Beyond the solar surface: an observational study of solar coronal loops

Abstract

This thesis presents an observational analysis of structures in the solar corona. The plasmafl in the corona is low (<I), therefore, the emitting material is intimately linked to the magnetic fleldlines. The initial aim of this work, outlined in Chapter 3, is to take advantage of this relationship to calculate realistic magnetic reconnection rates from EUV observations. Then, Chapters 4-6 concentrate upon examining a colour-colour method to provide temperature profiles of coronal structures.
Firstly, TRACE 171 A and 195 A data are used to indirectly determine a magnetic reconnection rate for a flare loop system on the limb of the Sun. The magnetic reconnection rate is calculated by measuring the footpoint velocity of the loop system and following the method adopted by Isobe et al. (2002). The footpoint velocity is thought to be equal to the rate at which successive loops brighten, in the emission lines, during the postflare phase. A typical footpoint velocity of 1.5 kms'±0.7 kmr' is obtained and a reconnection rate of -.0.00 1-0.03 is determined.
Secondly, the double filter ratio temperature analysis method employed upon TRACE 171, 195 and 284 A coronal images is re-addressed. Two features from TRACE datasets published initially in Isobe et al. (2002) are reanalysed and a new TRACE
example is also included for comparison. Only one of the original examples along with the new dataset are easily shown to be present in all three EUV passbands. This work investigates two important changes to the original analysis. A new temperature colour-colour curve is employed which utilises revised TRACE response functions; these take into account the emission from Fe VIII which significantly affects the 195 A passband. Also three different background subtraction techniques are used. Subsequently, the results show that when using this method for these examples, it is very difficult to assign specific temperature values to individual data points.
Thirdly, an extensive investigation into the colour-colour curve is performed. Monte Carlo statistics based on a Poisson distribution are used to get an error estimate for the curve. Based on these error statistics, the probability of analysed data points corresponding to differing temperatures can be found. The flare loop system studied in Chapter 3 is investigated, using the SOHO/EJT data, applied to the colour-colour curve. Also TRACE 171 and 195 A images are analysed with the original single filter ratio technique, for comparison to the EIT data. Both methods appear ambiguous indicating
temperatures of either —0.3 M K or - 1.1 MK. Two possible reasons which could push the analysed points off the c-c curve are investigated: (I) the poor resolution of the EIT data could lead to viewing a multi-thermal structure; and (2) if the structure was cooling over the time period of the observations the temperature would change from one passband image to the next.
Fourthly, TRACE and CDS data of the same active region are analysed to try and determine the usefulness of the colour-colour method for temperature determination.
The TRACE data under investigation is bright in all three wavelengths, and two loop structures are chosen from within the active region for analysis. One structure appeared to be static over the half hour period of the observations, whereas the second structure chosen showed a significant decrease in brightness over the same time period. The results show that the CDS data depicts the coronal structures to be initially at around --1.45 MK, with an increase to 1.58 MK over the time of the observations.
The TRACE data seems to show that these structures are slightly cooler at around —1.3 MK, but there is an increase in filter ratio values over the same time period. The biggest discrepancy seen in these two observational datasets is that TRACE shows the temperature of structure 1 to be decreasing towards its apex, whereas CDS shows the temperature increasing. These results reflect the need for better instrumentation. Finally a summary of the conclusions and future work are presented.