verification and scientific simulations

Final network presentation
VERIFICATION AND
SCIENTIFIC SIMULATIONS
OF JWST/NIRSPEC
Bernhard Dorner, CRAL/MPIA
The research leading to these results has received funding from the European Community's Seventh
Framework Programme (FP7/2007-2013) under grant agreement n° PITN-GA-2008-214227 - ELIXIR
What's left to say?
Text
The research leading to these results has received funding from the
European Community's Seventh Framework Programme (FP7/2007-2013)
under grant agreement n° PITN-GA-2008-214227 - ELIXIR
PhD thesis presentation
Verification and science with
the JWST/NIRSpec Instrument
Performance Simulator
Verification and science
with the NIRSpec
instrument performance
simulator
Simulation of NIRSpec
exposures
ESR report XI
Bernhard Dorner, ESR CRAL
ELIXIR mid-term review, Paris, 3/11/2010
Bernhard Dorner, ESR CRAL
ELIXIR meeting Oxford, 10/12/2009
Bernhard Dorner, ESR CRAL
ELIXIR school EADS/Astrium, 02/06/2010
The research leading to these results has received funding from the European
Community's Seventh Framework Programme (FP7/2007-2013) under grant agreement n°
PITN-GA-2008-214227 - ELIXIR
EXTRACTION AND
PROCESSING OF NIRSPEC
SPECTRA WITH THE NIPPLS
SIMULATIONS OF NIRSPEC
MOS EXPOSURES
SIMULATION OF EXOPLANET
TRANSIT OBSERVATIONS
WITH NIRSPEC
Bernhard Dorner, CRAL/MPIA
Bernhard Dorner, CRAL/MPIA
Bernhard Dorner, CRAL/MPIA
The research leading to these results has received funding from the
European Community's Seventh Framework Programme (FP7/2007-2013)
under grant agreement n° PITN-GA-2008-214227 - ELIXIR
The research leading to these results has received funding from the
European Community's Seventh Framework Programme (FP7/2007-2013)
under grant agreement n° PITN-GA-2008-214227 - ELIXIR
The research leading to these results has received funding from the
European Community's Seventh Framework Programme (FP7/2007-2013)
under grant agreement n° PITN-GA-2008-214227 - ELIXIR
Bernhard Dorner, CRAL - Observatoire de Lyon
ELIXIR annual meeting, Madrid, 05/10/2011
The research leading to these results has received funding from the European Community's Seventh
Framework Programme (FP7/2007-2013) under grant agreement n° PITN-GA-2008-214227 - ELIXIR
...after all those meetings...
Bernhard Dorner, ELIXIR final presentation, 12 Nov 2012
Kudos to:
Pierre Ferruit, Bruno Guiderdoni,
Xavier Gnata, Laure Piqueras, Emeline Legros, Pierre-Jacques Legay, Arlette
Pécontal-Rousset, Aurélien Jarno, Aurélien Pons,
Jess Köhler, Jean-Francois Pittet, Werner J. Hupfer, Markus Melf, Peter Mosner,
Maurice Te Plate, Peter Jakobsen, Stephan M. Birkmann, Torsten Böker, Guido de
Marchi, Marco Sirianni, Giovanna Giardino,
Jeff Valenti, Tracy Beck, Camilla Pacifici, Stéphane Charlot, Enrica Bellocchi,
Santiago Arribas, Hans-Walter Rix
The research leading to these results has received funding from the European Community's Seventh
Framework Programme (FP7/2007-2013) under grant agreement n° PITN-GA-2008-214227 - ELIXIR
Bernhard Dorner, ELIXIR final presentation, 12 Nov 2012
Outline
I.
Software for NIRSpec
simulations
II. How to build and verify an
instrument model
III. Science part 1: Spectrographic
deep field
IV. Science part 2: Integral field
observations
V. Science part 3: Exoplanet transits
VI. Conclusion
Bernhard Dorner, ELIXIR final presentation, 12 Nov 2012
Once more: NIRSpec overview
•
•
•
•
Spectral range: 0.6–5 µm
•
•
•
Fixed slits (0.2", 0.4", 1.6")
•
ESA project, built by EADS/Astrium
GmbH
Field of view: >9 arcmin2
Multi-object capability: >100 targets
Configurable masks (MSA, 250,000
shutters, 0.2")
Integral Field Unit (IFU, 30 slices, 3x3")
Two HgCdTe arrays (SCA), each
2048x2048 pixel
see Bagnasco et al., 2007
Bernhard Dorner, ELIXIR final presentation, 12 Nov 2012
The Instrument Performance Simulator
•
Purpose:
‣ Study geometrical effects
‣ Verify instrument performance
‣ Generate realistic output data
•
•
•
Software developed by CRAL 2005–2011
(Gnata, 2007, Piqueras et al., 2008, 2010)
>110,000 lines of C++ code
End-to-end simulation of NIRSpec:
‣ Noiseless electron rates
‣ NIRSpec raw data cube
Bernhard Dorner, ELIXIR final presentation, 12 Nov 2012
Auxiliary software for simulations
•
Science data input interface:
‣ Direct object placement in slits
‣ Typical input file types
•
"NIRSpec IPS Pipeline Software" (NIPPLS):
‣ Spectrum extraction from NIRSpec exposures
‣ Uses IPS instrument model to find spectra
‣ Standard "long slit" reduction, but flexible for
custom tasks
‣ Also used for measured data
Bernhard Dorner, ELIXIR final presentation, 12 Nov 2012
Up next:
I.
Software for NIRSpec
simulations
II. How to build and verify an
instrument model
III. Science part 1: Spectrographic
deep field
IV. Science part 2: Integral field
observations
V. Science part 3: Exoplanet
transits
VI. Conclusion
Bernhard Dorner, ELIXIR final presentation, 12 Nov 2012
NIRSpec model data
•
Collection of measurements and
calculations for subsystems
•
Efficiencies:
‣ Mirrors, filters, detector
‣ Gratings, IFU
•
Geometries
‣ Disperser, MSA, detector
‣ Optical distortion
•
Wavefront errors
‣ Dispersers + IFU
‣ NIRSpec optical train, Telescope
(te Plate et al., 2007)
FORE WFE in telescope pupil
Bernhard Dorner, ELIXIR final presentation, 12 Nov 2012
Model verification
•
Why?
‣ Verification of model as a whole: remove
uncertainties, check data interplay
‣ Provide input for data processing and simulations
•
How?
‣ Compare model prediction with calibration
measurements (fixed slits and IFU, February 2011)
‣ Analysis done in NIPPLS
•
What?
‣ Instrument geometry and efficiency
Bernhard Dorner, ELIXIR final presentation, 12 Nov 2012
Geometry: reference data
Spatial: Trace
polynomials
Spectral: Argon
emission lines
Reference data tuples
(Pixeli , Pixel j , λref )
Bernhard Dorner, ELIXIR final presentation, 12 Nov 2012
Optimization: Forward
•
•
•
Total forward residuals:
Dispersers
Reference points
Median residual /
pixel fraction
Gratings
2233
1/15
PRISM
219
1/4
MIRROR
35
1/5.6
Total
2487
1/14
Instrument requirement (spectral): 1/4th pixel
Modeling approach works
Bernhard Dorner, ELIXIR final presentation, 12 Nov 2012
Total instrument throughput
•
Ratio simulated to measured, all dispersers
•
•
•
Consistent across bands
Divergence of slits
Some residual features
•
Mean: 0.82±0.05
(Calibration source?)
•
Final accuracy: 0-10%
absolute, 5% relative
Bernhard Dorner, ELIXIR final presentation, 12 Nov 2012
...nearly 14 billion years ago, expansion started...
I. JWST, NIRSpec, and simulations
II. How to build and verify an
instrument model
III. Science part 1: Spectrographic
deep field
IV. Science part 2: Integral field
observations
V. Science part 3: Exoplanet
transits
VI. Conclusion
issue: 6 – re
Deep field spectroscopy simulation
•
Sky scene from
6
NIRSPEC TARGET ACQUISITION
6.1
Overview
The purpose of the NIRSpec target acquisition (TA) procedure is to fine-tune the JWST
such that a given set of astronomical targets is imaged precisely onto the grid of shutters
MSA. The problem is illustrated in Figure 6.1-1. The TA procedure, i.e. the entire seque
events following the spacecraft slew to the target field, up to the completion of the point
correction that places the science targets accurately within their intended apertures, must
executed without human intervention. In particular, the NIRSpec on-board software1 mu
autonomously derive the positions of a set of bright reference stars in a dedicated, short,
undispersed exposure (the “acquisition image”) and from these, must derive the necessa
pointing corrections.
‣ Hubble UDF: Objects with band
photometry and derived redshift
(Coe et al., 2006)
‣ Model galaxy spectra from
simulations (Pacifici et al., 2012)
•
Simulation with
‣ Point sources, CLEAR, PRISM
‣ Noise for 945s exposure
•
Extraction with NIPPLS
Figure 6.1-1 The MSA projected onto the target field during ground preparation. Science targets are indic
and the brighter reference stars in red.
For this purpose, the on-board software uses a set of stored coordinate transformations b
various optical planes of the sky-OTE-NIRSpec optical train, which take into account th
magnifications and distortions of the various optical modules. The required accuracies o
transformations listed in this document have been derived and quantified in RD 2. Calibr
1
The on-board software consists of two components, the C++ based “Flight Software” and the JavaScript-
Bernhard Dorner, ELIXIR final presentation,
NovFor2012
“Activity Description12
Scripts”.
details, see RD 6.
Multi-object processed exposure
Bernhard Dorner, ELIXIR final presentation, 12 Nov 2012
Multi-object processed exposure
Bernhard Dorner, ELIXIR final presentation, 12 Nov 2012
Galaxy spectrum example
z=6.204, magH=26.9
Bernhard Dorner, ELIXIR final presentation, 12 Nov 2012
NIRSpec cubism
I. JWST, NIRSpec, and simulations
II. How to build and verify an
instrument model
III. Science part 1: Spectrographic
deep field
IV. Science part 2: Integral field
observations
V. Science part 3: Exoplanet
transits
VI. Conclusion
Bernhard Dorner, ELIXIR final presentation, 12 Nov 2012
An ULIRG in the NIRSpec IFU
•
Single Ultra-luminous infrared galaxy with velocity
field in integral field mode
•
•
Data:VLT/VIMOS observation of Hα + [NII] (from
Bellocchi et al. 2012)
Bellocchi et al.: Kinematic asymmetries of disks and post-coalescence mergers
For NIRSpec:
Scale to redshift z=1
VIMOS CONTINUUM FLUX
−10
0.57
arcsec
10
−10
−10
0
0
0
0
0
10
0
arcsec
10
4927
10
0
4888
4849
10
0
37
−10
4810
−10
4770
0.57
10
0.08
0
10
VIMO
10
0
−10
Sour
arcsec
0
−10
−10
Sour
F
VIMO
0 0
10 1010
10
0
0
−10
−10−10
−10
−
−10
−10
FLU
FN
VIMO
5 5
00
Sour
arcsec
arcsec
−10
−0.41
6
−0.91
−1.40
10
2.53
0.57
1.72
0.08
0
0.91
−0.41
0.10
−0.91
−10
VIMOS CONTINUUM FLUX
10
0
−10
−10
FLUX
MAIN COMP
VIMOS
CONTINUUM
FLUX
10
10
68
−0.71
−1.40
2.46
2.53
0.57
10
10 10
0 0
0
−5 −5
−10
−10−10
FLUX BROAD COMP
−5
−
−10
FLU
F
5 5
10 10
0 0
0 0
arcsec
1.81
1.72
0.08
1.16
0.91
−0.41
0.52
0.10
−0.91
−0.13
−0.71
−1.40
arcsec
arcsec
arcsec
0
0
−10
−10
0
−0.71
10
99
10
10
5
10
10
0
−10
0
0
10
0.10
−10
−10
130
0.91
0
10
−10
−10
arcsec
1.72
0
−10
arcsec
arcsec
2.53
0
00
2.46
2.53
1.72
1.81
0.91
1.16
FLUX
BROAD
COMP
FLUX
MAIN COMP
VIMOS
CONTINUUM
FLUX
5
10
10
0
00
−5
−10
−10
−5
−10
−10
FLUX
BROAD
COMP
FLUX
MAIN COMP
5
10
0
0
FLUX MAIN COMP
10
−10
−10
−10
arcsec
arcsec
arcsec
10
10
−1.40
arcsec
arcsec
0
10
−10
−10
arcsec
10
10
−0.41
0
0
10
0.08
−0.91
arcsec
arcsec
arcsec
this pattern. Despite they are to first approximation
NICMOS CONTINUUM
ACS CONTINUUM
antisymmetric, they also show clear disturbances and
metries. These could be explained by a bent disk a
other tidal motions, consequence of the past merger.
σ maps are more asymmetric with off-nuclear regions
culiar high velocity dispersion values (e.g., IRAS F0
0840), and showing asymmetrical structures in the n
regions (e.g., IRAS F21453-3511).
All of our sources show signs of a broad and blue-s
! MAIN COMP
v MAIN COMP
components (i.e., with velocity shift of ∆v = (50
km/s with respect to the main component) in the
part (nucleus) of the galaxy. They show mean velocit
persions ranging between (140 - 260) km/s, which a
4 times higher than those of the main components.
components cover a small area in disk objects (i.e.
kpc2 ) while larger areas (i.e., 2 - 6 kpc2 ) are involv
the post-coalescence
mergers, showing more comple
Bernhard
final
presentation,
12 Nov 2012
! BROAD
COMPDorner, ELIXIR
v BROAD
COMP
Source F21453−3511
NIRSpec IFU example
Observation with G140H (band 1, R2700)
Only electron rates (no calibration)
Radiant intensity / arbitrary scale
•
•
Bernhard Dorner, ELIXIR final presentation, 12 Nov 2012
NIRSpec IFU example
Observation with G140H (band 1, R2700)
Only electron rates (no calibration)
Radiant intensity / arbitrary scale
•
•
Bernhard Dorner, ELIXIR final presentation, 12 Nov 2012
Seeing the bright light
I. JWST, NIRSpec, and simulations
II. How to build and verify an
instrument model
III. Science part 1: Spectrographic
deep field
IV. Science part 2: Integral field
observations
V. Science part 3: Exoplanet
transits
VI. Conclusion
Bernhard Dorner, ELIXIR final presentation, 12 Nov 2012
Observation setup
•
•
•
Observation of total system brightness
NIRSpec: special square aperture S1600A1
Subarray readout (2048x32 pixels)
Secondary transit:
Self-emission (mIR),
reflection (VIS)
During eclipse:
Star only
III
II
I
Primary transit:
Atmospheric absorption
I
II
III
Signal
Bernhard Dorner, ELIXIR final presentation, 12 Nov 2012
Time
Instrumental effects
•
High sensitivity: Maximum stellar
brightness limits (gratings: magK ≈ 6–7)
•
Readout overheads: Reduction of
effective exposure time during transit
(up to 2/3)
•
Thorough noise discussion:
‣ Limited by photon and readout noise
‣ Other instrumental noise sources
negligible
Bernhard Dorner, ELIXIR final presentation, 12 Nov 2012
HD189733b eclipse
∆λ = 0.67 nm
Data points:
Waldmann et al., 2012
Bernhard Dorner, ELIXIR final presentation, 12 Nov 2012 (4 transits, R=175)
The bottom line
I. JWST, NIRSpec, and simulations
II. How to build and verify an
instrument model
III. Science part 1: Spectrographic
deep field
IV. Science part 2: Integral field
observations
V. Science part 3: Exoplanet
transits
VI. Conclusion
Bernhard Dorner, ELIXIR final presentation, 12 Nov 2012
Conclusion
•
IPS + NIPPLS: useful tools for verifying and
simulating NIRSpec data
•
Assembly and verification of as-built model:
Successful with FM1 data
•
First science simulations of high-z galaxies and
exoplanets: Confirm exceptional capabilities of
NIRSpec
Bernhard Dorner, ELIXIR final presentation, 12 Nov 2012
Conclusion: Network
•
•
ELIXIR: Over, but not dead
Very beneficial for simulation activities:
‣ Spectra for deep-field scenes (Camilla)
‣ IFU sources (Enrica)
•
Hopefully continuation and further
exploitation (still some work on the
software)
Bernhard Dorner, ELIXIR final presentation, 12 Nov 2012
What's next?
•
New old job at MPIA: NIRSpec calibration and
verification (next campaign in 2013)
•
Instrument model:Verify with FM2 data
‣ MSA operable
‣ Higher orders in optics
‣ Throughput
•
Continue science preparation with simulations
Bernhard Dorner, ELIXIR final presentation, 12 Nov 2012