Last modified: May 2001
The sections in this page are:
The VSOP Polarization Study Team (PST) has completed its initial investigation into polarization imaging and the impurity of the HALCA antenna/feed system. Based on two experiments conducted in late October 1997, the best estimates of RCP leakage into HALCA's LCP feed are approximately 10% at 5 GHz and 3% at 1.6 GHz. These values were determined using standard calibration procedures and existing software in AIPS. These findings are described in this memo (1.05 Mbytes gzipped postscript), and have been published in in Kemball et al., 2000, Publications of the Astronomical Society of Japan, 52, 1055.
HALCA receives only left circular polarization (LCP). However, the use of dual-polarization radio telescopes on the ground may enable images of the polarized intensity to be made. The VISC established a Polarization Study Team to investigate the feasibility of polarization imaging with HALCA. The Study Team determined that the polarization impurity of HALCA's feed is sufficiently small that useful polarization observations are possible. Proposers are reminded that, unlike most ground-based telescopes used for VLBI polarimetry, HALCA does not rotate about the source direction and cannot slew to calibrators, and this significantly complicates calibration of the polarization impurity. When combined with HALCA's small aperture and limited polarized (u,v) coverage (only LCP) it is clear that VSOP polarization imaging proposals should be considered with great care. Further details are available from Section 9.10 of the VSOP Proposer's Guide.
The default observing mode for polarization experiments is for ground radio telescopes to record one, 2-bit sampled, 16~MHz channel each in both LCP and RCP, however `double-rate' recording (i.e. two, 2-bit sampled, 16~MHz channels each in both LCP and RCP) is possible with VLBA/MkIV systems and will be considered if a significantly strong scientific case is made.
When performing VLBI polarization observations, one usually thinks in terms of a VLBI experiment in which all telescopes record simultaneously both the left- and right-circularly-polarized (LCP and RCP) signals incident from a compact radio source. Information about the distribution of the total intensity emission and possibly the circularly polarized emission is derived from the sum (Stokes I) or difference (Stokes V) of the LxL and RxR (``parallel-hand'') correlations. Information about the distribution and position angle of linearly polarized emission is derived from the sum (Stokes Q) and difference (Stokes U) of the RxL and LxR (``cross-hand'') correlations. With modern ground-based arrays (e.g., the VLBA) that were designed with polarization observations in mind, such experiments are becoming more routine. However, for several reasons, polarization observations with HALCA will be considerably more difficult than the ground-based case.
First, HALCA receives only one hand of circular polarization (LCP). Unlike the real total intensity image, the complex linear polarization image requires both cross-hand correlations for each baseline to symmetrically sample the (u,v) plane. Since only one of the two cross-hands will be available from HALCA correlations (ground telescopes will observe both LCP and RCP), only half of the (u,v) plane will be filled with polarized visibilities on the space baselines. While formation of polarization images is still possible via complex deconvolution techniques developed to accommodate this difficulty in ground-based VLBI polarimetry, proposers should be aware that full-resolution space VLBI polarization imaging (like total intensity space VLBI imaging) will have stricter dynamic range limitations than is typically realized in full-polarization ground-based VLBI.
The second, and more fundamental issue connected with space VLBI polarization observations is calibration of HALCA's instrumental contribution to the polarization visibilities, i.e., its feed impurity. Like circularly polarized feeds on ground radio telescopes, HALCA's LCP feed's response to RCP is not identically zero. Since this `leakage' of RCP will correlate with the RCP signal from ground radio telescopes at a level proportional to the total intensity correlation of the radio source, it must be calibrated and removed. For ground arrays, this calibration has traditionally been achieved by making observations of a source of known (usually zero) polarization as it diurnally rotates in the field of view of the telescopes. Since the instrumental contributions (assumed constant) of each telescope rotate differentially in phase with respect to those of the others as well as to the source contribution, the instrumental contribution of each feed may be solved for. More recently, it has become possible to solve for both the instrumental and the source polarization (with some assumptions about the polarized intensity distribution), thus eliminating the need for a calibrator of known polarization.
Unlike ground telescopes used for VLBI polarimetry, HALCA cannot slew to sources of known polarization and it does not rotate about the source direction. Therefore, it will be difficult to calibrate HALCA's feed impurity by conventional means. Additional difficulties may arise if the polarization impurity is significantly variable in a non-predictable way on the timescale of the observations. Such variations could be caused, for example, by changes in the shape of the HALCA antenna as it comes out of eclipse. Furthermore, the polarization properties of the antenna may also be a function of the angle between the source and the Sun.
Possibilities for achieving reliable polarization calibration for HALCA are currently under investigation by a Polarization Study Team established by the mission. It is expected that the ground array instrumental polarization will be calibrated independently, then simultaneous solutions for HALCA's polarization impurity and the source structure on space VLBI scales will be obtained and tested for consistency with the ground-only results.
In addition to the fundamental issues described above, there are several logistical issues to be considered when planning space VLBI polarimetry with HALCA.
First, since HALCA polarization observations will require recording of both RCP and LCP at all capable ground telescopes, the default observing mode at these telescopes will be to record one, 2-bit sampled, 16 MHz IF channel each in both RCP and LCP (please refer to Section 3 and Table 2 for a description of the available HALCA recording modes) and one of the HALCA IF channels will be discarded. This is necessary to avoid doubling the tape requirements at these stations by increasing the total bit-rate from 128 Mbit/s to 256 Mbit/s. However, the larger bit rate will be considered, and the full HALCA bandwidth used, if a sufficiently strong scientific case is made. Another option--a mode in which the ground telescopes record both 16 MHz IF channels with 1-bit sampling in both polarizations--yields better sensitivity than the single channel mode while maintaining the 128 Mbit/s bit-rate limit at the ground telescopes. Unfortunately, this mode is currently not available due to limitations in the current FITS specification for correlator output data. Should this change, and this mixed mode become available, it will likely become the default for polarization observations.
Second, it is usually necessary to obtain a calibration of the electric vector position angle for polarization observations. In practice, this reduces to calibrating the phase difference between the RCP and LCP systems at the reference antenna and is usually achieved by including observations of a highly-polarized compact source (e.g., a BL Lacertae object) for which the correct position angle of the integrated emission is known or measurable with time granted by an independent single-dish telescope or the VLA. Since this property of the reference antenna can be measured with ground telescopes alone, the observational constraints of HALCA have no direct bearing on obtaining this calibration. In fact, in typical HALCA imaging observations, there are usually several gaps in the tracking station coverage during which the ground array could observe such calibrators. However, since scheduling of the ground array will be undertaken by the VSOG, proposers should explicitly indicate their requirements in this regard on their cover sheets (with supporting text in their justification), including a list of acceptable calibrators. Proposers should assume responsibility for separately obtaining the supporting single-dish observation, although it is possible that appropriate ground telescopes may be scheduled as part of the HALCA observation which would be similarly available during the HALCA tracking gaps, and thus could be properly scheduled for this purpose. Also, if the 256 Mbit/s recording rate is desired, proposers should be aware that very little tape may be available for periods when HALCA is not observing, i.e., observations of the proposed target source are given priority by default to maximize the space-baseline coverage of all observations.
Several papers related to VSOP polarization observations have been published. The January 2000 VSOP Symposium proceedings contain the papers "Multi-Frequency VSOP Polarization Observations of the BL Lacertae Object 1803+784" by D.C. Gabuzda, and "Polarization-Sensitive VSOP Observations of Bright Quasars and gamma-Ray AGN" by G.A. Moellenbrock, D.H. Roberts and J.F.C. Wardle. The Polarization Study Team's report, described above, has been published in Kemball et al., 2000, Publications of the Astronomical Society of Japan, 52, 1055.