Astron. Astrophys. 336, 1039-1055 (1998)

Appendix A: calculation of radiative losses

In order to obtain an estimate of the energy involved in the observed EUV brightenings, we have calculated the radiative energy losses for each of the brightenings. The calculation of the radiative energy losses from narrow-band (but not single line) EUV intensities requires a set of assumptions about densities, abundances, and temperatures in the observed plasma. Interpretation of the radiative energy loss as a measure of the total energy released in a heating event involves a second set of assumptions. Conduction is potentially a substantial energy loss mechanism in the corona. Although a consistent treatment of conduction losses in the corona has not been completed, it is common to assume that the conductive losses are approximately equal to the radiative losses in the corona. The large temperature gradients observed in the transition region provide evidence that conduction is not a significant process in the transition region. Even if the assumptions break down for some particular case, the following analysis is useful simply as a study of the intensity variations observed in the respective EUV bandpasses.

The radiative energy losses are obtained from the product of the radiative loss function of Cook et al. (1989) and an emission measure derived from the EUV intensity. To obtain an emission measure, a synthetic spectrum is calculated by assuming a density and temperature using the ISOTHERMAL routine (Newmark 1996) of the CHIANTI spectroscopy data base (Dere et al. 1997). The synthetic spectrum is folded into the instrument parameters of the relevant EIT sector to obtain the count rate observed by EIT for a unit emission measure. The instrument calibration used for this calculation is a modification of the instrument parameters presented in Delaboudinière et al (1995) with the filter throughputs of Song (1995). Further work on the SOHO EIT calibration is in progress with the data obtained from the EIT Calibration Sounding Rocket.

The densities assumed in the synthetic spectrum calculation are 1.5 cm^-3 for the corona (Fe xii) and 1.0 cm^-3 for the transition region (He ii). Since the EIT bandpasses are dominated by resonance lines, the spectra are relatively insensitive to the input densities. The temperatures assumed are the temperatures of maximum ionization fraction for the dominant ion in each bandpass: 1.5 K for Fe xii and 5.0 K for He ii. The contribution of lines formed at different temperatures within a given bandpass will be underestimated in this approach if the true differential emission measure is not strongly peaked at the assumed temperature. Such a contribution is minimal in the He ii and Fe xii bandpasses for quiet Sun plasma. With the isothermal assumption, one makes an overall underestimate of the energy of an event.

For some of the transition region lines, including He ii 304 Å, the differential emission measure, falls outside the trends set by the majority of transition region lines. The variation from this trend in differential emission measure can approach an order of magnitude. Thus, although the uncertainty of the emission measure calculation is on the order of a factor of two, there remains a caveat that should be kept in mind when interpreting our results. This caveat cannot be resolved with EIT data alone. A SOHO "Joint Observation Program" with CDS and SUMER may provide a more complete picture of the energy release of the EUV brightenings in the transition region.

© European Southern Observatory (ESO) 1998

Online publication: July 27, 1998
helpdesk@link.springer.de