The ALMA Project

1
The ALMA Project
Robert Laing (ESO)
2
What is ALMA?
Atacama Large Millimetre/Submillimetre Array

• Aperture synthesis array optimised for millimetre
  and sub-millimetre wavelengths.
• High, dry site, Chajnantor Plateau, Chile
• North America (NRAO) Europe (ESO) Japan
  (NAOJ) Chile
• EU/NA 50 dishes with 12m diameter. Baselines
  from 15m to 14km.
• ALMA Compact Array (ACA) provided by Japan
• 12 7m dishes in compact configurations
• 4 12m dishes primarily for total-power

3
What is ALMA (2)?

• Low-noise, wide-band receivers.
• Digital correlator giving wide range of spectral
  resolutions.
• Software (dynamic scheduling, imaging, pipelines)
• Will eventually provide sensitive, precision
  imaging between 30 and 950 GHz in 10 bands
• 350 GHz continuum sensitivity about 1 mJy in
  one second
• Angular resolution will reach 0.05 arcsec at 100
  GHz
• Resolution / arcsec ? 0.2 (?/mm) / (D/km)
• Primary beam / arcsec ? 17 (?/mm)

4
History

• Evolved over many years of US, EU and Japanese
  mm/submm array projects
• Bilateral (US/EU) ALMA project started 2002
• Trilateral agreement ( Japan) 2005
• Major construction in Chile 2006 ?

5
(No Transcript)
6
ALMA Sites in Chile
Array Operations Site (AOS) - 5050m
Operation Support Facility (OSF) - 2900 m
Santiago Central Office
7
Location Chajntantor Plateau at 5000m in
northern Chile
ALMA
8
The Chajnantor plateau
9
(No Transcript)
10
Work in progress
AOS Technical Building
A small problem
11
Another small problem
12
Continuous reconfiguration from compact to
extended configuration
13
Road to AOS
AOS Site (43 km)
OSF Site (15 km)
Width of road 14 to 19 meters
14
Antenna Transporters
15
Key antenna specifications

• 12m diameter
• 25 µm rms surface accuracy (goal 20 µm) measure
  using tower and interferometric holography
• 2 arcsec rms absolute pointing 0.6 arcsec rms
  offset
• Tracking speed for on-the-fly mapping 1 deg/s
• Fast switching required between target and
  calibrator (1.5o in 1.5s)
• Three prototypes (Vertex/RSI, EIE/Alcatel,
  Mitsubishi) tested at VLA site all meet
  specification as far as can be tested at lower
  altitude.

16
The three prototype antennas at the ATF
Mitsubishi Antenna
Vertex Antenna
AEC Antenna
12 Meter Diameter, Carbon Fiber Support Structures
17
Antenna Procurement

• US 25 x 12m
• Vertex RSI / General Dynamics (first 2007Q2 last
  2011Q4)
• Europe 25 x 12m
• Alcatel / EIE / MT Aerospace (first 2008Q3 last
  2011Q4)
• Japan 4 x 12m
• MELCO (first 2007Q4 last 2008Q3)
• Japan 12 x 7m
• Call for tender 2007

18
Mosaics and wide-field imaging

• Basic problem of extremely small fields, limited
  by primary beam and short spacing coverage
• Combination of a mosaic of pointings with the
  main array and single-dish data can be used to
  sample a larger range of spatial scales.
• Pointing errors severely limit the image fidelity
  unless scales around 10m uv distance are properly
  sampled.
• ACA 7m antennas fill in short-spacing coverage
• Four 12m antennas used to supply total power data
  (beam- switching using nutator on-the-fly
  mapping)

19
Sampling of spatial scales (compact configuration
and ACA)
1 Mpc corresponds to 125 arcsec at z 1

• ? Primary beam Minimum ?/D
  Resolution ?/D
• GHz arcsec arcsec
  arcsec
• 12m 7m 180m
  ACA 180m
• 35 170 291 116
  199 10
• 110 56 99 37
  64 3.1
• 230 27 46 18
  31 1.5
• 345 18 31 12
  21 1.0
• Also combine with 12m (single-dish) observations

20
ALMA bands

1. 31.3 45 GHz
2. 67 - 90 GHz
3. 84 - 116 GHz NRAO
4. 125 163 GHz
5. 163 211 GHz
6. 211 275 GHz HIA
7. 275 373 GHz IRAM
8. 385 500 GHz
9. 602 702 GHz SRON
10. 787 950 GHz

Under construction (NA/Europe) Under construction
(Japan) Development study (Japan) EU FP6 proposal
(6 antennas) Not yet funded
21
Sensitivity in 1 minute
RMS for 2 polarizations, each with 8GHz
bandwidth elevation of 50o. Brightness
temperatures are for a maximum baseline of 200m
50 antennas Median PWV 1.5mm Best 5 PWV
0.35mm ALMA Memo 276 Receivers may exceed
specification Sensitivity calculator available
at

• ? ?S ?TB
• GHz mJy K
• 35 0.019 0.0003
• 110 0.033 0.0004
• 345 0.14 0.0018
• 409 0.31 0.0040
• 675 3.8 0.049
• 0.46 0.0059
• 850 5.9 0.080
• 1.1 0.014

http//www.eso.org/projects/alma/science/bin/sensi
tivity.html
22

Front-end assemblies
23
Front End Design

• Diameter 1 m
• External optics top of dewar

• 10 Cartridges plugged from bottom
• Each cartridge contains one frequency

24
Band 7 noise performance
Band 9
Band 7
25
Front-end technologies

• SIS mixers (4K)
• 2 linear polarizations
• 8 GHz bandwidth per polarization
• 1-5 cartridges assembled for all 4 initial bands
  meet or exceed specifications
• First cryostats delivered integration in
  progress
• Tests at ATF 2007

26
Local oscillator/Back end

• Primary reference signal (27 142 GHz)
  distributed on optical fibre from a single master
  oscillator (rubidium standard/H maser) by
  frequency modulation of an optical carrier.
• Line length correction
• Frequency multiplied as needed at the antenna
• RF signals initially mixed to 4-12 GHz IF band
  then with second LO to 2-4 GHz channels
• 3-bit digitization and analogue total power
  detection
• Digitized signals returned along 120 Gbit/s
  optical fibre
• Use of ALMA as a phased array for VLBI is not
  precluded, but is not currently funded

27
Correlator

• Channel bandwidth 31.25 MHz 2 GHz
• Tunable FIR filter bank to subdivide bandwidth
  into 32 sub-channels
• FXF architecture
• Maximum 4096 x (4/N) x (2/P) spectral
  points/channel, where N 1, 2 or 4 is the number
  of channels and P2 for full polarization 1 for
  parallel hands only.
• Maximum spectral resolution 3.8 kHz.
• Output to archive and pipeline data-processing.
• First quadrant complete and under test
• ACA correlator has essentially the same modes,
  but FX architecture.

28
Correlator first quadrant
First of four Correlators at NRAO
29
ALMA calibration

• Instrumental gain calibration - hot and cold
  loads
• Specification 3 absolute amplitude calibration
  below 300GHz 5 at higher frequencies
• Primary amplitude calibration astronomical
  sources
• Phase calibration is difficult (tropospheric
  water) requires fast switching between target
  and calibrator (20s cycle at highest
  frequencies)
• Correction for phase variations on shorter
  timescales by measuring water-vapour emission at
  183 GHz.
• Polarization (specification lt0.1 instrumental).

30
WVR tests on the SMA

• 183 GHz WVR compared with 200m baseline at 230
  GHz
• DC offset only

ALMA radiometer team (Cambridge/ Onsala) and
Submillimeter Array
31
ALMA computing main requirements

• Control of the ALMA system antennas, front and
  back-end electronics, correlator
• Image and quick-look pipelines operating in near
  real time
• Dynamic scheduling according to scientific
  priority and atmospheric conditions
• Archiving of raw, calibrated and associated data
• Off-line data processing package is CASA (AIPS)
• Observing tool (Phase 1/2)
• Pipeline

32
Project status

• Rebaselining complete cost and schedule
  major reviews Oct 2005 Jan 2006. Descope to 50
  antennas.
• Prototype systems integration (ATF). First
  fringes between prototype antennas 2006
• First production antenna delivered 2007Q2
• First interferometry at AOS 2009
• Commissioning and science verification 2009-10
• Early Science (open call) 2010
• Full operations 2012

33
Early Science definition under discussion

• At least 16 antennas fully commissioned (more in
  process of integration)
• Receiver bands 3, 4, 6, 7, 8, 9
• Interferometry in single field or pointed mosaic
  mode
• Significant (TBD) range of spectral modes,
  including Tunable filter bank
• Circular and linear polarization (not mosaic)
• Single-dish mosaic (position and beam-switch) and
  OTF.
• 2 subarrays operational
• Formal proposal call

34
Science Verification

• Happens before Early Science
• Main goals
• Test ALMA modes end-to-end (includes projects
  from user community)
• Feedback to commissioning team
• Early access to ALMA data for the community
• Modes fully commissioned before science
  verification
• Open call for proposals, fast, not using formal
  machinery review for scientific value
  (external) and feasibility
• Data public immediately
• Projects executed by commissioning
  team/Operations
• ALMA Public Images pretty pictures

35
ALMA Key Science Drivers

• High-redshift galaxies
• Image CO and CII in Milky Way at z 3
• Starbursts to z gt 10
• Protostars and planet formation
• Image disks (a few AU)
• Disk structure, temperature, composition, tidal
  gaps

36
ALMA Science

• Galaxy, star and planet formation are the key
  science drivers
• Aside from non-thermal emission, most of what
  ALMA will see comes from elements heavier than H
  and He (except recombination lines, LiH).
  Therefore probe stellar products.
• Temperatures are lt stellar surface the Cool
  Universe
• Continuum thermal emission from dust
• Lines molecular rotational transitions
  (redshifted atomic)
• Heating via stellar UV, cosmic rays, hard photons
  from AGN hence the link to star and galaxy
  formation
• Non-thermal mechanisms include synchrotron (lower
  frequencies) and Compton scattering
  (Sunyayev-Zeldovich).

37
Star and planet formation

• Stars form from the gravitational collapse of
  dense cores within the innermost regions of
  interstellar molecular clouds, usually hidden by
  hundreds of magnitudes of visible extinction.
• Protostars and proto-planetary systems are best
  studied at mm and sub-mm wavelengths. ALMA will
• study the structure and dynamics of molecular
  clouds in detail
• resolve the structure of protostars and
  proto-planetary systems
• determine the chemical composition of material
  from which solar systems will form

38
Initial conditions for star formation
Low-mass starless core L1554. Spectra suggest
infall ALMA will resolve this spatially.
Simulation of polarization from protostar at
0.85 mm B-field crucial for star formation
MAMBO mosaic of the ? Oph cloud (Motte et al.).
Mass function incomplete Below 0.1 solar
masses. ALMA will image clumps of planetary mass
in this system, as well as pre- stellar
condensations.
39
Protostellar envelopes and outflow
Four stages in the evolution of a low-mass
protostar
ALMA will image infall, rotation and outflow in
Class 0 and 1 protostars
40
Outflow in HH211
Colour H2 (2.12µm) McCaughrean et al. (1994)
Contours CO (Gueth Guilloteau 1999)
What is the role of the magnetic field in
driving or collimating outflows? Physics and
chemistry of the interaction between the jet
and its surroundings?
41
Fragmentation and multiplicity
The formation of binary and multiple star
systems is not understood. Neither is the effect
of multiplicity on planet formation. Only ALMA
has the spatial resolution to study this problem.
Binary protostar imaged at 2.7mm with BIMA
42
High-mass star formation
W3(OH)
Colour 3.6cm Contours 1.3mm
Serpens
Only ALMA can resolve the complex physics and
chemistry
43
Chemistry of star formation
Biogenic molecules?
44
Protoplanetary disks
How do disks evolve from gas-rich to tenuous
dust debris? How many disks have
gaps indicating planet formation? What is the
chemical composition of disks and how does this
influence planet formation?
45
Debris disks
e Eridani debris disk observed by SCUBA. ALMA
will resolve the detail, and add dynamics
from spectral lines.
46
ALMA can image gaps in protoplanetary disks and
test models of planet formation
Model
Simulated ALMA image
47
Gas in galaxies out to z ? 5

• Mass of H2
• Kinematics
• Efficiency of star formation as a function of
  redshift

48
Dust and molecules in high-redshift galaxies
Large amounts of dust and CO already present at
z 4.69 Dynamics and mass assembly at high
redshift
CO and optical emission in a nearby merger (the
Antenna). Dust and CO is missed by HST.
49
Molecular Gas in Galaxy Disks
ALMA will image normal galaxies in CO and CII to
z 3 (top-level science requirement
50
Molecular absorption lines
Molecular absorption lines are weak, but
essential to probe the ISM in high-redshift
galaxies in detail
51
Gas in Galactic Nuclei
Cen A HCO absorption
Arp 220 local starbursts and their relation to
AGN
Cen A dust and molecular lines in the nearest
radio galaxy ALMA should image the AGN torus
Cen A HST
52
Characterising the sub-mm galaxy population

• (Sub-)mm emission from Lyman break galaxies,
  EROs,
• Follow-up of wide-area (sub-)mm surveys e.g.
  SCUBA2 -in continuum and lines
• Follow-up of surveys at other wavelengths
• Deep surveys

53
Known luminous sub-mm galaxies

• Currently, we just see
• the most luminous
• examples
• lensing
• AGN

54
Complementarity
S / µJy
Frequency / Hz
Standard spiral galaxy at z 2 (5s/1 hour)
Combes 2006
55
The sub-mm conspiracy
Spectral energy distribution of a dusty galaxy
The effect of redshift on the SED dusty galaxies
are easily detected at high z
56
Sub-mm galaxy surveys
A2125 MAMBO HDF SCUBA 850µm
cm and mm continuum from Carilli et al.

SCUBA sources in the HDF
Bolometer surveys suffer from confusion. High
resolution and sensitivity is needed to resolve
sources. Estimate z from broad-band, but measure
by CII or (better) CO lines. ALMA is needed for
follow-up of wide-area surveys (Herschel,
SCUBA2, APEX, Spitzer ) and deep, narrow-area
surveys
57
ALMA as a redshift machine
Redshift coverage for CO transitions as a
Function of rotational quantum number
J Feasible to measure z of galaxy detected in
dust continuum alone Atomic lines
redshifted into ALMA bands at high z e.g. OI
63, 145µm OIII 88µm NII 122, 205µm
58
CO transitions at different temperatures
59
CO rotational transitions
Temperature and density-dependent Low density in
Milky Way High density in starbursts Higher
excitation at high z (CMB) (Combes 2006)
60
Deep surveys with ALMA

• Feasible over small areas only because of small
  primary beam
• Avoids confusion
• Examples
• Broadband continuum survey 4 x 4 arcmin2 at 290
  GHz) 130 pointings 30 min each rms 20µJy, 100
  300 sources
• Continuum, 4 x 4 arcmin2 at 90 GHz, 16 pointings
  4 hr each rms 1.5 µJy
• Line, 50 kms-1 spectral resolution, 4 centre
  frequencies, 4 mJy km s-1 for 300 km s-1 line, 1
  CO line for z gt 2, 2 for z gt 6
• Then repeat at 200 GHz (6 days)

61
Optical submm comparison
HDF optical (Lanzetta)
ALMA deep field simulation (Guilloteau Wootten)
z lt 1.5 z gt 1.5
62
Gravitational lensing
Cloverleaf quasar Simulated
ALMA image The same cluster in in CO
(IRAM) 350 GHz
continuum optical R band
Cluster z
0.2 Image intrinsically faint,
high-redshift galaxies and determine their
dynamics
63
Sunyayev-Zeldovich Effect
Simulation (Kitayama, Tsutsumi et al. )
SZ effect RXJ13471145 NRO 150GHz data (Komatsu
et al. 2001)
64 (C1)
64SD
90 arcsec
0.22 mJy/beam
64
Imaging the Event Horizon of Sgr A
Kerr (spinning) black hole
Schwarzschild black hole
GR ray Simulated
VLBI tracing 0.6mm
1.3mm (Falcke et al. 2000)
(interstellar scattering)
65
Conclusions

• ALMA is really happening
• Lots of exciting science will be possible
• Be ready for 2010

http//www.eso.org/projects/alma/publications http
//www.strw.leidenuniv.nl/7Ealma/drsp.html