Astron. Astrophys. 319, 274-281 (1997)

2. Observations and method of analysis

The Australia Telescope Compact Array (ATCA) is located at latitude south, near Narrabri, NSW. It consists of 6 antennas, each 22 m diameter, located along an east-west line. For our observations the shortest baseline was 153 m, the longest 6 km. The telescope operated simultaneously at two wavelengths, 12.6 and 21.7 cm, with bandwidths of 128 MHz broken into 32 channels. Each antenna measures two orthogonal linear polarizations at each wavelength, from which the four Stokes parameters are derived.

Jupiter was observed on 10 days, 12-16 and 22-26 July 1995. The antenna locations were the same for the two, 5-day sessions. The five day gap between the sessions was chosen so that the UV coverage would be approximately uniform as a function of Jovian longitude (e.g. the CML's at transit would be spaced by about ). The Earth's declination as seen from Jupiter was jovigraphic (- jovicentric) and Jupiter's distance was 4.63 AU (giving at that distance). Observations continued for about 11.5 hours per day, the entire time that Jupiter was above the elevation limit of the antennas. Given Jupiter's period, all longitudes were observed each day, and an identical longitude range was observed near rise and set. With Jupiter at DEC - at that time, aperture synthesis produced a good degree of north-south resolution as well as excellent east-west resolution. The synthesized beam was about at 13 cm and at 22 cm. All four Stokes parameters were measured. The data were reduced in the Miriad system (Sault et al. 1995) using conventional editing and calibration (both gain and polarization), and multi-frequency synthesis was used to incorporate the 16 independent channels across the 128 MHz bandwidth. Multi-frequency synthesis significantly reduces the sidelobe level, particularly at 22 cm.

Despite our polarization calibration, for an array with linear feeds, a residual error in the XY phase and absolute feed ellipticity (both typically for the ATCA) allows a small fraction of linear polarization to corrupt the circular polarization, and vice versa (e.g. Sault et al. 1996a). Jupiter has strong linear polarization and weak circular polarization , so the effect is not important for linear polarization, but very important for circular. At present we have little confidence in our images in circular polarization and do not present them here. We remark that only the locations on Jupiter with a significant longitudinal magnetic field component along the line of sight should show circular polarization.

© European Southern Observatory (ESO) 1997

Online publication: July 3, 1998
helpdesk.link@springer.de