## 4. Temperature dependence of the distribution functionTo study the dependence of the parameters of the distributions we included more data in the chromospheric and transition region temperature ranges. The additional data sets in the range between K and K were SUMER raster scans covering 60"120" made in Jun. 1996 in the lines CI 1311 Å and CII 1334 Å, and a time series in the OI line at 1152 Å made in Mar. 1996. In the range between K and K SUMER raster scans covering 90"100" were available, made in Sept. 1996 in the lines SiIV 1402 Å, OIII 703 Å, OIV 1401 Å, and OV 629 Å. A lognormal fit was made to all distributions. We normalized the shape of the fitted distributions with respect to the number of pixels and additionally rescaled the radiance to a common range by dividing by the mean value. Fig. 6a shows the fitted lognormal distribution functions of all available data in the temperature range between K and K. The transition region lines show significantly different distribution shapes. These lines have their most common value in a lower radiance regime but their tails extend further out than for the chromospheric and coronal lines indicating that there is a greater probability for relatively high radiance values. This could be due to the large number of short-time scale enhancements seen at these temperatures. The chromospheric lines have a sharper peak at higher radiances, and their tails drop off more steeply. Finally, the coronal lines exhibit an intermediate behaviour. Fig. 6b depicts one of the corresponding fit parameters, namely the strength ( from Eq. (4)), which measures the height and position of the most probable radiance value. The larger this parameter, the higher the peak and the more "outward" shifted and symmetric is the shape of the lognormal function. This figure qualifies the results already visible in Fig. 6a.
Figs. 7a and b show the corresponding position
(
© European Southern Observatory (ESO) 2000 Online publication: October 24, 2000 |