W. D. Cramer1, J. Raeder1, F. R. Toffoletto2, M. Gilson1,3, B. Hu2,4 Plasma Sheet Injections into the Inner Magnetosphere: Two-Way Coupled OpenGGCM-RCM Model Results* W. D. Cramer1, J. Raeder1, F. R. Toffoletto2, M. Gilson1,3, B. Hu2,4 1Space Science Center, University of New Hampshire, 2Department of Physics and Astronomy, Rice University, 3Pattern Technologies Inc., 4CG Services September 27, 2017 *also see Cramer et al., 2017

Magnetospheric Convection Dungey cycle [Dungey, 1961] Reconnection at dayside magnetopause connects IMF to Earth’s magnetic field, which gets dragged into tail Return flow sunward from tail to inner magnetosphere back to dayside Main source of inner magnetosphere plasma, some becomes trapped Steady convection when dayside, tail reconnection rates match Steady adiabatic convection not possible, however [Erickson and Wolf, 1980] Adiabatic condition: PV5/3 (“flux tube entropy”) is conserved along drift paths (V=flux tube volume) Pressure increase of inward drift too large for observed magnetic fields (“pressure balance inconsistency”), would stop convection

Plasma Sheet “Bubbles” Possible resolutions of pressure balance inconsistency: plasma sheet bubbles, gradient/curvature drift “Bubbles”: depleted flux tubes (low relative pressure, density, and flux tube entropy) Interchange unstable with neighboring flux tubes Move inward until flux tube entropy matches background Brake and oscillate around equilibrium Likely cause of depletion: transient/local reconnection Reduces flux tube volume Plasma sheet entropy profile Xing and Wolf, 2007

Plasma Sheet Injections Narrow (1-3 RE) channels of earthward-moving plasma Flow Bursts Short-duration (1-2 min) high-speed flows (generally V>400 km/s) Bursty Bulk Flows (BBFs) Segments of moderate-speed flows (generally V>100 km/s) with flow burst(s) Usually accompanied by dipolarization front Primary means of plasma transport into inner magnetosphere during active times Yang et al., 2011

Model Components OpenGGCM CTIM (Coupled Thermosphere-Ionosphere Model) 3-D stretched cartesian grid Solves semi-conservative MHD equations for single fluid Plasma energy, not total energy, conserved Solves ionosphere potential on 2-D spherical grid using conductances, field- aligned currents Main inputs: solar wind plasma parameters, interplanetary magnetic field CTIM (Coupled Thermosphere-Ionosphere Model) Models chemical and photo-chemical reactions Determines conductances Main inputs: FACs, auroral precipitation, solar EUV RCM (Rice Convection Model) 2-D ionosphere grid representing footpoints of flux tubes Solves motion of flux tubes due to potential, magnetic induction and drift Main inputs: outer boundary conditions, ionosphere potential

Model Coupling Methodology OpenGGCM -> Ionosphere Sends MHD density, pressure, auroral precipitation, FAC Receives Ionosphere Potential RCM -> Ionosphere Sends RCM FACs, auroral precipitation Blends with MHD values RCM -> OpenGGCM Convert flux tube content to pressure and density (and vice-versa) Receives MHD pressure, density at boundary Flux tube volume-weighted averages along field line Sends RCM pressure, density Nudge MHD P, n values (RCM influence greater in inner region) RCM feedback slowly ramped up after MHD initialization period

Storm, Quiet Simulations ID Solar wind Driver Date Minimum Dst* Q1 --- 1/19/2014 I1 CIR 8/5/2013 -41 I2 6/6/1998 -46 I3 5/2/2010 -71 I4 2/27/1997 -86 M1 CME 5/1/2013 -54 M2 5/29/2010 -60 M3 7/15/2012 -105 M4 10/28/2001 -137

Injections in Simulation Data Calculation of radial velocity on current sheet at boundary Field lines traced from near-Earth points Current sheet marked as farthest field line distance Calculate flux tube volume MHD quantities (v, n, P, B) interpolated to current sheet crossings Values interpolated to spherical boundaries LLBL filtered out using T/n (as in Tsyganenko and Mukai [2003]), v(azimuthal), and proximity to boundary

Snapshot of Injections at R=10 Inward injections (velocity maxima) frequently associated with entropy minima (bubbles)

Bubbles and Injections (Storm M.P.)

Bubbles and Injections (Detail)

Bubbles and Injections (Quiet Period)

Injections collocated with bubbles Storm-times: ~70-80% of injections at R>8 Quiet-time: significantly lower % between R=6-10

Plasma Transport by Bubbles Storm-times: ~80% of transport at R>8 Quiet-time: significantly lower % below R=8

Discrete Injections Identify individual injection time series Only BBFs shown (w/ time of peak velocity) R-dependent flow burst cutoff speed

Injection Peak Velocity Average peak injection velocity vs. R (all injections)

BBF profile vs. Theory/Data Superposed epoch of BBFs relative to peak velocity BZ: similar dipolarization profile Flux tube entropy: observation of bubble Peak velocity occurs at dipolarization front Yang et al., 2011

Discussion/Conclusions Storm-time injections (fast or otherwise) are usually associated with bubbles beyond 8 RE Quiet-time: less frequent association below 10 RE Bubbles responsible for ~80% of storm-time plasma transport across 8 RE Quiet-time: % transport much lower below 8 RE Peak inward velocity of discrete injections decreases approx. linearly from 12 to 6 RE Model reproduces BZ, PV5/3 profile for BBFs Peak velocity coincides with dipolarization front