AXAF

The Advanced X-ray Astrophysics Facility (AXAF) is NASA's premier X-ray astronomy facility for the coming decade, with the very highest recommendations of both the Field and Bahcall National Academy of Sciences decadal reviews. It is currently scheduled for launch in late 1998, may operate for a substantial portion of a decade, and will generate data that will doubtless remain a resource for several future decades. Although multiple, interchangeable imaging and spectroscopic detectors can occupy the focal plane, all have small angular fields of view, and are fed by a precision-pointed, rather than scanning, telescope, designed to take advantage of superb quality X-ray focusing optics. For all of these reasons, one traditionally does not consider AXAF a "survey" instrument, and thus connections to SDSS might not be obvious at first glance. However, AXAF's tremendous sensitivity in fact implies that huge numbers of previously uncataloged X-ray sources will be discovered during the lifetime of the project. As pointed out in the discussion of ROSAT, many aspects of these sources can be elucidated only if an optical counterpart can be identified, and without these identifications, a significant segment of AXAF science will not be exploited. Here we briefly provide some quantitative data on this issue, and demonstrate that there is a good match between AXAF and SDSS in the area of optical identifications. We are grateful to H. Tananbaum and F. Seward of the Center for Astrophysics for much of the following analysis.

Each year of AXAF observing time will contain roughly 10³ scientific observations of 2x10⁴ s duration. A reasonable assumption is that 50% of the time will be used for spectroscopy, 33% for imaging with the AXAF CCD Imaging Spectrometer (ACIS), and 17% for imaging with the High Resolution Camera (HRC). This implies that imaging will consist of 330 ACIS fields ( 15'x15' ), totaling 20 deg² yr^-1 , and 170 HRC fields ( 30'x30' ), covering 40 deg² yr^-1 . In a 10 yr lifetime, about 600 deg² will be imaged. If we conservatively adopt the off-axis sensitivities, the limiting fluxes, while variable across the field, will be of order 3x10^-14 erg cm^-2 s^-1 , about 20 times fainter than the ROSAT All-Sky Survey.

Although these AXAF images cover <2% of the celestial sphere, their extraordinary X-ray sensitivity implies the detection of a huge number of serendipitous sources, unrelated to, but contained within, the fields of each intended target. Point sources will as usual consist of a mixture of galactic and extragalactic; for simplicity here we discuss only QSOs/AGN, which will probably predominate at high galactic latitude (e.g., in the SDSS survey region) at these flux levels. The "typical" QSO has log(f_x/f_v)~0 (Stocke et al. 1991), implying V=20.4 for objects at the limiting X-ray flux (where most of the objects are expected, given the steep log N/ log S curve). There are about 80 QSOs deg^-2 at this magnitude, or 5x10⁴ discovered by AXAF in its lifetime; >10⁴ of these will fall within the SDSS survey region, and almost all of those will be readily identifiable, regardless of redshift, due to their distinctive colors in SDSS 5-color space. A large fraction will have SDSS spectra already on hand, before any correlation between AXAF and SDSS data is even attempted.

Perhaps more interesting are the analogous values for QSOs with more extreme values of X-ray to optical luminosities. The Einstein Observatory uncovered a non-negligible number of QSOs with log(f_x/f_v)~1, implying V~23 at the AXAF X-ray flux limit. Note that this is 3 magnitudes fainter than the faint QSOs in the ROSAT All Sky Survey, posing a severe challenge indeed for optical identifications. The surface density of QSOs at this faint magnitude is still uncertain, but virtually all estimates place the value at >= 250 deg^-2 , implying 4x10⁴ potentially X-ray detectable QSOs in the joint AXAF/SDSS region. Even at this faint level, a huge number of objects will stand out in the SDSS photometric data base as highly certain identifications, easily extracted in an automated, totally homogeneous manner. Although these are of course too faint for SDSS spectra, spectroscopic followup of a limited subset with 4-6m class telescopes, when coupled with the excellent statistics of the large sample, should yield accurate and unprecedented data on the X-ray properties of very faint QSOs.

References

Stocke, J.T., Morris, S.L., Gioia, I., Maccacaro, T., Schild, R., Wolter, A., Fleming, T.A., & Henry, J.P. 1991, ApJSuppl 76, 813.