Event study on pre-substorm phases and their relation to the energy coupling between solar wind and magnetosphere

Abstract

On 11 November 1976, after a magnetically quiet period with the interplanetary magnetic field (IMF) directed northward, a sudden southward turning of the IMF immediately led to a world-wide intensification of convection which was observed to start almost simultaneously at stations within the auroral zone and polar cap. The two-dimensional equivalent current system over the northern hemisphere had a typical two-cell convection pattern with a maximum disturbance of ΔH = −300 nT observed on the morningside in the westward electrojet region. This enhancement of activity ended after 35 min in a localized substorm onset in the midnight sector over Scandinavia. The recordings made in this area indicate large fluctuations of various ionospheric parameters starting several minutes before the substorm onset. Two subsequent stages can be resolved: (1) high-energy particle precipitation recorded by balloon X-ray detectors and maximum ionospheric current density increase, while the electrojet halfwidth shrinks and the total electrojet current becomes weaker; (2) the maximum ionospheric current density stays constant and the high-energy particle precipitation decreases, while the auroral brightness increases and the total electrojet current and its half-width show a growing trend prior to the final breakup. A suggestion is made that the time interval of these two stages should be called “trigger phase”. A short discussion explains the trigger phase observations in a magnetospheric scale. The energy coupling between solar wind and magnetosphere during the pre-substorm phases is discussed by utilizing the energy coupling function ϵ defined by Perreault and Akasofu ( Geophys. J. R. Astr. Soc. 54, 547, 1978). The ϵ values appear to be on substorm level during the period of enhanced convection. A good correlation between ϵ and the growth of the Joule heating rate (estimated from the AE data) is found in the beginning, but during the last 20 min before substorm triggering ϵ is high while the Joule heating rate decreases. The behaviour of ϵ during the two stages of the trigger phase suggests that the start of the trigger phase is purely internally controlled while the length of the trigger phase and the final substorm onset may be influenced by the variation in ϵ.