Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.naic.edu/~tghosh/FOT/FEB07/arecibo_proposal_sc_justi_2.pdf Дата изменения: Thu Feb 1 20:13:40 2007 Дата индексирования: Mon Apr 11 17:21:15 2016 Кодировка:

Full Stokes 327-MHz Continuum Observations of the GALFACTS Pilot Region
Wasim Ra ja, Avinash Deshpande, Tapasi Ghosh, Chris Salter & Andrew R. Taylor Background and Scientific Justification Several existing surveys have mapped the polarized intensity of the extended Galactic emission at various frequencies using single dish as well as synthesis imaging. Derived images of the net polarization percentage, position angle and the apparent rotation measure (RM) have revealed rich structure over the range of angular scales probed. In almost all of such studies made so far, the spectral signatures of the polarized intensity have been studied to seek only a single RM value per image pixel, which would correspond to a RM in the foreground of the most dominant polarized emission component along the relevant line of sight. It is expected however that the extended polarized emission is well spread out in depth along each sight-line, in principle over the entire extent of the Galaxy, and not merely across the sky-plane (i.e. ., the two angular coordinates). It has been argued and shown that the Faraday modulation signatures in the spectral domain for polarized emission can be used to distinguish between the polarized components originating from different Faraday depths [Burn (1966), Ramkumar & Deshpande (1999) and Brentjen & Bruyn (2005)] providing a distribution of the polarized emission along the sight-lines. Recent feasibility study by Deshpande, Salter & Ghosh (A1765) have assessed the feasibility of performing Faraday Tomography with the Arecibo dish using an appropriate combination of observing frequency, bandwidth and spectral resolution. While their results are encouraging, the data consisted of only 1-D cuts in RA and DEC plane. The picture that is expected to emerge when the full potential of this method is realised, would consist of a polarized intensity data cube (quite like the spectral-line data cube) with two dimensions being the sky coordinates and the third being the RM. The spectral resolution (inverse of it) determines the unaliased RM-range that can be probed, while the bandwidth (i.e. the frequency span) determines the resolution attainable in RM. We wish to draw attention to the available continuum polarization data sets from the pilot study of GALFACTS with L-band Wide receiver (A1863) and also the ALFA receiver (A1947), where a 100 в 10 patch of the sky centred around RA 07h 08m and DEC 110 30' was imaged. The band-averaged Stokes Q and U image, obtained from these pilot observations, also reveal rich structure that has no counterpart in Stokes I (see for example, the several images at http://www.ras.ucalgary.ca/GALFACTS/). Several dark patches and canal-like structures in the polarized intensity are most likely due to depolarization of otherwise highly polarized emission along sightlines where there is rapid variation of Faraday rotation. This effect would persist even when viewed at high spectral resolution. The underlying distribution of the polarized emission and the magneto-ionic medium would be apparent in the images resolved across Faraday depth. The L-band data (even with 400 MHz bandwidth) offers resolution of only about 140 rad/m2 in RM. But by combining these data with a suitably lower frequency measurements in full Stokes (spanning a wider wavelength-square range), the resolution in RM can be improved significantly. With this view, we propose to observe the same region in full-Stokes at 327-MHz, which will improve the RM resolution to about 5 rad/m2 , allowing us to perform detailed Faraday tomography. The advantage of choosing this region, over any other interesting region, is that L-band data which would have required a large amount of observing time is already available. We note that the 327-MHz data would have much less angular resolution compared to that at L-band, and would be an additional source of depolarization. However, we will still be able to probe the polarized emission coherent over transverse length scales that are relatively large (for example, about 15 Parsecs or greater in the transverse direction, for a region at 4 kpc from us and using the angular resolution at 327 MHz). Although we will miss the details at arc-minute resolution attainable with interferometric images, the single dish mode will sample the crucial low spatial frequencies (corresponding to extended emission) which are otherwise not present in the synthesized images. Many pointings of the quarter degree beam at 327 MHz will contain many discrete sources each, given that we will be heavily confusion limited. The proposed tomography in fact will allow us to overcome the confusion, since such sources would be separated/resolved along the RM axis, much as expected for the diffuse component.


The expected images revealing the distribution of extended polarized emission along the RM axis, are eventually to be translated to that along physical line-of-sight distances. Possible reversals in the galactic magnetic fields and other non-uniformities make such translation difficult, particularly in the absence of any RM independent distance information. In general, certain discrete sources (e.g. pulsars, SNRs) within the Galaxy, for which independent distance measurements exists, can be used for calibration. In any case, the results from the proposed study would provide instructive assessement of several of these issues. Future Faraday Tomographic studies extended to wider regions of the Galactic plane, would provide unprecedented guidance and constraints for the modeling of the Galactic distributions of the large-scale magnetic field and free-electrons, thermal as well as non-thermal. Observation & Time request We propose to observe the same 100 в 10 test region (as in A1863 and A1947) centered around RA 07h 08min & DEC 110 30' recording full Stokes data in the meridian-nodding mode at 327 MHz using the Arecibo Telescope. The scans in elevation/declination would span 1.2 deg (with 0.1 deg extra at the two ends), so as to sample the 1 degree wide strip without any significant edge-effects due to turn-arounds. With a scanning speed of 1.5 deg/min in elevation, and with 12 seconds turn-around time at each end, one "up" plus one "down" would then take 2 minutes (0.5 degrees in R.A.). Sampling at 6 arc-mins in RA would require a total of 5 days to obtain a full coverage of the region. Hence we request for a total of 5 x 1.5 = 7.5 hours of telescope time, including time to also observe suitable (polarized, as well as unpolarized) calibrator sources, and other overheads. We plan to sample as much of the 50-MHz band of the 327-MHz Gregorian receiver, and would like to cover at least 30 MHz of RFI-free bandwidth with two WAPP boards, each sampling a 25 MHz bands with suitable overlap. We would record 2048 correlation channels (3-level, 32-bit output) in full Stokes per 25-MHz band at every 2-millisecond intervals. We also plan to use a correlated winking noise-CAL (switching at 25 Hz, LOW-CAL) for the dual-polarization receiver calibration during all our measurements. An rms noise of about 400 micro-Jansky/beam is expected in the Stokes-I image, after all the scans are combined using the well-known basket-weaving technique. References http://www.ras.ucalgary.ca/GALFACTS Burn, B. J., 1966, MNRAS, 133, 67 Ramkumar, P. S. & Deshpande A. A., 1999, JAA, 20, 37 Deshpande, A. A., Salter, C. & Ghosh T., 2002, Arecibo Observing Proposal A1765 Taylor et al., 2003, Arecibo Observing Proposal A1863 Taylor et al., 2004, Arecibo Observing Proposal A1947 Brentjen, M. A. & de Bruyn, A. G., 2005, A&A, 441, 1217