SLIDE 1
MHD simulations of Coronal Mass Ejections in the global corona
- P. Pagano1
D.H. Mackay1, A. Yeates2
1University of St Andrews, 2Durham University
MHD simulations of Coronal Mass Ejections in the global corona P. - - PowerPoint PPT Presentation
MHD simulations of Coronal Mass Ejections in the global corona P. - - PowerPoint PPT Presentation
MHD simulations of Coronal Mass Ejections in the global corona P. Pagano 1 D.H. Mackay 1 , A. Yeates 2 1 University of St Andrews, 2 Durham University 2nd SolarNet Meeting, Palermo February 2nd 2015
SLIDE 2
SLIDE 3
Ejection of Flux Rope
The ejection of a flux rope is believed to be the progenitor of CMEs. It is also a component of the flare standard model. Cheng et al., 2011
SLIDE 4
Boundary conditions of Space Weather
At ∼ 4 R⊙ CME are blown in the solar wind Magnetized plasmoid the Solar Wind can deflect the ICME The CME plasmoid can rotate The ”Bz” component of the magnetic field (perpendicular to ecliptic) is crucial for the impact of the ICME on the Earth-magneosphere
SLIDE 5
Boundary conditions of Space Weather
Space Weather forecast: arrival time and properties of CMEs For Space Weather forecast, we need: efficiency on computation accuracy on the injection of the CME in the solar wind
SLIDE 6
Life of flux Rope: formation
Patsourakos et al., 2013 Formation of flux rope: accumulation of free magnetic energy Flux rope formation Slow formation: days or weeks Quasi-static evolution. (t >> τAlf ) Magnetic evolution: β << 1 everywhere
SLIDE 7
Life of flux Rope: ejection
Flux rope ejection: release of energy Flux rope ejection Fast ejection: flux rope travels out of the corona in ∼ 2 hours Highly dynamic evolution. (t ∼ τAlf) Full MHD: plasma is locally compressed. (β ≥ 1)
SLIDE 8
Strategy: Pagano et al.,2013. We couple two models.
Model the life span of Flux Rope
Global Non-Linear Force Free Field (GNLFFF) evolution model Flux rope formation Decribes a magnetically dominated evolution Models the evolution of corona for weeks Computationally efficient: magnetofrictional technique MHD Simulation with the MPI-AMRVAC code Flux rope ejection Accounts for plasma and magnetic field Models multi-β domain
SLIDE 9
Ejection of the flux rope: 3D MHD Simulation
MPI-AMRVAC: KU Leuven MHD
∂ρ ∂t + ∇ · (ρ v) = 0, (1) ∂ρ v ∂t + ∇ · (ρ v v) + ∇p − ( ∇ × B) × B) 4π = +ρ g, (2) ∂ B ∂t − ∇ × ( v × B) = 0, (3) ∂e ∂t + ∇ · [(e + p) v] = ρ g · v − n2χ(T) − ∇ · Fc, (4) ∇ · B = 0 (5) p γ − 1 = e − 1 2 ρ v2 −
- B2
8π , (6)
- g = − GM⊙
r 2 ˆ r, (7)
SLIDE 10
What we achieved so far... Pagano et al.,2013, A&A, 554, A77
it is possible to couple the GNLFFF model with the MHD AMRVAC code we follow the life span of a flux rope from formation to ejection the stress accumulated during the formation justifies a flux rope ejection
SLIDE 11
What we achieved so far... Pagano et al.,2013, A&A, 560, A38
we study the role of the gravitational stratification on the early progation of a CME we identify the parameter space where ejections are more likely
SLIDE 12
What we achieved so far... Pagano et al.,2014, A&A, 568, A120
Non ideal MHD simulation of flux rope ejection Study of the role of thermal conduction and radiative cooling Synthesis of SDO/AIA
- bservations
SLIDE 13
CME MHD simulation of the global corona
Coupling MHD simulation with Global code The Global code uses a series of Magnetograms as boundary conditions Accounts for flux emergence and flux cancellation at the solar surface Predicts the formation of most flux ropes
SLIDE 14
A vec implementation
MHD ( A)
∂ρ ∂t + ∇ · (ρ v) = 0, (8) ∂ρ v ∂t + ∇ · (ρ v v) + ∇p − ( ∇ × ∇ × A) × (∇ × A) 4π = +ρ g, (9) ∂ A ∂t = v × ( ∇ × A), (10) ∂e ∂t + ∇ · [(e + p) v] = ρ g · v − n2χ(T) − ∇ · Fc, (11) p γ − 1 = e − 1 2ρ v2 − ( ∇ × A)2 8π , (12)
- g = − GM⊙
r 2 ˆ r, (13)
SLIDE 15
Advantages of the A formulation
∇ · B = 0 Direct coupling with GNLFFF and back Mind that:
Stencil of the numerical model Boundary conditions in terms of A
SLIDE 16
⊲⊲ ⊲ ⊲⊲ Computational efficiency of the GNLFFF model Accuracy and generality of MPI-AMRVAC MHD simulations
SLIDE 17
—-
t tCPU ∼ 15 t tCPU ∼ 0.0005
From magnetograms to GNLFFF From GNLFFF to MHD Feasible approach for Space Weather Forecast.
GNLFFF ∼ 30K more efficient than MHD
SLIDE 18
PRELIMINARY SIMULATION Magnetic configuration from the evolution of actual magnetograms We can study the reaction of the global magnetic field to the ejection The post ejection magnetic field can become initial conditions for the GNLFFF model
SLIDE 19
- A-MHD 2D Test
- A MHD
- B MHD
SLIDE 20
- A-MHD 3D Spherical Test
Initial Condition
SLIDE 21
- A-MHD 3D Spherical Test
Final evolution
SLIDE 22
- A-MHD 3D Spherical Test
Difference A-MHD/MHD
SLIDE 23