Muniwin aperture selection

Affiliation
American Association of Variable Star Observers (AAVSO)
Fri, 02/17/2017 - 03:33

Hi all,

I've been testing Muniwin as an inexpensive replacement for MaxIM DL for student light curve projects. The algorithms used by Muniwin are well documented and reserached and the software appears to be exactly what I'm after. However, I'm a little confused about the way it automatically derives a range of suggested 'best apertures'. It seems to offer a suspiciously small 'best aperture' as part of its light curve generation and aperture algorithm - usually a candidate aperture that is significantly less than the standard “3-4 x FWHM” approach. As part of this routine, it also appears to remove some of the data points depending on the chosen aperture (eg, various magnitude values disappear from the light curve table when selecting larger apertures). I gather this is a good thing - ie, the software doing its job to sift through outliers and such, semi-automatically. But I'm still left wondering.

I note that Muniwin does include detailed descrptions of its algorithms in the help file, but I've been unable to fully reconcile those with the above observations.

Advice and/or clarification gratefully received.

Regards,

Paul

Affiliation
American Association of Variable Star Observers (AAVSO)
Hi Paul,

Hi Paul,

I use Muniwin for high precision exoplanet transit photometry.  A great strength of Muniwin is the ability to see the impact of aperture choice on scatter.  Muniwin does not choose an aperture.  You choose a set of apertures in the Project Settings.  As part of the process of plotting a light curve Muniwin displays a chart of scatter vs aperture choice.  I generally see that ~FWHM x 2 gives me minimum scatter.

As for dropping data points for some aperture choices, I don't see that behaviour at all.  It does not remove outliers.  There are plots you can choose from which you remove outliers manually.

Regards,

TG Tan

http://pestobservatory.com/

 

Affiliation
American Association of Variable Star Observers (AAVSO)
Thanks TG,

Thanks TG,

I do appreciate the abiltiy to see changes to the scatter by altering the Aperture, but I'm still confused about the aperture choices it offers after computing a light curve - with the bottom of the curve (ie, best choices) usually represented by smaller apertures than one would expect. Should I be performing some preliminary inspection of my data using the Quick Photometry tool and defining / fine tuning the aperture ranges in the Project Settings (ie, setting the minium Aperture to better represent my FWHM characteristics)?

Re: dropped data points, I'll post a screen shot. It definitely occurs in my data.

Regards,

Paul

 

Affiliation
American Association of Variable Star Observers (AAVSO)
Screen shots

Two screen shots showing the effect of increasing the aperture. A few data points disappear, and this is reflected also in the table and output report.

Affiliation
American Association of Variable Star Observers (AAVSO)
Dropped points

Muniwin will drop points if any pixel within the aperture is higher than the linearity limit that you set.  It is possible that there is a hot pixel near either V or C that gets included with an increased aperture.  I would check that particular image for this.

TG

Affiliation
American Association of Variable Star Observers (AAVSO)
Setting apertures

My FWHM is usually 3-4 pixels.  I define the range of apertures to be 1,2,3..to 12.  This way I have a choice that is much wider than needed to pick the optimum.  I have never seen a situation where the optimum is 3-4 x FWHM.  My optimum is usually about 6.  I have seen a prescription for 1.5 x FWHM, but I think this may be for consistently sharp images.  My photometry is usually done with slight de-focus, and stars are also sometime a little bit oval due to imperfect guiding.

I don't see what manual inspection with Quick Photometry does for you.  Surely what you are after is minimum scatter, at that is what the muniwin plot of scatter vs aperture gives you - regardless of the theory!

TG

 

Affiliation
American Association of Variable Star Observers (AAVSO)
Thanks TG - very useful

Thanks TG - very useful information, particularly the likelihood that there is a hot pixel or two at play in my dropped points. The move from the very 'visual' mode of aperture adjustment in MaxIM DL is taking a bit of getting used to, but I am starting to appreciate how the plots in Muniwin provide confirmation of the best settings.

Paul

Affiliation
American Association of Variable Star Observers (AAVSO)
aperture size

Are the aperture sizes Paul and TG describe radii or diameters?

Steve Howell describes the "curve of growth" phenomena for selecting aperture sizes in a paper and in his book.  VPHOT also gives a visual display of signal/noise vs. aperture size.  For any given frame, the best aperture may be different than for the next field or even the next frame.  Most books give you general guidelines as to which aperture to use, much like the rule of thumb of 2-3 pixels per fwhm, seeing being 2 arcsec, etc.

It is usually a good idea to drop a magnitude estimate if the measuring aperture includes a bad pixel.  Otherwise, you have to "fix" the bad pixel by interpolation using its neighbors, an action that can have serious consequences in the result.  That said, it is easier to identify the bad pixels if you are oversampled, so there are always tradeoffs in photometry.

Muniwin is an excellent program.  You might also look at AstroImageJ as another freeware program, or talk to the AAVSO about gaining student access to VPHOT.  There are another half-dozen programs out there, like AIP4WIN, MaximDL, Mira, etc. where the vendor might also be willing to support students if asked.

Arne

Affiliation
American Association of Variable Star Observers (AAVSO)
Radius vs diameter

Thanks Arne,

All now makes sense.

I've been using MaxIM DL for years with students, who tend to use a more 'visual' method of setting aperture based on a given screen stretch. The more quantitiative approach used in Muniwin has seen me revisit the basics as part of the classroom activity. I'm embarrased to admit that the inability for my eyes and brain to distiguish the words "radius" and "diameter" in the documentation, help files and software interface was the issue here.

Thanks,

Paul

 

Affiliation
Astronomical Society of South Australia (ASSAU)
Growth Curve

Hi Arne, Paul, all

IRIS has a growth curve plot feature that is useful for selecting measurement apetures. The IRIS tutorial available during the DSLR photometry course has some good guidelines.

I have played with AstroImageJ and considered using it for DSLR photometry, at least after the calibration steps (which IRIS does a reasonable job of). Do you any anyone else know of a good tutorial for using it and in particular, for producing magnitudes rather than fluxes (the default I think)?

David

Affiliation
American Association of Variable Star Observers (AAVSO)
Radius vs diameter

Ahh.  I'm guilty of sloppy talk.  The aperture sizes I was talking of were radii.  FWHM of course is 'somewhat like' a diameter.  So for FWHM=4 pix, I use usually aperture radius=6, or diameter=12.  So indeed 3 x FWHM!