“The Dark Side of the SDSS”

1. “The Dark Side of the SDSS” Bob Nichol ICG, Portsmouth Thanks to all my collaborators on SDSS and other teams

2. Outline • A very brief overview of Dark Energy • A very brief overview of the SDSS • SDSS searches for the “Dark Side” • SDSS SNe • ISW effect • Cosmic magnification • Baryon acoustic oscillations (BAO) • Future experiments

4. Largest oscillations that are causally connected WMAP: Universe at 380,000 yrs

5. SNe and CMB force us into a Universe ~75% DE and ~25% DM. But is this true? What is DE? SN CMB (DARK) ENERGY (DARK) MATTER

6. Understanding Dark Energy (The billion dollar question) To confirm DE we need to observe it in as many ways as possible, but there are only two broad avenues: • Geometrical tests (distances, volumes) • Growth of structure (cluster counts) To determine what DE is, we can make progress on two simple questions: • Is DE just a cosmological constant (w(z)=-1)? (Push observations to higher redshifts) • Is DE a new form of stress-energy with negative effective pressure or a breakdown of General Relativity at large distances? (Study DE using different probes) As we don’t know much, all observations are important

7. DGP Cosmologies Fairbairn & Goobar 2005 Also, Sawicki & Carroll (2005) & Koyama (2006) show there are noticeable differences in the evolution of structure in DGP models. This maybe testable! DGP model for 5D gravity

8. SDSS DR4: 849k spectra, 6670 sq degs Done 07/2005: ~700,000 redshifts, 8000 sq degs Extension (2005-2008): Legacy, SNe, Galaxy

9. SDSSII SNe SurveyExploring DE & SNe at an epoch when DE dominates • Type Ia supernovae (SNe) • spectroscopically confirm and obtain “well-measured” light curves of ~200 SN Ia from z = 0.05 to ~ 0.4 • bridge low-z (z<0.05; LOSS, SNF) and high-z (0.3<z<1.0; ESSENCE, SNLS) sources • understand and minimize systematics of SN Ia as distance indicators • SN Ib/c, II, rare types • Other transients Riess et al. (2004) compilation Astier et al. (2005)

10. Survey Area N S Use the SDSS 2.5m telescope • September 1 - November 30 of 2005-2007 • Scan 300 square degrees of the sky every 2 days • discover supernovae and obtain multi-color light curves Follow-up ARC HET MDM WHT Subaru (NTT)

11. Photometric Typing • Color-type SN candidates using nightly g r i data: • make template light curves from multi-epoch spectra (Peter Nugent) and other sets of spectra of well-observed historical SNe (SUSPECT database) • Ia, Ia-pec, II-P, II-L, IIb, Ibc, Ibc-hypernova • fit for redshift, extinction, stretch for Ia • Able to type with >90% efficiency after ~2 - 4 epochs SN2005hy Ia II Ia II

12. Team of 15 “hand-scanner” visually inspected 144,000 objects selecting nearly 10k SN targets!

13. Results from 2005 • 126 spectroscopically confirmed SN Ia • 13 spectroscopically probable SN Ia • 6 SN Ib/c (3 hypernovae) • 10 SN II (4 type IIn) • 5 AGN • ~hundreds of other unconfirmed SNe with good light curves (galaxy spectroscopic redshifts measured for ~25 additional Ia candidates) • Focused primarily on Ia <z> = 0.21

14. 2005 spectroscopically confirmed + probable SN Ia

15. Lambda = 0.74 Preliminary No reddening =0.27 w) = 0.1 from SDSS+ESSENCE+WMAP+LSS (statistical errors only, constant w, flat Universe) SDSS data on host galaxies will allow study the scattering in this relation in greater detail Important training set for next generation of imaging surveys e.g. DES will detect 3000 SN by ~2010

16. Late-time Integrated Sachs Wolfe (ISW) Effect • DE also effects the growth of structure i.e. Poisson equation with dark energy: • In a flat, matter-dominated universe (CMB tells us this), then density fluctuations grow as: • Therefore, for a flat geometry, changes in the gravitational potential are a direct physical measurement of Dark Energy as they should be non-evolving if DE=0

18. Experimental Set-up See also: Nolta et al, Boughn and Crittenden, Myers et al, Ashfordi et al

19. WMAP vs SDSS WMAP W band temperatures across 50% of SDSS area Density of Luminous Red Galaxies (LRGs) selected from the SDSS

20. Searching for a detection LRG selection to z~0.8 (Eisenstein et al. 2001) 5300 sq degrees Achromatic (no contamination) Errors from 5000 CMB skies Compared to a null result >95% for all samples Low redshift sample contaminated by stars Individually >2s per redshift slice 4 redshift shells (not significant overlap) ISW and the SDSS Overall, we have detected signal at 5s Yellow: “smoothed clean”, Black: “Clean”, Red: Q, Blue: W, Green: V

21. ISW Predictions • Halo model. Biasing of b=1,2,3 & 4 for LRGs • Plus SZ on small scales • Data prefers DE model over null hypothesis at the >99% confidence for all combinations • The measurement is very sensitive to n(z) assumed and m Scranton et al 2003

22. dg/dz is a very powerful probe Only probe of DE clustering (Hu & Scranton 2004; Pogosian 2004) and highly complementary to geometrical measures of DE (SNe etc) Circa 2006 (SDSS) 8000 sq degrees (≥3s per redshift) Tighter redshift intervals (> 5 bins) Beyond ASTRO-F all-sky out to z~1.5 (>4.5s detection if there!) UKIDSS+VISTA all-sky (LRG selection to z>1) QSO catalogs (z out to 3) Cooray, Huterer, Baumann 2003 Future ISW directions

23. Dark Energy Survey (DES) • 5000 sq deg multiband survey of SGP using CTIO, • 40 sq deg time domain search for SNe • The survey will study the expansion history of the universe and the growth of density perturbations using four distinct techniques: • 4000 sq deg survey in collaboration with the SPT • weak lensing study • galaxy angular power spectrum distance measurement study • SNe Ia distance measurement study Each will independently constrain the dark energy eqn of state ~10% ISW with DES+Planck is as good as SNAP for non-constant w (Pogosian et al. 2005)

24. Cosmic Magnification Gravitational magnification increases flux received from galaxies and hence allows us to see fainter galaxies, resulting in an increased apparent galaxy number density. But, it also magnifies the solid angle of the projected lensed sky which results in a decrease in the apparent galaxy number density. Therefore a competition between the two! more solid angle more flux

25. more sources come in than diluted less source come in than diluted Effects cancel

26. Hunting for quasars Quasi-stellar sources: by definition they look like stars! Traditional approaches have used UVX approach to finding quasars, i.e., quasars are “very blue” so can be isolated in color-color space using simple hyper-planes (see Richards et al. 2002). However, there is significant contamination (~40%), thus demanding spectroscopic follow-up which is very time-consuming.

27. Probabilistic approach • Use Kernel Density Estimation (KDE) to map color-color space occupied by known stars and quasars (“training sets”) • Use cross-validation to “optimal” smooth the 4-D SDSS color space and obtain PDFs • Fast implementation via KD-trees (Gray & Moore) • ~16,000 known quasars and ~500000 stars • Using a non-parametric Bayes classifier (NBC)

28. F stars Stars additional cut 95% complete 95% pure QSOs

29. Now fully consistent with LCDM 13.5 million galaxies 195,000 quasars 8 detection Blue points= data Black line = best fit Red line= best fit + alpha Grey shading= 1sigma

30. Baryon Oscillation • Gravity squeezes the gas, pressure pushes back! They oscillate • When the Universe cools below 3000K these sound waves are frozen in Courtesy of Wayne Hu

31. Cosmic Microwave Background • Effect of this sound wave already discovered in relic light of the early universe • That was the Universe at 400,000 years. Can we see these sound waves today?

32. 500 Million Light Years A slice of the SDSS Credit: SDSS 700,000 light years

33. The Correlation Function The correlation function is the probability of finding pairs at a given separation, above that of a random distribution. Excess of galaxies separated by 500 million light years

34. What does it mean? • We have detected the sound wave in the Universe at two very different epochs (400,000 yrs after Big Bang and present-day). This is important because our theory of gravitational structure formation predicts that such features should have been preserved. Detecting the sound wave in the galaxies is the “SMOKING GUN” that our theory is correct. • Better yet, the sound wave is an object of fixed size, a “standard ruler” or “cosmic yardstick”. This means that we can measure its apparent size anywhere in the Universe, and determine how far it is away because we know its true size.

35. Looking back in time in the Universe CMB SDSS GALAXIES FLAT GEOMETRY FLAT GEOMETRY CREDIT: WMAP & SDSS websites

36. Looking back in time in the Universe Looking back in time in the Universe CMB SDSS GALAXIES FLAT GEOMETRY OPEN GEOMETRY CREDIT: WMAP & SDSS websites

37. Looking back in time in the Universe Looking back in time in the Universe CMB SDSS GALAXIES FLAT GEOMETRY CLOSED GEOMETRY CREDIT: WMAP & SDSS websites

38. UNIVERSE IS FLAT TO 1% PRECISION

39. WFMOS A quantum leap in spectroscopic efficiency. Thousands of fibres over a 1.5 degree field-of-view on an 8-meter class telescope: over an order of magnitude increase in mapping efficiency of 2dF KAOS purple book (Seo, Eisenstein, Blake, Glazebrook) z~1 survey with 2 million galaxies with twice LRG volume 1% accuracy Will get w to <5% and w’ to <20%

40. Conclusions • SDSS continues for 3 more years and has been successful in finding many hundreds of SNe. • The quality and quantity of SDSS data has provide several complementary detections of dark energy and dark matter, e.g., the ISW effect provided direct physical evidence that DE exists • Detected cosmic magnification and consistent with LCDM. Powerful new probe of the Universe • SDSS has detected the baryon oscillations in the local Universe, the “missing link” between CMB and LSS. Now have a “standard ruler”