Wing Filters and Mira Variables

Mon, 06/19/2017 - 18:02

I really enjoyed the photometry forum discussion we had Satuday over lunch at the SAS Telescope Symposium.  I would like to follow up the part of the discussion about additional filters for observing LBVs.  I mentioned the Wing Filter system as one specifically designed for observing Miras that may be fruitful to  deploy and crowd source.

I'm attaching a copy of Wing's JAAVSO paper to this post.  Here is a link to it in the ADS:

http://adsabs.harvard.edu/abs/1992JAVSO..21...42W

Also for reference, I'm attaching a copy of the paper I presented at the 2016 SAS Telescope Symposium which outlines and makes suggestions for any spectroscopy campaign for Miras.  The web page outlining that work is here:

http://go.uis.edu/BarberObservatory/research-programs/mira-variable-spectroscopy

Neither spectroscopy or the Wing system is ready out-of-the-box to use for an expanded crowd effort.  But I think they are worthwhile to pursue because LPVs are part of the AAVSO core strength and there is also potentially a lot of good science that a well-designed community effort could uncover.

Affiliation
American Association of Variable Star Observers (AAVSO)
wing filters

The 1992 paper by Bob Wing uses 3 filters: the 71 and 75 filters from his 8-color set, plus a new one at 1024nm (call it "102") that is a replacement for his 104 filter.  Astrodon used to produce a subset of the Wing filters in the 1.25" size, but I don't see them in his recent catalog.  I bet if a group order were placed, Don would go ahead and make another batch.

While these filters would do a far better job of monitoring the spectral changes in a mira atmosphere than do the broad Rc and Ic filters, the issue in my mind in their use is two-fold:  you need to calibrate any comparison stars used, and because mira variability is long, this would be a longer time-span project than the normal campaign.

Note that the 1024nm filter is very red, and the CCD throughput will be very small.  At the same time, however, a typical mira is 10,000 times brighter at 1024nm than at V-band!

Arne

Affiliation
American Association of Variable Star Observers (AAVSO)
Optec Filters

Does anyone out there have (or know who has) the Wing filters that were supplied by Optec for the SSP photometers? 

Correcting Errors In that 2016 Presentation

Bil Rea corresponded with some other astronomers and learned I had made two mistakes in the paper we published for the 2016 SAS Telescope Symposium (posted above).

I have uploaded a replacement copy of that paper that notes the errors and makes revisions.

The errors were my fault.  

In section 2 where we wrote about the TiO bands it was incorectly stated that the TiO is strongest and maximum brightness.  In fact TiO bands are strongest when the star is coolest which is at MINIMUM brighness.

Also in section 2 the end last sentence in the second paragraph is confusing.  Not sure what I was writing there.  But to iron it out:  Keenan hypothesized in his 1969 paper that the AlO bands grew weaker because they were filling in with emmission (which mostly happens when mass loss increases).  I did not follow that up and can't tell you with 100% certainty if Keenan's hypothesis was proven or disproven or never adressed.

I had a wild idea that

I had a wild idea that probably won't work for a number of reasons.  I'm going to spit-ball it here beceause someone might have a better idea inspired by this.  Please take this as a proposal for an experiment and not a tested recipe.

  One could do spectrscopy of both a target and one of the Wing Standards.  After carefully extracting the 1D spectra the same way (very important) and putting them both in units of counts/sec/pixel (also very important) you could sum up the pixels in the wavelengths corrresponding to the Wing Filter band pass.  Bassically you are calculating the instrumental flux in the Wing filter from the spectrum.  By comparing the instrumental flux sum over the filter band for the standard to the known magnitude of the standard you can figure a counts/sec -> magnitude transformation.  You then apply that transformation to the instumental flux sum for your target to get the target's magnitude.

I can imagine about one hundred ways that could end up going horribly wrong.  It doesn't take much imagination.  Think difference in seeing or airmass or slit/aperture illumination or guiding or flexure of the telescope....  So I'd be very suspicious of my results and would want to try it several ways to verify the them.  Maybe start with using one standard to measure the magnitude of another standard.  This might be worth exploring because while it might not be a winning recipe per se, experimenting with it might lead to a good recipe that could be used for specroscopically monitoring these stars using a Wing-like system.

Related: if one could propose a decent way to use Wing's indices to inform measurement of equivalent widths for the critical features that would be just as good.  The trick with equivalent widths is that you need to "know" (or resonably estimate) the continuum flux if the feature you are measuring was not present.  Estimating the contiuum is a tricky bussiness in M-type stars because the features are broad and overlap.  It is a daunting task to guess what the continuum flux is at a given wavelength if no absoprtion was present.