How to get a light curve from my observations?

[rogerg]

Hi all,

I'm struggling with the next step of occultation work - getting a light curve from my FITS files.

I have two occultations recorded from the weekend, both recorded as FITS files via CCDSoft/TheSky.

So, I have a bunch of FITS files which I know contain the target and several comparison stars, but I'm confused and currently unable to process them in to a light curve.

I've tried using CCDSoft, treating the occultation as a stationary variable star, my problem there is it insists that it must plate-solve the images to be able to further process them. This is problematic because in one occultation there's too few stars for a successful plate-solve and in the other there's so many it gets confused and comes to the wrong conclusion as to where it's pointing. Frustrating! So it seems I can't go further with CCDSoft. Frustrating because from what I understand there's no need for the plate-solve, all I need to graph is the variable/occulted object against 1+ comparison objects.

Anyone able to help? Suggest a program that can do this?

I've been reading through citizensky.org and aavso.org in an attempt to work out what software people use but have ended up more confused and haven't found a concise list of what software is recommended (have found many dead links to software).

I've read about "API4WIN" and apparently it's home page is here (http://www.stargazing.net/david/aip4win/), but I can't find information about it - what it does, if it'll help me, what to download.

Btw, I did record one of the two occultations as an AVI on a second camera and have successfully got a light curve from that using Tangra. But Tangra doesn't do FITS.

:help:

Thanks,
Roger.

[Terry B]

I use AIP4WIN but it isn't free. It isn't expensive and comes with a good book.
Otherwise Iris is free and will measure a light curve for you.
http://www.astrosurf.com/buil/us/iris/iris.htm

instructions are here
http://www.astrosurf.com/buil/iris/tutorial15/doc38_us.htm

[rogerg]

I've never had any luck with Iris in the past, been too hard to use. But that tutorial you linked to looks quite promising. I will try that.

I don't mind paying something, for AIP4WIN, if I know it's going to do what I want, but I haven't confirmed it will (couldn't find specs on capabilities of AIP4WIN).... but if you use it to get a light curve from a sequence of FITs files then perhaps that's the confirmation I need :)

Thanks for your reply.

Roger.

[[1ponders]]

AIP4 WIN is a bit clunky in some respects, but its fairly straight forward and simple to use. Add to that, the book should be on evey astrophotographers bookshelf (jmho :P ). Well the first half of the book anyway :lol: The rest loses me :)

[rogerg]

I've just become aware of "MPO Canopus" and remembered I have a license to Astrometrica.

This is all getting very confusing, steep learning curve. Not really understanding what each piece of software does WRT generating light curves and what type of "photometry" this occultation stuff is exactly ... :rolleyes:

I think I'm going to have to spend a lot of time reading and tinkering with:
- Astrometrica
- MPO Canopus
- IRIS
- CCDSoft
... in order to work out what exactly it is that I need. Hmm. And not give up part way through :) Then again, I guess the next one I try might just work easily and I can ignore the rest :)

[lepton3]

Roger,

You may also want to have a look at the program ImageJ. There are a number of useful plugins for Astronomy and CCD image processing. It is a free download, and quite user friendly.

There is a helpful (and reasonably short) tutorial at http://www.astro.physik.uni-goettingen.de/~hessman/ImageJ/Book (http://www.astro.physik.uni-goettingen.de/%7Ehessman/ImageJ/Book)

Ivan

[pvelez]

Hi Roger

I have MPO Canopus and am also trying to work out how to generate a lightcurve or 2.

Canopus is good, subject to:

1. Its full featured - which is good - but also confusing at first

2. for some reason, the latest version has decided not to plate solve on my lappy - still works well on a desktop.

I have AIP4Win - looks ok although I suspect it only generates a file of observations for export to Excel from which you generate your curve. I agree with Ponders - its a great text.

Pete

[[1ponders]]

Hi Pete, AIP4WIN can generate its own light curve graph or export to excel. But I'll admit its been a while since I last used it so my skills with it are rusty. There are certainly better programs around now. :)

[AstroJunk]

Hi Roger,

Here is a free and mostly painless way using VirtualDub.

1. Convert all your images to bmp, tif, jpg (bit not FITS!)

2. Name them all in sequence as follows something.001.bmp, spmething.002.bmp etc

3. In VDub, select open video file and change the files of type to image sequence

4. Select the first of the sequence, ensuring that the 'automatically load linked segments' is checked - this will load all your images in to vdub in order.

5. Select Save as AVI

6. Run it through LiMovie :thumbsup:

[rogerg]

Thanks for the continuing input everyone. I've got several leads to go on now, so will just make some time to work through them.

Much appreciated :thumbsup:

[rogerg]

an update..

I had some success with IRIS in the last couple of days. The attached light curve is from my Charon occultation. The prolonged dip near the start might be the Charon event.

It's taken a lot of hours to get this far in the last few days! lots to learn. More to go before I submit the results in the next couple of days.

If the IRIS results are correct I'll likely stick with using IRIS, as it was quite easy in the end (after a few hours working through the tutorial and doing some tests..)

Roger.

[Terry B]

Well done. Iris is a good choice. Maybe try taking a time series of a known variable star to give you a data set as a test. A good test is a SXPHE type star. They have a fast variation over about an hour.
A list of them is here
http://www.aavso.org/vsx/index.php?view=results.submit2
A good example in the evening sky now would be V0462 Hya or KZ Hya

[AstroJunk]

It's a shame that no one else has bagged this one. It makes analysis much easier when the event is known to have occured, and is vey useful for callibrating systems.

DaveGee had a miss, so from my basic calculations that means that the duration from your site could be upto about 50 seconds.

I would be more convinced by the curve if the tail end wasn't so wayward, but that may be a processing artefact - looking promising though.

[rogerg]

I completely agree. The first dip is quite consistent and sustained, but the erratic nature of the rest leaves you wondering.

I don't know, I'm happy to have something :)

There's been a lot of batting around in emails about if there is good data or not, I'm kind of "over it" and just want to get the data to Bruno so he can decide for himself :)

[rogerg]

Here's a new graph from a fresh analysis from iris using all 110 files I have (not just the 90 I could be bothered renaming previously) to show some more of the data beyond the end.

:shrug:

just thought I'd add it for completeness.

[AstroJunk]

What is you exposure time per frame? And does Iris give any magnitude comparisons to check? The worst thing about Pluto/charron events is their massive uncertainty. I think OW rather understates that.

This is a really hard one to confirm. Massively faint, low mag drop and no correlation to other observations (because there weren't any). A very bold attempt Roger and don't be dissapointed if it doesn't get accepted - but don't give up either :)

Thankfully, 90% of events will be easy...

[rogerg]

5 sec/frame (should have done shorter, but oh well :) )

Insure about magnitude comparison accuracy to be able to state. The numbers iris give are probably that comparison but I haven't told it the magnitudeof any stars for it to have a baseline.

[AstroJunk]

5 seconds is pretty good for that magnitude :thumbsup:

Might present a bit of an issue though, as the expected max duration could only have been 10 frames, or ignoring Dave's miss, 14 frames dead center:sad:

[Ron Garrett]

Isn't LiMovie the software of choice for occultations?

Ron

[rogerg]

My question was regarding getting a light curve from my FITS files, as I don't record video. LiMovie handles video.

I have ended up using IRIS with success.

Thanks,
Roger.

[rogerg]

Out of interest, seeing as I've been reminded about this thread.. I've done some more testing and definitely should have used shorter exposures than 5 seconds. For the upcoming June 4 occultation I'll be using 1 second exposures. At that exposure time I get 18 images per minute.

Roger.

[higginsdj]

I take it that you haven't done done photometry before.... How linear is the CCD for your target and comp stars for the exposure time you have chosen? Linearity falls off at both ends (over and under exposure) of most CCD camera's. Can I assume you used the 12"? 5 seconds in a 12" scope isn't going to produce a very high SNR for a magnitude 14 target meaning that whatever occultation you may record will be well inside the noise and linearity for such exposures is also questionable.

You have an option to stack your images (say 5-6 consecutive images) and then re-measure. Yes, your timing of the occultation will suffer but you should be able to get 0.05mag accuracy (or close to it) in the photometry and get around the extensive noise in your data.

(To achieve 2% photometry in my 14" I have to integrate for ~ 60 seconds on mag 14 minor planets)

[rogerg]

Thanks for replying David ...



I don't understand sorry ... an indication of my level of experience yes :) I have briefly about linearity of CCD's, seems from what you say I need to understand it.



12" @ ~F/7.5 yes, 5 second exposures gave the target (Pluto) an ADU of about 1150 (varies between exposures of course). That's compared to a backgorund noise level that I'd estimate to be about 200 ADU.



Yeap, stacking, sure... but if 5 seconds isn't sufficient I'd just increase the exposure time, I'd think at these "short" exposure times increasing the exposure time would be more suitable than stacking?




Wow, OK ... not the impression I was getting from the occultation people out there that use video cameras for these things! (not even 5 second exposures). Interesting.

Thanks,
Roger.

[AstroJunk]

I think the thread has begun to stray away from occultation timing and into photometry. Two very different outcomes...