Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://www.atnf.csiro.au/research/radio-school/2010/lectures/Brooks-Synthesis10.pdf Äàòà èçìåíåíèÿ: Thu Sep 30 09:02:53 2010 Äàòà èíäåêñèðîâàíèÿ: Mon Feb 14 00:50:46 2011 Êîäèðîâêà: Ïîèñêîâûå ñëîâà: star trail

Special Considerations for Millimetre Observing

Kate Brooks CASS Radio Astronomy School 28th September 2010


Why the need for a talk on millimetre astronomy?
· Everything about interferometry is more difficult at higher frequencies · Atmospheric opacity increases Tsys and attenuates the signal · Fewer bright millimetre sources to use for calibrators · Instrument stability is more difficult to maintain · Antenna accuracy is less · Field of view (antenna primary beam) is smaller


Talk Outline
· · · · · · Motivation for doing millimetre observations Capabilities of mmATCA Recap of observing strategies at centimetre wavelengths Highlight the differences for observing at higher frequencies Effects of the atmosphere Methods used by ATNF telescopes for atmospheric correction

Acknowledge presentations and contributions from Shari Breen, Cormac Purcell, Crystal Brogan, Debra Shepherd & Maxim Voronkov


Motivation for millimetre astronomy

K

Q

W
5 antennas only

· Sensitive to thermal emission from dust, very dense ionized gas & molecular lines · Probe of cool gas and dust in:
· · · · · Molecular clouds Dust in dense regions Star formation in the high-red shift Universe Recombination line emission from ionised gas Thermal emission from dense, compact regions (young stars)


mmATCA with CABB
- 8 GHz CABB range over which two 2-GHz bands can be positioned
2 x 2-GHz bandwidth with 2048 channels (1 MHz resolution) with full stokes 83.5 GHz 3-mm Up to a maximum separation of 6 GHz of the band centres 30 GHz 7-mm
38 GHz 44 GHz

106.0 GHz

50 GHz

Note: For the 7-mm system the centre frequency of the 8GHz CABB range cannot be placed between 38 and 44GHz. 16 GHz 12-mm 25 GHz


Observing restictions for mmATCA
· Stick with compact hypbrid array configurations (at least initially)
· In most cases observations at 3mm using the 750X, 1.5X or 6X arrays will be severely hampered by atmospheric phase noise.

· The best time for millimetre observing is the APR observing semester from April 01 to September 30 (also referred to as the "winter season"). · For 3mm observing there is a 60-65% chance of weather suitable for 3mm observing during the winter season. Good conditions start in May and the weather starts falling apart in October. The height of summer is unsuitable. · Observations at 12 and 7 mm may also be scheduled in the OCT observing semester, between October 01 to March 31.
· If applying for time in this semester then it is recommended you request your observations to be scheduled in the months October or March or during the nightime of the remaining months (if your source LST range permits).


CABB observations of NGC6334I at 3mm
Work by Brooks, Titmarsh, Voronkov, Urquhart done as part of the CABB science commissioning

3mm continuum (contours) and C34S (2-1) (grayscale)


Calibration for cmATCA observations ( 8.6 GHz)

· · · ·

Delay Bandpass Flux Scale Phase & Gain


Delay Calibration - centimetre
· Bright "Startup" Calibrator
Source 0823-500 1253-055 1934-638 RA 08:25:26.869 12:56:11.166560 19:39:25.026 DEC -50:10:38.49 -05:47:21.524580 -63:42:45.63 20cm 6.45 10.13 14.95 13cm 5.53 9.1 11.59 6cm 3.30 13.82 5.83 3cm 1.41 18.07 2.84

Delay Bandpass Flux Scale Phase

Delay Calibration (set delays to zero)


Bandpass Calibration - centimetre

Delay Bandpass Flux Scale Phase

· Bright "Bandpass" Calibrator · Correction of the different response across the band · Observer at least once during each observing run for ~10mins
Source 0823-500 1253-055 1934-638 RA 08:25:26.869 12:56:11.166560 19:39:25.026 DEC -50:10:38.49 -05:47:21.524580 -63:42:45.63 20cm 6.45 10.13 14.95 13cm 5.53 9.1 11.59 6cm 3.30 13.82 5.83 3cm 1.41 18.07 2.84


Flux Calibration- centimetre

Delay Bandpass Flux Scale Phase

· PKSB1934-638 is the "Primary Flux Calibrator" · Observer at least once during an observing run for ~10mins · Flux model in MIRIAD for 1934-638 between 408 and 8640 MHz
Source 1934-638 RA 19:39:25.026 DEC -63:42:45.63 20cm 14.95 13cm 11.59 6cm 5.83 3cm 2.84

Startup, Flux & Bandpass t Calibration using 1934-638


Phase & Gain Calibration - centimetre
· · · ·

Delay Bandpass Flux Scale Phase

Calibrator nearby to science target Sometimes called "secondary calibrator" or "gain calibrator" Must be unresolved (point source) for all baselines Should be ideally brighter than 0.5 Jy with an angular separation to the target of less than10 degrees · Calibrators vary and so always have a couple of backups if the calibrators are close to the recommended flux limit · Typically observe for ~2 min every 40 min (more frequently for higher frequencies) · Always finish with a scan on the secondary calibrator
CALIBRATOR ­ TARGET - CALIBRATOR The most up-to-date calibrator list is available by position or flux-limited online Search: http://www.narrabri.atnf.csiro.au/calibrators/


Calibration for mmATCA observations ( 16 GHz)

· · · · · ·

Delay Bandpass Flux Scale Phase + Pointing + Atmosphere


Delay & Bandpass Calibration - millimeter

· Bright "Startup" Calibrator and bright "Bandpass" calibrator
Source 0823-500 1253-055 1934-638 RA 08:25:26.869 12:56:11.166560 19:39:25.026 DEC -50:10:38.49 -05:47:21.524580 -63:42:45.63 20cm 6.45 10.13 14.95 13cm 5.53 9.1 11.59 6cm 3.30 13.82 5.83 3cm 1.41 18.07 2.84 12m 0.55 17.37 1.03 7mm ? 16.1 0.38

Delay Bandpass Flux Scale Phase Pointing Atmosphere
3mm <0.12 14.5 0.03

Too faint to use for mm observing
Source 0521-365 0537-441 1253-055 1921-293 2223-052 2227-088 RA 05:22:57.98465 05:38:50.362 12:56:11.166560 19:24:51.055957 22:25:47.259291 22:29:40.084346 DEC -36:27:30.850920 -44:05:08.94 -05:47:21.524580 -29:14:30.121150 -04:57:01.390730 -08:32:54.435410 20cm 12.09 4.54 10.13 9.06 5.95 1.00 13cm 6.60 3.92 9.1 10.45 5.96 0.84 6cm 3.84 2.99 13.82 11.85 6.12 1.32 3cm 3.00 3.04 18.07 13.59 6.28 1.42 12m 2.59 8.35 17.37 10.98 7.48 4.14 7mm 3.9 10.2 16.1 9.5 8.8 4.7 3mm 4.1 9.2 14.5 7.0 6.6 5.4

1921-293 and 1253-055 are the favorites


Flux Calibration - millimetre
Source 1934-638 RA 19:39:25.026 DEC -63:42:45.63 20cm 14.95 13cm 11.59 6cm 5.83 3cm 2.84 12m 1.03

Delay Bandpass Flux Scale Phase Pointing Atmosphere
7mm 0.38 3mm 0.03

· For ATCA 12-mm & 7-mm flux calibration
· 1934-638 is a suitable flux calibrator with CABB (just!)

· For ATCA 3-mm flux calibration (and backup for 12-mm and 7-mm)
· Uranus is the preferred flux calibrator

· Flux model for Uranus is available in miriad · Uranus is above an elevation of 30 degrees only for a few hours each day · Flux calibration can be "boot-strapped" using calibrators 1921-293 or 1253-055 · Search for more flux calibrators is underway


Phase & Gain Calibration - millimetre

· Calibrator nearby to science target (the closer the better) · In principle > 1 Jy with an angular separation to the target of less than 5 degrees · Fewer bright calibrators at higher frequencies · In practice ~ 0.8 Jy with an angular separation up to 15 degrees · Typically observe:
· 2 min every 10 min at 90 GHz · 2 min every 15 min at 40 GHz · 2 min every 20 min at 20 GHz

Delay Bandpass Flux Scale Phase Pointing Atmosphere


Pointing Calibration - millimetre
· Global pointing model for ATCA antennas achieves an uncertainty of ~10 arcsec · Primary Beam ~ 1.22 /D · Antenna response to flux is reduced if off target · Additional pointing correction required for observations at 3 mm, 7 mm and (maybe) 12 mm · Pointing should be corrected once per hour or if the antenna has slewed to a different part of the sky · Use pointing scan (see Max's talk for details)
1.3 GHz
Primary Beam

Delay Bandpass Flux Scale Phase Pointing Atmosphere



2.2 GHz
21 arcmin

4.5 GHz
11 arcmin

8.0 GHz
6 arcmin

16 GHz
2 arcmin

30 GHz
95 arcsec

90 GHz
32 arcsec

36 arcmin


Atmospheric Calibration - millimetre
· Atmospheric opacity is significant below 1 cm
· Raises Tsys and attenuations the source signal · Loss of visibility amplitude (co-herence)

Delay Bandpass Flux Scale Phase Pointing Atmosphere

· Opacity varies with frequency, altitude and weather conditions
· Leads to significant phase fluctuations

· Opacity arises in the troposphere (lowest layer of the atmosphere, 7 ­ 10 km high


Atmospheric Calibration - millimetre

Delay Bandpass Flux Scale Phase Pointing Atmosphere

total optical depth

optical depth due to H2O vapor optical depth due to dry air

22 GHz
Water Vapor

43 GHz
Dry air

100 GHz
Water + Dry air


Atmosphere adds noise and reduces signal


The less air and the less humidity the better

ALMA site ­ Chajnantor at 4600m - PWV 1mm


How to gauge dry observing weather

Plane trails - HUMID


How to gauge dry observing weather

Curly Hair - HUMID


How to gauge dry observing weather
ATCA seeing monitor A two-element interferometer with an east-west baseline of 230m, tracking the 30.48GHz beacon on the geostationary satellite OPTUS-B3, at an elevation of 60 degrees. The seeing monitor works by taking the difference between to successive phase measurements, computing the standard deviation of this difference, and converted into a path length in microns. Results from the seeing monitor are updated in real time and displayed on MONICA in the control room. This gives a measurement of the rms phase fluctuation caused by water vapour in the atmosphere.


Correcting for the atmosphere - Tau
Extinction is expressed in nepers () and actual attenuation is given by e Transmission is inversely related to absorption as 1-e- The attenuation at an arbitrary zenith angle is given by the "sec z" term: Attenuation = e- = e-osecz = e-o/sin(el) = ewhere, 0 = zenith optical depth A = number of air masses (secz) el = elevation angle z = zenith distance (90 ­ el)
oA -


Atmospheric Model Outputs for tau
Using Miriad task "opplt" For For For For For 22 GHz: tau = 0.4 to 40 GHz: tau = 0.2 to 50 GHz: tau = 0.6 to 98 GHz: tau = 0.7 to 115 GHz: tau = 1.35 0.2 0.1 0.3 0.35 to 0.7 x x x x x 1.5 to 1.2 1.2 to 1.1 1.8 to 1.3 2.0 to 1.4 3.8 to 2.0 (20%) (9%) (26%) (30%) (48%)

These were made with humidity = 70%, pressure = 1013 hPa and temperature = 300K. For mmATCA observing stay above 35 degrees


Simplified Sky Noise Equation
The measured antenna temperature from an astronomical source can be represented by: TON = Trec + Tamb (1 - e-) + Tsourcee- + TCMBe- where, Trec = the receiver temperature Tsource = the source temperature Tamb = ambient temperature · TCMB = the contribution to the signal from the cosmic background signal


Simplified Sky Noise Equation
Background Emission (TCMB)

Source (Tsource)

Atmosphere (Tamb and )



Receiver Noise (Trec)


Correcting for the atmosphere ­ System Temperature
· The term "system temperature" is used to define the noise of the whole system. Tsys = [Trec + Tatm (1 - e-) + TCMBe-] In good observing conditions the typical values for Tsys are: 12 mm ~ 60 K 7 mm ~ 120 K 3 mm ~ 220 K For ATCA the primary contribution to Tsys is from the atmosphere


Consider single dish (Mopra) for a moment

TON

TOFF


Consider single dish (Mopra) for a moment

Output from ASAP: TON ­ TOFF / TOFF = Quotient Spectrum


Consider single dish (Mopra) for a moment
Output from ASAP: TON ­ TOFF / TOFF = Quotient Spectrum Using: TON = Trec + Tamb (1 - e-) + Tsourcee- + TCMBe- TOFF = Trec + Tamb (1 - e-) + + TCMBe-

Quotient Spectrum = Tsourcee- / Tsys Therefore to calibrate the flux scale for the spectral line produced by ASAP a measurement of Tsys and attenuation e- are required.


Measuring the Tsys for Mopra 12 and 7 mm
· For the MOPS system on Mopra a switching noise diode is used to measure the change in Tsys · We are continually tracking Tsys variations on one hot/cold load measurement taken at the start of the semester · The noise diode is used to calculate the contribution of the receiver and the antenna to the Tsys measured at the feed aperture · The noise diode adds a fixed and known contribution to the system · The noise diode computes a new Tsys value for each 2-second integration cycle · There is one Tsys value for each of the MOPS 2.2 GHz IF bands


Measuring the attenuation e- for Mopra 12 & 7 mm
· For the MOPS 7- and 12-mm systems the atmospheric attenuation is measured via 2 independent methods: 1. Skydip measures the power of the atmosphere emission as a function of airmass while tipping the telescope from high to low elevation
· Assuming that for each skydip procedure the values for Trec, Tatm and TCMB are constant then:
Tsys(el) = A - (B)eo/sin(el)

· Time consuming and no good in bad (unstable) weather

2. Atmospheric Model
· Inputs are: Temperature, Pressure, Humidity · This is the standard way to observe · Requires no observing time (offline correction in ASAP)


Measuring the attenuation e- for Mopra 12 & 7-mm


Measuring the Tsys and attenuation e Mopra at 3 mm

-

for

· For the MOPS 3-mm system the Tsys and the atmospheric attenuation is measured via the "Paddle" method · This involves placing an ambient load (paddle) in front of the receiver horn for 30 sec. · Assume that the temperature of the ambient load is equivalent to the temperature of the atmosphere obtain a value for Tsys corrected for atmospheric contributions · CABB output temperature scale is automatically corrected each time a paddle measurement is made
Psky Pabs

Tsys

eff

= (300 K) ____Psky____ Pabs - Psky


What happens for ATCA
· At 3-mm
· Both the system temperature and the correction for tau is determined using the paddle method · The paddle measurement should be executed every 15min · The correction for tau is interpolated between paddle measurements · The CABB output temperature scale is automatically corrected to reflect the Tsys and tau values (1 value for the full 2GHz band)

· At 12 and 7 mm
· The system temperature is determined by the noise diode · The correction for tau is determined offline (in miriad) using an atmospheric model


Calibration limitations ­ Flux uncertainty
· If you have data taken in good weather then expect the following uncertainties in your final flux measurements:
· For 12 mm - 5 to 10 % · For 7 mm ­ 10 to 20 % · For 3mm ­ 30 to 50 %

· Be aware of these limitations and the sensitivity limitations of ATCA when you are preparing your observing proposal · While observing unstable weather is your enemy.


Is all this extra effort worth it?
· One good thing for millimetre observing is that there is much less radio frequency interference · Sensitive to thermal emission from dust, ionized gas & molecular lines · Probe of cool gas and dust in the nearby and distant universe · Observing with mmATCA gives you the skill set to use and work with the data products from the first of the next-generation astronomy mega-structures ­ ALMA


Thank you
CSIRO Astronomy & Space Science Australia Telescope National Facility Kate Brooks Millimetre Astronomy Research Scientist


Measuring Tsys using the paddle method!
Power measured from blackbody paddle: P
load

= C[Iload + Irx]

Compare power from blank sky and known load: P P
load sky

=
sky

J(Tatm)(1-e ) + Ibge + IRx I
load

­P

+ IRx - J(Tatm)(1-e ) - Ibge - I

Rx


Measuring Tsys using the paddle method!
P P
load sky

­P

=
sky

J(Tatm)(1-e ) + Ibge + IRx I
load

- J(Tatm)(1-e ) - Ibge i.e.



Assume:

Tload = Tatm

Iload = J(Tatm)

P P
load

sky

­P

=
sky

J(Tatm)(1-e ) + Ibge + IRx J(Tatm) e - Ibge



Measuring Tsys using the paddle method!
P P
load sky

­P

=
sky

J(Tatm)(1-e ) + Ibge + IRx J(Tatm) e - Ibge


Tsys

=

J(Tatm)e ­ J(Tatm) + Ibg + IRxe J(Tatm) - I Tsys J(Tatm) - I
bg bg

P P
load

sky

­P

=
sky

Measured

Assumed : 300 K & < 1K


Calibration to TA* scale
Tsys = P P
load sky

-P

sky

(J(T

load)

­ Ibg)

Assumed 300 K (Ambient temperature)

<1 K From CMB

TA* =

(P

on-source

­P

off-source)

P

Tsys

off-source