Consensus:

• During growth phase there is a stress balance between the total pressure gradient and the magnetic tension in the x direction.
• Onset: initial localized brightening of arc, followed later by expansion.
• Every substorm expansion phase includes dipolarization of B.
• Any explanation for substorm expansion must account for plasma pressure reduction in near-Earth (~ 6-10 Re) plasma sheet.
• Many (at least >20%) substorms associated with an external IMF change that should lead to reduction of convection.
  

Open Questions:

[Vasyliunas]

• What is the mechanism for pre-dipolarization:
  Does it start with X-line?
  Does it start close in with "instability"?
  Does external triggering play a significant frole in substorm process?
  Do instabilities at the inner edge (ballooning, CCI, etc.) lead to changes of the large-scale magnetic field, and if so, how?
  Do the BBF's play any special role in substorms?

[Shiokawa]

• How is growth-phase stress balance broken - by X-line formation, or by near-Earth (<10 Re) process?
• Is magnetic flux to the near-Earth region carried mainly by BBF's or by other processes?
  

[Otto]

• What is the source of E_y to enable dipolarization?
• What makes aurora?
• What is the coupling between magbetosphere and ionosphere?

[Henderson]

• Is there more than one "type" of substorm? (i.e. are there different causes?)
• Must lobe reconnection occur in every substorm?
• Are all substorms externally "triggered"?

[Lyons]

• Is the fraction of substorm onsets that are "caused" by IMF changes that lead to reduction of convection closer to 50% or to 100%?

[Ridley]

• Does the ionospheric electric field increase or decrease?
• What is the role of ionospheric conductance?
• What is the role of pseudo-breakups in magnetotail "stability" or substorm development?

[Smith]

• What is the effect of "unequal" ionospheres (north/south, winter/summer, B_y modulation) on the substorm processes in the magnetotail?

[Lui]

• Does magnetic reconnection always produce significant energization of particles in the magnetotail?
  

Personal statements (points about which there is no consensus but, in the opinion of the individual in question, they should be):

[Cheng]

• Complex structures of auroral breakup correspond to plasma turbulence in the near-Earth plasma sheet during expansion phase.
• Substorm onset is initiated by KBI which leads to other instabilities, to form turbulence and eventually lead to transport of plasma pressure.
• Plasma pressure transport (reduction) process in near-Earth plasma sheet leads to dipolarization of B and change of E fields.

[Henderson]

• Magnetic storms are not equal to summation of substorms.
• Magnetic storms are not equal to SMC's (steady magnetospheric convection, "convection bays").
• Auroral substorm morphology is more complex than the "classical" picture of Akasofu.

[Lui]

• The substorm expansion onset process is intimately associated with a strong current sheet in the near-Earth region.
• Magnetic field description is equivalent to current description, by Maxwell's equations.
• There are multiple activity sites with multiple scales in the magnetotail during substorm expansion.
• There is a wide range in the spatial scale and magnitude (energy release) of magnetotail disturbances, ranging from pseudobreakups to large substorms covering many local time zones.

[Otto]

• Magnetotail lobe reconnection occurs in every substorm.
• Current sheet thinning requires divergent flux transport.

[Shiokawa]

• substorm is the release of energy stored in the magnetotail as tail-like magnetic flux (during the growth phase) to the dipolar field region (during the expansion phase).
• During expansion phase, significant magnetic flux is carried from NENL and piled-up in the near-Earth tail.

[Vasyliunas]

• Growth phase produced by flow that is enhanced in the dayside and reduced in the magnetotail.
• A pre-dipolarization phase is unavoidable.
• New X-line formed during every substorm, at what stage (before or after onset) not yet clear.

[Kan]

• Brightening of an auroral arc can result from enhanced global convection driven by enhanced dayside reconnection during the growth phase.
• Dipolarization in the near-Earth plasma sheet (~6-10 Re) is the cause of the substorm expansion onset.
• Dipolarization can result from impeding the enhanced sunward convection on the earthward side of the dipolarization region in the near-Earth plasma sheet.
• The cause of impeding the enhanced sunward convection is not yet understood.
  

[Swift]

• The major need are coordinated and adequately financed efforts to develop global-scale kinetic models of the magnetosheath, magnetosphere, ionosphere system. Computer technology and algorithms have advanced to the point where such simulations are feasible. Only few percent of the support for observations directed into kinetic modeling efforts will do more resolving the outstanding issues than continued debate based on more observations.

[Bristow]

• Reconnection in the tail is necessary to allow tail magnetic flux to replenish dayside flux removed by subsolar reconnection. This is because steady convection cannot occur fast enough when dayside reconnection is rapid and the nightside x-line is in the distant tail. Hence, while tail reconnection may or may not cause expansion phase onset, it is a necessary part of a substorm.
• Steady convection may be possible if the distant tail x-line is near enough to the earth and the dayside driving rate is low enough so that tail particle energy can be dissipated in the ionosphere. Under such conditions, substorms are not necessary for replenishment of dayside flux.
• The development of the Harang discontinuity during growth phase and its subsequent disappearance at expansion onset appear to be a nearly universal feature of substorms.