5 Discussion and conclusion

A series of SOHO/SUMER UV line spectral measurements were made during the passage of an ECH across the disk to search for evidence of higher chromospheric or transition region temperatures in areas of CH radio enhancements. Spectral analysis shows there is no significant electron temperature increase in the enhancement region relative to the quiet sun for temperatures between approximately 10000 K and 230000 K. However, significant changes were found within the ECH between radio-enhanced and non-radio-enhanced regions. UV line emission was greater in the radio enhanced regions as compared with the ECH outside the enhancements. We need more observations in lines formed at 10000 K to arrive at firm conclusions regarding the physical conditions in radio enhancements at the levels of UV line formation.

Paradoxically, the only correlation of increased ECH emission (relative to the quiet sun) with radio enhancements was found with H$\alpha$ and Fe XII emission. The compact radio enhancements coincided with the footpoints of coronal plumes as observed in Fe XII emission, but not with the plume structure above the base. H$\alpha$ emission is formed between 7000 K and 11000 K (Avrett 2000), while the Fe XII emission is formed at 1400000 K. No increase above quiet sun levels was detected in lines formed between these temperature extremes. The radio enhancements cannot be caused directly by the mechanism which causes the Fe XII increase, since the radio brightness temperature is 10000 K, and modeling of the coronal 17 GHz radio contribution shows it is negligible (Gopalswamy et al. 1998, 1999a,b). The coronal plasma at 1000000 K above the radio enhancement formation layer must be heated by some form of energy flowing along magnetic flux tubes, such as wave flux, which would explain the correlation with radio enhancements.

Previous ECH microwave observations have led to speculations that the radio enhancements are caused by increased cross-field heat conduction or reconnection in the upper chromosphere, which would cause a temperature increase at the radio formation layer (Gopalswamy et al. 1998). The SUMER observations presented here provide no evidence of such a temperature increase relative to the quiet sun in the form of increased UV line emission. A higher transition region filling factor as cause of the ECH enhancement was also suggested (Gopalswamy et al. 1999a or b), but this would also cause increased UV line emission, which is not observed. Other spectral studies in the UV and FUV have shown higher inter-network chromospheric line emission, 6 to 7% greater than in non-hole regions, at several locations in ECHs for Ca II K, Mg II K and H I Ly$\alpha$(Bocchialini & Vial 1996), but radio images were not available for comparison. Analysis of network chromospheric line profiles showed a smaller enhancement, 1 to 2%. The area covered by the FUV observations reported here was relatively small, and therefore this increase would not be detectable. Since the radio and H$\alpha$ enhancements occur within the network where fibrils are present, our results are consistent with these observations. However, ECH UV line observations with larger coverage of the disk would allow more accurate statistical comparisons between the enhancement and non-hole regions.

No well defined radio enhancements were found without corresponding H$\alpha$ enhancements. This suggests that the cause of the radio enhancement may be in the middle to upper chromosphere, which is part of the H$\alpha$ formation region. This is consistent with comparisons of polar radio enhancements and He II 304 Å emission, formed at 80000 K, which show a lack of correlation between the 17 GHz radio and EUV bright regions (Nindos et al. 1999). The authors thus concluded that 17 GHz microwave enhancements are formed below 80000 K. In addition, studies of 17 GHz polar emission scale heights show that radio brightenings lie below the 304 Å limb, supporting a formation temperature below 80000 K (Nindos et al. 1999). These results agree with studies of 87 GHz CH enhancements which show no correlation with 304 Å emission (Pohjolainen 2000). Since the H$\alpha$ enhancement is due to increased fibril and bright point emission, the microwave enhancement is likely formed by bright radio fibrils and points. The ECH radio enhancement fibrils are bright either because they have a larger filling factor relative to non-hole fibrils or because the radio emission is formed at a higher temperature. In the ECH radio enhancements, the integrated bright emission in radio and H$\alpha$was approximately equal, suggesting a fibril filling factor increase in both cases. However, both phenomena might also be caused by a temperature increase, for example from 10000 K to 12000 K or higher at the radio formation height. Since the formation of the H$\alpha$ line is complicated, we cannot predict with certainty how the line brightness would be affected. The lack of enhancement in upper chromospheric lines seems to weigh against a filling factor increase. Larger fibrils should cause more upper chromospheric line emission, unless there is a regulating mechanism which fixes the temperature and line-averaged density in that region, regardless of the energy passing through it.

High-resolution images of radio-enhanced CH regions, which could be made using the VLA, might reveal the structure of the ECH radio enhancements at 17 GHz. If the cause is bright fibrils, observations at 2 $^{\prime\prime}$ resolution or better would resolve them. The enhancements may be caused by the combination of concentrated flux tubes and reduced ECH pressure. Fibrils, which form along flux tubes, may have a slightly different structure in high-field ECH regions. Since the coronal pressure is lower, the pressure scale length between the center of the fibril and the corona may be longer, causing both the 17 GHz radio and H$\alpha$ emission to form at a higher temperature. The density at 7000 K to 12000 K might be slightly higher in the enhancements, causing the radio and H$\alpha$ emission to form at higher temperatures, since the optical depth is proportional to the square of the density. The densities in regions with temperature above 12000 K might be similar to non-hole regions, explaining the lack of enhancement in upper chromospheric and transition region lines. Modeling of fibrils in ECHs may help to explain the radio and H$\alpha$ enhancements and lack of ECH UV line emission increases. Another explanation for the enhancements may be nonthermal phenomena. Brightness temperature fluctuations of 300% observed in another ECH were suggested to have a non-thermal cause (Gopalswamy et al. 1999a). However, the factor of three increase in microwave brightness could also be caused by a temporary factor of 2.2 increase in density.

Acknowledgements

The SUMER project is financially supported by the Deutsches Zentrum für Luft- und Raumfahrt (DLR), the Centre National d'Études Spatiales (CNES), the National Aeronautics and Space Administration (NASA), and the European Space Agency's (ESA) PRODEX programme (Swiss contribution). The work presented here was supported by AFOSR Grant No. USAF F49620-00-1-0012 and NASA Grant No. NAG5-9600 and carried out at the NASA Goddard Space Flight Center. The radio images were provided by the Nobeyama Radio Observatory, and the H$\alpha$ images were provided by the Big Bear Solar Observatory, operated by the New Jersey Institute of Technology. The EIT and MDI data are by courtesy of the EIT and MDI consortia. We would like to thank Philippe Lemaire, who assisted in planning the SUMER observations, and Stuart Jordan of NASA/GSFC who provided useful comments on the manuscript. SOHO is an ESA/NASA collaborative mission.