The Super/Nova Acceleration Probe (SNAP)

1
The Super/Nova Acceleration Probe (SNAP)

• Natalia Kuznetsova
• Lawrence Berkeley National Lab

Cosmo06 September 24 - 29, 2006 Tahoe City, CA
2
SNAP 101

• Space-based 2-m class telescope dedicated to
  performing precision measurements of the dark
  energy equation of state parameter w through
• A wide field lensing survey
• Discovery and follow-up of 2,000 type Ia
  supernovae

SNAP will be very data-rich, producing data
useful not only for precision cosmology
studies, but for many other applications
2
3
Focal plane
Visible
NIR
Spectrograph port
3
4
Physics with SNAP Deep Large Space Surveys
SNAP Deep Survey Area

• The SNAP surveys will have an unprecedented
  combination of depth, solid-angle, angular
  resolution, temporal sampling, and wavelength
  coverage
• Hubble Deep Fields illustrate the impact of a
  deep space survey.
• SNAP SN survey 5,000 x HDF.
• SNAP mAB 27.7 per filter (30.4 co-added) every
  4 days
• SNAP lensing survey 106 x HDF, 500 x COSMOS!
• mAB 28.1 co-added

SNAP Lensing Survey Area
4
5
Physics With SNAP Supernovae

• SNAP s homogeneous SN dataset over the
  redshift range up to z 1.7 will have carefully
  controlled systematics
• The quality, not the quantity, of SN observations
  is the primary factor for dark energy accuracy
• SNAP will have photometric measurements of 2,000
  type Ia SN in 9 broadband filters, as well as
  their spectra at maximum light

2000 SNe Ia
DE discovery
redshift z
5
6
Physics with SNAP Weak Lensing

• Weak lensing (WL) provides an independent and
  complementary measurement of cosmological
  parameters
• Space-based WL measurements are particularly
  helpful at small scales, where the shot noise is
  small due to the large surface density of
  resolved galaxies

Courtesy Jason Rhodes
6
7
Ancillary Science From SNAP

• Galaxy structure formation
• Galaxy clusters
• Gamma-ray burst afterglows
• Reionization history
• Transients/variables
• Stars
• Solar system objects
• Strong gravitational lensing
• .

7
8
Simulating a Dark Energy Mission

• We have created a sophisticated simulation that
  allows one to simulate a dark energy mission
  (space- or ground- based)
• It is a collaborative project written in
  object-oriented Java
• Basis for future data processing pipeline

8
9
Studies with SNAPsim

• SNAPsim is easily configurable for studying
  various choices of mission parameters
• Examples of studies done using SNAPsim include
• SNAP exposure time - cadence trade study
• SNAP detector-noise requirements
• Calibration error propagation
• Spectroscopic measurement requirements
• Alternative instrumentation suites
• SNAP primary aperture trade study
• Ground-based missions
• SNAP telescope blur requirements
• Weak gravitational lensing mission simulation

9
10
SNAPsim Physics

• Type Ia, II supernova spectra, varying stretch
• Zodiacal background
• Cardelli-Clayton-Mathis model dust
• Atmosphere effects for ground-based missions
• Sophisticated fitting algorithms for lightcurve
  and cosmology fitting

filter 7
filter 8
filter 9
10
11
Lightcurve Redshift Series
Z 0.8
Z 1.2
Z 1.6
Optical Bands
Rest frame B
NIR Bands
Rest frame V
11
12
Extracting Cosmology

• The final step of the simulation is extracting
  the cosmological parameters
• The plot is an example of the cosmology to be
  obtained from SNAP results only (no CMB priors)

courtesy Eric Linder
12
13
NIR Detector RD

• SNAPsim is used extensively for SNAPs
  instrumentation and RD work
• For example, a recent study has investigated the
  effect of varying NIR detector parameters on the
  output physics
• The idea is to find out what combination of
  detector specs (dark current, read noise, quantum
  efficiency) produces optimal science at the
  lowest cost

Infrared Sensors
m Error Contours
Total Noise (e)
QE
courtesy Matt Brown
13
14
IR Detector Trade-Off Study (2)

• m error vs. redshift for visible only and visible
  NIR detectors

Matt Brown et al. , proc. of 2006 SPIE symposium
on Astronomical Telescopes and Instrumentation
14
15
Simulating a Ground-Based Observatory
Atmosphere transmission

• SNAPsim is also capable of simulating a
  ground-based observatory
• As an example, we simulate a somewhat idealized
  8-m class telescope in the Southern hemisphere,
  with NIR detectors
• We then look at the lightcurves for z 1.2 and z
  1.4 supernovae for a GOODS South target and an
  equatorial pole one

Atmosphere emission
Natalia Kuznetsova, Larry Gladney, Alex Kim
15
16
Simulating a Ground-Based Observatory (2)

• Examples a (somewhat) idealized 8-m ground
  telescope (with IR), observing a target in the
  GOODS South field and an equatorial one
• Equatorial pole target gets a worse S/N, but
  there are no holes in the lightcurve

GOODS South target
Equatorial pole target
16
17
Spectrograph Simulation

• Pixel-level simulation using shapelets to create
  fake spectra of both point and extended objects

courtesy Richard Massey
17
18
Spectrograph Simulation (2)
Same magnitude SN and galaxy no noise
SN spectrum
Reconstructed SN spectrum (z 1.7)
?
y
Host galaxy spectrum
(a few slices from slicer mirror)
courtesy Alain Bonissent
18
19
Pixel Scale for Weak Lensing

• Also using shapelets to simulate space-based,
  pixel-level images
• Initial result SNAP nominal pixel scale of 0.10
  arcsec/pixel is in the optimal well
• This pixel scale is optimal for both supernova
  and weak lensing studies

Contribution of intrinsic shear variance to the
weak lensing power spectrum error
SNAP nominal
courtesy Will High
19
20
Self Calibration in Supernova Surveys

• Filter zeropoint uncertainties affect precision
  of cosmological parameters.
• Fitting for all SN distance moduli ?
  simultaneously allows for a degree of self
  calibration which yields a noticeable
  improvement in the final precision (Kim Miquel,
  Astropart. Phys. 24 (2006), 451).
• We show this effect by simulating an SNLS-like
  survey and comparing the results against the
  usual SN by SN fit we fit for s(WM) with w-1.

courtesy Lorenzo Faccioli also see poster in
hallway
20
21
Conclusions

• SNAP is specifically targeted at controlling
  systematic uncertainties
• Our sophisticated mission simulation, SNAPsim,
  enables us to pursue such a tight control of
  errors
• Numerous R D, trade-off, and physics studies in
  progress, not only those presented in this talk
• For more info, please go to http//snap.lbl.gov

21