The X-ray astrophysics community is currently analyzing ASCA data with plasma spectral models that are inadequate for the task. The situation for AXAF non-grating spectra can be significantly improved, but only with concerted efforts. Major efforts are now underway to produce plasma spectral models which will be far more accurate and comprehensive than previous models. What the astrophysics community does not yet recognize is the fact that the atomic data compiled for the plasma spectral models even from the most recent sources is entirely theoretical, with little to no experimental verification.
The situation is drastic, perhaps especially for CCD-resolution spectral interpretation (see Section 4). Despite tremendous effort and much progress, spectra obtained with AXAF for supernova remnants, clusters, cooling flows, and galaxies will be analyzed with models that we have little confidence in. Much greater confidence will be gained if we can benchmark the plasma codes with very high signal-to-noise grating data for a small set of stellar coronae.
Just three stellar grating calibration targets -- Procyon, Capella, and HR 1099 -- will do an excellent job of providing spectra from most of the ionization states that emit X-rays in collisionally ionized plasmas (see Section 5). We are planning to use the calibration targets to begin the ``X-ray Emission Line Project'' (XELP), a database of observed spectral lines consisting of identifications, line intensities and comparisons to standard models.
Our chief concern, and a matter of some time urgency, is that the exposures we need may be longer than the required exposures for instrument calibration purposes alone.1 We recognize that this project is outside the scope of what is traditionally considered to be ``calibration;'' however, we believe that ``calibrating the models'' falls within the scope of the AXAF Science Center's role to maximize the scientific return of AXAF. Observational validation of widely used atomic models is a basic, necessary, and feasible step to take before astrophysics can be done with any confidence. What we are proposing is that the ASC lead the community in carrying out this step.
If these data sets are taken as part of the in-flight calibration, then they would be made immediately public. We will make the results available in stages as we produce them (e.g. beginning with a list of identified strong lines), such as through a website, as well as producing a publication quality catalog. Other groups could also use the public data sets to assess the atomic physics models.
Dedicating a total of 550 ks (see Section 6) soon after launch for this project may have large implications for public relations with the instrument teams, the Science Working Group, and the Guest Observers. Given a clearly articulated motivation -- and an immediate release policy -- this could demonstrate ASC leadership. The immediate release policy of the recent Hubble Deep Field observation contributed to fruitful collaborative (and competitive) efforts unimaginable if the data had been proprietary. We feel that our proposed project is important enough to deserve serious consideration by the AXAF Science Center.