SCAN-MAP POLARIMETRY WITH SCUBA: THE CRAB NEBULA

Although SCUBA has had an imaging polarimetry mode since 1998, it has always been difficult to observe large sources. The problems of chopping onto nearby emission are exacerbated in polarimetry because of potentially chopping onto an area with differently oriented polarization, and thus corrupting the source measurement in both degree and direction.

In January 2000 we made the first successful observations with scan-map polarimetry, imaging a source substantially bigger than the SCUBA field-of-view. We used the `EmersonII' scanning technique, where different sized chop throws in both RA and Dec are used to reconstruct the image in Fourier space. To minimise sky variations, the maps were made as rapidly as possible, by using the minimum number of chop throws (four) and scanning at twice the usual rate (giving 6" sampling).

The test source was the Crab Nebula, which has highly polarized synchrotron emission, and is about 6 arcmin in diameter. The results of an hour's observing are shown in the figure: about 200 5-sigma vectors were obtained when binned to beamwidth spacings. To demonstrate reproducibility, we binned the data to 1 arcmin resolution and compared it to the 9 mm polarimetry results of Flett & Henderson (1979; MNRAS 189, 867). The vector directions were found to be the same within a few degrees, and the level variations e.g. to the north-west are also consistent.

The results have not yet been scientifically analysed, but magnetic fields are thought to be very important for the evolution of the Crab, particularly in the wispy filaments that envelop the ionized gas. Radio and far-infrared polarimetry have both suffered from problems, respectively Faraday depolarization and possible contamination by dust emission. The SCUBA data show the highest percentages and may give the best high-resolution picture so far of the magnetic field. There is remarkably uniform polarization across the main nebula, running along the long axis of the structure (the jet axis). A major deviation occurs where the vectors appear to circle the cavity north-west of the bright core - confirming a link between the physical and magnetic structures.

Scan-map polarimetry at 850 microns will be available in semester 01A, but potential observers are cautioned that stable sky conditions are required over timescales of about an hour (somewhat longer than with jiggle maps), and that observing and data reduction are not yet automated. It is not recommended to attempt scan-map polarimetry of very faint (<< 1Jy/beam) or very extended (<< 10 arcmin) clouds, due to the long times over which the sky would have to be stable. The polarized fluxes (S x p) in the Crab Nebula are about 0.5 Jy/beam x 10% or 50 mJy; this is equivalent to attempting to detect 2.5% polarization in a 2 Jy/beam dust cloud, for example. However, since a fractional change in sky transmission causes more error for less polarized sources, getting good S/N may be more difficult than for the Crab.