250 likes | 439 Views
ATCA – Calibration at mm wavelengths. Rick Forster University of California, Berkeley Hat Creek Radio Observatory Berkeley-Illinois-Maryland Association Topics: Basics of interferometry The atmosphere Amplitude & phase calibration Atmospheric & instrumental effects
E N D
ATCA – Calibration at mm wavelengths Rick ForsterUniversity of California, Berkeley Hat Creek Radio Observatory Berkeley-Illinois-Maryland Association Topics: Basics of interferometry The atmosphere Amplitude & phase calibration Atmospheric & instrumental effects Antenna effects ATCA synthesis workshop - May 2003
ATCA Bands Band l cm GHz mJy/b beam L20 1.5 0.19 33 S13 2.3 0.24 22 C6 5.6 0.22 10 X3 8.6 0.21 5 K1.2 20.5 0.53 2 W 0.35 87.5 2.70 0.6 ATCA synthesis workshop - May 2003
A few definitions Flux density S = B dW W/Hz m-2 Brightness B = 2k TB/l2W/Hz m2W Effective area Aeff = hpD2m2 Received power VA2 = Aeff S Dn Watts Janskys/K S/TA = 3510/hD2 Jy/K V2 or the total output power is TSYS = TR + TSKY + TA K (100+200+.04 K) V (volts) ATCA synthesis workshop - May 2003
V2 Amb ( ) Tamb - Tsky Tsys = V2sky • Tsky T Sky V2amb – V2sky Tsys Tsky Tamb ~ LO t s Digitized IF (V) RF G PB Sampler Delay k=1380 Jy m2/K 2 k Tsys Tsys=300 K IF n s n = Jy A=380 m2 h=0.3 A h · bw t bw=128 MHz ) a = sin-1(n/s) t=10 sec Tsys – ‘Paddle’ or Chopper Wheel method Tsys sets the flux scale and the noise level! a = sin-1(0.2) ~ 12o n=0.2 Jy ATCA synthesis workshop - May 2003
Basic Interferometry _ tg= B s ^ tg = geometrical delay x Tsys = system temperature Tant = antenna temperature Ny–Nn r correlation coefficient = ^ s _ N B Digitized IF from two antennas Ant 1 r12 Correlator 1x2 Ant 2 ATCA synthesis workshop - May 2003
Correlation Function Correlation Coeffieient 64 lag channels (+/- 32) Fourier Transform 32 freq channels (amp & pha) Ampl (Jy) & Phase Visibility V = Aeif Tsys ~ 200 K r ~ 5 10-4 Jy/K ~ 150 S = r Tsys Jy/K ~ 15 Jy ATCA synthesis workshop - May 2003
Atmosphere Lay, 1997 absorption delay TA = Tsou ( e-t ) Tsky = Tatm (1-e-t ) Tsys = TR+Tsky+TA Tsys = TR+Tatm(1-e-t)+Tsou(e-t) T’sys = TRet +Tatm(et -1)+Tsou T’sys = Tsyset for t = 1, et = 2.7 ATCA synthesis workshop - May 2003
Opacity of the Atmosphere – O2 & H2O O2 118 GHz H2O 180-190 GHz O2 50-70 GHz 84.9-87.3 H42a SO SiO 88.5-91.3 HCN HCO+ CH3OH HC3N H2O H2O 22 GHz O2 150 50 100 Frequency (GHz) ATCA synthesis workshop - May 2003
DF ~ Da a = 1.0 (Gaussian) a = 5/6 (Kolmogorov) Log DF D saturation a = 1/2 (Lorentzian) Log Distance (D) DF Power density 10 min 3 hrs 30 sec Log time Butler & Desai, 1999 Structure Function (DF vs D) a ~ 0 Lay, 1997 “Frozen Turbulence” Model (DF power vs t) a = 5/6, 1km layer, 5km/s wind, 500m D Longer baselines shifts peak right Stronger winds shifts peak left ATCA synthesis workshop - May 2003
Amplitude Phase M87 data calibrated with 3C273 80m 45m 15m ATCA synthesis workshop - May 2003
Baseline Solution – C-array, March 2003 Phases AFTER Correction Phases BEFORE Correction ATCA synthesis workshop - May 2003
Absolute flux calibration Test source 1310+323 (QSO) 0909+013 (QSO) cal test source MWC 349 (HII) Use a primary flux calibrator (planet or HII region) to check the value of Jy/K. Use your phase calibrator to monitor temporal gain changes. Use a test source to help quantify the effects of atmospheric decorrelation in the image plane. ATCA synthesis workshop - May 2003
k = 1380 Jy m2/K Tsys = 560 K 4 MHz = 14 km/s 64ch = 0.2 km/s A = 28 m2 h = hahc = .70x.88 = .61 bw = 100/32 = 3.125 MHz t = 6m = 360 s 2 k Tsys n = Jy A h · bw t Passband – 32 visibility channels over 100 MHz Channel BW = 3.125 MHz (~11 km/s) n = 2.6 Jy (each 11 km/s channel) s/n = 12/2.6 ~ 4.6 phase noise ~ 11 deg (ave 100 MHz ~ 6 times better) At 0.2 km/s channel n = 8x2.6 ~ 21 Jy ATCA synthesis workshop - May 2003
Az El 30 min ATCA synthesis workshop - May 2003
A few final suggestions • Calibrate Tsys often • Check Jy/K on a flux cal • Mosaic sources larger than ½ fwhm • Use offset pointing for extended sources • In a pinch total power correction might help • Choose nearby calibrator if baseline is uncertain • Choose integration time so S/N > 5 on calibrator • Use a guest calibrator to gauge atmospheric effects • Don’t push your observations beyond the seeing limit • Avoid unnecessary PB calibration, & check sensitivity • Monitor the atmosphere – abort 3mm in severe conditions • Fast-switching might be worth a try in reasonable conditions ATCA synthesis workshop - May 2003