[GEM] THE GEM MESSENGER, Volume 28, Number 46

[Newsletter Editor]

***************************
     THE GEM MESSENGER
***************************

Volume 28, Number 46
Oct.08,2018

Announcement submission website: http://aten.igpp.ucla.edu/gem/messenger_form/

============================================================
Table of Contents

1. GEM Magnetotail Dipolarization and Its Effects on the Inner Magnetosphere Focus Group: 2018 Workshop Report

============================================================

------------------------------------------------------------
1. GEM Magnetotail Dipolarization and Its Effects on the Inner Magnetosphere Focus Group: 2018 Workshop Report
------------------------------------------------------------
From: Christine Gabrielse, Matina Gkioulidou, David Malaspina, Slava Merkin, Drew Turner (cgabrielse at ucla.edu)

Panel Session on the topic of Dipolarization and Global Modeling

The Dipolarization FG held a panel discussion session to focus on how magnetotail dipolarization is currently captured in global models and how those models need to be developed to better simulate dipolarization and its effects (in the inner magnetosphere and ionosphere) based on what observations are telling us about the nature of the system.  Approximately 70-80 members of the greater GEM community were in attendance.  The panelists included: Katie Garcia-Sage, Colby Lemon, San Lu, Yann Pfau-Kempf, Jimmy Raeder, and Misha Sitnov. Christine Gabrielse chaired the panel and guided the conversation with comments and questions.  

Prior to the panel session, panelists were sent the following three questions to consider and respond to as guidance for the topics that were to be discussed during the session:
1) Given our current modeling capabilities, discuss which kinds of models are best at capturing which aspects of dipolarization events and their effects in the magnetosphere.
2) What determines the dipolarization scale size in different models? (e.g., physical description, boundary conditions, model input parameters, ionosphere conditions, etc.?)
3) The transition region is where both inertia and energy dependent drifts are important. No existing models treat that region correctly.
(a) How do we move forward?
(b) Or, more specifically, address the question of dipolarization front deceleration: (i) How do various models treat dipolarization deceleration as they approach the inner magnetosphere? (ii) What processes are decelerating the fronts in the models? (iii) What inner magnetosphere processes are missing (e.g. plasmasphere, complex ionospheric conductivity models) and does excluding these processes lead to different deceleration predictions?
(c) And/or address: (i) What are the relative roles of ExB, energy-dependent drifts and particle trapping in transport and energization in the transition region? (ii) To what extent are these processes adiabatic for particles of different energies? (iii) What is their overall contribution to the ring current build up?

Types of Modeling: Which are best for what questions?
Christine started the panel discussion by reviewing some of the responses she had received from the panelists concerning question 1.  The panel then moved into open discussion on that topic.  There was general consensus that the relevant physics are global in nature, and in particular that the role of the ionosphere and small-scale physics are both relevant and not properly being captured by any of the models. Models must capture both the plasma sheet and dipolar inner magnetosphere correctly plus the feedback loop provided by the non-idealized ionosphere. 

San stressed that a combination of models, such as global MHD with embedded PIC and global hybrid models is our best current approach for capturing both global and critical small-scale processes. Concerning small-scale physics, Misha raised the point that we still don’t have a good sense of where in the tail the reconnection X-line typically forms and whether the models are capturing even that correctly.  He also stressed that with empirical models, such as TS07, we can much more accurately capture individual events.  Yann introduced the Vlasiator model, and stressed that the location of inner boundary conditions and 2D limitations in the global hybrid model are still a major limitation for accurately capturing magnetotail reconnection, dipolarization, and substorm activity.   Jimmy focused on the differences between global MHD and other models, stressing that global MHD has a “lack of knobs” that is both limiting in one sense but more trustworthy in another sense.  Jimmy also stressed the importance of the ionosphere and also the transition region in and around GEO, where reconnection fronts (dipolarization fronts and the associated BBFs) start to decelerate and deflect in the inner magnetosphere; he stressed that once these plasma “bubbles” start to slow down and disperse, the fluid picture no longer applies, so it is difficult to say how well MHD model results showing that represent reality.  There was also general consensus that data-model comparisons are very important and we need to continue developing those capabilities and approaches. 

>From the audience, Andrei Runov asked about the nature of the X-line in the magnetotail: did the panelists think it was a global scale feature?  The panelists consider X-lines in the tail to be fragmented and spread out throughout the tail between around X_GSE of -15 and -20 RE.  Misha Sitnov thinks the typical X-line lies further downtail, more like -30 RE or beyond, and that models that include too much resistivity will get this closer to Earth.  Andrei also stressed the tailward side of the picture, that is, those reconnection jets that are ejected tailward from an active X-line.  From recent ARTEMIS results, the reconnection jets observed at lunar orbit (-60 RE) are still localized in nature, which is further evidence that the X-lines in the magnetotail are also localized.  

Dipolarization Scale Sizes:
Christine next steered the panel to question 2.  Katie stressed that the resolution in global MHD tends to break down in the ionosphere, which might fundamentally limit the scale sizes of features in the magnetosphere.  She also mentioned that ion composition and ionospheric outflow are not well captured in global models currently but might play a key role in scale sizes of magnetotail dipolarization via instability leading to reconnection, reconnection scale sizes, and the global magnetotail properties.  Colby also agreed that the grid resolution in the ionosphere in the RCME model was also a major limiting factor.  Two grid points in the ionosphere in the model map to a very large region of the magnetosphere, meaning that the model might not be able to capture localized features in a stretched magnetotail.  Colby stressed that RCME seems to be doing a good job capturing the Y (i.e. cross-tail, azimuthal) scales of flow channels (BBFs) but is concerned about how well they are capturing the X (downtail) scales.  

>From the audience, Shin Ohtani asked about time scales: at 500 km/s velocity, it takes only a few minutes to go from 20 RE to GEO, which is similar to the Alfvén speed travel time from the reconnection site to the ionosphere, so does the ionospheric feedback really matter?  Colby responded that was a good but unresolved question.  Jimmy disagreed, saying the speed is much faster down to the ionosphere.  Bob Lysak mentioned that models often don’t capture the density along field lines correctly, but that with the current best estimates, the travel time for information down to the ionosphere was a few minutes.

That discussion transitioned into the importance of Pi2 waves.  Joachim Birn mentioned that oscillation in the transition/stopping region is on the scales of the ionospheric travel time (Pi2 period timescales).  He again stressed the importance of the transition region and how many of our challenges currently fall back into that region around GEO.  Yann showed a movie from Vlasiator, and stressed that with a perfectly conducting “ionosphere” at 5 RE, the speeds were too fast in their simulations.  They were seeing peak flows around 2000 km/s.  He also stressed that with 2D simulations, all of the reconnection in the system was forced into the XZ plane.  The Vlasiator simulations take some time to get reconnection after initialization, and they are actively investigating how the addition of oxygen ions to the plasma sheet will affect that delay time.

Larry Lyons introduced another question of the audience. He asked, “Do dipolarizations only occur in thin current sheets?”  He stressed that with streamers being observed under a variety of different conditions, is a thin current sheet a necessary condition to get dipolarization fronts and BBFs in the plasma sheet?  Misha stressed that the problem is multiscale and that no, a thin current sheet is not a necessary condition.  Tail reconnection, dipolarization fronts, and BBFs may develop in a thick current sheet.  San agreed with Misha’s point and stressed that the thickness of a dipolarization front is determined by ion kinetic physics and that from observations, the width of a front is complicated and might have to do with the scale of the responsible X-line or with the conductance in the ionosphere or both.  

San then showed results from the ANGIE3D global hybrid model.  Slava Merkin asked: what determines the scale size of the X-line?  San didn’t know but stressed that it was not resistivity but likely an inherent property of the X-line itself, perhaps due to a non-uniform magnetotail.  Shin asked why dipolarization fronts moved dawnward, and San replied that it was just a result of ExB drift.  Mostafa asked if there is a correspondence between sizes of X-lines and dipolarization fronts/flows?  Can larger X-lines produce smaller flows or vice versa?  There was some disagreement and discussion between whether or not dipolarizing flux bundles should get smaller as they move inwards.  Joachim brought up that when reconnection starts, an X-line might be extended in the tail due to solar wind driving conditions and the distribution of resistivity in the model, but over time the active X-line narrows to a few RE due to entropy reduction and the system becoming unstable to ballooning.  He discussed how the tearing mode and ballooning mode can either compete or act in concert, and ultimately, that the cross tale scale depends on the region of the outflow where the BBFs go to. 

Jimmy brought up an analogy to seismology and terrestrial earthquakes.  He stressed that an earthquake in one place on the planet can trigger another earthquake 1000s of miles away.  He thinks that one active X-line can similarly trigger reconnection elsewhere in the plasma sheet.  The formation of the active X-line changes the entire environment in the tail; it is a disruptive event.  This is of course all driven by changes in the solar wind too, which further complicates the picture.  He pointed to auroral arcs as evidence that there is likely no preferred scale size for X-lines and the dipolarization fronts they spawn.

The Transition Region: How do we Move Forward?
Christine next turned the discussion to question 3.  Misha kicked off the discussion on that and stressed that we do have a comprehensive picture of the transition region from a collection of many, many years of observations throughout it.  He argues that with data mining, relying on observations from many, many similar cases, we have full coverage of the region.  From his empirical model, which employs data mining, he finds that the transition region expands downtail from ~-8 to -18 RE during substorm dipolarization.  From here, Andrei asked how Misha defines a substorm, to which Misha replied with the AL index.  This sparked a debate on how to define substorms.  

Katie changed the subject to stress that plasma pressure in the inner magnetosphere has to be captured correctly to properly model the transition region.  This requires that plasma sheet models be coupled to accurate inner magnetosphere models.  She again stressed the important role of the ionosphere, and how that can help dictate how far into the inner magnetosphere a dipolarizing flux bundle can travel and the properties of its rebound and oscillations as it comes to rest there.  

Larry Lyons brought up that we had not discussed the ground based observations point of view.  He asked how we can connect where reconnection is occurring in the models to what we are seeing in the aurora with streamers.  He stressed that in the aurora, much of it is east/west aligned, which corresponds to azimuthal drifts in the inner magnetosphere, and streamers are the only features that can correspond to dipolarizing flux bundles and BBFs.  San agreed and mentioned that localized reconnection and dipolarization fronts may be the consequences of dayside streamers loading small, localized portions of the tail.  Jimmy agreed and stressed that models might be capturing the east/west features but that we just haven’t focused on analyzing them.  Jimmy stressed too that we had to be careful, because there is a filter effect with the ionosphere too.  Not everything seen in the aurora/ionosphere is reflecting what is happening in the magnetosphere.

***From this panel discussion, we established a GEM challenge: modelers are challenged to simulate three different cases: storm-time substorm, isolated substorm, and magnetotail reconnection during steady magnetospheric convection.  From the simulation results, how well can a given model capture the observed similarities and differences between these different cases?  How will models be constrained so that they do not start reconnection prematurely?  This challenge will be further developed and fully defined at the mini-GEM meeting at AGU 2018 and will be conducted in partnership with the focus group on mesoscale aurora, polar cap dynamics, and substorms.***

Contributed Talks Session:

The Dipolarization FG held a second session immediately following the panel, chaired by Drew Turner, to allow for contributed talks. Also attended by about 70-80 GEM members, the session had ten contributed talks and excellent discussion:

1. Chih-Ping Wang presented on “RCM simulations of entropy reduction caused by plasma bubbles from different MLT locations”. He showed that the earthward transport of the simulated plasma bubble qualitatively explains the two-point THEMIS observation of a BBF event. He showed that the simulated entropy reduction caused by a plasma bubble varies significantly with the bubble’s initial MLT and background convection. A plasma bubble starting at 23 MLT results in an entropy reduction that extends closer to the Earth and azimuthally wider than does a bubble starting at 1 MLT.  

2. Ryan Dewey presented on "Dipolarization effects at Mercury and comparisons to Earth". He used MESSENGER observations at Mercury to identify dipolarizations in Mercury's near magnetotail, and discussed the statistical characteristics of these events. He showed that dipolarization fronts are short-lived (~2 s) enhancements of the northward component of the magnetotail field (~30 nT) and are associated with fast sunward flows, energetic particle acceleration, and thermal plasma heating/depletion. He discussed that these signatures are analogous to those at Earth, however, he showed that dipolarizations are most frequently observed in the post-midnight plasma sheet at Mercury, opposite to that at Earth.

3. Joachim Birn presented an MHD simulation of dipolarization braking. He showed that dipolarization penetration is deeper if it starts from farther out in the tail. He showed that a decrease in density was related to a decrease in energy flux at lower energies. He discussed the dipolarization stopping by an excess of pressure gradient force and pileup of the pre-existing medium.

4. Brian Swiger presented a talk entitled, “Do different substorm strengths accelerate keV electrons the same?” He showed that from X=-6 to -25 RE, for all electron energies between ~5-52 keV, the average flux increase was greater for larger AE events.

5. Andrei Runov presented THEMIS and LANL observations in the near-Earth plasma sheet and at GEO, respectively, during events of prolonged, extreme solar wind/IMF driving. Events with IMF Bz <-10 nT during longer than 5 hours were selected. THEMIS measurements indicate that the magnetotail responded by a set of thinning-dipolarization events with a duration of 1 hour, which resemble the sawtooth events. The dipolarizations were accompanied by ion and electron injections in energy ranges ~50 to 500 and ~20 to 200 keV, respectively. Dispersionless and dispersed injections in these energy ranges were also detected by LANL spacecraft at GEO.

6. Sasha Ukhorskiy presented on ion acceleration and transport from the tail to the inner magnetosphere, the effects of trapping, adiabaticity, and the role of charge. (See Ukhorskiy et al., 2017.) Recent analysis showed that the buildup of hot ion population in the inner magnetosphere largely occurs in the form of localized discrete injections associated with sharp dipolarizations of magnetic field, similar to dipolarization fronts in the magnetotail. Because of significant differences between the ambient magnetic field and the dipolarization front properties in the magnetotail and the inner magnetosphere, the physical mechanisms of ion acceleration at dipolarization fronts in these two regions may also be different. He discussed an acceleration mechanism enabled by stable trapping of ions at the azimuthally localized dipolarization fronts, and showed that trapping can provide a robust mechanism of ion energization in the inner magnetosphere even in the absence of large electric fields.

7. Anton Artemyev discussed regimes of ion energization during injections: adiabatic vs. nonadiabatic acceleration. The canonical approach for the guiding center theory was proposed, and using this approach the particle equations of motion were rewritten in the coordinate frame with vanishing inductive electric field (a non-inertial coordinate system).  Using these equations of motion, Anton discussed three regimes of plasma acceleration: the hot plasma in a large background Bz field, the cold plasma in a small background Bz field, and the intermediary plasma/background Bz field. He referenced Zhou et al. [2018] to discuss mass dependence on energization, with more massive particles (e.g., O+) able to gain the most energy. He showed that ions of different charges at ~5-6 keV will gain a similar amount of energy, but that ions with greater positive charge (e.g., O+6 vs. O+) at ~20 keV can gain more energy.

8. Xiangning Chu discussed broadband waves on plasmapause induced by deep penetration of dipolarization front. He showed that most plasmapause observations with broadband waves are centered around pre-midnight, similar to the distribution of flows/dipolarization fronts. He also found parallel electron fluxes around the same time. He found that AE was larger when the waves were observed at the plasmapause than when no waves were observed at the plasmapause.

9. Shin Ohtani presented on “Spatial structure and development of dipolarization in the near-Earth region”.  By statistically comparing the relative timing of dipolarizations at two satellites, he found that the dipolarization region expands earthward as well as away from midnight at r <= 6.6 Re.  The expansion velocity was estimated at several tens of km/s, noticeably slower than outside geosynchronous orbit.  He suggested that this earthward expansion of the dipolarization region can be attributed to a two-wedge current system with a R2-sense wedge moving earthward and a R1-sense wedge staying outside of geosynchronous orbit.

10. Tetsuo Motoba reported on "A near-Earth dipolarization event observed by MMS (r ~13 Re)". In the course of the dipolarization, MMS observed multiple dipolarization fronts (DFs, < 1min), energetic particle injections (> 70 keV), and oscillating flows. The injected energetic ions were field-aligned accelerated with pitch angle asymmetry, while no apparent pitch angle asymmetry was found for the energetic electrons. The MMS-GOES and MMS-ground comparisons revealed good correlation between the dipolarizations at MMS and GOES and between the oscillating flows and low-latitude Pi2 pulsations, respectively.


========================================
The Geospace Environment Modeling (GEM) program is sponsored by the Division of Atmospheric and Geospace Sciences (AGS) of the National Science Foundation (NSF).

To broadcast announcements to the GEM community, please fill out the online request form at:

http://aten.igpp.ucla.edu/gem/messenger_form/

To subscribe to the newsletter, please go to the web page at:
http://lists.igpp.ucla.edu/mailman/listinfo/gem
(IMPORTANT: Do not use this web page to post announcements.)

For any other questions, please contact Peter Chi, GEM Communications Coordinator, at <pchi at igpp.ucla.edu>

URL of GEM Home Page:  http://aten.igpp.ucla.edu/gemwiki
Workshop Information:  http://www.cpe.vt.edu/gem/index.html
========================================