Forum Springer Astron. Astrophys.
Forum Whats New Search Orders

Astron. Astrophys. 357, 743-756 (2000)

Previous Section Next Section Title Page Table of Contents

6. Discussion and conclusions

6.1. Quiet Sun

Our observations and analysis indicate that most of the emission from transition region temperatures ([FORMULA] K) comes from fine structures, which appear to be low lying loops across the network boundaries. This is consistent with the type of model proposed by Feldman (1983) and Dowdy et al. (1986). Feldman et al. (1999) published large images produced by SUMER in the C iii (977 Å) and S vi (933 Å) spectral lines corresponding to temperatures of around [FORMULA] K and [FORMULA] K respectively. They observed low lying loop structures in C iii emission apparently randomly and sparsely distributed in the cell centre, but intense and clumped across the network boundaries. The signal to noise was not as good for the S vi emission, so the fine structures were not so obvious. Our results confirm their conclusions on the morphological nature of the quiet solar transition region. We have extended their work with better observations at [FORMULA] K and we provide a detailed, quantitative investigation into some of the physical properties (electron density and temperature, filling factor, plasma motions) of these unresolved fine structures. We are not able to draw any conclusions about whether or not the coronal emission is also confined in loop structures, since the Mg x emission which we observe is too weak. Other observations, from TRACE, suggest that the coronal emission from the quiet Sun is also predominantly in loop structures. It is difficult to conceive how transition region models based on the idea of a transition region as a thin, continuous interface between the photosphere and the corona could reproduce these observations.

There are a few different models in the literature which attempt to explain the down-flows seen at transition region temperatures. Hansteen (1993) and Hansteen et al. (1996) considered nano-flares occuring at the top of coronal loops. Following the suggestion of Peter & Judge (1999), the numerical model by Teriaca et al. (1999b, 1999c) extends these ideas and proposes that the pattern of Doppler shift as a function of temperature can be explained by the occurence of nano-flares in magnetic loops at a temperature of around [FORMULA] K. This new insight into transition region dynamics explains the net redshifts of 11 km s-1 found for lines in the quiet solar atmosphere formed at around [FORMULA] K. Their simulations predict intensities for C iv, O vi and Ne viii lines. There is a tendancy for the higher intensities for the C iv spectral line to be more redshifted. In order to reproduce our observations, the nano-flare process would need to be continuously taking place in the cool transition region loops. We note that this model predicts that the plasma is not in ionisation equilibrium. It would be interesting to know if such numerical models can reproduce the specific behaviour of line-shift versus intensity which we have determined.

Teriaca et al. (1999d) claim that their model is also able to explain the distribution of non-thermal velocities as a function of temperature. Indeed, they propose that the larger range of values detected for active regions could be explained in terms of a higher frequency of occurrence and/or energy deposition of nano-flares in the middle-high transition region. Again it would be interesting to learn if such models could predict the strong correlation which we have found between intensity and non-thermal velocities.

Fig. 8 (top) shows that the electron density does not change markedly between darker plasma and structure. Intepreting darker regions as cell plasma and bright regions as structured network plasma, means that the electron density values at transition region temperatures are similar in the network and cells. A similar behaviour has been recently noted by Del Zanna & Bromage (1999) in coronal hole and quiet Sun observations obtained by the Coronal Diagnostic Spectrometer on SOHO using O iv lines. Earlier measurements by Dupree et al. (1976) using C iii lines observed by Skylab show a similar trend. Our measurements disagree with those by Vernazza & Reeves (1978), who report higher densities in network areas relative to cell centres as obtained from C iii lines. Our results suggest that the differences in intensities observed between cell centres (dark areas) and network (bright structured plasma) are mainly due to differences in filling factors in these different areas.

6.2. Dynamic activity

A comprehensive review of the characteristics of explosive events from HRTS observations of C iv was provided by Dere et al. (1989) and Dere (1994). It is of interest to compare our results with theirs. The size, maximum velocities and lifetimes we obtain are comparable. The profiles are also usually asymmetric. Dere (1994) suggests that the explosive events are caused by magnetic reconnection near the edges of strong network field regions. He said that `the basic picture is that small magnetic bipoles emerge in the supergranular cell centres and are transported to the boundaries where opposite polarity flux elements are driven together by photospheric flows and forced to cancel'. This type of scenario does indeed seems to have been confirmed by more recent SOHO observations. Dere found that the explosive events appear to occur low down, just above the photosphere.

The numerical model by Sarro et al. (1999), Erdelyi & Sarro (1999) allows for an energy release to take place in a semi-circular flux tube, at a height just below the transition region. This generates upwards and downwards flows along the loop as well as a rise in temperature and sound waves which develop into shocks. As a result, new transition regions moving at opposite velocities are created. The signature of these flows can be associated with distorted line profiles with red and blue components. If there is an ensemble of filamentary quiescent loops present, with just one loop undergoing an energy injection, then a central profile with perturbed wings would be observed, consistent with observations.

A sophisticated numerical model of explosive events in a 2D environment is presented by Roussev et al. (1999). Preliminary results are encouraging, but detailed spectroscopic predictions are still required. Innes & Toth (1999) present a model for explosive events, based on compressible MHD simulations of the evolution of a current sheet to a steady Petschek, jet-like configuration. They are able to reproduce the broad line profiles with extended wings in the temperature range [FORMULA] K to [FORMULA] K. They also explain how jets flowing outward into the corona are more extended and appear before jets flowing towards the chromosphere.

These numerical models, combined with a comprehensive analysis of observational evidence, provide a real opportunity for understanding the nature of the transition region fine structures.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 2000

Online publication: June 5, 2000