Subject: Transforming GRIM photometry to the standard sytem

From: rhoads@astro.Princeton.EDU

Submitted: Wed, 29 May 1996 02:28:02 -0400 (EDT)

Message number: 15 (previous: 14, next: 16 up: Index)

I've spent the past couple days trying to calibrate some GRIM photometry
to published standard star magnitudes, and thought I would summarize my
results here, both in hopes that they may be useful to others and that
others who have done this exercise might comment on consistency of the
results.  Please note that is my first serious attempt at this game, and
is based on relatively few standard star observations. 

<H3> Summary of results: </H3>
For the night of 1996 April 29 (UT), I found the following based on
observations of 5 standard stars in K', 3 in J, and 4 in 2.22 and 2.36
micron CO index filters:

  observed  mK' = K - (22.209 +- 0.006)
or          mK' = K - (22.295 +- 0.017) + (0.061 +- 0.008) * X
or          mK' = K - (22.284 +- 0.015) - (0.022 +- 0.017)*(J-K)
			+ (0.059 +- 0.007) * X
	    mJ  = J - (22.985 +- 0.005)
or          mJ  = J - (23.086 +- 0.009) + (0.068 +- 0.004) * X
            m(2.36)-m(2.22) = CO + (0.309 +- 0.005)
or          m(2.36)-m(2.22) = CO + (0.259 +- 0.013) + (0.037 +- 0.006) * X

Here K, J, and CO are tabulated in Elias et al (1982, AJ 87:1029), 
mJ, mK', etc are instrumental magnitudes defined as -2.5*log10(ADU/second),
and X is the airmass of the observation.  The counts are total counts
for the star, measured in a big aperture.  To get good agreement between
counts from requested 1.5 and 3 second exposures, I found I had to
assume true exposure times of (1.5 - 0.185) and (3 - 0.185) seconds.

For 1995 December 11, a K band zero point based on one Elias standard gives
	    mK' = K - (22.191 +- 0.007)
and a CO index calibration gives
	    m(2.36)-m(2.22) = CO + (0.373 +- 0.010)   .
I also took UKIRT standard observations on 951211, which are
consistent with the above K band zero point at about the 2% level. 
(The faintest of three observed UKIRT standards, at K = 14, was
discrepant at more like 10%.)

So, the zero point was pretty consistent night to night in K' band, but
appears not to have been in the CO index; I am not sure if that is indicative
of true variations or of a problem with my calibration.  Any feedback on
whether these coefficients look reasonable would be appreciated.
I also got some rather larger color and extinction coefficients, of order 0.2, 
for 951211 (on the basis of a less extensive color and airmass range than for
960429).  I don't think they are right.

The airmass corrections for 960429 are in essence based on a single standard
observed at airmass ~ 3.8, and the K band color correction is similarly
based on a single red standard (BD+3_2954).  That star was not observed
in J so no color term for mJ could be established reliably.
The photometry was done in large software apertures around defocused
images of bright standards.  I didn't do a very careful job of aperture
corrections and estimate ~ 1% errors in stellar photometry from that. 
I assumed 2% overall photometric errors for all stars, based on scatter
among different exposures for the same star and filter. 
The coefficients were fit using the IRAF PHOTCAL package.

	<H3> Comments on the literature, finding charts, and other tools: </H3>
I get my standard stars from the lists of Elias et al (AJ 87:1029, 1982,
which defined the CIT standard system) and Casali & Hawarden (UKIRT/JCMT
newsletter, August 1992, p. 33, which defines the UKIRT system and gives
transformations from UKIRT to CIT magnitudes).  I generally defocus the
telescope by 300 to 500 focus steps before observing the Elias standards
in broadband filters; otherwise they saturate the chip even in < 2 second
exposures.  The UKIRT standards don't require defocusing.  Finding
charts for Elias standards are rarely an issue (the standard is the
brightest star around), though proper motions matter in some cases.  

	I have written an MC Tcl script that will point the 3.5 meter to
any of the Elias faint standards and take a 5 exposure sequence in a cross
pattern; interested people may look at the source code which is
available on in /export/apotop/visitor1/peyton/dostan.tcl .
The coordinates in that script have been modified for 44 years of proper
motion (i.e. to 1994) in the handful of largest proper motion stars in
the catalog.

	For UKIRT standards, a handy source of many finding charts is
Landolt 1992, AJ 104:340.  I was able to find charts for UKIRT faint
standards numbers 3, 5-10, 12-14, 24-26, 29, 32, and 35, and there may be
others (I did not do an exhaustive search).  That article is available
through the ADS Article server, too, so you can get postscript versions of
these charts online.  (Cf. .)

	K' band magnitudes for some Elias standards, plus a
transformation from K' to CIT K magnitude, are given in Wainscoat and
Cowie 1992 (AJ 103:332). 

	I would be interested to hear about other reliable lists of
near-infrared standard stars, and in particular about any fields
containing a few reasonably well separated stars with a range of colors
and well measured photometry on a standard system.
						    -James Rhoads
APO APO APO APO APO  Apache Point Observatory 3.5m  APO APO APO
APO  This is message 15 in the apo35-grim archive. You can find
APO  the archive on
APO  To join/leave the list, send mail to
APO  To post a message, mail it to