WFC3 Post-Observations Ground System Design
Post on 14-Jan-2016
27 Views
Preview:
DESCRIPTION
Transcript
February 13, 2001 WFC3 IPT PDR 1
WFC3 Post-Observations Ground System Design
M. Giavalisco
H. Bushouse
W.King
February 13, 2001 WFC3 PDR 2
Introduction
• Purpose : Obtain and manipulate all WFC3 SI data from
observations through final storage and dissemination to the
observer
Process data through steps of :
• Generic conversion (OPUS)
• Data reduction and Calibration (CALWF3)
• Associations (CALWF3)
• Archiving (DADS)
February 13, 2001 WFC3 PDR 3
WFC3 Layout
February 13, 2001 WFC3 PDR 4
WFC3 Detectors
UVIS Channel
• 2 2k x 4k EEV Marconi CCDs
• 200 – 1000 nm
• High QE
• 15 micron pixel
• 0.04 arcsec/pix
• FOV 160 x 160 arcsec
IR Channel
• 1k x 1k HgCdTe array
• 850 1700 nm
• High QE
• 18 micron pixel
• Scale 0.13 arcsec/pix
• FOV 123 x 139 arcsec
February 13, 2001 WFC3 PDR 5
WFC3 and Other HST S.I.
February 13, 2001 WFC3 PDR 6
WFC3 Limiting Flux
February 13, 2001 WFC3 PDR 7
WFC3 Pipeline Routine Science Data Processing
• Assembles data from WFC3 into scientifically useful data sets.
• Processes the science data using standard calibration tools and reference
files.
• Calculates error arrays.
• Stores raw data sets in the Archive.
• Offline Tools/files for Observers to re-calibrate.
Engineeringheaders
Associations
Reffiles
Calfiles
CALWFC3 SOFTWAREArchive
MISSIONSCHEDULE
February 13, 2001 WFC3 PDR 8
Design Guiding Principles
• Maximize Reuse of ACS and NICMOS Pipelines
• Treat UVIS and IR channels as similarly as possible
• Leverage on the experience from and solutions
adopted for WFPC2, NICMOS, STIS, ACS
• Incorporate DRIZZLING to enhance effective spatial
resolution, remove geometric distortions
February 13, 2001 WFC3 PDR 9
CALWFC3 Structure
• Use ACS structure for pipeline backbone: CALWF3 built from ACSCCD, ACS2D, ACSREJ.
• Replace MAMA branch with CALNICA for IR channel.
• Implement (minor) modifications and upgrades when required by lack of specific capabilities of ACS’s pipeline.
February 13, 2001 WFC3 PDR 10
Major Pipeline Building Steps
• 1) Required changes for reuse of ACS/NICMOS pipelines:
– 1.1) Flatfield implementation (due to geometric distorsions)
– 1.2) IR subarrays (UVIS is a copy of ACS WFC)
– 1.3) IR overscan pixels (UVIS overscan pixels - built into ACS pipeline)
– 1.4) Binning (currently removed from ACS pipeline, but straightforward to replace)
– 1.5) Associations - use ACS as baseline
– 1.6) WFC3 Keywords (based on ACS and NICMOS)
– 1.7) Adopt future NICMOS and ACS science upgrades, as applicable
February 13, 2001 WFC3 PDR 11
1.1 Flatfield Implementation
• Build capabilities in the SW to handle different
type of flat fields (pix-to-pix, delta-flat, low-
frequency, etc.), as needed for WFC3
• Account for the effect of geometric distortions
during flat fielding operation (due to variations
in pixel area)
February 13, 2001 WFC3 PDR 12
1.2 Subarrays
• Must use IR subarrays for photometric calibrations of NICMOS standards
• UVIS subarrays to follow ACS model : center, size variable in parallel direction
• IR subarrays to be of size 512,256,128,64,32 on a side, centered on middle of array
• Enable the NICMOS SW to handle OVERSCAN pixels
• Must be able to correct for geometric distortions for different size/position subarrays
• Must handle relevant calibrations for specific size/position subarrays
February 13, 2001 WFC3 PDR 13
1.3 Overscan Pixels
• Both physical (to remove bias effects) and virtual (to remove
amplifier effects) overscan for UVIS - use ACS solutions
• Physical overscan (“reference”) pixels only for IR
• Used to remove itinerant baselines, biases varying by quadrant,
amplifier
• Necessary to avoid corruption of edge pixels if binned; DQI
flag to be set to acknowledge this (no problem if binning is
1x,2x,3x, and OVSC size is multiple of these numbers)
February 13, 2001 WFC3 PDR 14
1.4 Binning
• Needs to be put back into ACS pipeline
• Implemented via binned reference files
• Will support 2x2, 3x3 for UVIS
• Prevent mixing of overscan and “good” edge pixels => set DQI flag
February 13, 2001 WFC3 PDR 15
1.5 Associations
• Inputs will follow ACS example: science imsets from CALWF3, pixel shift from jitter file/guide star system, “pixel fraction”, pixel weight (bad mask) file
• The implementation of DRIZZLING will associate dithered images, no mosaics
• DRIZZLING can also do 2 UVIS imsets => 1 UVIS image, geometric distortion corrections as well
February 13, 2001 WFC3 PDR 16
1.6 WFC3 Keywords
• Reuse ACS, NICMOS keyword sets• Keep separate set of keywords for UVIS, IR channels• Resolve contradictions in definitions using ACS
conventions• Keep only critical engineering snapshot keywords
(about NICMOS) in science headers, remainder in .SPT headers for monthly trending of other engineering keywords
February 13, 2001 WFC3 PDR 17
1.7 NICMOS Science Upgrades
• The pipeline will incorporate the improved CR rejection algorithms (proper treatment of correlated errors)
• The upgrades Wwll be meshed with improvements in WFC3 MUX, detector design to track systematics
February 13, 2001 WFC3 PDR 18
Major Pipeline Building Steps
• 2) Desired changes for improved CALWF3 pipeline
• 2.1) Specific dithering strategies for DRIZZLING implementation -> dither patterns
• 2.2) Geometric distortions corrections: e.g. DRIZZLING, but other options possible- (will monitor ACS solutions/progress)
• 2.3) Adopt optimized CR rejection strategies, either with CR-REJECT or during DRIZZLING (will monitor ACS solutions/progress)
February 13, 2001 WFC3 PDR 19
2.2 DRIZZLING & Dithering Strategies
• DRIZZLING will improve angular resolution (but make error analysis more complicated)
• Allow either standard TRANS line, box/spiral patterns, or custom ones
• DRIZZLING can also paste together two halves of UVIS image
February 13, 2001 WFC3 PDR 20
2.3 Geometric Distortions
• Expected geometric distortion (~10% across the WFC3 FOV)
• Need to correct for registration and calibration, and when taking calibrations for the different modes of the instrument.
• Need to account for geometric distorsion effects to properly flat-field point vs. extended sources
• Correction done in CALWF3 using DRIZZLING
• Cubic polynomial description in X,Y w/ cross-terms
• Coefficients kept in IDC, SIAF files
February 13, 2001 WFC3 PDR 21
Conclusions
WFC3 Pipeline Will :
• Take raw instrument science data and convert it into
FITS files suitable for science analysis
• It will support WFC3 Thermal Vacuum ground testing
and verification at NASA/GSFC
• Take advantage of maximal reuse of ACS and
NICMOS pipelines.
• It will build on ACS pipeline structure with CALNICA
replacing the MAMA code, including DRIZZLING
top related