I. Status of the Data Archive
Data release to the US ROSAT Data Archive has been continuing. As of this writing, the US data archive contains data released by the US through August 1995 as well as data released by MPE through June 1995. A major effort is now underway to automate the release of data to the public archive, which should ensure more timely release and availability of ROSAT data.
II. Release of Reprocessed Data
Observations made in the early part of the ROSAT mission are presently being re-processed with the most up-to-date version of SASS and redistributed to the principal investigators. The PIs retain data rights to the reprocessed data for a period of 6 months, after which time the data are released to the ROSAT archive. The first set of reprocessed data will become publicly available sometime in the fall of 1995. Archive users should note two important points:
- the reprocessed data will REPLACE the earlier versions of the processing.
This means that the original (REV0) version of the dataset will no longer be
readily available from the ROSAT archive.
- the reprocessed data will enter the archive in RDF FITS format, not in the REV0 fits format previously used, independent of whether the data were reprocessed at MPE or GSFC.
At the present rate of data reprocessing, it is expected that reprocessing will be completed by March 1996.
III. The ROSAT RESULTS Archive
Probably the most important development to impact ROSAT archive users is construction of the ROSAT Results Archive, which began this spring. As users of the data archive know, certain ROSAT data products ("results") were deemed potentially unreliable and were withheld from public distribution pending reprocessing and data screening. "Results" include background maps, extracted source lists, source spectra and source lightcurves. The RESULTS ARCHIVE is now being developed to make these "results" available in a sensible way to the general public so as to allow access to a reliable source list and to alert the archive user to potential problems in derived counting rates, spectra, etc. The ROSAT RESULTS ARCHIVE is being constructed by joint effort of the data centers in Germany, the UK and the US. Each site will be involved in the flagging of the data and releasing the data to the public.
B. Phase 1 - Results Flagging and the Q-SOURCE file
In this initial stage (Phase 1), most of the effort is being directed toward construction of a reliable source list. The source list we produce will represent a summary of detected source characteristics for ROSAT pointings with the PSPC and HRI. Similar lists for the PSPC have been recently released (the "WGACAT" list produced by White, Giommi and Angelini and the "ROSATSRC" list produced by Voges et al. at MPE). The aim of the ROSAT RESULTS ARCHIVE effort is to produce a reliable list, i.e., a list in which each source has an indicationas to whether the derived source characteristics (position, brightness, etc.) are deemed reliable or not.
To do this, the source lists produced by SASS processing are screened by automatic and visual methods and obvious problems are flagged. The SASS source lists, along with the flags, are copied to a "quality-screened" source file which is then released to the public. We emphasize that this "quality-screened" source file (sometimes called the Q-SOURCE file) contains ALL the original SASS data in addition to quality flags and other useful information generated as part of the data flagging. The FITS version of the Q-SOURCE file is named according to the following convention:
where <inst> = p, f, or h is the instrument, <ror> is the 6-digit ROSAT observation request number, and <pointing> the 3 character pointing extension (N00, A01, etc.) Users will have access to the Q-SOURCE file from the US archive via the usual methods (anonymous ftp from legacy.gsfc.nasa.gov or by e-mail request to the NDADS ARMS server).
Data flags are divided into 2 types: "Sequence Flags" which pertain to an entire sequence and "Source Flags" which apply to individual sources in a data sequence.
Sequence Flags are logical flags which indicate to a user whether any of a number of conditions are thought to exist for a given sequence. Sequence flags are stored as logical header keywords in the first binary table extension header in the Q-SOURCE FITS file. Table 1 lists the keyword names and the condition which triggers a "True" value. If a condition is not thought to exist for a given sequence, the keyword value is set to "False".
Table 1. Sequence Flags KEYWORD NAME "True" condition ------------ ----------------Source Flags are logical flags which indicate to a user whether any of a number of conditions are thought to exist for a given source in a given sequence. Each source has a set of flagged quantities. Source flags are stored as logical columns in the first binary table extension header in the Q-SOURCE FITS file. Table 2 lists the column names and the conditions which trigger a "True" value. If a condition is not thought to exist for a given source, the column value is set to "False".
ASP_SUSP obvious aspect problems (sources appear multiple or have obvious ellipticity) EXT_EMIS presence of extended emission over most of the field FALS_DET one or more detected sources deemed false DEFERRED data screening not done MISS_SRC one or more obvious sources not detected by SASS BKG_SUSP Field background may be incorrect (due to inclusion of high background intervals or extended emission) NO_SRC source list is missing (no sources detected by SASS or source detection stopped)
Table 2. Source FlagsAs part of the HRI data flagging, images with smaller pixel scales (5 arcsec/ pixel and 2 arcsec/pixel vs. 8 arcsec/pixel image produced by SASS) are generated. These images better sample the HRI point-response function and therefore offer potentially more spatial information than the standard SASS image. Presently we plan to release these images to the public HRI archive so as to allow users access to these images as well as the standard SASS image. In addition to the above products, a list of obvious sources which were not detected by SASS will also be compiled.
BINTABLE COLUMN NAME "True" condition ----------- ----------------
CONFUSED close sources confused detect DET_STRUCT source near detector structure EXTENDED source appears extended FALSE_DET false detection INTEN_SUSP source brightness suspect LTC_SUSP source lightcurve suspect MULT_DET complex emission split into individual sources NOT_CHECKED source not checked POS_SUSP source position suspect ASSOCIATED another source nearby may influence derived source quantities VARFLG_SUSP source variability flag suspect WITHIN_EXT source within extended emission UN_UNIQUE source matched with another source which was identified in a smaller detect cell (HRI only)
C. Phase 2 and beyond
Additional data products and software tools are needed if users are to take full advantage of the wealth of data in the ROSAT archive. These products and tools include:
1) the merging of add-on and follow-on observations to maximize mean exposure
in each field;
2) unique identification of each detected source and correlating with known sources from other catalogues;
3) extraction of upper limits from pointed data;
4) extraction of information for sources previously not detected by SASS;
5) generation of light curves for all detected sources found in the HRI.
These products and tools will be generated during phase 2 of the RRA
Present plans call for phase 2 to start some months after phase 1.