The solar outer atmosphere is highly dynamic and inhomogeneous, even in regions where the magnetic fields can be ignored. Acoustic waves, both traveling and evanescent, in a range of frequencies and wavelengths result in a complicated and rapidly evolving pattern of beats in which the compression waves are modified by three-dimensional radiative transfer that has yet to be modeled in detail. In addition to these waves, there are the more gradual undulations associated with adjustments of the mean upper photospheric structure in response to granulation and other scales of convective overshoot discussed in Section 2.5. In this section we first concentrate on the dynamics of the nonmagnetic atmosphere, and then we discuss its temperature structure. The term nonmagnetic does not imply that there are regions in the outer atmosphere that are entirely free of field (see Section 4.6), but rather that in these regions the magnetic field is weak and does not inhibit the plasma motions in flows or waves. And although the focus here is on the nonmagnetic outer atmosphere, we discuss its dynamics in comparison to the magnetically dominated regions.
At photospheric heights, the oscillatory properties of the centers of supergranules and of the magnetic network appear to be rather similar: both are characterized by the broad p-mode spectrum with periods near 5 min (e.g., Kulaczewski, 1992). Higher up in the atmosphere, however, a clear difference is observed. The power spectrum of intensity fluctuations in chromospheric lines observed over cell centers are dominated by 3-min oscillations (Von Uxkull et al., 1989, Deubner and Fleck, 1990, and Lites et al., 1993b), which lose their coherence with increasing height, until only occasional coherent wave trains are observed at transition-region temperatures above ~50,000K. Deubner and Fleck (1990) argue that the 3-min periodicity reflects at least in part a trapping of modes between the temperature minimum and the transition-region temperature rise, while small-scale inhomogeneities in the structure and vertical extent of the cavity cause a broad distribution in the power spectrum. It is not necessary to have a resonating cavity to see these waves, however, because any disturbance will excite waves at the local cutoff frequency.
The evolving wave patterns result in a patchy undulation of the intensity in the chromo-spheric domain, similar to what is seen in the photosphere. Apart from these relatively large patches, much more localized phenomena are also observed. Rutten and Uiten-broek (1991), for instance, review the properties of theCaII K2V grains or bright points. They point out that narrow-band spectroheliograms taken in the H and K line cores show, apart from the many clusters of small bright elements in the magnetic network, the so-called internetwork bright points, cell flashes, or cell grains (referred to in Section 2.7) that are observed within supergranular interiors. These grains are 1,500 km or smaller in size, and persist for ~100s each time they brighten. They tend to recur several times with 2- to 5-min intervals between successive brightenings. Rutten and Uitenbroek (1991) proposed that these features are related to the inner-cell bright points seen in the ultraviolet continuum near 1,600 A (which originates near the classical temperature minimum, which appears, as we pointed out in Section 2.7, not to exist as such in nonmagnetic environments) and they argued that these grains are a hydrodynamical phenomenon of oscillatory nature. They speculate that these grains are a direct result of the acoustic processes in the nonmagnetic chromosphere, which is now supported by simulations (Section 2.7).
An even lower coherence of waves at still higher formation temperatures is found in the early work of Vernazza et al. (1975). Using raster spectroheliograms obtained with the Apollo Telescope Mount on board Skylab, taken 1 min apart with a resolution of 5 arcsec, they found substantial power in the short-term fluctuations that occur synchronously in a few spectral lines. The average lifetime of the brightenings is 70 s, and their mean occurrence is spaced by 5.5 min, but no strong periodic oscillations are seen. The brightenings occur in the central regions of the network cells as well as in the boundaries.
Athay and White (1979), in contrast, found that the C IV k 1,548 line intensities, observed by OSO-8 with an effective aperture of 2" x 20", show relatively frequent periodic oscillations in the 3- to 5-min range. The periodic oscillations have a short coherence length and a tendency to be mixed with prominent aperiodic fluctuations. Athay and White propose that this mixing prevented other studies from uncovering these periodic variations. They interpret the much more frequent low-amplitude aperiodic intensity fluctuations as sound waves whose periodicity and coherence are destroyed by the variable transit time through an irregularly structured chromosphere (see the review by Deubner, 1994).
Over the magnetic network, the oscillations are much more stochastic in nature. The power spectra for the strong network are dominated by low-frequency oscillations with periods of 5 up to 20 min. Interestingly, Deubner and Fleck (1990) find that the spectra of areas of atmospheric emission intermediate to the cell interior and the strong network are intermediate to the abovementioned spectra, displaying a clear 3-min component; this suggests that magnetic and acoustic energy dissipating processes coexist within the same resolution element.
During the past two decades, evidence has been found - mainly from carbon-monoxide infrared spectra - that there are also very cool parts in solar and stellar outer atmospheres, with temperatures down to below 4,400 K, which have been interpreted as being caused by a thermal bifurcation of the chromosphere (e.g., Heasly et al., 1978, for a Boo; Ayres, 1981; Ayres etal., 1986; and Solanki etal., 1994). Initial numerical modeling suggested that the strong CO-line cooling could cause such a thermal bifurcation (e.g., Kneer, 1983; Muchmore and Ulmschneider, 1985), but more recent simulations by Anderson (1989) and Mauas et al. (1990), which incorporate many CO lines and the CO dissociation equilibrium in time-independent models, yield no evidence for an instability.
The CO lines go into emission some 300 to 1,000 km above the solar limb (Solanki et al., 1994; Uitenbroek et al., 1994), so that the cool material is truly superphotospheric. The formation height of the CO lines apparently samples a substantial height interval because of the presence of power near 3 min - characteristic of chromospheric oscillations -as well as 5 min - characteristic of photospheric oscillations - in CO spectra (Uitenbroek et al., 1994). Ayres (1991a) proposed that a distribution of cold columnar regions with a collective surface filling factor of less than some 20% (causing increasing shadowing of cool areas in front of others as one looks progressively nearer to the solar limb) would be consistent with the observational constraints (see the discussions by Athay and Dere, 1990, and Ayres et al., 1986). The response of the atmosphere to acoustic waves makes these warm and cool regions transient, resulting in evolving patterns across solar and stellar disks.
Was this article helpful?