2. Observations and imaging
We have four sets of observations between July 1995 and November 1997, measured using two different radio interferometers. Three observations were done with the Australia Telescope Compact Array (ATCA) at cm and one with the Very Large Array (VLA) at cm. During that time the jovicentric value of changed from -2.9o to +0.03o. Table 1 gives some of the relevant parameters.
Each of the three ATCA observations were made over ten days, with 11 to 12 hours per day, in two five-day sessions separated by a five-day gap. This observing scheme optimizes the Fourier plane coverage at many longitudes. This is needed because the ATCA, unlike the VLA, has limited instantaneous sampling of the Fourier plane (it is an east-west array), The VLA observations were made for about 8 hours per day on four days, 6, 7, 11, and 12 May, thus covering more than two complete Jovian rotations. All of the data sets are of excellent quality except that of November 1997 where the Fourier coverage was incomplete due, in part, to instrumental problems.
This paper is based on two-dimensional images constructed every 20o of central meridian longitude (CML) from data recorded at the given . A standard construction technique was used to ensure the images from the different epochs were directly comparable. After normal calibration, a 350 K black body disk (corresponding to Jupiter's thermal emission) and background confusing sources were subtracted from the data. The data were adjusted so the resultant images would represent Jupiter at a standard orientation and distance of 4.04 AU. We used standard radio interferometric imaging followed by maximum entropy deconvolution. The resulting images were then convolved to a consistent resolution of about 70% of the point-spread function width (i.e. the images are mildly superresolved) and the residuals folded back in. Although all four Stokes parameters were measured, here we use only the images of total intensity, converted to brightness temperature.
2.1. Sample images showing the east-west asymmetry
Fig. 1 shows two images from the ATCA in 1995 when and two from the VLA in May 1997 when . The two images of each pair were made from data separated by 180o of Jupiter rotation, i.e., and 200o. The region on the east (i.e. left) limb when is on the west (right) limb at 200o. CML is awkward to use because it is defined with respect to the observer, not with respect to longitudes on Jupiter. It is better to use , the system III longitude on Jupiter itself. To convert from CML to , we add 90o to the CML for east limb data and subtract 90o from west limb data. As seen in Fig. 1, the region at is on the east limb when and on the west limb when .
The 1995 ATCA images on the left, when , show a distinct east-west asymmetry. For example, the brightness of the region at , on the west limb in the top image, is lower by about 300 K than it is on the east limb in the bottom image.
In contrast, the VLA images on the right, when was nearly 0o, have almost no east-west asymmetry. The brightness at is approximately the same on both the east and west limbs.
The radiation in front of the disk of Jupiter is visible in both the ATCA and VLA images. In the top images at the northern end of the magnetic dipole is tilted away from the observer, so the equatorial radiation belt appears concentrated to the north of disk center. At the northern end of the dipole is tilted toward the observer so the equatorial radiation is larger to the south of disk center.
The brightness temperature of the radiation in front of the disk is only about 400 K, i.e. much smaller than above the limbs. This "limb brightening" effect is due to the optical thickness being larger above the limbs. In particular the optical thickness is largest at limb passage of locations where the observer is located exactly in the plane tangent to the local magnetic equator. This phenomenon is discussed below in terms of and the warp of the magnetic equator as described by the magnetic declination .
2.2. Brightness vs. in the dipole equator
We have images similar to those of Fig. 1 every 20o of CML, from which we have taken cuts along the "dipole equator". Here we define the "dipole equator" to be the magnetic equatorial plane of a dipolar approximation to the Jovian field. We take this dipole as having its north pole tilted from the rotation axis towards by 10o. Our objective is to compare brightnesses east vs. west as different Jovian longitudes traverse the limbs. To do this we have converted the east and west halves of the equatorial cuts from CML to and then have joined them in the middle. Fig. 2 shows the results. (Leblanc et al. (1997) showed a similar figure from ATCA observations at 13 cm, without the conversion from CML to .)
In the right panel, the VLA observations at , the brightness east and west of center is almost the same at all . In contrast, in the left panel, the ATCA observations at , there is a distinct east-west asymmetry: the brightness of most regions at east limb passage is larger than at west limb passage, and the peak brightness occurs at different longitudes on the two limb passages: and respectively.
The difference in the two sets of observations can be attributed solely to , and we will now show how the asymmetries are related to and the magnetic declination .
© European Southern Observatory (ESO) 1999
Online publication: June 6, 1999