2018 THEMIS SCIENCE NUGGETS

Global-scale Dawn-Dusk Millihertz Auroral Oval Oscillations

by Kan Liou¹ and David G. Sibeck²
¹The Johns Hopkins University Applied Physics Laboratory, Laurel, Maryland, USA
²GSFC, NASA, Greenbelt, Maryland, USA

Introduction

The auroral oval, a region where the aurora occurs frequently, maps magnetically to the entire equatorial magnetosphere and observations of the aurora are often used to infer magnetospheric processes. We use global auroral images acquired with the Ultraviolet Imager (UVI) onboard NASA's Polar satellite to study the motion of the magnetosphere and show, for the first time, that the entire magnetospheric tail can move east-west in harmony like a windsock flapping in wind. The characteristic period of the flapping motion may be a major source of global long-period ULF waves, adding yet another possible driver for global mode ULF waves.

Figure 1. (A) From top to bottom, the panels show: (a) solar wind density (N_p), (b) the x- and (c) y-component of the solar wind velocity (V_x and V_y), (d) dynamic pressure (P_d), (e) three components of magnetic field (B_x, B_y, B_z), (f) IMF latitude and longitude, (g) IMF clock angle (θ_cl = sin(B_y) cos^-1(B_z/[B_y²+B_z²]^1/2) and defined for -180° < (θ_cl ≤ 180°) and cone angle ((θ_cn = tan^-1[(B_y²+B_z²)^1/2/|B_x|], blue dots), (h) and the total magnetic field (|B|). The data is based on OMNI HRO. (B) Auroral keograms derived with the Polar UVI image data for a number of hourly sectors from 22 (top panel) to 01 MLT (bottom panel). The y-axis is the magnetic latitudes.

Observations

Figure 1A shows solar wind plasma and magnetic field observations convected to the subsolar bow shock from NASA GSFC's OMNIWeb for the interval from 15:00 to 21:00 UT interval 6 January 2000. Both the solar wind density and dynamic pressure were small and relatively stable. The x component of the solar wind velocity was slightly faster than usual (~550 km/s) and increasing. The y component of the flow speed was near zero from 17:07 to 17:30 UT. It then increased (up to ~80 km/s) and became more variable. The z component (red) of the interplanetary magnetic field (IMF) turned northward from ~ −7 nT to 7 nT at 1600 UT and then fluctuated between negative and positive thereafter, with a period of ~ 1 hr. We are interested in the northward turning at ~17:08 UT. This northward turning was associated with a field rotation, possibly a rotational discontinuity, because the total magnetic field strength did not change significantly (see Figure 1Ah). The overall geomagnetic activity associated with this event was moderate (Kp ~ 3). A small geomagnetic storm (minimal Dst ~ − 40 nT) commenced at ~12:00 UT, accompanied by a substorm (AE ~1500 nT). During this event, both the storm and substorm subsided.

Figure 1B shows auroral keograms at the dawn (01–07 MLT) and dusk (18–22 MLT) sectors from 16:30 to 2000 UT (the auroral region from ~08 to 16 MLT was outside of the field of view of the UVI). Each of the auroral keograms, which show time varying latitudinal profiles of aurorae at fixed local times, is derived from a sequence of auroral images acquired by Polar/UVI. The response of the aurora to the northward IMF turning at 17:08 UT is a poleward motion of the auroral oval near dusk (Figures 1Bh and 1Bi) starting around 17:15 UT. In the postmidnight sector the auroral intensity increases and this bright auroral region begins the north-south latitudinal oscillation. Between 17:00 and 20:00 UT, the primary frequency of the oscillation in the peak auroral intensity latitude was ~3 mHz.

An interesting feature in the auroral pulsations is that all the auroral pulsations in the dawn sector are in phase; they do not propagate. This is also true for the auroral pulsations in the dusk sector. However, the dawn and dusk pulsations are 180° out of phase, as indicated by the four vertical dashed lines in Figure 1B. The nodal point is around midnight (not shown). This dawn-dusk antiphase exists throughout the event period.

Figure 2. A schematic drawing (view from the north) of the global-scale motion of the separating magnetopause shear layers and vortex shedding in the distant magnetotail (adopted from Hones Jr. et al. [1981]).

Conclusion

Figure 2 illustrates our interpretation of the observations. The interaction of the solar wind flow with the Earth's magnetic field results in the formation of the magnetospheric cavity. As an analogy, we consider a uniform flow around a sphere or a cylinder. For small Reynolds numbers, symmetrical laminar flows form downstream from obstacles. When Reynolds numbers increase, vortices appear in the wake behind the obstacles. If the Reynolds number exceeds a critical value, shear instabilities will arise and the vortices will break away periodically from either side of the object and create a so-called von Karman vortex street further downstream. In a uniform flow, the vortex streets can start on one side or the other of the obstacle due to small fluctuations in the symmetric upstream flow. In the solar wind-magnetosphere system, discontinuities in the solar wind can serve as the small fluctuations needed to cause alternating east-west vortices. Perhaps such an effect occurs on the dawn-dusk flanks of the magnetosphere. This is reasonable because open lobe magnetic field lines, which couple to the solar wind plasma flow, are confined in the north-south direction. If the sphere/cylinder is not mounted rigidly, it can begin to resonate with the frequency of the vortex shedding when this matches the resonant frequency of the object. This is because the vortices are low-pressure regions and the object tends to move toward the vortices. In the magnetosphere, the resonance frequency can be the eigenfrequency of magnetic field lines as flapping requires the entire magnetotail to sway back and forth. If this global-scale low-frequency oscillation exists, it is expected to cause the motion of the plasma sheet as seen in the auroral data. Although the K-H instability can also produce flow vortices at large Reynolds numbers, Karman vortices have wavelengths and scales much larger than K-H vortices. A large-scale (the entire magnetosphere) coherent and periodic motion of the magnetotail is required to account for the periodic latitudinal motion of the entire auroral oval. Indeed, such a large-scale (20–40 R_E) vortex has been reported by Hones Jr. et al. (1981).

Reference

Liou, K., & Sibeck, D. G. (2018). Dawn-dusk auroral oval oscillations associated with high-speed solar wind. Journal of Geophysical Research: Space Physics, 123, 600– 610. https://doi.org/10.1002/2017JA024527

Hones Jr., E. W., J. Birn, S. J. Bame, J. R. Asbridge, G. Paschmann, N. Sckopke, and G. Haerendel (1981), Further determination of the characteristics of magnetospheric plasma vortices with Isee 1 and 2, J. Geophys. Res., 86(A2), 814–820, doi:10.1029/JA086iA02p00814.

Biographical Note

Kan Liou is a member of the principal professional staff at the Johns Hopkins University Applied Physics Laboratory. His research interests focus on auroral activities and their implications to solar wind-magnetosphere coupling.

David G. Sibeck is NASA GSFC's Mission Scientist for the THEMIS and ARTEMIS missions. He studies the solar wind-magnetosphere interaction and in particular the Earth's magnetopause and magnetosheath.

Please send comments/suggestions to

Emmanuel Masongsong / emasongsong @ igpp.ucla.edu