&

Photometric Redshiftswith Poisson Noise

Christian Wolf

Quenching Star Formation in Cluster Infall

Kernel regression‘correct error’

z/(1+z) ~ 0.006

λ / nm

QE (

%)

5 Mpc

5 M

pc

Principal Dataset:COMBO-17 + Spitzer + HST

I. Declining Star Formation

• COMBO-17 covered redshift 0..1

• Mapped luminosity evolution of galaxy population

• Build-up of stellar mass in red sequence

• COMBO-17 photo-z and spectroscopic surveys get identical results

• SF decline *not* driven by mergers(Wolf et al. 2005)

Hopkins et al. 2006

Faber, Willmer, Wolf et al. 2007Bell, Wolf et al. 2004Wolf et al. 2003

Declining Star Formation with Density

• COMBO-17 studied A901ab/902 (z=0.17)

• Dusty red galaxies = red spirals,see transformation in action, ongoing SF across the “gap”

• Slow quenching of star formation, *not* driven by mergers or starburst

Hopkins et al. 2006S0

SpE

frag

Koo

pm

ann

& K

en

ney

20

04K

enn

ey

et a

l. 2

008

Wolf et al. 2005, 2009& press release 2008

Old Red & Dusty Red Galaxies

Aspects of Transformation

Type Blue spiral Red spiral S0 galaxy

SFR-(280-U)rest

clumpiness

(U-V)rest

EB-V stars

EB-V SF

Examples

Physics of Star Formation Decline?

• Discussed origins of SF decline– Merger-triggered starbursts & exhaustion– Ram-pressure stripping of cold gas– Starvation of cold gas supply– Tidal effects from cluster potential– Galaxy-galaxy interactions– Gravitational heating in structure formation*

• Approaches– Detailed physical models challenging– Need better account of what’s there

• Are S0s made from present spirals?– Disk fading, light profiles, central SF– Outside-in truncation of SF disk,

dependence on cluster?– Cluster infall time scales from SFH

templates in fossil records

• Trends with redshift?– Higher-redshift clusters in comparison– Expect inversion of SF-density relation…

• Future observations– H imaging of A901/2 (OSIRIS via ESO)– Colour gradients in SDSS group galaxies– AO-assisted IFU study of intermediate-z

galaxies (SWIFT@Palomar)– Analysis of HIROCS clusters z~0.6 to 1.6

• Long-term approaches– AO & MCAO-assisted line imaging and

IFU spectroscopy of z~1..2 clusters (ESO VLT: MUSE & post-MAD devices)

– HST-based line imaging with WF3 NIR (J-band) narrow-bands of z~1 clusters

II. Photometric Redshifts and Classification

• COMBO-17– Many narrow bands– Template-based classification

Classes galaxy, star, qso, wd Reaches z/(1+z) < 0.01

Accurate rest-frame SED(and old red : dusty red)

λ / nm

QE (

%)

Wolf 2001

rms 0.008 7%-20%

outlier

rms 0.006

Abd

alla

et a

l. 20

07

Photometric Redshifts 2010+

• Trends– Before CADIS & COMBO-17

little science published from photo-z (only “experimentation”)

– Scientific value demonstrated!– Now: Strong drive towards

huge / all-sky surveys:• Dark Energy Survey,

PanStarrs

• VST, VISTA, LSST, IDEM …

– Beat Poissonnoise…

– But: impact of systematics?!

• Applications– Extremely large stand-alone

photo-z surveys– Search for interesting

spectroscopic targets

• To learn about– Cosmology & dark energy via

BAO and gravitational lensing– Galaxy growth embedded in

dark matter haloes– Galaxy evolution over time and

with environment– New phenomena

The Two Ways of Photo-”Z”eeing

• Model input: – Educated guesses– I.e. template SEDs and for

priors evolving luminosity functions etc.

• Photo-z output: – Educated guesses*– “What could be there? What is

it probably not?”

• Needed for– Pioneering new frontier: depth

in magnitude or z beyond spectroscopic reach

• Model input: – Statistical Distributions– I.e. empirically measured n(z)

distributions for all locations in flux space

• Photo-z output: – Statistical Distributions– “What is there? And in which

proportions?”

• Needed for– Extrapolating to wide area:

known mag/z territory with spectroscopic description

*You will never obtain a reliable estimate of an n(z) distribution from a technique that involves more than zero pieces of information which do not represent a true distribution but a best-guess approximation instead…

The Two Ways of Photo-”Z”eeing

• Model input: – Educated guesses– I.e. template SEDs and for

priors evolving luminosity functions etc.

• Photo-z output: – Educated guesses*– “What could be there? What is

it probably not?”

• Needed for– Pioneering new frontier: depth

in magnitude or z beyond spectroscopic reach

• Model input: – Statistical Distributions– I.e. empirically measured n(z)

distributions for all locations in flux space

• Photo-z output: – Statistical Distributions– “What is there? And in which

proportions?”

• Needed for– Extrapolating to wide area:

known mag/z territory with spectroscopic description

*You will never obtain a reliable estimate of an n(z) distribution from a technique that involves more than zero pieces of information which do not represent a true distribution but a best-guess approximation instead…

• Work continues on– Best templates– Best luminosity functions– Best code debugging– An engineering problem

taken care of by PHAT

• Work continues on – How to get n(z) correctly in

presence of errors?

• Problem solved:– Wolf (2009), MN in press– n(z) with Poisson-only errors

Kernel regression‘added error’

Kernel regression‘correct error’

ANNz Template fitting

Reconstruction of Redshift Distributionsfrom ugriz-Photometry of QSOs

Left panel: The n(z) of a QSO sample is reconstructed with errors of 1.08 Poisson noise(works with any subsample or individual objects as well) using “2-testing with noisy models”

State of the Art vs. Opportunity

• Current methods– Precision sufficient for pre-2010

science– But insufficient and critical for

the next decade

• Current surveys– VVDS and DEEP-2: 20%..50%

incomplete at 22..24 mag– SDSS (bright survey) 3…4%?

• Propagates into bias non-recov=0.2 (20% incomplete)

|z|=0.2 ! and w ~ 1 !

• Poisson-precision results need Adoption of Wolf 2009 method

removes methodical limitations Complete(!) “training set”

removes data limitations– Incompleteness systematics

• Issues – Lines in the NIR, weak lines at

low metallicity, what else?– Goal 1..5% incompleteness– Provide sub-samples with <1%

incompleteness– Fundamental limits? Blending?

Vision: The PRATER Project

“Photometric Redshift Archive for Training External Resources”

– Gaussian-precision photo-z code & residual risk quantification from model incompleteness/size moves all attention to model

– Residual outlier risk incomplete

– Noise floor on n(z) 2n(z) 1/Nmodel,z-bin

– Best model is (i) most complete and (ii) largest

Build: a world reference repository for post-2010 photo-zeeing

– Best-possible photo-z quality by design (method, completeness, size)

– Quality limits understood and communicated quantitatively

– Dynamic and growing with time

– Customers submit small “photo-zeeing” jobs over the web or ask for mirror installation in large applications (e.g. PanStarrs)

How to Make It Happen

• Collaboration with critical mass, expertise & telescope access

– O. LeFevre, Marseille (VIMOS)– J. Newman, S. Cruz (DEEP-2)– T. Shanks, Durham (XMS)– P. Prada, Granada (SIDE/GTC)– G. Dalton, Oxford (FMOS)

• Clarify incompleteness sources

• Coordinate multi-instrument plan to gather required data

– VIMOS upgrade at ESO-VLT– FMOS (NIR) at Subaru 2010– XMS at Calar Alto 2015?– SIDE at GTC 2015?

• First demonstrator:– Using zCOSMOS for SDSS

• 5-year goal:– Ingest best-possible VIMOS &

FMOS completeness– Perfect performance to 23 mag

• Drive next-gen instrumentation– Supported by ESA/ESO Report

on Fundamental Cosmology– Dark Energy Task Force (U.S.)

• 10-year goal:– Reach as deep as possible with

newest spectrographs, possibly space-based post-2020 (IDEM)

Summary

I. Cosmic decline in star formation• Over time and environment: not(!) driven by major mergers

• Star formation fading in cluster infall makes S0-like galaxies

• Interplay of fading process with structure formation…???

II. Photo-z’s or n(z|c) for future surveys • Method devised for Poisson-precision results (if model complete)

• Completeness of deep spectroscopy is primary quality limit

• Window of opportunity: PRATER project as world reference

• Community concerned and supportive, considering instrumentation

III. Stellar explosions• …see avalanche in available data, but follow-up and interpretation?

• Dark energy from SNe Ia: extinction correction dubious, metallicity?

III. Stellar Explosions

• Host galaxies SNe II : L-GRBs– If GRB Z* cut-off, then solar

• Cosmic SFR : SNR history– SN Ia delay time 0-4 Gyr

• Archival event search– GALEX UV flash before SN II

red giant progenitor– Radio: BBC, SWR, …

Stellar Explosions: Work 2010+

• Ages & Z* of SN subtypes

– Links to progenitor evolution

– Galaxy proxy, residual trend

• SN Ia light curves: 3-p family– Host galaxy age, dust & Z*?

– Cosmic calibration drift?

• Early light curves & flashes– Stacking, short cadence, SDSS,

PTF, SkyMapper, Calan 2010+

– Explosion conditions

Con

ley

et a

l. 20

08

Hamuy et al. 1993/5