The Case for Massive Spectroscopy in the South

Josh FriemanMS-DESI Meeting, LBNL, March 2013

Some details in DESpec White Paper arXiv: 1209.2451 (Abdalla, etal)

2

Motivation• What is the physical cause of cosmic acceleration?

– Dark Energy or modification of General Relativity?• If Dark Energy, is it Λ (the vacuum) or something else?

– What is the DE equation of state parameter w?• The DE program would be substantially enhanced

by a massive redshift survey that optimally synergizes (overlaps) with the DES imaging survey and by a larger redshift survey of LSST galaxies, capitalizing on those investments. Maximum overlap requires a southern site.

3

• Probe dark energy through the history of the expansion rate:

• and the growth of large-scale structure:

• Weak Lensing cosmic shear Distances+growth Imaging

• Supernovae Distances Imaging• Cluster counting Distances+growth Imaging

Spectroscopy• Baryon Acoustic Oscillations Distances and H(z) Imaging Spectroscopy• Redshift Space Distortions Growth Spectroscopy

What can we probe?

Distance vs.Redshift

Growth of DensityPerturbations

4

Massive Spectroscopic Surveys in the Southern Hemisphere

• 8-million Galaxy Redshift Survey in ~250 nights– Uniformly selected from deep, homogeneous DES+VHS

imaging over 5000 sq. deg. (2018+)• 15-million Galaxy Redshift Survey in ~500 nights

– Uniformly selected from deep, homogeneous LSST imaging of additional 10,000 sq. deg. (2021+) or from DES extension

• Photometric+Spectroscopic Surveys over same sky– Enable powerful new science beyond what either can

provide alone• Deep, uniform multiband imaging from DES, LSST

– Enable efficient, well-understood selection of spectroscopic targets, control systematic errors

5

Massive Spectroscopy of DES and LSST Targets Enables New and Improved DE Probes

• Weak Lensing and Redshift-Space Distortions– Powerful new test of Dark Energy vs Modified Gravity

• Galaxy Clustering– Radial BAO for H(z) and improved DA(z)

• Photometric Redshift Calibration– Determine DES and LSST N(z) from angular correlation, improve

DE constraints from all methods in these imaging surveys• Galaxy clusters

– Dynamical masses for DES/LSST clusters from velocity dispersions, reduce the major cluster DE systematic

• Weak Lensing– Reduce systematics from intrinsic alignments for DES/LSST

• Supernovae– Reduce systematics from host-galaxy typing for DES/LSST

6

Slide from Enrique Gaztanaga

RSD, BAO determine galaxy survey density~1500 per sq deg to z~1(for nP>1)

slide from Enrique Gaztanaga

7

Weak Lensing and Redshift Space Distortions

• Powerful test of Dark Energy vs. Modified Gravity• RSD from MS-DESI

– Measures degenerate combination of growth f and bias b• Weak Lensing from DES and LSST

– Helps break degeneracy• RSD and WL over same sky

– RSD, shear-shear, plus galaxy-shear correlations

MacDonald & Seljak, Bernstein & Cai, Cai & Bernstein, Gaztanaga, etal, Kirk, et al (in prep)

8

• Constraints much stronger if imaging and spectroscopy cover same sky: galaxy-shear cross-correlations

Weak Lensing and Redshift Space Distortions: DETF FoM for 15,000 sq. deg. surveys

LSST WL MSDESI RSD+BAO255 113

Combine:Different sky 1126Same sky 4240

Assuming GR Not Assuming GRM. Eriksen

9

• Constraints much stronger if imaging and spectroscopy cover same sky: galaxy-shear cross-correlations constrain bias

Weak Lensing and Redshift Space Distortions: Jointly Constraining DE and Gravity

Assuming GR

LSST WL MS-DESI RSD+BAO

Combined: Different Sky Same Sky

10

Cai & Bernstein

Relative Gain in Modified Gravity Figure of Merit from Same Sky

Note WL data not available in the north

LSST WL

MS

DE

SI R

SD

DES WL

11

2-parameter Modified Gravity model

Assuming GR Modified Gravity constraints

Kirk etal

Different SkySame Sky

Different SkySame Sky

12

DES and LSST Photo-z Calibration

DES-BigBOSS Joint Working Group Report

Angular Cross-Correlation of Imaging with Redshift Survey

Requires same sky coverage of imaging and spectroscopy, improves with overlap area

Photo-z systematics could otherwise limit DES, LSST Dark Energy reach

13

ClustersNumber of clusters above mass threshold

Dark Energy equation of state

• Spectroscopy of DES/LSST Clusters • Determine Cluster velocity dispersion (dynamical mass) using 10’s of redshifts per cluster calibrate mass-richness relation: complement WL, SZ, and X-ray estimates

Mohr

14

Slide from Sarah Hansen

DETFFOMgain for clusters

15

Dark Energy Spectrograph Concept for MS-DESI

• 4000-fiber optical spectrograph system for the Blanco 4m• Mohawk robotic fiber positioner

– Based on Echidna system, has demonstrated requisite pitch• High-throughput spectrographs• Fibers tile full 3.8 deg2 DECam Field of View• Fiber positioner rapidly interchangeable with DECam imager

– Maintain wide-field imaging capability for the Blanco• Use much of the DECam infrastructure installed on Blanco

– Prime focus cage, hexapod, 4 of the 5 optical corrector elements, shutter

• DESpec White Paper– arXiv: 1209.2451 (Abdalla, etal)

DECam Prime FocusCage Installed on Blanco Telescope

DECam Prime FocusCage Installed on Blanco Telescope

+DESpec

Saunders, etal

High, dry site: 80% useable nights, 0.75” site seeingNext door to LSST and Gemini.

18

Blanco Telescope

Cerro Tololo

19

Dark Energy Spectroscopic Survey

• Redshift Survey optimized for– Baryon Acoustic Oscillations– Redshift Space Distortions

• Target DES+VHS Galaxies (from grizYJHK colors, fluxes)−19 million Emission Line Gals (to z~1.5, BAO) 1200/sq deg−4 million Luminous Red Gals (to z~1.3, RSD) 300/sq deg

• Strawman Survey Design−2 exposures each field to reach target density and high

completeness (1500 successful redshifts per sq. deg.)−20-min cumulative exposure times to reach requisite depth

(40 min for QSOs)−~750 total nights to cover 15,000 sq. deg.

DES deepmultiband image

Excellent spectrosocopictarget sourceand WL shapes

Deep targetingimages with WLover large area not availablefrom Northernhemisphere

21

Target Selection Simulations

Deep, homogeneous parent catalogs from DES, VHS, LSST enable efficient selection and sculpting of redshift distributions

ELG goal:1200/deg2

LRG goal:300/deg2

22

Target Selection Simulations

Deep, homogeneous parent catalogs from DES, VHS, LSST enable efficient selection and sculpting of redshift distributions

ELG goal:1200/deg2

LRG goal:300/deg2LRG goal:300/deg2

23

Conclusions• Two hemispheres better than one • Southern hemisphere has critical science advantages:

– DES and LSST photometric surveys for DE synergy (WL+RSD, clusters, photo-z cal) and deep, uniform target selection (Cf. SDSS): Figures of Merit increase by factor 2-3 with same sky

– Synergy with other southern facilities as well (SPT, SKA, …)• MS-DESI on the Blanco would capitalize on existing,

installed, tested DECam infrastructure– Reduce cost and technical and schedule risks– Fiber system interchangeable with DECam maintains Blanco

imaging capability into the LSST era and provide world-class imaging plus spectroscopy facility for the astronomy community