Best software for very accurate differential photometry work?

Tue, 10/11/2016 - 20:10

Hi, I am a complete beginner with photometry.  I just ran my first photometry run with Maxim DL and a CCD camera a few nights ago.  It looks like all that Maxim DL does is plot the magnitudes of each star (target star, reference star, and check star) independently on a graph.  It also allows you to export your data to CSV or AAVSO format reports.  That's apparently all Maxim DL does.  I have a few questions:

First Question : I heard that Maxim DL's magnitude readings are not very accurate.  Can anyone confirm that?  Is there more accurate software out there that I should use instead of Maxim DL to take photos and report magnitudes with?

Second Question : I would like to do extremely accurate (as accurate as possible) differential photometry to study variable stars and especially exoplanets.  Can anyone make a recommendation on what is the best software available for analyzing the .fit photos?  My objectives for this software are (in the order of importance):

1) Windows based
2) As accurate as possible for extremely sensitive differential photometry
3) Versitile and capable of excellent analysis
4) Easy to learn
5) Easy to use
6) Good documentation and support
7) Widely used by professionals and ameteurs

Someone recommended Mira Pro, but that is the only recommendation I have received so far.  I would be interested in other's opinion of Mira Pro as well as any other software that might be better.  Thanks.

Ed

Affiliation
American Association of Variable Star Observers (AAVSO)
best software

Hi Ed,

For high precision work, the software used is less important than the observation methodology.  I recommend that you look at Bruce Gary's book to get oriented.

The main problems with exoplanet photometry are to get sufficient number of photons with as long exposure as possible, while introducing the least amount of noise.  The longer exposures are necessary to remove scintillation, a major contributor with amateur-sized telescopes.  The larger number of photons to get down to S/N=1000 or more usually requires defocusing so that the star image covers more pixels, none of which are saturated.  Defocusing can be a problem in crowded fields, as neighboring stars start to contaminate your measuring aperture.  The "least amount of noise" gets into how you calibrate your images and what filters you use.

So my recommendation is to learn how to take the best images for exoplanet photometry before worrying too much about whether one software package does better than another in the measurement step.

Arne

Affiliation
American Association of Variable Star Observers (AAVSO)
best software

After taking into account Arne advice you can use MuniWin for both variable star and exo planet transit.

I also used ASTROMETRICA but it will take more time as you have to measure every image. I read good feedbacks from AstroImageJ.

Pierre

Affiliation
American Association of Variable Star Observers (AAVSO)
Best software

I prefer AIP4Win from Wilman-Bell for observations. A friend of mine ran AIP against Maxim DL 5 and thought AIP had slightly better results. However, I concur with Arne. The procedure is way more important than the software. Bruce Gary's book is still the best resource available for exo-planet searches and very helpful for general photometry as well. Here is the link. I also bought a kindle version for my phone, kindle and computer. http://brucegary.net/book_EOA/x.htm

Paul Temple

Affiliation
American Association of Variable Star Observers (AAVSO)
Autoguiding very important

As an exoplanet transit observer, I found autoguiding, particularly off-axis autoguiding, to be very important to produce nice exoplanet transit light curves with lower chance of systemic errors by reducing drift over the several hours of transit observation.  Locking the exoplanet star to a small region of the image reduces potential problems with flats and other distortions in the image plane that could produce false mmag changes in the light curve, which can destroy a 5-30 mmag transit.  Having good flats also helps as well.  I use AIP4WIN and AstroimageJ, but I think most photometry software will work fine.

-Paul

BPAD

Affiliation
American Association of Variable Star Observers (AAVSO)
Best software for very accurate differential photometry work?

I especially agree with what Paul has said in regards to your question.  All of the advice about defocusing to get better signal-to-noise and avoiding scintillation problems are first-rate advice.  Working on observational techniques to get great results for exoplanet transits is a wonderful teacher.  In my experience, you quickly find that it doesn't really matter that you have a large, beautiful telescope that is equipped with a great set of filters and the supersensitive CCD if you are not able to eliminate some of the pesky systematic errors.  What Paul has said about reducing or eliminating drift entirely over lengthy data runs is critical to getting the best results.  I have seen trials where observations were set up so that a star would track across the remnant of a dust annulus that had been removed by the flat fielding process.  Often times, the result light curve would look like the star had experienced a low amplitude transit event with a duration equal to the time it took to cross the flat fielding artifact.  In comparison, I have seen many light curves done with modest telescopes that essentially keep targets on the same pixels during a time series run that reveal transits with a depth of 1-3 mmag.  I am absolutely convinced that careful technique and attention to basics will do more to improve observational quality than just about any other improvements you make regardless of the cost.

With regards to the software question, I have seen several programs mentioned along with AstroImageJ.  These are all really good programs but for exoplanet work, I would really recommend using AIJ.  The AIJ software was written by an exoplanet researcher specifically to do work on high-quality transit detection for exoplanet survey follow-up.  There is a modeling section in the recent versions of AIJ that allows you to do a quick fit of the suspected transit in order to estimate stellar and planetary properties.  In addition, there are a large number of ways that you can process the light curve that you generate in order to detrend using a large number of variables.  It is possible to detrend the data by position, total counts, X and Y position, a meridian flip,...  There is an active AIJ form online with lots of questions and problems being addressed as well as lots of practical tips for how to utilize various functions.  The AIJ software is free to download and works well on various platforms.  You can even automate the program so that it will process the data and produce a real-time light curve while the observations are being done.  My only complaint about AIJ is that I will likely never have enough time to get proficient with all of the many capabilities it possesses.

Anyway, I hope this is all helpful.

Mike Joner

Affiliation
American Association of Variable Star Observers (AAVSO)
Sign up for the next offering

Sign up for the next offering of the Exoplanet CHOICE course. You will learn AstroImageJ along the way. I am just finishing (in week 4 of 4) the course, and I can tell you from personal experience these last four weeks that Dennis Conti has created a rapid way to learn the methodology.

Affiliation
Astronomical Society of South Australia (ASSAU)
Tutorial material for use of AstroImageJ

Hi Brian, Michael, all

I would quite like to use AstroImageJ for DSLR photometry because of its cross-platform (it's Java!) nature and apparently capable feature set.

I've played with it but what I'd really like is some succint tutorial material for using AstroImageJ.

I'm happy to calibrate, stack, and extract B,V,R channels with IRIS but would like a more capable system for the photometry steps than IRIS offers. I do like its growth curve feature as a means to determining measurement etc apetures but AIJ's use of the astronometry.net API to plate solve, multiple comp star selection etc, seems enticing.

I note that Michael mentions an active online forum and Brian mentions the exoplanet course as a good way to learn it. I planned to sign up for the exoplanet course but time "got away from me".

Advice?

David

Affiliation
American Association of Variable Star Observers (AAVSO)
I use Muniwin

I use Muniwin, or more precisely, C-Munipack in the linux version.  I do mostly exoplanet work so high precision (not necessarily accuracy) is the main game.  Muniwin was built for time series photometry and great tools for aperture optimisation, selection of comparison stars, identification of variables in the field etc.  More info on my website; http://pestobservatory.com/photometry/

Lots of folk in the exoplanet community use AIJ and there is good support for it.  But AIJ does not deal well with images if the tracking/ guiding is not good enough to keep the target within a few pixels through the run.

As others have said, good tracking is the ideal, but with my equipment this is not possible.  I deal with this through meticulous flat fielding.  A big strength of Muniwin is that it will match stars even if images are very jumpy between exposures. Another strength vs AIJ is use of multiple apertures, so enabling optimisation.

Plusses for AIJ - exoplanet transit modeling, plots, support.

TG

Affiliation
Astronomical Society of South Australia (ASSAU)
Muniwin, AIJ

Thanks TG. PEST looks cool.

IRIS also does a good job of stellar matching with significant movement between frames. So good for wide field DSLR photometry. I have no problem with using 2 tools, as mentioned above.

So, can anyone recommend a good short tutorial/reference for differential photometry using an ensemble  of ref stars with AstroimageJ, to yield instrumental mags?

David

Affiliation
American Association of Variable Star Observers (AAVSO)
AIJ W/out Accurate Auto-Guiding

Chiming in really late here, but hopefully with information that some will find useful(?).

The way to use AIJ for images that are not auto-guided (or not very well) is to open a series of images as a stack (File -- Import -- Image Sequence -- select virtual stack for larger time series), and then perform the Align Stack procedure.

In essence, use the Set Aperture button to enter aperture, gap and annulus figures, and make them quite large -- large enough to include the star as it drifts -- so crazy-large numbers like 40-50-60 are not out of the question, though smaller numbers will work better for image sets w/out much drift.

Now under the Process tab, select "Align stack using WCS or apertures".  De-select any check boxes, and click on OK.  Now place the large apertures around 3 - 4 isolated stars, and press enter.  AIJ then chugs through the stack, and creates an output file called "Aligned".  You can close the image stack, import the aligned stack, and proceed with your analysis.  If there are spoiled images (clouds, etc...) in which AIJ can't recognize the selected stars, it will abort the process.  Thus, it's a good idea to "play" the image sequence as a time-lapse and note any bad images before trying to align the stack.  Any bad frames can be eliminated from the Main menu (Image -- Stacks -- Delete Slice).

Remember to re-set your apertures either manually, or by clicking on a star, then use Analyze -- Plot Seeing Profile to establish default apertures (2 X FWHM), or whatever sizes you prefer.

Clear skies,

Brad Vietje

Newbury, VT

www.nkaf.org

Affiliation
American Association of Variable Star Observers (AAVSO)
Dennis Conti's tutorial for AIJ

There is nothing like taking his CHOICE course to help you plow through the written material and AIJ's features. So far another session of the course hasn't been announced. It must be a huge amount of work to run the course, but it was popular so I imagine it will be run again soon. Those of us who just finished the course might be able to answer specific questions about AIJ by email. You can email me directly pretty easily by following the links on my profile, and I'm sure it would be educational for me to attempt to answer questions.

Affiliation
Astronomical Society of South Australia (ASSAU)
Thanks Brian

Thanks Brian. I may yet take you up on that.

I'll try to sign up for the next course and indeed, intended to sign up for the first. Never mind. :)

David

Affiliation
American Association of Variable Star Observers (AAVSO)
another tutorial on AstroImageJ

One of my graduate students, Andrew Lipnicky, started using AstroImageJ on a recent run at Kitt Peak. He wrote down things as he went, carefully enough that he was able to create a short tutorial. His project was photometry of a cataclysmic variable star, not an exoplanet transit, but in many ways the issues are the same (dealing with hundreds of images taken over a few hours).

It might help you to read his guide. Probably it can't hurt :-)

This web page contains links to the tutorial (in PDF) and to some AstroImageJ macros that Andy wrote.

http://spiff.rit.edu/richmond/wiyn/technotes/tech_23/tech_23.html

If you like it, and it helps you, please let Andy know! His E-mail address is awl6964 followed by rit dot edu.