The Evershed effect seen in sunspot penumbrae is characterized by spectral line shifts whose sign is compatible with a material outflow. The magnitude of these shifts, obtained using visible lines, range between 1-2 km s-1 at low spatial resolution and 6 km s-1 at high spatial resolution. The flow seems to be concentrated in the dark penumbral filaments. For more information on the Evershed effect we refer to the reviews by Thomas (1994), Martínez Pillet (1997) and Solanki (1997).
The Evershed effect is usually explained by means of the siphon flow mechanism which was originally proposed by Meyer and Schmidt (1968). In their model the gas pressure difference between the two footpoints of the loop is due to the difference in field strength between these footpoints (the flow is directed from the footpoint with weaker field to that with stronger field). Originally it was thought that the flow starts inside the penumbra and ends in strong-field flux-tubes outside the sunspot. Recently, however, Westendorp Plaza et al. (1997) presented new observations which indicate that most of the flow returns into the solar interior in the penumbra, in agreement with the mass-balance arguments of Solanki et al. (1994, 1999). These observations, coupled with the fact that the field strength decreases from the inner penumbra to the outer boundary of the sunspot, place severe constraints on a possible siphon flow of the type proposed by Meyer & Schmidt (1968). Thus, although Thomas & Montesinos (1993) and Montesinos & Thomas (1997) could reproduce certain features of the observations with flows starting and ending in the penumbra or ending in the field-free surrounding, they had to assume that the background penumbral field is independent of the radial distance from the sunspot centre. They analysed the case of a magnetic flux tube embedded in a magnetic environment (corresponding more closely to penumbral conditions).
In this paper we address this problem on the basis of observations at 1.56 µm and 2.2 µm (Rüedi et al. 1998). We also discuss the implications of these observations for the model proposed by Schlichenmaier et al. (1998) in which the flow is accelerated locally in the penumbra through the interchange instability and does not require strong magnetic field concentrations at the downstream footpoints to accelerate it.
© European Southern Observatory (ESO) 1999
Online publication: July 26, 1999