5. Discussion and conclusions
Westendorp Plaza et al. (1997) claim that most of the Evershed flow already returns to the solar interior inside the penumbra. This could explain the sudden disappearance of the Evershed flow at the penumbral boundary reported by many observers. However, it means that the field strength of the upstream footpoint is larger than that of the downstream footpoint, which conflicts with the requirements for a siphon flow. Our data defuse this problem since they show that the cool outflowing gas in the penumbra has the same (low) field strength everywhere. A model of the type proposed by Schlichenmaier et al. (1998) also appears to be compatible with our observations, however.
Our observations suggest that the penumbra is composed of at least two different kinds of flux-tubes: those of one type are cool, almost horizontal and possess a low field strength while the others are hotter and can have a larger range of inclination and field strength. The temperature sensitivity of the Ti lines enables us to see the cool structures, unaffected by averaging over different spatial structures such as dark and bright filaments or poor seeing. Due to its low magnetic field strength and low temperature this component is difficult to isolate using visible observations.
There remains the problem of force balance between the two
components, in particular for the traditional siphon flow models. It
could be achieved if the cool gas is denser than the warmer gas,
particularly in the inner penumbra. Recall that the
© European Southern Observatory (ESO) 1999
Online publication: July 26, 1999