Reversed two-cell convection in the Northern and Southern hemispheres during northward interplanetary magnetic field

Abstract

[1] This article presents a case study of large-scale ionospheric convection in the Northern and Southern Hemispheres under strongly northward interplanetary magnetic field (IMF) conditions on 9 November 2004. Using a comprehensive data set from both ground-and space-based instruments, the study shows the formation of reversed two-cell convection in both the Northern and Southern Hemispheres that lasted for nearly 2 hours. Examination of the concurrent satellite energy-time spectrograms of precipitating particles reveals that reverse convection occurs in the region filled mostly with the boundary plasma sheet (BPS) type precipitating electrons except that the electron number flux is much smaller than that in the normal BPS. We have named this region the northward B z boundary layer (NBZBL), which we interpret as a consequence of double-lobe reconnection. This interpretation is corroborated by the global MHD simulations, which show that the NBZBL consists of mostly closed field lines, resulting from double-lobe reconnection in both the hemispheres, together with intermittent presence of overdraped open field lines, resulting from single-lobe reconnection in one of the hemispheres. In addition to reversed two-cell convection, the distribution of field-aligned currents (FACs) shows clearly the presence of a pair of the northward B z (NBZ) currents near the central polar region in both the hemispheres. Intense downward Poynting flux with a peak value around 100 mW/m 2 is also seen in the high-latitude polar region, which tends to surround the upward leg of the NBZ currents. Finally, the potential drop between the two reverse-convection cells exceeds 100 kV, which is far larger than the values reported in any previous studies of reverse convection under northward IMF conditions. The unusually large reverse potential drop in this case is attributed in part to the strong NBZ component of 35-40 nT and in part to the unusually large solar wind dynamic pressure that is about five times its nominal value.

Magnetosphere‐Ionosphere Convection Under the Due Northward IMF

“…If there is a viscous cell in the ionosphere, it would be a candidate. A slightly similar structure can be seen in Lu et al (2011), but it is not so clear. Especially in the case of Bhattarai et al (2012; in the case of Bz = 5 nT), potential of the viscous cell is larger than that of the reverse cell.…”

“…In the previously reported simulation, there is no case reproducing these wavy normal circulations (Bhattarai et al, 2012;Lu et al, 2011;Song et al, 1999). However, this part of potential includes normal circulation pattern (clockwise in the evening).…”

“…In such circulation, open field line sections may appear in the polar cap convection. This process is explained in detail by Lu et al (2011). This process is schematically shown in a rectangular frame in Figure 1.…”

Undulations in MeV solar energetic particle fluxes in Earth's magnetosphere associated with substorm magnetic field reconfigurations

“…P dyn , Dst and Dst ∗ are also superposed for reference. Since the signature of the SSC in the 1‐min Dst index indicates the arrival of the pressure increase at the magnetopause, the time of the OMNI P dyn is corrected before using it to determine Dst ∗ [ Lu et al , 2011]. These corrections ranged from −10 to +10 min with a median of 0 and quartiles of −2 and +6 min (negative values indicate that OMNI P dyn is early relative to the signature in the 1‐min Dst ).…”

“…The magnetospheric particle response to the stretching and dipolarization is represented by the >0.8 MeV electron fluxes from the GOES satellite that observed the earliest dipolarization. For a more accurate measure of the disturbance ring current, the contribution of magnetopause currents is removed from the 1‐min Dst index as follows [ Burton et al , 1975; Lu et al , 2011]: where P dyn is in nPa and the third term is the contribution of the quiet day magnetopause and ring currents in nT. In the original paper [ Burton et al , 1975], this correction was applied on a 2.5‐min cadence using interpolated data, so it justifiably can be applied to the USGS 1‐min Dst index.…”