Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://www.stecf.org/conferences/adass/adassVII/reprints/micola.ps.gz Äàòà èçìåíåíèÿ: Mon Jun 12 18:51:47 2006 Äàòà èíäåêñèðîâàíèÿ: Tue Oct 2 03:08:28 2012 Êîäèðîâêà: Ïîèñêîâûå ñëîâà: ï ï ï ï ï ð ï ð ï ð ï ð ï ð ï ð ï ð ï ð ï ï ð ï ï ð ï ï ð ï ï ð ï ï ð ï ï ð ï ï ð ï ï ð ï ï ð ï ï ð ï ï ð ï ï ð ï ï ð ï ð ï ï ð ï ð ï ï ð ï ð ï ï ð ï ð ï ï ð ï ð ï ï ð ï ð ï ï ð ï ð ï ï ð ï ð ï ï ð ï ð ï ï ð ï ð ï ï ð ï ð ï ï ð ï ð ï ï ð ï

Astronomical Data Analysis Software and Systems VII
ASP Conference Series, Vol. 145, 1998
R. Albrecht, R. N. Hook and H. A. Bushouse, e
Ö Copyright 1998 Astronomical Society of the Pacific. All rights reserved.
ds.
Constructing and Reducing Sets of HST Observations
Using Accurate Spacecraft Pointing Information
A. Micol and B. Pirenne
Space Telescope -- European Coordinating Facility, ESA/ESO, Garching
D­85748, Germany
P. Bristow 1
Space Telescope -- European Coordinating Facility
Abstract. The implementation of ``On­The­Fly'' Re­Calibration at the
ST­ECF and CADC goes some way towards alleviating the problem of
obtaining good and timely calibration of HST exposures. However, the
data access paradigm is still to consider each exposure individually, re­
calibrate them and o#er the results to users, who subsequently process
the data further.
We describe here techniques to automatically group together HST
WFPC2 exposures for cosmic ray removal, co­addition and combination
into mosaics with minimal resolution loss. We show that the execution
of these tasks has been made essentially automatic.
The ST­ECF archive now o#ers the possibility to select ``associa­
tions'' of datasets and the automatically combined final products. A
further spin o# of this project is that more reliable pointing information
for all exposures is provided.
1. Introduction
In 1995 the ST--ECF introduced, along with the CADC, the re­calibration on­
the­fly (OTF) of archived HST exposures. By making use of the best current
calibration software and reference files, OTF guarantees that archive users al­
ways retrieve from ST--ECF and CADC archives the current best products (i.e.,
calibrated data).
Pushing the concept further, ST--ECF and CADC are working in the direc­
tion of providing the astronomers with an active archive system, able to relieve
the astronomers of time consuming and boring tasks, therefore increasing the
productivity of archive researchers. In this framework the ST--ECF embarked
upon a project aimed at building associations of WFPC2 exposures, recovering
the otherwise lost information of which observation strategy (e.g., CR­SPLIT,
dithering) a WFPC2 proposer decided to employ.
1 Now at NATO/SACLANT Undersea Research Centre
45

46 Micol, Pirenne and Bristow
When looking at an association the archive users immediately see what the
shifts between all the exposures were without having to compute them manually.
Upon request the archive system can produce a cosmic ray­free mosaic out of
the association. Indeed a reliable automatic pipeline has been put in place.
All this is possible thanks to the reliable HST pointing information provided
by the jitter files which are a product of the Observatory Monitoring System
software at STScI.
2. Jitter Files as a Source of Reliable Pointing Information
After some tests it was clear that:
. the world coordinate system (WCS) keywords available in the header of the
exposure FITS files are unreliable. For example, cases where two exposures
with identical WCS keywords were found to be clearly shifted.
. Cross correlation to compute shifts among exposures cannot be easily au­
tomated, since not all the exposures have enough features or a good signal
to noise ratio, and because of the presence of cosmic rays.
Since October 20th 1994 the Observatory Monitoring System subsection of
the HST pipeline (STScI) has generated the so called jitter files. Jitter files are
normally used to monitor the telescope pointing stability and the trends in the
telescope and instrument performance as the orbital environment changes. By
correlating the HST Mission Schedule and the time­tagged engineering teleme­
try data stream downloaded from HST it is possible to reconstruct with great
accuracy the pointing sequence of any given scientific observation.
Since the initial production of jitter files, di#erent formats have been adopted.
The current format jitter files contain:
. a table of pointing and environment measurements, e.g., right ascension,
declination, magnetometer readouts, etc., sampled every 3 seconds.
. a 2D histogram showing the number of times HST was pointing in each
element of a 64x64 grid with each pixel 2 milliarcsec in size.
2.1. Jitter Pointing Accuracy
Due to relative errors in Guide Stars coordinates (about 0.3 to 0.4 arcsec),
the Fine Guide Stars alignment uncertainty (up to 50 milli­arcseconds [mas]),
thermal breathing of the telescope (around 15 mas) and other minor e#ects, the
absolute accuracy in the pointing ranges between 0.5 and 2 seconds of arc. On
the other hand the relative accuracy is better than 10 mas within the same HST
visit, that is, if the telescope didn't have to re­acquire a guide star during the
sequence of observations.
Problems arise if the observation is taken in parallel mode. While the veloc­
ity aberration is corrected for the primary instrument, the di#erential velocity
aberration causes the secondary instrument aperture to experience a drift on
the sky. The e#ect can be as high as 50 mas for a full orbit observation, de­
pending on the relative position on the focal plane of the primary and secondary
instruments.

Constructing and Reducing Sets of HST Observations 47
3. WFPC2 Associations: Computing the Shifts
Shifts among exposures can hence be computed with good accuracy using the
jitter information in the case when all the observations were made during the
same visit and WFPC2 was the primary instrument. The procedure consists of:
. Computing the right ascension, declination and roll angle averages and
standard deviations from the jitter table, along with some telemetry key­
words.
. Assigning a jitter quality flag to each exposure depending on the telemetry
keywords and on the standard deviations. The flag can assume the value:
-- `P' for exposures with small standard deviations and no suspect key­
word values,
-- `G' for exposures with not so small standard deviations or with bad
keyword values like GUIDEACT not equal to 'FINE LOCK' etc.
-- `B' for exposures with missing jitter information or with standard
deviations too high or having the SLEWING flag on, etc.
. Grouping together all the `P' and `G' exposures
-- taken within the same HST visit
-- in the same filter
-- by the same PI
-- where the distance between each pair is less than 25 PC pixels
-- and the di#erence between the roll angle of two exposures cannot
account for a shift bigger than a tenth of a PC pixel on the full
image.
. Computing the shifts #x and #y in PC pixels
. Identifying those exposures in the association which are well registered
within a tenth of a PC pixel and placing them in sub groups for cosmic
ray removal.
4. WFPC2 Association Pipeline
Once we can compute the shifts we may recover the observation strategy (CR­
SPLIT, POS­TARG) adopted by a PI. By making use of the computed shifts and
rejecting all the exposures not flagged as `P' (see above), it is possible to run an
automatic pipeline which provides not only on­the­fly re­calibrated observations
but also o#ers cosmic ray­free images from all the exposures found to be well
aligned within the association.
The overall mosaic of the association can also be requested. In this case the
`drizzle' software (Hook & Fruchter, 1997) is used to build the mosaic of all the
cosmic ray­free images preserving photometry, minimising loss of resolution (for
sub­pixel shifts) and correcting for geometric distortion using the polynomial
model of Trauger et al.

48 Micol, Pirenne and Bristow
We plan to use a simpler, less CPU demanding, `shift­and­add' technique
for those associations where the shifts are simple integer numbers of pixels.
5. Conclusions
The great reliability of the jitter information has allowed the ST--ECF to:
. Reconstruct the WFPC2 PI's observation strategy (CR­SPLIT, POS­TARG)
. Build associations of WFPC2 exposures
. Compute the shifts among exposures within an association
. Build an automatic pipeline able to clean cosmic rays from aligned expo­
sures and to co­add them with shift­and­add or drizzle techniques.
The ST--ECF archive users can now concentrate on their science while
the bulk work of re­calibration, removal of cosmic rays, co­addition of WFPC2
frames is taken care of by the ST--ECF archive system.
The same system will be installed at the Canadian Astronomy Data Centre
(CADC).
More information and access to the associations can be found at the fol­
lowing URLs: http://archive.eso.org/archive/hst/wfpc2 asn/ (ST--ECF)
and http://cadcwww.dao.nrc.ca/ (CADC).
Acknowledgments. We would like to thank M. Lallo and J. Baum (STScI)
who helped a lot in understanding the jitter file secrets; D. Durand and S. Gaudet
(CADC) for the always fruitful collaboration; D. Shade (CADC) and D. Durand
who helped in tuning the WFPC2 pipeline; R. Hook (ST--ECF) and A. Fruchter
(STScI) for providing us with details of the drizzle method.
References
Lupie, O., Toth, B.A., & Lallo, M., ``Observation Logs'', STScI, 1997
Bely, P.Y., & Toth, B.A., ``Line of sight jitter reconstruction from guide stars
motion'', STScI Report, TR­88­01, 22 January 1988
Micol, A., Bristow, P., & Pirenne, B., ``Association of WFPC2 exposures'', 1997
HST Calibration Workshop, STScI, S. Casertano, et al., eds.
Hook, R., & Fruchter, A., 1997, in ASP Conf. Ser., Vol. 125, Astronomical Data
Analysis Software and Systems VI, ed. Gareth Hunt & H. E. Payne (San
Francisco: ASP), 147
Micol, A., Dolensky, M., & Pirenne, B., ST--ECF Newsletter No 25, in prep.
Dolensky, M., Micol, A., & Pirenne, B., ``Browsing the HST archive with Java­
enriched Database Access'', this volume
Dolensky, M., Micol, A., & Pirenne, B., Rosa, M., ``Enhanced HST Pointing
and Calibration Accuracy: Generating HST Jitter Files at ST­ECF'',
this volume