Space, time and velocity association of successive coronal mass ejections

¹ Instituto de Geofísica, Universidad Nacional Autónoma de México, Ciudad de México, Mexico
e-mail: alara@igeofisica.unam.mx
² The Catholic University of America, Washington DC, USA
³ GSFC/NASA, Greenbelt, MD, USA
⁴ Smithsonian Astrophysical Observatory, Cambridge, MA, USA
⁵ Centro de Geociencias, Universidad Nacional Autónoma de México, Querétaro, Mexico

Received: 4 June 2019
Accepted: 25 January 2020

Abstract

Aims. Our aim is to investigate the possible physical association between consecutive coronal mass ejections (CMEs).

Methods. Through a statistical study of the main characteristics of 27 761 CMEs observed by SOHO/LASCO during the past 20 years.

Results. We found the waiting time (WT) or time elapsed between two consecutive CMEs is < 5 h for 59% and < 25 h for 97% of the events, and the CME WTs follow a Pareto Type IV statistical distribution. The difference of the position-angle of a considerable population of consecutive CME pairs is less than 30°, indicating the possibility that their source locations are in the same region. The difference between the speed of trailing and leading consecutive CMEs follows a generalized Student t-distribution. The fact that the WT and the speed difference have heavy-tailed distributions along with a detrended fluctuation analysis shows that the CME process has a long-range dependence. As a consequence of the long-range dependence, we found a small but significative difference between the speed of consecutive CMEs, with the speed of the trailing CME being higher than the speed of the leading CME. The difference is largest for WTs < 2 h and tends to be zero for WTs > 10 h, and it is more evident during the ascending and descending phases of the solar cycle. We suggest that this difference may be caused by a drag force acting over CMEs closely related in space and time.

Conclusions. Our results show that the initiation and early propagation of a significant population of CMEs cannot be considered as a “pure” stochastic process; instead they have temporal, spatial, and velocity relationship.