EMU: Evolutionary Map of the Universe

Ray Norris13 September 2011

Goals of this meeting:

• update people on what’s happening• get input into future directions• tap into local expertise• hope to get plenty of round-table discussion as well as

presentations

if you are not yet part of EMU and you would like to be, please ask me!

ASKAP=Australian SKA Pathfinder

• A$170M (=€120m) project now under construction in Western Australia

• Completion early 2013

• 36*12m antennas

• Antennas have a 92-pixel phased array feed (PAF)• 30 sq. deg FOV!

ASKAP Design Specifications

• Number of antennas 36 (630 baselines)• Antenna diameter 12 m (3 axis)• Maximum baseline 6 km• Cont. Angular resolution 10 arcsec• Sensitivity 65 m²/K• Frequency range 700 – 1800 MHz

• Focal plane phased array 188 elements (92 dual pol)

• Field of view 30 deg²• Processed bandwidth 300 MHz• Number of channels 16 384

Current major 20cm surveysNVSS75% of skyrms=450μJy

EMU75% of skyrms=10μJy

EMU+WODAN100% of sky

Increasing sensitivity

Increasin

g area

Perth

Sydney

Brisbane

Melbourne

Adelaide

Darwin

Alice SpringsWESTERNAUSTRALIA

NORTHERNTERRITORY

SOUTHAUSTRALIA

QUEENSLAND

NEW SOUTHWALES

VICTORIA

TASMANIA Hobart

ACT Canberra

Murchison RadioAstronomy Observatory

INDIAN

OCEAN

CORAL

SEA

SOUTHERN OCEAN

Perth

Sydney

Brisbane

Melbourne

Adelaide

Darwin

Alice SpringsWESTERNAUSTRALIA

NORTHERNTERRITORY

SOUTHAUSTRALIA

QUEENSLAND

NEW SOUTHWALES

VICTORIA

TASMANIA Hobart

ACT Canberra

Murchison RadioAstronomy Observatory

INDIAN

OCEAN

CORAL

SEA

SOUTHERN OCEAN

Australia – WA – Midwest – Murchison

Perth

Sydney

Brisbane

Melbourne

Adelaide

Darwin

Alice SpringsWESTERNAUSTRALIA

NORTHERNTERRITORY

SOUTHAUSTRALIA

QUEENSLAND

NEW SOUTHWALES

VICTORIA

TASMANIA Hobart

ACT Canberra

Murchison RadioAstronomy Observatory

INDIAN

OCEAN

CORAL

SEA

SOUTHERN OCEAN

Murchison

gazetted towns: 0

population: “up to 160”

Geraldton

Perth

Sydney

Brisbane

Melbourne

Adelaide

Darwin

Alice SpringsWESTERNAUSTRALIA

NORTHERNTERRITORY

SOUTHAUSTRALIA

QUEENSLAND

NEW SOUTHWALES

VICTORIA

TASMANIA Hobart

ACT Canberra

Murchison RadioAstronomy Observatory

INDIAN

OCEAN

CORAL

SEA

SOUTHERN OCEAN

Murchison

gazetted towns: 0

population: “up to 160”

Geraldton

Configuration and uv-coverage

• Antenna configuration• 2-km core with 30 antennas

• 6 antennas further out

−50°

−10°

Gupta et al. 2008, ATNF ASKAP Memo 21

36 Antenna Array Configuration

Spectral Line:Inner 30 dishes only, resolution 30 arcsec

(all at 1.4 GHz)

Continuum: uniform weighting:resolution 10 arcsec

Natural weighting: resolution 20 arcsec

Sensitivity of EMU to a 1 Mpc cluster halo:

Antennas

Antennas built by CETC54 (China)

Delivered and assembled on site

Antenna 1 delivered late 2009

Antennas

Antennas 2-6

ASKAP – PAF: 188 element unit

First 2 full-size ASKAP PAFs

Data Transport

• Direct burial of 300km fibre (Boolardy-Geraldton) started Oct 2010

• Now complete(June 2011)

• Includes 3 repeater huts

• NBN network Perth-Geraldtonalso completed June 2011

BETA: Boolardy Engineering Test ArrayBETA construction on schedule

ASKAP current status

• Construction is on schedule• 36 antennas in place by end of 2011

• fibre all in place

• eVLBI to NZ conducted in July 2011

• Parkes PAF is performing well• BETA will be available end of 2011/early 2012

PAF performance

• Performing very well• Tsys < 50K at low freq• Tsys ~ 100K at high freq

• can be fixed

• End of 2011:• 36 antennas on site• all hardware on site for 6-antenna BETA array

• Early 2012: BETA commissioning starts

• 2012-2013 ASKAP Design enhancement• Use experience from Mki PAF to develop MkII PAF• Better PAFs at lower cost

• March 2013: Science-ready ASKAP available• minimum of 12 antennas equipped with PAFs

• 2013… Antennas continue to be equipped with PAFs as funding permits

• A good target would be: • All antennas equipped with upgraded PAFs by Dec 2013

ASKAP – Possible schedule

The Science38 proposals submitted to ASKAP

2 selected as being highest priority

8 others also supported

• EMU all-sky continuum (PI Norris)

• WALLABY all-sky HI(PI Koribalski & Staveley-Smith)

• COAST pulsars etc• CRAFT fast variability• DINGO deep HI• FLASH HI absorption• GASKAP Galactic• POSSUM polarisation• VAST slow variability• VLBI

• Deep radio image of 75% of the sky (to declination +30°)• Frequency range: 1100-1400 MHz• 40 x deeper than NVSS

• 10 μJy rms across the sky• 5 x better resolution than NVSS (10 arcsec)• Better sensitivity to extended structures than NVSS• Will detect and image ~70 million galaxies at 20cm• All data to be processed in pipeline• Images, catalogues, cross-IDs, to be placed in public domain• Survey starts 2013• Total integration time: ~1.5 years ?

Complementary radio surveys

• Westerbork-WODAN• using Apertif PAF on Westerbork telescope

• will achieve similar sensitivity to EMU

• will observe northern quarter of sky (δ>+30°)

• well-matched to EMU

• LOFAR continuum surveys• lower frequency

• covering Northern half(?) of sky

• valuable because yields spectral index

• Meerkat-MIGHTEE• Potentially deeper over smaller area, but will be limited by

confusion until Meerkat Phase II (2016?)

ATLAS=Australia Telescope Large Area Survey

Slide courtesy of Minnie Mao

The role of ATLAS as a testbed for EMU

ATLAS • has the same rms sensitivity (10uJy) as EMU

• has the same resolution (10 arcsec) as EMU

• covers (only!) 7 sq. deg.

• has (2000) 16000 galaxies

We’re using ATLAS to test many aspects of EMU• Imaging (dynamic range, weighting)

• Source extraction

• Cross-identification

• Source database and VO server

• Science!

Science Goals

Redshift distribution of EMU sources

Based on SKADS (Wilman et al.2006)

• Evolution of SF from z=2 to the present day,• using a wavelength unbiased by dust or molecular emission.

• Evolution of massive black holes • and understand their relationship to star-formation.

• Explore the large-scale structure and cosmological parameters of the Universe.• E.g, Independent measurement of dark energy evolution

• Explore an uncharted region of observational parameter space• almost certainly finding new classes of object.

• Explore Diffuse low-surface-brightness radio objects• Generate an Atlas of the Galactic Plane• Create a legacy for surveys at all

wavelengths (Herschel, JWST, ALMA, etc)

Science Goals

To trace the evolution of the dominant star-forming galaxies from z=5 to the present day, using a wavelength unbiased by dust or molecular emission.

Science Goal 1:

SFR measurable (5σ) by EMU

Arp220z=2

M82z=0.4

Milky Wayz=0.1

With 45 million SF galaxies, can stack to measure SFR to much higher z

HyperLIRGz=4

Measured radio SFR does not need any extinction correction

Region occupied by unidentified sub-mJy radio sources?

(Fro

m H

opk

ins

et a

l 200

4, B

arg

er e

t al 2

000

)

What dominates SFR at each z?

Present Day

Redshift 0 1 2 3 4 5

13.7 6 5 4 3 2 1

Time since Big Bang (Billions of years)

Science Goal 2:

• To trace the evolution of massive black holes throughout the history of the Universe, and understand their relationship to star-formation.

EMU will detect 25 million AGN

Other questions:• How much early activity is obscured from optical views?• Can we use trace the evolution of MBH with z?• When did the first MBH form?

• We will detect rare objects, such as• high-z AGN• composite AGN/SF galaxies• galaxies in a brief transition phase from quasar-mode to radio-mode accretion.

• Norris et al. 2008, arXiv:0804.3998

•S20cm= 9mJy• z = 0.932

• L20cm= 4 x 1025 WHz-1

F00183-7111 (ULIRG with L=9.1012 Lo)

Merger of two cool spirals:

• SB just turned on - AGN just turned on

• radio jets already at full luminosity, boring out through the dust/gas

• Almost no sign of this at optica/IR wavelengths

• see Norris et al. arXiv:1107.3895

20kpc 1 kpc

P=6.1025 W/Hzz=0.327

z=0.22. From Mao et al 2010,MNRAS, in press.

• Radio surveys are unbiased by dust & sky lines

• How similar are the cosmic webs of AGNs and SF galaxies?

• Did they have a common origin?

• We can use head-tail galaxies as probes of clustering to high z.

• We should detect Integrated Sachs-Wolfe (ISW) effect directly, so testing the scale of Dark Energy at z > 1

Science Goal 3: To use the distribution of radio sources to explore the large-scale structure and cosmological parameters of the Universe.

Integrated Sachs-Wolfe EffectF

rom

http

://ifa

.ha

wa

ii.e

du/c

osm

ow

ave

/su

per

void

s/

~10°

Science Goal 3: Cosmoology and Fundamental Physics

• EMU (70 million galaxies) has fewer objects than DES (300 million) but samples a larger volume

• => complementary tests

EMU

Dark Energy

See Raccanelli et al. ArXiv 1108.0930“Current error ellipse” is based on Amanullah et al., 2010, ApJ, 716, 712, plus Planck data

Modified Gravity

See Raccanelli et al. ArXiv 1108.0930

6.1 mJy at 20 cm < 5 µJy at 3.6µm

Norris et al 2007, MNRAS, 378, 1434; Middelberg et al 2008, AJ, 135, 1276; Garn & Alexander, 2008, MNRAS,391,1000; Huynh et al.,2010, ApJ, 710, 698; Norris et al. 2011, ApJ, in press

Science Goal 4: To explore an uncharted region of observational parameter space, almost certainly finding new classes of object.

• SERVS=>1400 hours on Spitzer• 3.6 μm median stacked image • Median of 39 images • Flux density = 0.21 +0.14 μJy. • Pixel size = 0.6 arcsec.

SERVS Warm Spitzer observations of IFRS

3.6 μm Flux density = 0.21 +0.14 μJy.

SERVS=>1400 hours on SpitzerMedian of 39 3.6 μm imagesNorris et al. 2011, ApJ, in press.

What is the density of new discoveries in parameter space?

• ATLAS pushed the boundaries by only a factor of a few, yet discovered two new classes of objects (PRONGS, IFRS).

• What happens when we push the boundary by a factor of 40?

Lim

it of

con

vent

iona

l rad

io-te

lesc

opes

SKA

path

finde

rs

New WG: Discovering the Unexpected(WTF: Widefield ouTlier Finder)

Instead of hoping to stumble across new types of object, we will systematically mine the EMU database, discarding objects that already fit known classes of object based on their:

• morphology

• spectral index

• polarisation

• SED in optical/IR

• etc

Objects that remain will be either• processing artefacts (important for quality control)

• statistical outliers of known classes of object (interesting!)

• New classes of object (WTF)

Science goal 6: Produce the most complete catalogue of the Galactic Plane to date.

Much deeper and higher res than any other survey:

• CGPS: arcmin, few mJy, 73° of Northern plane

• SGPS: arcmin, 35 mJy, most of S plane

• MAGPIS: 6 arcsec, 1-2 mJy, 27° of N plane

• EMU: 10 arcsec, down to 50 μJy, most of plane

• all of plane when linked to WODAN

• Build a complete census (and possibly discover new types of):

• all phases of HII region evolution

• the most compact and youngest supernova remnants

• radio-emitting Planetary Nebulae to constrain galactic density and formation rate

Helfand et al 2006, AJ 131, 2525.

Science goal 7: Radio stars

HR diagram for 420 radio stars (Gudel, 2002)

Goals• Increase # of known radio stars by 10-100

• Discover new types and define typical populations (unbiased)

• Identify trends and correlations• current samples too small

Algol: Mutel et al 2009

•Understand stellar magnetic activity

•Understand coherent emission mechanisms

New EMU leadership structure

Ray NorrisProject Leader

Ray NorrisProject Leader

Andrew HopkinsProject ScientistAndrew HopkinsProject Scientist

Ilana FeainProject Scientist

Ilana FeainProject Scientist

Nick SeymourProject ScientistNick Seymour

Project Scientist

ScienceWorking Groups

ScienceWorking Groups

Design Study Working Groups

Design Study Working Groups

Design Study Working Groups

• Observing Strategy (Shea Brown)• Commissioning, BETA, Observing Process (Ilana Feain)• Simulations (Emil Lenc)• Data processing Pipeline (Tom Franzen)• Compact Source Extraction (Andrew Hopkins)• Extended Source Extraction (Tom Franzen)• Quality Control (Lisa Harvey Smith)• Automated cross-IDs (Loretta Dunne)• Galaxy Zoo Cross-IDs (Julie Banfield)• Data Requirements (Ray Norris)• Data Access and VO (Minh Huynh)• Redshifts (Nick Seymour)• WTF (Ray Norris)

EMU Roadmap

Proposed changes to design study WGs:

• more, smaller, groups• WG chairs to lead their WG

actively• EMU memo series• Progress will be marked by

milestones• Major goals: written

reports/papers• Management team will regularly

review progress of WG

Science Working Groups

Currently unchanged:• Evolution of Galaxies (Nick Seymour)• Cosmology (Matt Jarvis/Shea Brown)• Clusters (Melanie Johnston-Hollitt)• Galactic Plane (Mark Thompson)• Radio Stars (Gracia Umana)• Diffuse Low Surface Brightness objects (Shea Brown)• Stacking (Jose Afonso)

Terminated:• Simulated Radio Sky (=> simulations)• Application to existing data (=> ATLAS)• Collaboration with other SKA Pathfinders (=> SPARCS)

Science WGs

• Specific requests to science WGs

• Each WG to produce (at least) a paper/report on the science that will be done with EMU

• WG chairs need to involve members, not do things single-handedly!

Redshifts WG(WG chair Nick Seymour)

• Few of our 70 million galaxies will have spectroscopic redshifts

• Not enough photometry for good photometric redshifts

BUT• Have upto 9-band photometry for ~50% of EMU sources

• SkyMapper, SDSS, VHS, WISE

• Also have radio data that can constrain algorithm• polarisation, spectral index, morphology. radio/Kband ratio, FRC

• Expect to be able to produce rough “statistical redshifts” for most sources if we have a good training set

• Use 16000 ATLAS +3000 COSMOS sources to train algorithms• Proposing large spectroscopy programs on all ATLAS sources• Experimenting with different algorithms • Expect to classify in z bins: (e.g.) 0-0.5, 0.5-1, 1-2, 2-4, 4-8

Fraction of EMU sources detected at optical/IR

Solution: “statistical redshifts”

• Photometric redshifts are poor-mans spectroscopic redshifts• Statistical redshifts are poor-mans photometric redshifts.• Useful if you want to know the statistical properties of a

sample.

E.g. median z of EMU is 1.1• select polarised sources=> median z = 1.9• select steep-spectrum polarised source => median z =2+(?)

Radio/MIR ratio also correlates with z, so even a K-band non-detection is valauble

EMU Survey Design Paper (PASA, in press, http://arxiv.org/abs/1106.3219)

Western Australia