Infrared imaging polarimetry of Cepheus A:
The source of the IR reflection nebula

L. S. Casement and I. S. McLean

A grayscale image of Cep A in the K' band showing the infrared reflection nebula, the location of GGD37, the compact radio source HW-2 and the infrared "peaks" reported by Lenzen et al. 1984. North is up and east is to the left. The field of view is approximately 3' x 1.5'.

Deep 1-2.5 um (JHK) images and linear polarization observations of the infrared reflection nebula in Cepheus A have been obtained using a two-channel infrared camera system with 0.675" per pixel and a large field of view 3' to determine the exact location of the illuminating source which is completely hidden by extinction. Approximately 1,000 independent polarization vectors are used to locate the center of light to an accuracy of 1.5" at J, H, and K. The polarization pattern is complex, well-resolved and exhibits many features characteristic of multiple scattering models in a dusty nebula with an evacuated, bipolar cavity. For this reason, fitting all the polarization vectors to a centrosymmetric pattern is shown to be flawed, and can actually cause errors in positional estimates. A more careful analysis shows that the illuminating source is the same at J, H, and K, and that it is clearly coincident within 3 sigma errors with the compact radio source HW-2 observed at 6 cm, but yet it is apparently displaced 5" south of the FIR source detected at 20 um.

Reference: Casement, L.S. & McLean, I.S., ApJ, May 10 1996 issue
J image of Cep A. The 0,0 position corresponds to HW-2. The image scale is 180 x 90 arcsecs with a plate scale of 0.675 arcsec/pixel. North is up and east is to the left. Polarization vectors are superimposed after binning 4 x 4 pixels for clarity; 100% polarization is equivalent to 10 arcseconds. A total of 48 minutes of integration time was used in this image which 45 seconds of on-chip integration time represented by the contours. There is a small region of missing data in the lower right corner due to an imperfect mosaic.
H image of Cep A, same attributes as Figure 1. A total of 28 minutes of observing time was used with 15 or 20 second on-chip integrations. There is a small region of missing data in the upper right corner due to an imperfect mosaic.
K' image of Cep A, same as for Figure 1. A total of 76 minutes of integration time is shown with the flux levels representing 6 seconds on-chip. There is no missing data in this mosaic. Note the reduced extinction compared to Figure 1.