophotometric reference stars are:
  • Brightness. Example: Magnitude <V=4 for Alpy + DSLR and 8" telescope.
  • Air mass could be fulfilled with similar RA and DE, probably alternating observations
  • Suggest one reference with DE=90-[altitude of site location] to calibrate instrument at air mass close to one.
Quick Reference: http://www.eso.org/sci/observing/tools/ ... anlis.html

Perhaps, we could compile a list with additional reference targets.

Thilo
Peter Somogyi
Posts: 420
Joined: Sun Jul 13, 2014 8:56 am

Re: VV Cep spectroscopy reference star ?

Post by Peter Somogyi »

Thilo,

A reference star does not need good SNR - though in case of Alpy where you have UV, it might be more demanding towards 3550A edge case (A-type stars drop there accidentally).
It also depends on resolution and coverage and project requirements which is the better.

As for dslr + Alpy: if you use it with the barlow, the barlow makes such a poor snr to DSLR that you're far better with switching to a DSLR-priced ccd (like with the Orion G3 I did - though its fits header date format needed workarounds, but it's just software manipulation).
Anyway Alpy can be used without the barlow + very special adapters if you manage to do a hard mod. taking out the mirror (I did it, snr got much better). You also need focal reduction for your scope as Alpy is optimized for f/5.
But I suggest continuing this discussion at another forum topic.

Back to VV Cep: which technique and how frequent spectra required?

Cheers,
Peter
Thilo Bauer
Posts: 37
Joined: Sun May 31, 2015 4:45 pm
Contact:

Re: VV Cep spectroscopy reference star ?

Post by Thilo Bauer »

Peter,
Peter Somogyi wrote:A reference star does not need good SNR...
Here I would go back to the question for what purpose a reference star will be used, esp. one having a bad signal.

And yes, you are right. The Alpy DSLR barlow lens decreases S/N, as spectra are oversampled in both wavelength and spatial resolution along the slit, at least for the prime focus of my Cassegrain. This could be considerably improved.

On the other side, noise of a Canon EOS 60D is considerably low and close to 3 e-/pix. Thus 2x2 pixel binning will work well. The most important thing is false adaptation of the Alpy to my Cassegrain using f/9 at prime focus. In this case, and with typical amateur telescope mount with tracking errors, seeing and remaining tracking errors of the mount are the most important sources to limit the signal through the slit.

Therefore, I will switch to use the Vixen focal reducer of my Vixen VC200L to go with the Alpy 600, soon. A focal ratio of f/6.4 not only will improve resolution, but decrease the probability, that the star will move from the slit. I think this will improve both, resolution and signal - to something more reasonable for my Canon DSLR. I can live with that for a while. One of the great advantages is color to my spectra.

Thilo
Olivier Thizy
Posts: 370
Joined: Sat Sep 24, 2011 10:52 am
Location: in the french Alps...
Contact:

Re: VV Cep spectroscopy reference star ?

Post by Olivier Thizy »

Thilo,


After our object spectrum has been extracted (ie: after image preprocessing, geometry corrected, sky background removed, and signal extracted colmumn by column) and calibrated in wavelength, we have to apply an instrumental correction to adjust the overall profile to correct from:
-atmospheric extinction (depending on your local air parameters & airmass of course)
-telescope optical response to wavelength
-spectrograph optical response to wavelength
-CCD sensor efficiency to wevelength

As I mentionned, we should certainly (well, let's say I personally should!) split the first one as the 3 last ones should be constant over time and certainly withint few nights if nothing is changed in the equipment/configuration.

But you can also record a "reference star" which is a star with minimum lines (typically A or B-type), observed at the same air mass as the target to integrate the same atmospheric extinction.

This is where I proposed to use 7 Cep as "reference star".


There would another reference star if you intent to calibrate your spectrum in absolute flux. This would be a "flux calibrated standard".


I hope this helps,

Cordialement,
Olivier Thizy
Post Reply