Mid-Infrared Flux Standard Stars

Because of the high thermal background in the 8-25 micron wavelength range and the resultant relatively low sensitivity of ground-based mid-IR instruments, mid-IR observations are flux calibrated using a small number of bright standard stars. The most extensive effort to bring a set of infrared standards into a unified calibration scheme has been led by Martin Cohen. Cohen's basic product is a continuous spectrum of each star that is carefully tied to observational data. The spectra can be used to flux-calibrate both the T-ReCS imaging and low-resolution spectral data by smoothing to the appropriate resolution or integrating over the filter bandpass. (The calibration spectra are of relatively low spectral resolution, so further effort is needed to establish reliable calibration methods for R ~ 1000 spectroscopy.) Gemini recommends adopting Cohen's standards and calibration methods for mid-infrared observations.

Cohen's work is described in a series of papers beginning in 1992. The paper which establishes a large set of standards scattered over the sky, "Paper X", Cohen et al. 1999, AJ 117, 1864 , includes a summary of the previous papers. The numerical spectral data from the papers were tabulated in the AAS CDROM series. Cohen's primary and secondary standards (see below) along with the set of standards in Paper X are the main calibration stars which are used for T-ReCS (and Michelle) observations.

Cohen's calibration is based on two primary standards: Alpha Lyr (A0 V) and Alpha CMa (A1 V). Alpha Lyr is defined to be 0 mag between 1 and 20 microns but at longer wavelengths allowance must be made for emission from dust. For this reason Cohen favors Alpha CMa as the primary standard over all wavelengths. Kurucz model atmospheres fitted to observational data of these two stars define the absolute flux calibration (Cohen et al. 1992, AJ, 104, 1650, Paper I in the series).

Cohen extends his system to ten bright secondary standards which are tied to the primaries via well-documented ratio spectra obtained from several ground-based, airborne, and space-based platforms (Papers II, IV, VI, and VII). The secondary standards are K and M giants which exhibit broad CO and SiO absorption features in the M band and near 8 microns, so approximating their spectra with blackbodies would introduce calibration errors of 10-20%. Using Cohen's absolute spectra, errors should be under a few percent.

In Paper X (Cohen et al. 1999, AJ 117, 1864), spectral templates for 422 stars are presented, potentially defining a dense all-sky network. The templates can be obtained from the electronic version of the paper. However, many of these templates have not been verified observationally, so these stars must be used with caution.

Many other 8-20 calibration systems exist. For a list of references, see Chapter 7, "Infrared Astronomy," by Alan Tokunaga, in the recently published "Allen's Astrophysical Quantities, 4th ed." (ed. A. Cox, Springer-Verlag:New York, 2000). Two useful summaries are Tokunaga 1984, AJ, 89, 172, and Hanner & Tokunaga 1991, in "Comets in the Post Halley Era" (eds. R.L. Newburn, Jr. et al., Kluwer) p. 67.

To help with the planning of observations, the basic properties of Cohen's 2 primary and 10 secondary standards are listed in the following table. Also listed are a few other stars which are often used as mid-IR standards. Most of the K and N magnitudes are taken from papers I (1992, AJ, 104, 1650) and IV (1995, AJ, 110, 275). In Cohen's system, the 0 magnitude flux density is 655 Jy at K and 35.21 Jy at N.

In addition to using these stars as flux calibration standards, some of them can be used to calibrate the PSF. Note, however, that it is not certain that any star which can be used as a flux calibration standard can also be used as a PSF standard, and Gamma Cru is an example of this. Those of the Cohen standard stars of spectral type K or earlier are likely to be suitable as PSF standards. Those of M spectral type are suspect for use as PSF standards and may possibly exhibit variability in the mid-infrared.

Notes to table:
(a) Beta Peg is variable at about the 10% level but is useful for spectral calibration.
(b) Gamma Aql appears in Cohen et al. Paper X as one of the 422 stars with templates.
(c) Gamma Crux may be used as a flux calibration star, but it definitely is not useful as a PSF calibration star since the N-band PSF is found to be 50% larger than those of other standards taken at nearly the same time.