Combined geometrical modelling and white-light mass determination of coronal mass ejections

Context. We use forward modelling on multi-viewpoint coronagraph observations to estimate the 3-dimensional morphology, initial speed and deprojected masses of Coronal Mass Ejections (CMEs). The CME structure is described via the Graduated Cylindrical Shell (GCS) model, which enables the measurement of CME parameters in a consistent and comparable manner. Aims. This is the first large-scale use of the GCS model to estimate CME masses, so we discuss inherent peculiarities and implications for the mass determination with a special focus on CME events emerging from close to the observer's central meridian. Further, we analyse the CME characteristics best suited to estimate the CME mass in a timely manner to make it available to CME arrival predictions. Methods. We apply the method to a set of 122 bright events observed simultaneously from two vantage points with the COR2 coronagraphs onboard of the twin NASA STEREO spacecraft. The events occurred between January 2007 and December 2013 and are compiled in an online catalogue within the EU FP7 project HELCATS. We statistically analyse the derived CME parameters, their mutual connection and their relation to the solar cycle. Results. We show that the derived morphology of intense disk events is still systematically overestimated by up to a factor of 2 with stereoscopic modelling, which is the same order of magnitude as for observations from only one vantage point. The overestimation is very likely a combination of projection effects as well as the increased complexity of separating CME shocks and streamers from CME fronts for such events. We further show that CME mass determination of disk events can lead to overestimation of the mass by about a factor of 10 or more, in case of overlapping bright structures. Conclusions. We conclude that for stereoscopic measurements of disk events, the measurement of the initial CME speed is the most reliable one. We further suggest that our presented CME speed-mass correlation is most suited to estimate the CME mass early from coronagraph observations.