sub-mm/mm observing techniquesthe evlaaug 14, 2006 the evla project sean dougherty national research...

Sub-mm/mm Observing Techniques The EVLA Aug 14, 2006

The EVLA Project

Sean DoughertyNational Research Council

Herzberg Institute for Astrophysics

Rick Perley & Michael Rupen (NRAO)

Peter Dewdney (HIA)


What is the Very Large Array (VLA)

SMD – Basic Radio AstronomySub-mm/mm Observing Techniques Aug 14, 2006

What is the Very Large Array (VLA)?

Receivers @ 92,21,6,3.6,2,1.3,0.7cm

World’s ‘largest’ array.

100 MHz total bandwidth

27 “movable” antennas

25 m diameter each

Total area = 130-m dish

Y configuration (“2D” array)

Longest baseline = 36 km

Completed in 1980


Moveable Antennas Railroad and “trains”

The VLA – the Scalable array


Why upgrade the VLA?

• The VLA is the world’s premier imaging radio telescope:– fast, sensitive, flexible, productive

• If it’s so good – what’s the problem?

• Astronomy today requires a more powerful and flexible radio telescope than the VLA.

– more sensitivity

– more frequency coverage

– more spectral flexibility

– better imaging….

• No significant technical upgrades since completion – 1970s technology severely limits scientific capability.

• Modern electronics and signal processing vastly increase the VLA’s scientific capabilities.


The EVLA Project – leveraging the VLA

• Builds on the existing infrastructure – antennas, array, railroad, people

• Implement new technologies – Receivers– Electronics – Data transmission – Correlator

• Goal of ten Times the “astronomical capability” of the VLA

– Sensitivity, Frequency coverage, Image Fidelity, Spectral Capabilities

– On a timescale and cost far less than required to design, build, and implement a new facility.


EVLA: order-of-magnitude improvements

Parameter VLA EVLA Factor

Point Source Sensitivity (1-, 12 hours) 10 Jy 1 Jy 10

Maximum BW in each polarization 0.1 GHz 8 GHz 80

# of frequency channels at max. bandwidth 16 16,384 1024

Maximum number of frequency channels 512 4,194,304 8192

Coarsest frequency resolution 50 MHz 2 MHz 25

Finest frequency resolution 381 Hz 0.12 Hz 3180

(Log) Frequency Coverage (1 – 50 GHz) 22% 100% 5

• The EVLA performance is vastly better than the VLA • EVLA cost is less than ¼ the VLA capital investment• No increase in basic operations cost


How is sensitivity improved?

• Recall minimum detectable flux:

1

2e

syss A

Tk

• Reduce Tsys

– Lower Trx - better receivers

– Lower Tspill – new feed designs

• Increase Ae via antenna efficiency

• Improves both continuum & spectral line observations

• For continuum, increase – 100 MHz to 8 GHz

54

39

34

31

26

EVLA

1541.3

1162

303.6

506

6020

VLAcm

Tsys


Frequency Coverage

• Continuous frequency coverage from 1 to 50 GHz

– a key EVLA requirement

match instrument to science, not science to instrument!

Blue - current VLA Green - EVLA

Yellow letters and bars show band names and boundaries.

Two low frequency bands (74 and 327 MHz) omitted


Point-Source Sensitivity Improvements : 1-, 12-hours

Red: Current VLA Black: EVLA Goals


Bandwidth, Spectral and Time resolution

• Combination of 2:1 bandwidth ratios and huge number of spectral channels• instantaneous spectral indices, rotation measures, uv-coverage

• instantaneous velocity coverage

53,300 km/s vs. current 666 km/s at 45 GHzlines at arbitrary redshift

• Spectral flexibility• 128 independently tunable sub-bands (vs. 2 currently)• “zoom in” on the lines of interest

• Temporal flexibility• Fast time recording: initially 100 msec; 2.6 msec possible• Pulsars: 1000 phase bins of 200 μsec width, 15 μsec possible

pulsar searches, timing, etc. with an interferometer!

• Spectral/temporal capability due to the WIDAR correlator


The WIDAR correlator

• Designed and built in Canada at HIA– $15M USD – initially enabled Canada’s participation in ALMA via the North

American Partnership in Radio Astronomy (NAPRA)

• 8-GHz bandwidth in each polarization covered by 4 x 2-GHz bands

• Each of these 2 GHz bands covered by 16 sub-bands – each 128 MHz wide

• 16,384 channels at 8 GHz bandwidth

• 4,194,304 channels possible

• 2 MHz – coarsest frequency resolution

• 0.12 Hz – finest frequency resolution

• Time sampling (up to 20 s)

• And lots more……………………………………………………


EVLA Design Driven By Four Science Themes

Magnetic Universe Obscured Universe

Transient Universe Evolving Universe

Measure the strength and topology of the cosmic

magnetic field.

Image young stars and massive

black holes in dust enshrouded

environments.

Follow the rapid evolution of

energetic phenomena.

Study the formation and evolution of

stars, galaxies and AGN.CO at z=6.4

Sgr A*


Key EVLA Correlator Capabilities

Deep ImagingPolarization

8 GHz Bandwidth (dual polarization). Full polarization processing. Wide-field imaging.

Narrow spectral linesWideband searches

16,384 channels at max. bandwidth (BW). >106 channels at narrow BWs. Spectral resolution to match any linewidth. Spectral polarization (Zeeman Splitting).

Eight 2 GHz wide bands input. Each input band decomposed into 16 tunable sub-bands of adjustable width Gives 128 independent sub-bands

Flexibility Many resources

High time resolution

1000 pulsar “phase bins”. “Single-dish” data output to user

instruments. Very fast time sampling (20 s).


Synergy with ALMA

• High-redshift star-forming galaxies– CO lines from star-forming galaxies

• Key science goal of ALMA• At redshifts of a few CO J=1-0 and 2-1 line in EVLA bands• At high redshifts (z~6) CO J=3-2 line in EVLA bands

– EVLA will detect synchrotron component out to z ~ 3 (normal) or z~5 (ultraluminous)• contribution of AGN

– EVLA will detect free-free emission from HII regions out to z~ 2– EVLA will be able to detect dust continuum out to redshifts > 10

• Young & proto-stellar objects– ALMA “bread and butter” – so where does the EVLA come in?– Defeat high dust opacity in the densest regions – opaque to 10’s of GHz– identify dust from free-free from synchrotron emission


Star-Forming Galaxies at High Redshift

• Enabled by EVLA sensitivity–Synchrotron emission: AGN, SNR

–Free-free emission: HII regions–Thermal dust emission

• Resolution 50 mas = 200 pc @ z=10

• EVLA+ALMA – similar sensitivity – dust+ionized gas+NT –SED over 3-orders of magnitude in frequency

– large range of redshift

Arp220 SED scaled to high redshifts.

Spitzer

non-thermal/AGN

ionized gas

dust


COJ=3-2 Z = 6.42 Peak ~ 0.6 mJy

Carilli, Walter, & Lo

Molecular lines in High-RedshiftStar-Forming Galaxies

•Currently:–50 MHz (z range of 0.001 at 50 GHz!) –Need to know a precise redshift or be lucky!–8 spectral channels = no frequency resolution

• No z searches• Very poor spectral resolution• Each line must be done independently (CO, HCN, HCO+, …)


• EVLA:

–8 GHz bandwidth @ 40-50GHz (z=1.4 -1.9 for CO J=1-0; z=3.8 to 4.8 for J=2-1)

–16384 spectral channels at maximum bandwidth

–Searches are a piece of cake!

–Other lines: HCN, HCO…..

Molecular lines in High-RedshiftStar-Forming Galaxies

Arp 220 @ z=8

Red line = EVLA in 8 hrs

CO J=1-0 @ 12.7 GHz J=2-1 @ 25.6 GHz J=3-2 @ 38.3 GHz


EVLA Setup for CO Z-Search

• 42-50 GHz band provides lowest redshift.

• z = 1.4 to 1.9 for J=1-0.

• z = 3.8 to 4.8 for J=2-1.

• v ~ 5.0 km s-1 (1 MHz).

• 200 km-s-1 galaxy would occupy ~40 channels.

• Interferometry• High resolution imaging.

2 , 1 , 0 .5 , ... G H z O bs e rv ing B andw id ths

16 S ub -B ands (s how n w ith de fau lt pos itions and w id ths )

1 2 3 4 ... 15 16

R ight P o la r iza tion L e f t P o la r iza tion

48-50 G Hz48-50 G Hz

46-48 G Hz 46-48 G Hz

44-46 G Hz44-46 G Hz

42-44 G Hz 42-44 G Hz

2-41-2 4-8 8-12 12-18 18-27 27-40 40-50L S C X U K Ka Q

G Hz

S k y F re q u e n c y B a n d s

8 tu n ab le IF B an d s

Note: S ub-bands can beseam lessly jo ined acrosseach observing bandwidth.

~30 H+ recom lines within 4 MHz band width (also He+, C+)

• Each line individually targeted

• Zoom in – 128 to 4 MHz

• Each of 62 spectra gets 256 channels

• = 15.6 kHz (1.6 km/s)

• EVLA imaging gives:

•Gas density

•Temperature

•B-field (Zeeman splitting is weak - 2.8 Hz/G)

•Improve SNR by “stacking”

2-41-2 4-8 8-12 12-18 18-27 27-40 40-50L S C X U K K a Q

GH z


R ig h t P o l'n Lef t P o l'n

8 M H z

25 km/s

H II

H eIIC II

(8 0 0 k m /s )

1 0 2 4 chan 's.R ec irc . fac to r = 1 6 v = 0 .8 k m /s

C o ntinuum S e tup C o ntinuum S e tup

2 GHz

Magnetic Fields in the ISM


Many Spectral Lines at once!

• Nobeyama obs of TMC-1. • 414 lines (8 to 50 GHz)• 38 species

– including “heavy” molecules– Slow rotators

• Some may show Zeeman splitting.

• EVLA can observe 8 GHz at once– an average of 80 lines – EVLA Correlator can “target”

many (~60) lines at once.

2-41-2 4-8 8-12 12-18 18-27 27-40 40-50L S C X U K Ka Q

G Hz


TA*

8 GHzKaifu et al., 2004.


EVLA Project Status

• Six (of 27) antennas currently withdrawn from VLA service, and being outfitted with new electronics.– Two antennas are fully outfitted, are now part of regular VLA

observations– Two others being outfitted with final electronics, and under test.

Available for astronomical use by late summer. – Two others in early stages of outfitting.

• Antennas will be cycled through the conversion process at a rate six per year, beginning in 2007.


New Capabilities Timescale

• The old correlator will be employed until the new correlator achieves full 27-antenna capability – mid 2009.

• Full band tuning available before 2009, on schedule shown here.


Major Future Milestones

• Test prototype correlator mid 2007– Four antenna test and verification system– Not available for science

• Correlator installation and testing begins: mid 2008– Capabilities will rapidly increase until mid 2009.

• Correlator Commissioning begins: mid 2009– VLA correlator turned off– New correlator capabilities will be much greater at this time.

• Last antenna retrofitted 2010• Last receiver installed 2012


Summary• The EVLA will improve the VLA capabilities more than tenfold

through up-to-date receivers, data transmission and the WIDAR correlator

• The project is on-track for completion in 2010 (antennas and correlator), and 2012 (for all frequency bands).

• The HIA-designed WIDAR correlator is an essential and critical component of the EVLA.

• Powerful new capabilities will begin to be available in 2008– just two years from now!


The EVLA: A North American Partnership

Project info: http://www.aoc.nrao.edu/evla/

sub-mm/mm observing techniquesthe evlaaug 14, 2006 the evla project sean dougherty national research...

Documents

techniquesthe evlaaug

vla sensitivity

vla evla cost

scalable array slide

t sys slide

vla capital investment

evla performance

hz3180 log frequency