On February 7, 1986 during a major geomagnetic storm the Bz component of the interplanetary magnetic field (IMF) turned strongly northward for several hours. Data from the Defense Meteorological Satellite Program F6 and F7 satellites and the HILAT satellite were used to study the evolution of the pattern of high-latitude precipitation and field-aligned currents in response to this change. Prior to the northward IMF period, the auroral zone was observed down to mid-latitudes and was very wide in latitude, and strong, large-scale ''region 1 / region 2'' currents were clearly present (ΔB~1700 nanoteslas). Following the northward turning, the equatorward boundary of the auroral zone on the nightside contracted sharply poleward and polar cap arcs were observed. The strength of the region 1 / region 2 currents decreased markedly and became immeasurably small at the time of the maximum contraction of the auroral oval. An NBZ current system was observed to grow and expand in the southern (summer) high latitude region over a period of more than 2 hours. Simultaneously, an irregular pattern of field-aligned currents was observed in the northern (winter) hemisphere. During the contraction, the latitudinal width of the auroral region mapping to the central plasma sheet (CPS) decreased dramatically while the width of the area mapping to the boundary plasma regions (BPR) in the magnetosphere increased greatly.

At the time of the maximum contraction the BPR extended up to a latitude of at least 87.1¿. The NBZ currents expanded with and were entirely located within the BPR precipitation. Polar cap arcs were observed in both regions of BPR precipitation and polar cap precipitation and were not correlated with the location of the large-scale field-aligned currents. There was no indication of the CPS intruding to high latitudes, and thus no evidence for bifurcation of the magnetotail. The boundary between the CPS and the BPR showed little change. If this implies that the boundary between open and closed field lines contracted slowly or not at all, then a significant portion of the observed BPR precipitation was observed along open field lines. We interpret the pattern of particle precipitation and field-aligned current and the implied plasma convection in terms of a distorted, two-cell convection pattern driven by merging of closed field lines with the IMF as proposed by Crooker [1988>. When the IMF turned southward again, the pattern quickly reversed. The BPR contracted; the CPS precipitation regions expanded; the equatorward boundary of the auroral oval moved to lower latitudes; the NBZ currents disappeared in less than 30 min; and the region 1 / region 2 currents reappeared. Again, the BPR/CPS boundary did not move as rapidly and thus may indicate that the changes are due more to a reconfiguration within the magnetosphere than a change in the portion of the magnetosphere that is open or closed.