THE MARSEILLE-ARTEMIS CATALOG OF LASCO CMES

摘要

We have developed a new automated method of detection of Coronal Mass Ejections (CMEs) on LASCO-C2 synoptic maps based on their morphological properties. The approach is based on an adaptive filtering and segmentation, followed by a merging with high-level knowledge step. The program builds a catalog which lists the CMEs detected given a Carrington Rotation together with their main estimated parameters: time of appearance, position angle, angular extent and average velocity. Our fi- nal catalog LASCO-ARTEMIS (Automatic Recognition of Transient Events and Marseille Inventory from Synoptic maps) is compared with existing catalogs (LASCO CME catalog, CACTUS catalog). We find that I) we detect many more events that the visual detection method but in good agreement with the automated CACTUS detection ii) our rate of events follows very well the pattern of solar activity like the LASCO CME catalog, which has been highlighted by a correlation study with the sunspot number. The total number of detected CMEs is heavily controlled by the sensitivity to small, faint and numerous events. Adapting the thresholding step, we show that a continuous distribution of CMEs must exist. Two classes of CMEs can be distinguished according to their velocity and their morphology : I) fast and huge CMEs disturbing the whole solar corona ii) slow plasmo?ds, or blobs, forming the slow solar wind.