ROSAT Status Report #94

[REQUEST@rosat.gsfc.nasa.gov]

                ROSAT/HEASARC Status Report No 94
			May 19th 1994


Important Changes to Data Set Naming Conventions
------------------------------------------------

Data processed before Jan 1993 was processed with the initial release
of SASS, which is known as Rev 0 processing.  After Jan 1993, a new
release of SASS was used to process data; this is known as Rev 1
processing.  Some improvements to SASS and the implementation of the
Rationalized Data Format (RDF) for the output data heralded the start
of Rev 2 processing in April 1994.  One significant difference between
the initial (Rev 0) processing and the later processings is the
handling of follow-on observations.  This change has necessitated an
expansion of the file naming convention from the relatively simple
scheme used during Rev 0 processing.

The observation naming scheme used during Rev 0 processing used a 2
letter code + 6 digit ROSAT Observation Request (ROR) number to
identification a given observation sequence.  For example a typical
sequence id would be RP123456 (for a US-processed PSPC observation) or
WF123456 (for an MPE-processed filtered PSPC observation).  As
additional data for this observation were obtained, they were combined
with previously obtained data and sent to the user (and the archives)
under the original 2 letter + 6 digit observation ID.

Starting with Rev 1 processing and continuing through Rev 2 processing,
there are different naming conventions for a given observation, based
on whether it's 1) an initial pointing, 2) an add-on or follow-up
observation, or 3) a mis-pointing. Archive users should understand
something about the change in naming convention in order to retrieve 
data from the archive.
 
The Rev 1 file naming convention is as follows:
		1) initial pointing or mis-pointing: 
			00 added to the 6 digit ROR number
			Eg: RP12345600
		2) add-on or follow-up observation:
			01...99 added to the 6 digit ROR number
			(depending on the number of add-ons)
			Eg: RP12345601

The Rev 2 file naming convention is as follows:
		1) initial ("normal") pointing:
			NOO added to the 6 digit ROR number
			Eg: RP123456N00
		2) add-on or follow-up observation:
			A01...A99 added to the 6 digit ROR number
			(depending on the number of add-ons)
			Eg: RP123456A01
		3) mis-pointing:
			M01...M99 added to the 6 digit ROR number
			(depending on the number of mis-points)
			Eg: RP123456M01
			
The Rev 2 convention shown above, which is the most clear, was not put
into effect for Rev 1 datasets. Because of the potential confusion
between normal observations and mis-points caused by the Rev 1 naming
convention, it was decided that BOTH REV 1 AND REV 2 DATA WILL BE
ARCHIVED USING THE REV 2 NAMING CONVENTION.  For example, a data set
which was processed by Rev 1 SASS and distributed to the PI with the
dataset ID RP12345601 will enter the archive as RP123456A01.


How will this effect archive users?  

1)  Unlike the Rev 0 data sets in which the entire observation was
"packaged" under one observation id (RP123456, say),  each individual
"pointing" is packaged under an individual name for Rev 1 and Rev 2
processed data.  For example, suppose there is an observation of the
galactic center having an ROR of 123456, which is made in 2 pointings
(initial pointing + 1 add-on).  This data will enter the archive as 2
discrete data sets, RP123456N00 and RP123456A01.  To get the entire
observation, the archive user will need to retrieve both datasets
(RP123456N00 and RP123456A01).

2) Because the Rev 1 and Rev 2 data sets have different output file
formats (the Rev 2 uses the RDF format, while the Rev1 did not), a
given observation ID may be archived as a Rev 1 processed data (in the
old-style, non-RDF format) at one time and then later (after
reprocessing) the data set will be replaced by an RDF version.

We regret the inconvenience caused by the change in naming scheme, but
point out that as data are re-processed, all REV 0 processed-data
will be replaced by REV 2 versions.  This means that eventually all
data in the ROSAT archive will follow the REV 2 naming convention.

Questions about data in the US Public Archive should be sent to
Dr. M. Corcoran at the address below.


Special Note About MPE-Processed Hri Archival Data
--------------------------------------------------

HRI FITS-formatted datasets produced by MPE have filenames ending in
".mt" (rather than ".tfits" or ".ifits" which is used for the PSPC data
produced by MPE).  Users should rest assured that these data are in the
standard MPE EXSAS FITS format similar to that used for the PSPC, only
the extension name has been changed (to protect the innocent?).
Previous versions of the user guide "archive_intro.txt" (available from
the legacy.gsfc.nasa.gov anonymous ftp account) were written before we
received any HRI datasets from MPE and were under the (mis)conception
that the ".ifits" and ".tfits" file extensions were to be used for both
PSPC and HRI data.  This was not the case, so the recent version
of the archive user's guide (available in rosat/doc/archive on
legacy.gsfc.nasa.gov) has been corrected to reflect this reality.


ROSAT CD Rom Volume II
----------------------

A sneak preview of volume II of the ROSAT Images CD ROM can be found
on the WWW under

http://heasarc.gsfc.nasa.gov/docs/rosat/cdroms/rosatcdroms.html

This CD ROM, produced by the HEASARC and US ROSAT Data Center, will be
distributed at the Minneapolis AAS meeting. Copies of Volume I and II can be
obtained by sending your mail address to ksmale@lheavx.gsfc.nasa.gov


ROSAT Catalog
-------------

The US ROSAT Data Center in collaboration with the HEASARC is currently
working on rationalizing the various ROSAT database tables into a 
single observation catalog. The current situation of multiple tables
with inconsistent information is confusing and results from the fact
that we receive the information from different sources (MPE, University
of Leicester, and the US processing team). By combining all these
tables into a single database we hope to provide a much cleaner
interface to gather information on the status of a particular 
observation and the resulting data. This is a major effort, that
will take several months. We hope to have it complete in the July/August
1994 timeframe. Until then, please be patient and tell us of
any inconsistancies you find in the current tables.  Comments,
questions or complaints should be sent to:


		Dr. Mike Corcoran
		US ROSAT PROJECT/HEASARC
		Goddard Space Flight Center
		Greenbelt, MD 20771
		corcoran@barnegat.gsfc.nasa.gov
		HEASRC::CORCORAN