A statistical study of traveling convection vortices using the Magnetometer Array for Cusp and Cleft Studies

Abstract

The Magnetometer Array for Cusp and Cleft Studies (MACCS) is primarily a longitudinal magnetometer array located in the eastern Canadian arctic at cusp latitudes. We present a statistical study of traveling convection vortices (TCVs) as observed by MACCS from the first year of operations, August 1992 to July 1993. For this period, data are routinely available from three stations, extending more than 2 hours in magnetic local time (MLT) and located roughly on a line of constant geomagnetic latitude at 75°. With the selection criteria used, we find an average of 38 TCV events per month. We characterize the events as “isolated” and “nonisolated”. The isolated events have the typical, accepted signature of a TCV while the nonisolated ones are events that satisfy the selection criteria but occur among relatively strong magnetic activity. The diurnal distribution of these events shows a prenoon‐afternoon asymmetry, but the distribution has a single peak at 10 MLT and then falls off smoothly in the afternoon hours. It does not show the previously reported minimum of event occurrence around magnetic local noon except for isolated events occurring during higher Kp values. The most likely explanation for this discrepancy is the wide longitudinal extent of MACCS. The majority of the events have propagation velocities between 3 and 11 km/s, which can be a factor of 2 larger than the events reported in previous studies. The majority of the events move away from noon, as is expected, with a small fraction moving toward noon. Most (80%) of the studied events exhibit all the characteristics of a typical TCV, while the remaining 20% show some form of irregularity. We find four different types of irregularities. The isolated and nonisolated events have many similar morphological and propagation characteristics. However, quantitative comparison of the statistical distributions of the two types of TCVs indicates that they belong to different populations. We also find that nonisolated TCVs may occur for somewhat higher Kp values. We therefore suggest that the two classes of TCVs have different source mechanisms.