Any satellite that "lives" a long time ages. This means that software used to process data has to evolve in order to deal with new "features" that develop. SASS (Standard Analysis Software System) is the MPE supplied software (including the SAO-developed HRI software) that converts level 0 spacecraft data to the data products that the PI receives. There have been seven SASS releases since the last newsletter and they include fixes for problems ranging from the spacecraft clock reset to being able to deal with ROSAT Days (RODAYS) beyond 2000. The other major activity occupying both the US and German (MPE) data centers is porting SASS and all the other support software to run on the Alpha class of DEC computers. This will enable processing to run faster and be less expensive to maintain.
The US ROSAT Science Data Center (USRSDC) has finished processing and distributing nearly all the AO5 data and is now in the midst of AO6. Note that there still might be some AO5 data sets that have to be processed by hand due to minor observation problems. At present all AO6 data that is at the USRSDC has been processed and hence there is no backlog. Currently observers can expect to get their data 6-8 weeks after the scheduled observation. It takes approximately 3-4 weeks to receive the data from the German Science Operations Center (GSOC) and a few weeks for MPE to pre-process and determine any observation interval corrections (determining which OBIs belong to which sequence for the aficionados in the audience). Finally the data are processed and distributed for US PIs by the USRSDC. These times include shipping delays, computer down times and any other delays inherent in the system. To check the status of your data, information on processing will soon be available on the US ROSAT Science Data Center (USRSDC) home page via the Web. Users will also be able to reach this page from the HEASARC and the ROSAT GOF pages.
Some PIs have been receiving reprocessed AO data. The original reason for the reprocessing was to fix bugs in the older versions of SASS and add improved data products. Some of the fixes include correction of the photon timing problem, better background determination, better source detection, better attitude determination through the use of the WFC star-tracker in the attitude solution, and better data products such as SIMBAD source correlations for the HRI. The other fundamental reason for the reprocessing is to provide a common data format for the public archive. The Rationalized Data Format (RDF) was developed jointly by the USRSDC and MPE (see ROSAT Newsletter Number 10, pg. 14 for more information) and is a true FITS format. The first RDF release was RDF 0.6 on May 1993 and after many iterations RDF 2.4 was used for the beginning of the reprocessing on May 1994. The ROSAT Results Archive (RRA) (see issue Number 11, pg. 11 for more information) came into being shortly thereafter and further requirements were imposed on the RDF such as including a light curve and adding various FITS keywords that help researchers make better use of the archival data. This resulted in further RDF releases. To date there have been 28 RDF releases and currently reprocessing is using RDF 3.4. This RDF has been agreed upon by both the USRSDC and MPE to be the final archival version.
Different SASS and RDF versions exist, and now to further confuse everyone there are different processing and reprocessing versions or revisions. REV0 processing refers to ROSAT DAY (RODAY) 28 (28 June 1990) through RODAY 852 (18 September 1992) excluding the survey. REV1 starts with RODAY 853 and goes through 1318 (21 December 1993). REV2 of course starts with RODAY 1319 and keeps on going. As a PI, all these reprocessings will only affect data that were REV0. If your observation was prior to RODAY 853 then you will get a new data set in the mail that was processed using REV2 and these new REV2 data will have a 6 month proprietary time period. In general most observations will not be that greatly affected. All later observations will be reprocessed for the purpose creating a uniform archive and these data will not be distributed to PIs since the difference between REV1 and REV2 is primarily a matter of formats.
There are two parts to the reprocessing: running the REV2 version of SASS and reformatting the data using RDF 3.4. The latest SASS (version 7_8) with RDF 3.4 will insure that data will be consistent on both sides of the Atlantic and can be used by the staff to populate the RRA. The USRSDC has been running 3 processing streams. We are reprocessing REV0 data to REV2/RDF3.4 and are currently up to RODAY 813. These data are being distributed to PIs. As part of this stream we will also reprocess REV1 data (RODAY 853-1318) but this will NOT be distributed to PIs. The second stream is a reformatting (applying RDF 3.4) of REV2 data that doesn't have the latest RDF format. This consists of RODAYS 1319-1780 and about 120 RODAYS have been done to date. The third stream is the processing of new incoming data that is processed and distributed to PIs with the standard proprietary time period in place. This should make everything clear!
In summary, AO6 data are flowing smoothly while reprocessing is being done to achieve a uniform archive and most importantly to extract data into a format that can be easily accessed by users. In the future if there are problems (with the gain, for example) then the user will probably be able to run project-provided software that will correct this problem. This will be possible because required data such as "raw coordinates" are now accessible in RDF 3.4. This is designed to alleviate the need for future reprocessings.
Select another edition of the ROSAT Newsletter
Please use the Feedback link if you have questions on ROSAT.