Delivery Note for Observation - 418/0128530801 (GRB020322)
===============================================================
XFTS Delivery Date to SOC : Wed Apr 23 17:41:17 2003
1 PCMS Processing
-----------------
Release note for pipeline 02000050/20021220.132037
The previous production pipeline (04000010/20020508.121539, prod2)
used SAS 5.3.2, that is xmmsas_20020507_1701.
The new pipeline started operations on 20/12/02. It uses SAS 5.4, that is
xmmsas_20021218_1802 (but see below). The pipeline is prod version
02000050/20021220.132037.
The task ppssumm-2.3.7 will not run under SAS 5.4 for reasons which are not
yet understood, thus ppssumm-2.3.7 from SAS 5.3.4 (xmmsas_20021111_1801) has
been used in its place.
Summary:
--------
This pipeline uses SAS 5.4, and is able to process data from after the late
2002 MOS & RGS cooling.
EPIC PN bad pixel detection is now sensitive to very hard bad pixels. PN event
lists have a better structure for CCD columns with odd offset values. The PN
response kernel and CTI correction for low energies have been updated.
EPIC source detection now includes enhancements resulting from the 1st XMM
catalogue developement work. These include larger background spline map source
cutouts (scut was 0.01, now 0.001), a tighter inter-EPIC source merge radius
(systerr was 4, now 2.5), position rectification using USNO A-2 (based on an
input systematic error of 1.5" and a minimum total error of 0.2", both 1 sigma
radii) with a keyword POSCORROK=Y/N indicating the quality of the
rectification solution (RA/DEC_CORR=RA/DEC and sources are assumed to have
a 1.5" 1 sigma radius systematic error for catalogue correlation if
POSCORROK=N, otherwise a systematic error of 0.5" is used). In addition,
source lists from images for which there was no low background interval are
now excluded from the source merging process, images in bands 1-5 no longer
have out of FOV events (this prevents inclusion of twice the background where
MOS CCDs overlap), and the exposure maps reflect the event flag and column
offset selections. The boxdetect Galactic coordinates are now correct, and the
ML detections have improved positions near CCD edges, a reduced tendency to
large negative count rates at very low fluxes, correct errors at low flux, and
now include more information in the source lists.
The OM source detection makes use of a much improved algorithm, and now has
improved detection and photometry. Mosaiced OM images now average over small
image gaps.
There are 2 new RGS product types: a FITS timeseries which aims to reflect the
particle background rate using the CCD nearest the beam (*RGS*FBKTSR*), and a
total (source plus background) FITS spectrum (*SBSPEC*) to allow the
background to be better dealt with (eg in xspec). The RGS spectral plots have
fixes for wavelength scale and mean value errors.
The observation summary html product now shows the proposed duration and the
actual duration of the observation. Image plots are made by a substantially
enhanced routine, and contain more information. The 4-character exposure
identifier and the attitude file type used are now recorded. As usual, there
have been many bug fixes and improvements made to increase the efficiency of
the software.
In this pipeline, for the first time, products are made which are not delivered
to the SOC. These are experimental, and require large scale evaluation. The
products are: EPIC simultaneous all-exposure source detection lists, OM USNO
A-2 position rectified source lists.
2 Data Product Screening Report
-------------------------------
XMM SSC data-product screening log
Version: 23-Apr-03
Revolution 418
Obs Seq. 0128530801
Target GRB020322
Seq_id 006389
** GENERAL COMMENTS
Last observation in revolution
According to VILSPA observation log: "GRB 020322 exposures performed with ID=0110980301.
Poinging at 18:00:49.4 +81:06:10.8
Exposures started manually due to problems in generation of new ToO POS on time."
OM partly with filter Grism1 or Grism2 or Magnifier
Medium bright source in center of EPIC images
Problems:
EPIC source detection
OM source detection
ROSAT overlay
EPIC comments refer, in general, to Energy-band 8.
** EPIC MOS1
Exposure U003:
PrimeFullWindow with Thin1 filter (duration 28436.6 s; 28379.0 s in image).
No flaring activity
High background not totally screened in images
Noisy area in external CCD in band 1
Bright pixels in outer corner of external CCD
Bright pixels in bands 1,8
Spurious detection(s) due to obvious bright pixel(s)
Spurious detection at inside corner (of CCD 4)
** EPIC MOS2
Exposure U003:
PrimeFullWindow with Thin1 filter (duration 28441.7 s; 28429.7 s in image).
No flaring activity
High background not totally screened in images
Suspected spurious detections due to warm pixels
** EPIC PN
Exposure U003:
PrimeFullWindow with Thin1 filter (duration 26304.2 s; 22943.9 s in image).
Flaring activity affects less than half of exposure time
Bright segments of columns in band 2
Bright pixels in bands 1,8
Spurious detection(s) due to obvious bright pixel(s)
Spurious detection(s) found in the wings of the bright source
** RGS1, RGS2
RGS1:
CCD 7 missing
Spatial map: extraction region NOT in center
(no spectrum visible)
Orders image: No spectrum visible within 1st and 2nd order
Spectrum plots: Noisy spectra
RGS2:
CCD 4 missing
Comments to spectra as RGS1
** OM
3 full frame sky image exposures (Filters UVW1,UVM2,UVW2 ; 4000,3500,4000 s duration each)
Scattered light central enhancement
Scattered light features
Reduced counts in some corners
small `wedges' in image corners
Source detection problems:
Several sources missing
** Catalogue cross-correlation products
No identifications for many XMM sources
Problem with cut values in ROSAT overlay
ROSAT image with very short exposure
Bright EPIC source not detected by ROSAT