Data-Driven MHD Modeling of
CME Events
Session Summary, day 2
Session Organizers: Yuhong Fan (HAO), George Fisher (SSL/UCB),
Mark Linton (NRL-DC), Brian Welsch (SSL/UCB)
Slava Titov: "Structural Analysis of the Coronal Magnetic Field: How Can It Be Used in Models of CMEs?"
Start: review of separatrix surfaces, and quasi-separatrix layers
(QSLs) as sites of reconnection;
Q := "squashing factor" quantifies rate of change of field-line
connectivity near a QSL
Q enables:
(1) localising preferred sites of reconnection, i.e., quasi-separators
(2) identifying building blocks, e.g., erupting and non- strands of flux
(3) determine evolving fluxes for each blocks
(4) relate observable structures to building blocks:
– H-alpha flare ribbons,
– EUV dimmings,
– X-ray sigmoids
Relation to observational features
t=32 (≈ 38 min after the CME onset)
t = 77 (~ 1 hr 32 min)
Roussev: Use Dynamic Flux Emergence Simulations
(w/Galsgaard & Archontis) to Drive Coronal Model
• This enables driving with more self-consistent boundary conditions.
- required rescaling β from photospheric to coronal values, and
decreasing peak field strength.
• Flux rope forms, and erupts; post-eruption field consists of two flux
ropes, which form a “double J” structure
– synthetic X-ray emission resembles observed sigmoids
• Topology matters: nulls & QSLs play important roles (cf., Titov)
• Footpoints of erupting rope do not remain stationary, but move
across surface as eruption proceeds.
• Flux rope does not remain intact: after eruption, two flux ropes are
formed, linking “core" of emerged field to external field
Magnetic Field Geometry at Later
t = 68 min
SHINE Workshop
Institute for Astronomy
University of Hawaii
June 25-29, 2012, Maui
Several “data inspired” simulations of actual
eruptions were presented: Zuccarello; Fan; Jin.
Models had varying degrees of thermodynamic realism.
Fan’s CME was slower than the observed CME:
– Fan: Rescaling must preserve height profile of B.
– Fisher: If you get a 1000 km/sec eruption, how do you change
parameters to achieve a 2000 km/sec eruption?
– Mikic: To increase eruption energy, confine the pre-eruption field
more strongly.
– Fisher: So, for instance, if you want to build a pipe bomb, you should
not use a cardboard tube, you should instead use a metal tube.
– Mikic: Yes, you want to prevent expansion for a while…
Meng Jin studied a CME shock with one-T and two-T models:
• 1T: 9 MK precursor far ahead of shock
• 2T: 3 MK peak e- T, 100 MK peak proton T (Mach #’s near 4)
Heat flux saturation (not in model) was discussed
Reinard: Charge-state predictions from MHD CME simulations can
be used to interpret in situ observations.
QFe derived from Breakout model
QFe derived from Flux cancellation
QFe observations for the two May 19-21, 2007 MCs

similar documents