Transcript SDSS-II Supernova Survey Josh Frieman Leopoldina Dark Energy Conference October 8, 2008 See also: poster by Hubert Lampeitl, talk by Bob Nichol.

SDSS-II Supernova Survey
Josh Frieman
Leopoldina Dark Energy Conference
October 8, 2008
See also: poster by Hubert Lampeitl, talk by Bob Nichol
1
SN Models and Observations
•SN cosmology based on a purely empirical approach (Phillips)
•SN observations over the last decade have strengthened evidence for cosmic
acceleration, but dark energy constraints now dominated by systematic errors
•SNe will be one of 3 dark energy probes pursued by JDEM
•Reaching JDEM level of precision for SNe will require improved control
of systematics
•Improved SN modeling, better empirical approaches to estimating SN distances,
and better data are all important weapons in the arsenal to reduce systematics
•Current empirical distance estimators are limited by the paucity of high-quality
input/training data. The situation is improving (CfA, CSP, KAIT, SNF, SDSS),
but we need better, homogeneous data at low/intermediate redshifts and a
systematic approach to ingesting them to build better empirical estimators.
Will current ground-based SN surveys deliver what we need for JDEM?
2
Cosmic
Acceleration
Discovery from
High-redshift
SNe Ia
SNe at z~0.5 are
25% fainter than in
an open Universe
with same value of
m
Technological Redshift
Desert:
Possible photometric
offsets between lowand high-redshift data
Desert
still there
10 years
later
 = 0.7
 = 0.
m = 1.
4
SDSS II Supernova Survey Goals
• Obtain few hundred high-quality SNe Ia light curves in
the `redshift desert’ z~0.05-0.4 for continuous Hubble
diagram
• Spectroscopic follow-up for redshifts, SN typing, and to
study diversity of SN features
• Probe Dark Energy and systematics in redshift range
complementary to other surveys
• Well-observed, homogeneous sample to anchor Hubble
diagram & train distance estimators
• Large survey volume: rare & peculiar SNe, probe outliers
of population to test SN models
5
Frieman, et al (2008); Sako, et al (2008)
Spectroscopic follow-up telescopes
R. Miquel, M. Molla
CfA team
P. Challis, G. Narayan, R. Kirshner
Searching For Supernovae
Search
g
Template
Difference
• 2005
– 118,693 objects scanned
– 10,937 unique
candidates
– 130 confirmed Ia
• 2006
r
– 14,430 scanned
– 3,694 candidates
– 193 confirmed Ia
• 2007
 13,613 scanned
 3,962 candidates
– 175 confirmed Ia
i
•Positional match to remove movers
•Insert fake SNe to monitor efficiency
B. Dilday
Redshift Distribution for SNe Ia
and counting
SDSS
SN
Lightcurves
Holtzman
et al (2008)
Well-sampled, multi-band light curves, including measurements before peak light
Spectroscopic Target Selection
2 Epochs
SN Ia Fit
SN Ibc Fit
SN II Fit
Sako etal 2008
Spectroscopic Target Selection
2 Epochs
31 Epochs
SN Ia Fit
SN Ia Fit
SN Ibc Fit
SN Ibc Fit
Fit with
template
library
Classification
>90%
accurate after
2-3 epochs
Redshifts
5-10%
accurate
SN II Fit
SN II Fit
Sako etal 2008
SN and Host Spectroscopy
MDM 2.4m
NOT 2.6m
APO 3.5m
NTT 3.6m
KPNO 4m
WHT 4.2m
Subaru 8.2m
HET 9.2m
Keck 10m
Magellan 6m
TNG 3.5m
SALT 10m
2005+2006
SDSS SN Ia Spectra
~1000 spectra taken over 3 seasons
Zheng et al (2008)
Fitting SN light curves I: MLCS2k2
• Multicolor Light Curve Shape (Riess et al '98; Jha et al '07)
• Model SN light curves as a single parameter family,
trained on low-z UBVRI data from the literature
• Assumes SN color variations are due to dust extinction, subject to prior
P(Av)
time-dependent model “vectors”
fit parameters
Time of maximum
distance modulus
dust law extinction
stretch/decline rate
MLCS2k2 model templates
Jha et al, 2007
∆ = -0.3: bright, broad
∆ = +1.2: faint, narrow
Fitting SN Light curves II: SALT2
Guy et al
• Fit each light curve using rest-frame spectral surfaces*:
light-curve shape
• Transform to observer frame:
color term
• Light curves fit individually, but distances only estimated globally:
Global fit parameters, determined along with cosmological parameters
*Not trained just on low-redshift data; distances are cosmology-dependent,
18
flat priors on model parameters
Light Curve Fitting with MLCS2k2 and SALT2
19
Monte Carlo Simulations
match data distributions
Use actual observing conditions (local sky, zero-points, PSF, etc)
Model Spectroscopic & Photometric Efficiency
Redshift
distribution for all
SNe passing
photometric
selection cuts
(spectroscopically
complete sample)
Data
Need to model
biases due to
what’s missing
Difficult to model
spectroscopic
selection
Extract AV Distribution from SDSS
(no prior)
Extract RV distribution from
SDSS SN data
RV 
AV
2
E(B  V )
• MLCS previously
used Milky Way
avg RV=3.1
• Lower RV more
consistent with
SALT2 color law
• Not conventional
dust
Preliminary Cosmology Results
w
open
Kessler, Becker, et al. 2008
Issues with rest-frame U band
epoch
•Data vs. SALT2 Model Residuals
•Similar Low-z vs. High-z discrepancy seen in MLCS
•MLCS trained only on Low-z, SALT2 model dominated by SNLS
•Similar differences seen in rest-frame UV spectra (Foley et al)
25
SN Ia vs. Host Galaxy Properties: I
Bright
SN Luminosity/Decline Rate
Faint
Smith
26 et al
SN Ia vs. Host Galaxy Properties: II
Is reddening
local to the
SN environment?
Color/reddening
Smith
27 et al
SN Ia vs. Host Galaxy Properties: III
Preliminary
Two SN Ia
Populations?
Implications
for SN
cosmology:
host-galaxy
population
evolution
Smith
28 et al
Future: Improved SN Ia Distances