Generating Light Curves - APT and Excel

From CoolWiki
Jump to navigationJump to search

Install APT

Heres some info that a previous group put together for installing and using APT Aperture photometry using APT.

If using Windows, you need to type java -mx512M APT in to the run window to start APT.

It sounds like the newer version is making this easier, but I haven't tried yet--Piper 18:52, 30 May 2010 (PDT).

APT Settings

For our practice light curves, we wanted to compare the various aperture settings as discribed in Don's IRAC notes File:IRAC notes.pdf The aperture and background settings we used were

aperture/background

3-3-7
3-12-20
5-5-10
5-12-20
10-12-20

We also chose to have APT subtract the background using setting B subtracting the mean background

Steps for we used to gather photometry data for test stars

  1. you must first download and unzip the images under Test Light Curve Images below
  2. open APT
  3. click on get image, navigate to the folder where you put your images
  4. open first test light curve image
    1. the first image is 0003
    2. there are 22 images
    3. see notes below for Reading Fits Names
  5. set your aperature and background numbers
    1. aperature controls the number of pixels used to determine luminosity (red squares)
    2. inner sky radius sets the inside of the area to use for background subtraction (green donut)
    3. outer sky radius sets the outside of the area to use for background subtraction
  6. click on More Settings in lower left
    1. we had to click on the maximize window button to be able to see whole window
  7. choose Model B, sky median subtraction, apply settings, then close window
  8. put your cursor over the star you want to measure and click
    1. our three test stars are shown on the previous page Link title
  9. click on Snap to center star
  10. click on Save Photometry data
  11. click on Show Photometry file to see the data you collected
    1. look in file to make sure you are using Model B
    2. look in file to make sure you are on the image you think you're on:)
  12. close photometry file window
  13. put your cursor over the second star you want to measure
  14. click on Snap to center star
  15. click on Save Photometry
  16. you can check the file again if you want
  17. repeat for third star

"""Open next image"""

  1. click on get image
  2. we chose option A to read in new image and clear old
    1. we weren't sure what advantages the other options might have, but A worked
  3. check heading under image to make sure you truly have the next one
    1. the second image is 0020
  4. you have to go back to step 5 and 6 to set aperture, background and Method B
    1. we think there's a way to set this, but never found it:(
  5. click, snap, and save for all three stars
    1. you don't have to check the file every time
    2. you should check the file occasionally to look for
      1. Method B
      2. 3 stars with correct RA and Dec for each image

Excel to Generate Light Curve and Standard Deviation

APT Photometry data are saved in a table named APT.tbl in the programs root directory. To open this file in Excel, open Excel, navigate to the APT directory, allow all files to show be shown and open APT.tbl. Excel will allow you to separate the data into cells automatically. If all data ends up in one column, choose that column and then find the "text to columns" option to separate.

You can use this spreadsheet to check to see that you

  • gathered data from all three stars for each image
  • did not duplicate any data
  • did not skip any images
  • subtracted the background using method B

As of now, we need to get the Julian Date by hand from the individual fits headers for each image set and insert it in to Excel.

Data can be graphed, Julian Day vs. Luminosity using Excel

Mean value and standard deviation can be found for each individual star by sorting by image name and then using the formulas function in Excel.

Test Light Curve Images

Mysterious IRAC images, along with a chart indicating 3 stars for which you should perform photometry. Construct light curves by plotting your photometry as a function of time (found in the image headers)

[1] Click here or paste this into browser ftp://anon-ftp.ipac.caltech.edu/outgoing/hoard/nitarp/test_lightcurve_data.tar.gz


Sample IRAC images from each channel for HAT-P-1b, TrES-2, and TrES-4 [2] Click here or paste this into browser ftp://anon-ftp.ipac.caltech.edu/outgoing/hoard/nitarp/sample_images.tar.gz


Reading FITS names

A note on IRAC image file names: Here's a sample image name...

SPITZER_I1_24745472_0002_0000_4_bcd.fits

This follows a general file naming convention, as follows:

"SPITZER" = in case you forgot which satellite you were using

"I1" = Instrument (I for IRAC) and channel number. In this case, channel 1 = 3.6 microns. Other channel possibilities include 2 = 4.5 microns, 3 = 5.8 microns, 4 = 8 microns. Note that, as described in the IRAC Data Handbook and the document on performing IRAC photometry that I circulated amongst you a while ago, these channel wavelengths are not the "true" (or isophotal) values that should be used during data analysis. The isophotal values are channel 1 = 3.544 microns, 2 = 4.479, 3 = 5.710, 4 = 7.844.

"24745472" = Unique AOR identifier for this observation (not for this IMAGE, but for all images comprising a single Spitzer visit to the target).

"0002" = Image sequence number in this observation; in this case, the second image of the sequence.

"0000" = Some observing modes obtain more than one exposure per "image". In such cases, this sequence number would increment upwards.

"4" = number of times this image has been reprocessed through successive (improved) versions of the data processing and calibration pipeline. All of the data I have sent to you (indeed, all of the IRAC data now available in the Spitzer archive) have been processed with the latest and greatest version of the pipeline, S18.7.0.

"bcd" = Basic Calibrated Data. The standard result from raw Spitzer data that have been run through the processing and calibration pipeline. This is the data product that we work with.

".fits" = It's a standard FITS image, can be loaded into DS9. The BCD images are flux calibrated in units of surface brightness per pixel, MJy/sr (mega-Jansky per steradian). The also include world coordinate solutions in their image headers, so can be displayed in the usual format (N up, E to the left) and have, for example, 2MASS catalogs overlaid on them, etc.

-Don --- Dr. Donald W. Hoard, Ph.D. Research Scientist, California Institute of Technology Science Applications Administrator, Infrared Processing and Analysis Center Science User Support Team, Spitzer Science Center

The Truth

Look here for what the light curves really should look like per Don