SpringerLink
Forum Springer Astron. Astrophys.
Forum Whats New Search Orders


Astron. Astrophys. 357, 351-358 (2000)

Previous Section Next Section Title Page Table of Contents

4. Discussion

As we have seen in the preceding section, highly asymmetric Stokes V profiles may arise from superposed atmospheric layers with different magnetic and flow properties. The conditions, however, for which large asymmetries and significant amplitudes are obtained are somewhat complex.

Canopy configurations produce very large asymmetries only for unrealistic spatial resolution; the amplitudes of these Stokes V profiles are invariably very small. When spatially averaged to a degree which would make the computed profiles comparable to observations, the Stokes V amplitudes become larger but the asymmetry values fall far below those of a one-lobe type. These statements, however, are only true if the run of temperature with depth corresponds to a quiet sun atmospheric model. A different temperature profile as, for example, that shown in Fig. 8 enhances the asymmetry to a degree which allows the canopy configuration to produce a one-lobe profile.

Embedded flux tubes are more promising because they produce very large asymmetries together with fairly large amplitudes even when spatially averaged and with a quiet sun temperature profile.

A vertical magnetic flux tube fans out with height, so that in some small region of the resulting canopy the condition for extreme asymmetry is met. However, for spatially unresolved tubes, the contribution of the central part of the tube dominates the aggregate Stokes V profile, which therefore is only moderately asymmetric. Strong internal velocities and shocks may deform the Stokes V profile considerably but do not lead to one-lobe profiles (Steiner et al. 1998). Interpreting the observed one-lobe profiles by superposed layers therefore requires either very extended canopies, which may form at the periphery of network patches, or inclined flux tubes, perhaps loops with an internal flow. Note, however, that the virtual absence of significant signals in Stokes Q and U indicates that the field cannot be strongly inclined ([FORMULA]o according to our calculations).

Two alternative possibilities come to mind for the origin of extremely asymmetric Stokes V profiles: Mixed magnetic polarity in the resolution element and magnetic micro-structuring of the atmosphere.

Mixed polarity: Superposition of (say, two) modestly asymmetric Stokes V profiles arising from patches of opposite magnetic polarity within the resolution element can lead to strongly asymmetric profiles if the values of internal velocity and Stokes V amplitude are nearly equal (e.g., Rüedi et al. 1992). We would then expect a whole spectrum of strongly asymmetric profiles, including double-humped profiles with two lobes of the same sign. One-lobe profiles would then represent a very special case, because one pair of corresponding lobes of the two superposed profiles has to cancel nearly exactly. We should therefore expect more double-humped than one-lobe profiles if this mechanism were a dominant source of one-lobe profiles. This is not the case, however, because we found 1051 one-lobe profiles in our data set but only 227 double-humped ones. On the other hand, abandoning the assumption of statistically uncorrelated components, it is possible to design small magnetic loops with appropriate velocity fields which, when their foot point regions are spatially averaged, will give rise to one-lobe Stokes V profiles. However, evidence against this scenario, and against the mixed polarity concept in general, is derived from the fact that a similar data set from a region of emerging magnetic flux with much more mixed polarity than the quiet sun data does show a smaller fraction of one-lobe profiles (Sigwarth 1999).

Magnetic micro-structuring: Sánchez Almeida et al. (1996) have shown that an atmosphere consisting of an ensemble of optically thin magnetic features embedded in a field-free gas with relative velocities between magnetic and non-magnetic components yields a large variety of asymmetric Stokes profiles (presumably including one-lobe profiles), depending on the choice of model parameters. Such an ensemble may be regarded as generalization of our canopy and flux-tube models insofar as now many interfaces contribute to the total asymmetry, each with a small amount.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 2000

Online publication: May 3, 2000
helpdesk.link@springer.de