The fifth reprocessing campaign started on 9/22/20 and will include all observations (~25,000 observations with
level 2 event files) taken from the beginning of the mission (July 1999) up to current data
(July 2020). We will reprocess ~900 observations during the first six months of the mission,
which were excluded in the previous reprocessing. All observations will be reprocessed from
the raw telemetry data.
There are three data sets, based on the observation date, the previous reprocessing
status and the applied improvements in REPRO-V.
- Period A: Early observations that were taken before Jan. 29, 2000.
These early data have not been reprocessed in the previous major reprocessing campaign (Repro IV).
- Period B: Data reprocessed in Repro IV (2000-2011).
While these observations have been processed with Repro IV, the improved calibration files for ACIS, HRC, and ASPECT will be applied.
- Period C: Recent observations that were taken in 2012-2020.
These observations have not been processed with Repro IV. In addition to the improved calibration
files, the interpolated spacecraft clock correlation files, definitive ephemeris, and the proper
time-dependent ACIS and HRC gains will be applied.
We plan to process roughly in order of reverse observation time, i.e., C, B, and A, so that the new
(recent 3 years) data are reprocessed first and the early 1999 data at the end.
Number of Observations per Repro V period
|1999-07-20 to 2000-01-29
||2000-01-30 to 2011-12-31
||2011-12-31 to 2020-07-20
||1999-07-20 to 2020-07-20
* CTI (Charge transfer inefficiency) ER (engineering request): ACIS data taken
for a calibration purpose with no astronomical target.
- ACIS detector gain improvement
A new Phase2 Response (DET-GAIN) was implemented in CalDB 4.9.1 (March 24 2020)
to fix the small gain depression (1-3%) at mid-chip (around chipx=512) of the
front-illuminated (FI) CCDs. This improves most ACIS-I observations
(except graded-mode observations). Note that the back-illuminated (BI) CCDs
(CCDID = 5 and 7) are not affected by this gain droop. More information can be
found in the
CalDB 4.9.1 release notes
Chandra calibration page
- ACIS time-dependent gain application
ACIS T_GAIN is regularly updated every three months from February 2012 to November 2015, upgraded every
six months thereafter. All ACIS observations (2012-2020) since the last repro will be improved.
The background information can be found in the 'CIAO why' page of
ACIS time-dependent gain.
- A full upgrade to the ACIS T_GAIN library.
In addition to the incremental upgrade, the entire T_GAIN files for all -120C epochs
(from January 29 2000) were upgraded with an improved statistical averaging technique
for data with decreasing counting statistics. Also, T_GAIN files were recompiled with
(CHIPX x CHIPY) of (32 x 128), which reduced the file size significantly without compromising
T_GAIN correction detailing . More information can be found in the
CalDB 4.9.2 release notes.
- ACIS bad pixel library upgrade
The Framestore shadow regions at the bottom of each chip are added to ACIS bad pixel
(BADPIX) CalDB library. It was implemented in CalDB 4.9.0 (December 17 2019) and the corresponding
s/w was fixed in DS 10.8.3 (March 24 2020). This applies to all ACIS chips except in CC-mode
observations for the data taken from the beginning of the mission. More information can be
found in the CalDB 4.9.0 release notes.
- ACIS header keyword updates
Unnecessary ACIS header keywords (CCD_ID and FEP_ID) were removed from the ACIS event file. It
was implemented in DS 10.8 (June 18, 2019).
- HRC-I gain map application
For the time-dependent HRC-I gain correction, HRC-I GMAP (Gain Map) is regularly updated every year
to compensate for the decline in HRC-I gain, typically a few % per year. This affects the last
quarter of each year where there is a 3-month lag between the calibration data and the actual
release into CalDB. The background information can be found in the CIAO why page
of HRC-I Gain Maps.
- HRC-S Degap improvement
The HRC calibration team has uncovered a small error in the determination of the plate edge positions
in the V-direction (along the length of the array). The fix (in HRC-S GAPLOOKUP table) was Implemented
in CalDB 4.8.3 (May 24, 2019) and applied in DS 10.9 (July 22, 2020). This improves
the energy scale for all LETG observations by correcting HRC-S plate corner positions. Further details
can be found in the
CalDB 4.8.3 release notes.
- HRC-S T_GMAP (Time-dependent Gain Map)
The HRC-S gain degradation has taken place since the beginning of the mission and more rapidly since 2015.
The proper correction for the time-dependent gain degradation was implemented in CalDB 4.9.2 (July 06, 2020)
and the corresponding s/w fix in DS 10.8.3 (March 24, 2020). More information can be found in the CIAO ‘why page’ of
HRC-S Time-dependent Gain and the
CALDB 4.9.2 release note.
- HRC event files with new columns
HRC event files retain more columns which were previously excluded. They include amp_sf
(amplifier scale factor), sumamps (scaled sum of three nearest amp signals) and raw coordinates
(crsu/v, au/v and rawx/y). It was implemented in DS10.8 (June 18, 2019).
GRATING (HETG and LETG)
- Grating response files
The new GARF (grating Auxiliary Response File) and GRMF (grating response matrix file) for multiple
orders are generated at the default position. It was implemented in DS10.8 (June 18, 2019) and affects
all grating observations for both LETG and HETG. Further details can be found in
the CIAO 4.12 mktgresp description.
- Aspect Camera Assembly (ACS) plate scale correction
Improvements in the aspect camera plate scale calibration and other algorithmic changes,
(March 24, 2020), can improve astrometric accuracy by up
to 0.10 arcsec. This is mostly relevant for data taken over the last five years. Further details can be found
- Off-axis angle correction
A new tool (axiscorrect) was implemented in DS 10.8.3 (March 24, 2020) to improve the boresight
correction. The OAA (off-axis angle) correction is up to 30” in recent data but much less for
early phases of the mission. Further details of the off-axis angle algorithm can be found in
summary and the
- Improved timing accuracy
A new set of interpolated clock correlation files will be applied. This update will
improve all data since the last REPRO-IV (Period C). Further details can be found in
a memo on
Absolute time calibration for the Chandra X-ray Observatory.
- Applying definitive ephemeris
The definitive orbit ephemeris data are not available until about a week after the observation. Instead, a
predictive orbit ephemeris is used in the pipeline processing. This is important in the high precision
timing analysis e.g., applying barycenter corrections to fast pulsars, and affects all data since the
last REPRO-IV (Period C). See also
Absolute time calibration for the Chandra X-ray Observatory.
- DOIs (Digital Object Identifier)
From DS 10.8.3 (March 24, 2020), Digital Object Identifiers (DOIs) replace the existing dataset
identifiers (DS_IDENT keyword) to provide a versatile and powerful mechanism for fast, efficient,
and durable access, citation, and cross-linking of Chandra observations in the scientific literature.
The Chandra DOIs follow the DataCite standard and are available in the headers of all FITS data products
that can be unequivocally associated with a single Chandra Observation Identifier (ObsID). For example,
the DOI of ObsID 23336 is defined as "10.25574/23336" where the prefix "10.25574" identifies
the Chandra Data Archive. A full description can be found at this
- Replacement of bad target names
The target names of observations taken before January 1st, 2018 that did not resolve into
correct sky coordinates, contained a typo, were uninformative or too generic, have been replaced
to increase the discoverability of these observations through the archival interfaces.
The new target names have been selected to reflect as closely as possible the original target
name and/or the scientific motivation of the observation. No other metadata that can be used to
identify an archival Chandra observation (ObsID, pointing coordinates, etc) has been modified.
More details can be found at this
- Improved sky fov (field-of-view)
An improved skyfov algorithm was implemented in DS10.8.1 (October 2, 2019) to increase the precision of
the (dithered) fov region by estimating the extent of the dither pattern with a convex hull solution
(method=convexhull), upgrading from the previous 8 point min and max solution (method=minmax). More
details can be found in the CIAO 4.12 skyfov
- Updated Header keyword
The old header keyword MJD_OBS was replaced with a new header keyword MJD-OBS. It applies to all
Chandra data products to bring into compliance with the FITS-3.0 standard. It was implemented in
DS 10.8.1 (October 2, 2019).
Previous reprocessing notes
Pipeline s/w release dates
DS 10.9 2020-07-22
DS 10.8.3 2020-03-24
DS 10.8.1 2019-10-02
DS 10.8 2019-06-18
CalDB release notes and dates