Why? When? How What to do What to worry about

Transcription

1 Tom Muxlow Data Combination Why? When? How What to do What to worry about

2 Combination imaging or separate imaging??..using (e-)merlin (e-)merlin covers a unique range of telescope separations, intermediate between that of the (E)VLA and VLBI Where surface brightness limitations allow: MERLIN will extend the resolution of the VLA VLBI extend that of MERLIN In VLA/MERLIN and MERLIN/VLBI combinations, the addition of the shorter-spacing data provide otherwise missing spatial frequencies to place the compact structure in context they also stabilise the de-convolution procedure Remember that the largest holes in the u-v coverage will always limit the size of extended region that can be imaged e-merlin coverage is significantly better

3 Combination imaging stabilising de-convolution 3C293 Global VLBI + MERLIN (Beswick et al) MERLIN/VLBI combination stabilises the de-convolution of the VLBI-only image and permits high resolution imaging over extended components Global VLBI

4 Separate images 3C293 Global VLBI + MERLIN (Beswick et al) VLA image shows low surface brightness lobes which are too extended to be successfully imaged by MERLIN or VLBI Separate images are appropriate to show the relationship between regions of low and high surfacebrightness

5 Extended components 3C401 VLA + MERLIN MFS (L-Band, beam 0.15 see steepspectrum emission in detail cf VLA A-array 5GHz beam 0.5 LAS = 24 arcsec Burns et al MERLIN User Guide (L-Band) Guidelines: Maximum component size which can be Detected Imaged (1 frequency) Imaged (MFS) 4 arcsec 2.5 arcsec 8 arcsec VLA image shows low surface brightness lobes which are too extended to be successfully imaged by MERLIN monochromatically (e-merlin will be much better!) Multi-Frequency-Synthesis MERLIN observations alleviate the condition, but the very extended tails show some evidence of breakup during de-convolution procedure (CASA MS-CLEAN will be better)

6 Combined and separate imaging Global VLBI + MERLIN (Britzen et al) VLBA 15 GHz Global VLBI+MERLIN 1.7 GHz WSRT 1.4 GHz Global VLBI 5 GHz MERLIN 1.7 GHz VLA

7 Amplitude calibration Data sets are usually processed separately and then combined MERLIN/EVN data combinations: Check visibility amplitudes on any common baselines (e.g. Jb Cm Kn) MERLIN/VLA data combinations: Check the flux densities on any common phase-calibration sources. Try full crossing-point analysis on an averaged combined test dataset (UVCRS) provide insights into interferometry, but also helps with imaging fidelity Beware variability between the combination epochs for both target and calibration sources

8 Data set alignment Check the epochs and equinoxes of the datasets agree Beware early archival VLA data sets nominally observed in B1950 the phase calibrator positions are quoted in equinox 1950, epoch If the data sets have been phase-calibrated, check the assumed phase-calibrator positions Old data sets may suffer from poor phase-calibrator positions If the data sets have been self-calibrated only, they are almost certainly not aligned Make each image separately and convolve down the higher resolution image to that of the other and check the positions on a compact component within the overall source structure beware blending

9 Data set alignment B1950 or J2000? Use J2000. Convert from B1950 where appropriate use high precision Old MERLIN and VLBI data will be equinox , epoch Check phase-calibrator positions note any offset which needs correcting Modify the epoch & position information in the header with AXDEF & PUTH to correspond with the correct J2000 position incorporating any required offset [Beware pops rounding errors] Run UVFIX to recalculate the u-v-w axes in the data set Do not run UVFIX on a B1950 data set it attempts to convert to J2000 but does not do so to the required precision

10 Pre-combination self-calibration alignment If data sets have only been subject to selfcalibration or still do not align after careful study of positions and offsets, consider a single pass of self-calibration using the higher resolution image to correct the lower resolution data 3C459 VLA 5 GHz Use uvrange to restrict solutions to only match model flux density 3C459 MERLIN MFS 5 GHz Thomasson et al 2003 beam 70mas

11 Pre-combination self-calibration alignment If data sets have only been subject to selfcalibration or still do not align after careful study of positions and offsets, consider a single pass of self-calibration using the higher resolution image to correct the lower resolution data 3C459 VLA 5 GHz Use uvrange to restrict solutions to only match model flux density 3C459 MERLIN MFS 5 GHz Thomasson et al 2003 beam 70mas

12 Data set weighting Statistical weights associated with each integration differ between arrays and may depend on the processing route Inspect weights with PRTUV have they been modified for sensitivity?? Source= 3C459 RA = DEC = IF = 1 Freq= GHz Ncor= 4 No. vis= Sort order= TB Source= 3C459 Freq= Sort= TB 1 RR 1 LL 1 RL 1 LR Vis # IAT Ant Su Fq U(klam) V(klam) W(klam) Amp Phas Wt Amp Phas Wt Amp Phas Wt Amp Phas Wt 1 0/23:12: /23:12: /23:12: /23:12: In order to assign similar weights to each array check sum of gridded weights in IMAGR task time messages for user 666 IMAGR1 16:30:38 GRDMEM: Frequency E+09 Hz IMAGR1 16:30:38 Field 1 Sum of gridding weights = E+07 IMAGR1 16:30:38 Field 1 Beam min = MilliJy, max = 1.0 Jy IMAGR1 16:30:38 Field 1 fit FWHM = x Milliarcsec, PA= 28.3 Modify weights with REWEIGHT in routine DBCON as required

13 Combining data with different characteristics A potential nightmare in AIPS Use CASA... Process each array dataset as multi-source data files retaining the characteristics of that particular data set Convert to single-source data files containing the fully calibrated, astrometrically aligned target data Do any epoch conversions, position offset corrections, u-v-w re-calculation before combining AIPS data sets must contain identical numbers of IFs and frequency channels before combination split out each IF and re-combine to form a nominally time sequential single IF data set with separate sub-arrays Multi-channel data sets are problematical. They should contain identical reference frequencies, channels numbers and channel bandwidths otherwise gridding errors will cause smearing in wide-field images consider SPECR and BLOAT in AIPS 2 IF data set IF data set 1 2 Nominal 5 day separation

14 Combining data with different characteristics A potential nightmare in AIPS Use CASA... Process each array dataset as multi-source data files retaining the characteristics of that particular data set Convert to single-source data files containing the fully calibrated, astrometrically aligned target data Do any epoch conversions, position offset corrections, u-v-w re-calculation before combining AIPS data sets must contain identical numbers of IFs and frequency channels before combination split out each IF and re-combine to form a nominally time sequential single IF data set with separate sub-arrays Use CASA clean to simultaneously grid & image Multi-channel multiple datasets data with sets different are problematical. characteristics... They should contain identical reference frequencies, channels numbers and channel bandwidths otherwise gridding errors will cause smearing in wide-field images consider SPECR and BLOAT in AIPS 2 IF data set IF data set 1 2 Nominal 5 day separation

15 Image plane combination Since the Fourier transform is a linear transform can use the equivalence principle which states that the transform of the sum of two distributions is equivalent to the sum of the separate transforms + FT=> Standard data-plane combination FT=> FT=> + => Image-plane combination

16 Image plane combination This is the standard historical method for deep field VLA+MERLIN combinations The VLA and MERLIN data sets are each Fourier transformed to separate dirty maps and dirty beams with appropriate image and pixel sizes The VLA dirty map and dirty beam are then re-gridded with HGEOM to the geometry of the MERLIN images Combination dirty maps and dirty beams are constructed with COMB prior to conventional de-convolution via an image-based algorithm VLA dirty beam MERLIN dirty beam VLA dirty beam re-gridded to MERLIN geometry

17 Image plane combination The sensitivity of the combination image may be maximised by altering the relative weighting in COMB to reflect the array sensitivities Test image performed from GOODS-N field (Muxlow et al 2005) 307µJy source J , which lies 320 arcsec from field centre. Test shows that images have the same for dynamic ranges ~300:1

18 Image plane combination GOODS-N MERLIN 18 days, 1 IF (1420 MHz, 31x0.5 MHz channels) VLA 42 hours, 2 IFs (1365&1435 MHz, 7x3.125 MHz channels) J VLA Beware confusion on short VLA spacings Beware VLA smearing at high resolution Sensitivity (rms µjy/beam) MERLIN ~ 5.9 VLA ~ 5 Combined ~3.2 (Equal weight) Beam 2 arcsec J Combined Beam 200 mas

19 Image plane combination GOODS-N MERLIN 18 days, 1 IF (1420 MHz, 31x0.5 MHz channels) VLA 42 hours, 2 IFs (1365&1435 MHz, 7x3.125 MHz channels) Original study postage-stamp images over a 10x10arcmin square field Extended to 15arcmin complete circular field now covers whole of 76m Lovell telescope primary beam and to the limit of VLA radial smearing at MERLIN resolution 76% more area & new sources (Nick Wrigley MSc project) J J Combined Beam 200 mas

20 Combining VLA/EVLA data sets VLA/EVLA can be configured in several configurations A D 27 antennas with maximum baselines of: 36km (A), 10km (B), 3.6km (C), 1km (D) Surface brightness sensitivity (SBS) changes by ~10 with each configuration step. Guideline 4 increase in data set size for each configuration step SBS decreases by ~5 though not always followed exactly but need more data in larger configurations

21 Combining VLA/EVLA data sets Laing et al C31 at L, C, & X-Band with configurations A D L L L C X X X Must select array configurations, uv ranges, and tapering scheme to produce images at the required resolution & surface brightness sensitivity No single image can provide all you need Beware core variation between epochs and possible calibration problems between configurations

22 Combining ALMA Configurations ALMA will have many configurations with maximum baselines from 250m km m antennas for imaging + a compact array for total power Full snapshot imaging capability for compact configurations some Earth-rotation synthesis required for extended configurations Scheduling Blocks for desired Science Goals may require multiple array combinations (including compact array) CASA to be optimised to enable such imaging as a standard Detailed strategies for such imaging yet to be determined...

23 How to combine interferometric and single dish data in CASA Ammonia dataset EVLA & GBT 27 pointing mosaic on the Galactic Barnard 5 s-f region in Perseus NH3 (1,1) line data at 23.7GHz EVLA Open Shared Risk Observations CASA multiscale clean EVLA Coverage GBT single dish image shown with EVLA mosaic area overlaid.

24 EVLA only 27 pointing EVLA mosaic Processed in CASA (spectral cube) Multiscale clean Robust=0.5 clean(vis="data.ms", imagename= "image", psfmode="clark", imagermode="mosaic", weighting="briggs", robust=0.5)

25 EVLA mosaic combined with GBT image 27 pointing EVLA mosaic Multiscale clean, robust=0.5 GBT spectral cube re-gridded to EVLA spectral resolution and then imported as a CASA image via FITS used as a single dish prior clean(vis="data.ms", imagename= "image", psfmode="clark", imagermode="mosaic", weighting="briggs", robust=0.5, model="single_dish_regrid.image")

26 Self combination... MERLIN/(e-)MERLIN monitoring of M82 transient Source appeared within one week, between April 25 th and May 3 rd 2009 At peak, the new source is ~100 fainter in brightness than SN2008iz Intensive monitoring programme launched astrometric position study phase-referenced + self-calibrated aligned on SNR population New Source SN2008iz M82 MERLIN MFS +VLA 5GHz M82 MERLIN 5GHz Combination image 80mas beam

27 Tentative superluminal motion detected Probably beyond the scope of interpolated phase referencing alone New Source SN2008iz E day=2 day=26 day=40 day=48 W M82 MERLIN 5GHz In first 50 days (May-June 2009) E W proper motion detected relative to other compact SNR in M82 (10 +/- 5mas)

28 Preliminary VLBI images VLBI radio structure shows extension in same PA as e-merlin superluminal motion Dec 2009 Feb 2010 Object now reported as having faded after ~2 years Candidate intermediate BH system with radio emission from relativistic jet Mar 2010 (A. Brunthaler Sep 2010

29 Self combination... Historical VLA/MERLIN monitoring of 3C120 Self-calibrated VLA images of 3C120 observed 1983 & 1987 Images aligned on peak core emission Image subtraction in greyscale shows outward movement of 4 arcsec knot 9mas (2.4mas/yr) v=3.7c VLBI find similar velocity for inner jet near core

30 Self combination... Historical VLA/MERLIN monitoring of 3C120 Phase referenced & self-calibrated MERLIN images of 3C120 observed 1980 & 1988 Image subtraction in greyscale shows outward movement of inner jet 19mas (2.4mas/yr) v=3.7c No motion detected in 4 arcsec knot

31 Self combination... Historical VLA/MERLIN monitoring of 3C120 Phase referenced & self-calibrated MERLIN images of 3C120 observed 1980 & 1988 Image subtraction in greyscale shows outward movement of inner jet 19mas (2.4mas/yr) v=3.7c No motion detected in 4 arcsec knot

32 Self combination... Historical VLA/MERLIN monitoring of 3C120 Phase referenced & self-calibrated MERLIN images of 3C120 VLA images subject to core variability (Peak =1.9Jy in 1983, 3.2Jy in 1987) observed 1980 & core + jet Image subtraction in greyscale shows outward movement of inner jet 19mas (2.4mas/yr) v=3.7c No motion detected in 4 arcsec knot Alignment point for peak in image when blended together at VLA resolution MERLIN images phase-referenced but beware variable reference sources!!

33 What to worry about? Calibration are the data sets consistent? Alignment are they aligned? They need to be!! Weighting what is the optimum scheme? Variability for both target and phase calibrators Smearing can all the data be used off axis? Data or image plane combination? Do you really need to do this? Will separate imaging answer the same scientific question? Life...