Multithermal Analysis of a CDS Coronal Loop

Abstract

The observations from 1998 April 20 taken with the Coronal Diagnostics Spectrometer CDS on SOHO of a coronal loop on the limb have shown that the plasma was multi-thermal along each line of sight investigated, both before and after background subtraction. The latter result relied on Emission Measure Loci plots, but in this Letter, we used a forward folding technique to produce Differential Emission Measure curves. We also calculate DEM-weighted temperatures for the chosen pixels and find a gradient in temperature along the loop as a function of height that is not compatible with the flat profiles reported by numerous authors for loops observed with EIT on SOHO and TRACE. We also find discrepancies in excess of the mathematical expectation between some of the observed and predicted CDS line intensities. We demonstrate that these differences result from well-known limitations in our knowledge of the atomic data and are to be expected. We further show that the precision of the DEM is limited by the intrinsic width of the ion emissivity functions that are used to calculate the DEM. Hence we conclude that peaks and valleys in the DEM, while in principle not impossible, cannot be confirmed from the data.