The Accelerating Universe and the Sloan Digital Sky Survey Supernova Search

1
The Accelerating Universe and the Sloan Digital
Sky Survey Supernova Search

• Jon Holtzman (NMSU)
• many collaborators (FNAL, U. Chicago, U.
  Washington, U. Penn., etc., etc.)

2
The Expanding Universe

• Recession velocities of astronomical objects can
  be measured using the Doppler shift
• Applied to galaxies, we find that all except the
  nearest galaxies are receding
• Recession velocities are proportional to the
  distance to objects --gt Hubble's law

3
Hubble's Law

• v H d (locally)
• To see that relation is linear only requires
  relative distances
• To determine the Hubble constant (H slope
  current rate of expansion), requires absolute
  distance measurements
• Hubble's law implies an expanding Universe

4
Cosmology and Einstein

• Einstein's theory of general relativity combined
  with assumption of homogeneous and isotropic
  universe is consistent with an expanding Universe
• Rate of expansion, however, changes with time
  depending on the contents of the Universe how
  much matter/energy there is
• With no matter, expansion rate is constant
• With matter, the expansion rate slows down with
  time
• Since Einstein didn't know about the expanding
  Universe, he also noted that an arbitrary term
  the cosmological constant -- could be added to
  the equations to allow for a non-expanding
  Universe

5
Expansion rate change with time for different
cosmological models note that different models
correspond to different ages of Universe
The figure above shows the scale factor vs time
measured from the present for Ho 71 km/sec/Mpc
and for Oo 0 (green), Oo 1 (black), and Oo
2 (red) with no vacuum energy the WMAP model
with OM 0.27 and OV 0.73 (magenta) and the
Steady State model with OV 1 (blue). The ages
of the Universe in these five models are 13.8,
9.2, 7.9, 13.7 and infinity Gyr. The recollapse
of the Oo 2 model occurs when the Universe is
11 times older than it is now, and all
observations indicate Oo lt 2, so we have at least
80 billion more years before any Big Crunch.
(from Ned Wright's cosmology page).
6
The Accelerating Universe

• Since we know there's matter in the Universe,
  everyone always expected that the rate of
  expansion has been decreasing the big question
  was always how fast the deceleration was, whether
  it would be enough to cause an eventual
  recollapse of the Universe, and what the inferred
  age of the Universe was
• But about ten years ago, observations of distant
  supernovae threw a very unexpected wrinkle into
  the picture

7
Supernovae as Cosmological Probes

• Certain types of supernovae type Ia --can be
  used as distance indicators
• Out to intermediate redshift (z1), SN are
  fainter than expected for decelerating (or even
  empty) Universe --gt they are farther away, so
  Universe has been expanding faster than expected
• Possible problem are SN at earlier times
  intrinsically fainter? Or is there gray dust?
• At highest redshifts (zgt1), SN are brighter than
  expected --gt probably rules out evolution.
• Universe was decelerating a while ago

8
Cosmological parameters (1)

• Supernovae constrain cosmological parameters via
  the redshift-distance relation
• Supernovae by themselves indicate the need for
  acceleration, but don't constrain cosmological
  parameters uniquely
• Multiple combinations of matter density and
  cosmological constant match SN data

9
Cosmological parameters (2)

• Complementary constraints on cosmological
  parameters can come from observations of objects
  of known size, or alternatively, from statistical
  power at some known size, via the
  redshift-angular size relation
• Such a size scale is imprinted on the
  matter/energy distribution because of acoustic
  oscillations in the growth of perturbations in
  the early Universe

10
Cosmological parameters WMAP

• Wilkinson Microwave Anisotropy Probe measures the
  cosmic microwave background
• Angular power spectrum measures acoustic peaks at
  recombination
• Size-redshift relation constrains cosmological
  parameters
• Hubble constant measurements constrain things
  further

11
Cosmological parameters BAO

• Acoustic oscillations are also imprinted on the
  large scale galaxy distribution (baryon acoustic
  oscillations), since this evolves from the
  initial density perturbations
• Feature in the galaxy power spectrum has been
  observed in SDSS galaxy sample (Eisenstein et al
  2005) typical redshift z0.35
• Location of peak places strong constraint on
  matter density

12
Cosmological Parameters summary

• Constraints from
• Supernovae
• WMAP
• BAO
• Hubble constant
• All observations together lead to concordance
  model

13
Dark energy

• What causes current acceleration?
• For lack of knowledge, call it dark energy
• Dark energy is usually parameterized by its
  equation of state
  
  
• Cosmological constant has w-1 and unchanging
  could result from vacuum energy but amplitude way
  off from simple expectations
• Other models, e.g. quintessence, has w that
  varies with time
• Major observational goal measure w and its
  evolution !

14
The SDSS Supernova Survey goals

• Existing SN surveys have targetted either nearby
  or very distant SN
• nearby SN via targetted galaxy search
• distant SN via small field blind search
• neither technique gets intermediate redshift
  objects
• SDSS telescope/camera has very wide field,
  moderate depth --gt ideally suited for
  intermediate redshift
• Calibration uniformity is also an issue
  cosmology results depend on comparing low and
  high redshift samples, which are taken with
  totally different instruments/techniques
• SDSS bridges the gap
• look for continuity in redshift-dist relation
• uniform calibration
• evolution of w

15
Supernovae as distance indicators

• Several types of supernovae
• core collapse supernovae (type II, Ib, Ic)
• binary star supernovae (type Ia)
• None are standard candles however, type Ia SN
  are standardizable based on light curve shape
• Nagging problem we don't exactly know what type
  Ia supernovae are!

16
SDSS SN search techniques

• SDSS uses dedicated 2.5m telescope at Apache
  Point Observatory with very wide (corrected)
  field, very large format camera (30 science
  2048x2048 CCDs)
• SDSS drift scans across sky in 2.5 degree strip
  two strips fill the stripe
• SDSS SN survey looks at equatorial stripe during
  Sep-Nov 2006-2008, alternating strips each clear
  night roughly 50 Gbytes per night

17
SDSS-SN Discovery

• Candidate SN identified after subtracting
  template images taken earlier as part of main
  SDSS survey
• Automatic and manual identification both play a
  part
• Biggest contaminator is moving (solar system)
  objects partly removed by time lag between
  filters!

18
SDSS-SN followup

• Identification as type Ia supernovae requires
  spectroscopic followup
• Candidates identified by color selection very
  effective using 5 colors, 2 epochs (90)!

19
SDSS-SN followup spectroscopy

• Multiple larger telescopes used for spectroscopic
  followup

20
SDSS-SN results

• 129 confirmed type Ia's from 2005, 193 more from
  2006!
• target redshift regime well sampled

21
Photometry results lots of light curves!

• Photometry extracted using scene-modelling
  software developed at NMSU
• Light curve fitting in progress using a variety
  of techniques
• MLCS 2K2
• Modifications for fitting in flux
• Systematic effects being explored through
  Monte-Carlo

22
Photometry results lots of light curves!

• Light curve fitting in progress using a variety
  of techniques
• MLCS 2K2
• Modifications for fitting in flux
• Systematic effects being explored through
  Monte-Carlo

23
SDSS-SN Cosmology

• No obvious departures from concordance cosmology
• No discontinuity in Hubble relation

24
SDSS-SN Cosmology (2)

• In conjunction with other measurements (e.g.
  BAO), should provide constrain on w at moderate
  redshift

25
Other projects SN Ia rates

• Understanding SNIa rates important for
  understanding of nature of Ia progenitors (which
  is important for using Ia's as cosmological
  probes!)
• Rate measurement requires accurate understanding
  of experiment efficiency
• detection efficiency obtained by inserting fake
  SN during initial selection
• Sample efficiency from sophisticated light curve
  simulations
• Total low redshift efficiency 0.83 /- 0.02
  (stat) /- 0.01 (sys)
• SDSS sample ideal large numbers, blind search,
  well-defined (reasonably) sample definition
• Will get better with 3 year sample possible
  extension to higher z with photometric Ias

26
Other projects photometry-only Ia's

• Significantly more likely Ia light curves than
  those for which we have followup spectroscopy
• Figuring out how to use these will help with
  cosmology statistics, rate evolution, etc.
• Potential importance for future projects/missions

27
Other projects host galaxies

• Studying relationship of Ia properties to host
  galaxy properties may shed light on Ia
  progenitors and potential systematics
• Large samples, but also spatial resolution,
  required
• SDSS SN provides good sample
• HST and SIRTF proposals for followup also
  submitted

28
Other projects self-contained cosmology

• Currently, Ia light curve training done from
  nearby sample, but this is non-homogeneous and
  may not have well defined photometry
• Large sample of low-z SDSS SN may allow for
  self-consistent light curve training and
  application

29
Future directions

• Complete 2005 SDSS analysis, prepare for full
  sample analysis
• Variety of projects underway to understand and
  use type Ia SN
• Note SN Factory and possible NMSU 1m contribution
• Many new projects under development to contribute
  to understanding of dark energy
• JDEM (Joint Dark Energy Mission) space mission
• Mission concepts SNAP, DESTINY, JEDI
• DES (Dark Energy Survey)
• SDSS AS2 (After Sloan 2) one of the selected
  projects is a study that will find BAO at higher
  redshift