|
|
We found, however, that the intrinsic position angle of the MWC 1080 system
implied by this ISM polarization
calculation does not agree with the position angle along which the polarization
varies in the Q-U plane. In addition, neither angle seems related
to the observed position of the outflow from this object (Poetzel et al. 1992). Closer inspection
revealed that both the magnitude and the direction of polarization vary, and
that observations close together in time are more similar than observations
close together in binary phase (Figure 2). This implies the polarization
variation in MWC 1080 is mostly random and not due to orbital effects.
|
BM Ori (theta-1 Ori B) is one of the Trapezium stars in the Orion Nebula.
It is a spectroscopic binary (the bright single object on the left side of the
image, not the resolved pair) whose primary star is a B-type zero-age main
sequence star. The secondary, which is shrouded in dust, is likely a pre-main
sequence object. Spectropolarimetric observations can, in principle, be used
to constrain the shape and characteristics of this dust envelope.
However, given our findings for MWC 1080 above, we must proceed with caution.
| |
Image credit
Figure 3
Enlarged image and caption
|
|
We also observed BM Ori with HPOL at
WIYN in 1998 and 2000. We
have better phase coverage of BM Ori than of MWC 1080, but due to difficulty
removing the nebular contamination from our data, we were unable to obtain an
ISM polarization estimate from the line polarization. We do, however, observe
clear polarimetric
variations with phase, indicating that the object possesses some intrinsic
polarization (Figure 3). Perhaps most interesting is the suggestion of a change
in both polarization magnitude and position angle at secondary eclipse, which
is so shallow in the light curve as to be nearly undetectable.
We can derive interstellar polarization estimates from the
behavior of this variation in the Q-U plane. However, this derivation
(and any analysis of the polarimetric behavior at secondary eclipse)
depends on the assumption that the variations are repeatable with phase.
Our data, which were taken at several separate epochs, cannot support this
assumption.
|
Implications of these results: Though the assumption of polarimetric
repeatability is justified in the study of evolved binary systems, our data
show that it is highly suspect in the case of young binaries. Individual
Herbig Ae/Be stars are known to be polarimetrically variable, and we find
(from our MWC 1080 data) that this intrinsic variability may be comparable to
or larger than any phase-locked variation due to orbital motion. This
uncertainty makes observation and analysis of such systems more difficult,
as we saw with BM Ori. Thus, in order to investigate disk properties in a
pre-main sequence binary system by observing orbital polarization, one must
observe an entire binary cycle (or, better, several complete binary cycles).
Though such a concerted effort would require high-quality polarimetric data
and large amounts of telescope time, it appears to be necessary in order to
separate orbital polarimetric variations from random ones.
| |
