| dc.description.abstracteng | This work presents the analysis of eleven CME events which were observed by instruments of the Sun-Earth Connection Coronal and Heliospheric Investigation (SECCHI) imaging suite on board the NASA Solar TErrestrial RElations Observatory (STEREO) mission, launched in October 2006. The selected CMEs were chosen such that their projected speeds exceed 1000 km/s. Five out of the eleven events drove interplanetary shocks with distinct signatures in white-light observations from both the coronagraph COR2 and the heliospheric imager HI1 on board STEREO. For each event geometrical modelling via the Graduated Cylindrical Shell (GCS) model was performed in order to test the assumption of self-similar evolution for the CME. The performance of the GCS model at heliocentric distances higher than 15 R☉ , corresponding to the outer field of view of COR2, was tested. The results show that the GCS is appropriate in reproducing the observed CME geometry at higher heliocentric distances mainly for events with projected angular half-width ω such that ω < 45° - PA_APEX, with PA_APEX being the position angle of the CME apex. The GCS results, moreover, showed that all the modelled CMEs decelerated during their propagation. The average acceleration was found to lie in the range -15.21 m/s^2 ≤ a ≤ -1.13 m/s^2. The observed CME morphology was preserved during the CME propagation.
The Stereoscopic Self-Similar Expansion Model (SSSEM) was used to infer the kinematics of the analysed CMEs, and, in the cases in which a shock was detected in time-elongation plots, of the shock itself. Arrival times and speeds at the Lagrangian point L1 were inferred by assuming propagation at constant speed and at constant deceleration and compared to in-situ Advanced Composition Explorer (ACE) measurements. The precision of the arrival time estimates is compatible with the accuracy of the models employed, with the smallest errors being as low as 1.3 hours. In six out of the nine events detected by ACE the inclusion of deceleration produced the most accurate results.
Models for the shock location were employed in order to determine the temporal evolution of the shock standoff distance Δ, the density compression ratio ρu/ρd and the upstream Mach number M for the five events in which both the CME and shock kinematics could be inferred.
The derived shock parameters were extrapolated to the location of ACE and compared to their values computed from in-situ plasma and magnetic field measurements. ACE data was available for four of the five shock events. The observed values lie in the ranges 15 R☉ ≤ Δ ≤ 38 R☉ , 1.89 ≤ ρd/ρu
≤ 2.87, and 1.8 ≤ M ≤ 5.8. The standoff distance estimates yield the most precise results when constant deceleration is included in the CME and shock propagation. In this case the errors are below 40% of the observed values, with the better predictions yielding discrepancies as low as 10%. For three out of the four shock events detected by ACE the precision in the compression ratio extrapolations for the linear and quadratic fits is comparable, of the order of 10%. For the other event the inclusion of deceleration reduces the error by 40%. The accuracy of the Mach number extrapolations is lower, ranging from 20% to ≈ 70% in the worst-case, and with the linear fit yielding more precise predictions in three out of four events.
The analysis presented in this work confirms that CME-driven shocks can be detected in remote-sensing observations by coronagraphs and heliospheric imagers, and that their kinematics can be derived independently from the CME kinematics via the application of inverse modelling techniques. The results also show that multipoint space observations are necessary in order to achieve a sufficient precision for the independent analysis of CMEs and CME-driven shocks when using the methods employed in this work. The analysis of their interplanetary evolution, as well as of the shock standoff distance, suggests that deceleration plays an important role in their propagation. More accurate results for the arrival time predictions and the standoff distance extrapolation are obtained when the effects of deceleration are accounted for. To the best of the author's knowledge this is the first time that the separate CME and shock kinematics, as well as the temporal evolution of the shock parameters in the inner heliosphere are derived from the application of inverse modelling techniques. Finally, the application of the methods presented here to data gathered from upcoming space-based mission such as the ESA Solar Orbiter and the NASA Solar Probe Plus, to be launched in 2018, will provide an unprecedented tool to further verify the models presented here and to determine the CME-driven shock parameters at locations unexplored so far. | de |