Content uploaded by Igor V. Sokolov
Author content
All content in this area was uploaded by Igor V. Sokolov on Apr 13, 2016
Content may be subject to copyright.
TRANSITIONING EEGGL TO THE CCMC.
Igor V. Sokolov1, Richard E. Mullinix2, Aleksandre
Taktakishvili2, Anna Chulaki2, Meng Jin3, Ward B.
Manchester1, Bart van der Holst1 and Tamas Gombosi1
1. CLaSP, University of Michigan, Ann Arbor MI
2. CCMC, Goddard Space Flight Center, Greenbelt MD
3. Lockheed Martin Solar and Astrophysics Lab,Palo Alto CA
+Thanks to Maria M. Kuznetsova and Spiro Antiochos
April, 11, 2016.
CCMC Workshop, April,11-16, 2016. Annapolis, MD, USA.
Transients in the Solar Wind and Their
Simulation in Real Time
2
We present and demonstrate a new tool, EEGGL (Eruptive Event
Generator using Gibson-Low configuration) for simulating CMEs
Coronal Mass Ejections).
CMEs are among the most significant space weather events.
Some of these effects may be efficiently simulated using the “cone
model” (as we heard) The cone model provides a capability to
predict the location, time, width and shape of the hydrodynamic
perturbation in the upper solar corona (at ~0.1 AU), which can be
used to drive the heliospheric simulation (with the ENLIL code, for
example).
At the same time the magnetic field orientation in this perturbation
is uncertain within the cone model, which limits the capability of
the geomagnetic activity forecast.
3
The new EEGGL tool recently developed at the CCMC in
collaboration with the University of Michigan, provides a capability
to simulate the magnetic field evolution at 1 AU too
Based on the magnetogram and evaluation of the CME initial
location and speed, the user may choose the active region from
which the CME originates and then the EEGGL tools provides the
parameters of the Gibson-Low magnetic configuration to
parameterize the CME.
The recommended parameters may be used then to drive the CME
propagation from the low solar corona to 1 AU using the global code
for simulating the solar corona and inner heliosphere.
At the CCMC The Community Coordinated Modeling Center (CCMC)
provides the capability for CME runs-on-request, to the heliophysics
community.
EEGGL is a new animal in the CCMC Zoo, which is well integrated
with the other animals (Donki, STEREOCat).
Eruptive Event Generator (Gibson-Low):
EEGGL
Demo for CME 2012-07-12
4
We demonstrate how the new tools are used to simulate a halo CME
2012-07-12 (https://kauai.ccmc.gsfc.nasa.gov/DONKI/view/CME/14/1)
StereoCAT is used to find CME Speed
5
StereoCAT (http://ccmc.gsfc.nasa.gov/analysis/stereo/) is developed at
the CCMC. By tracing the CME front, we find CME Speed=1300km/s.
StereoCAT finds CME Start time
6
CME start time = Snapshot Time – D/(CME speed) = 13:51
D
Snapshot time
StereoCAT guesses CME place of birth
7
Latitude is -20o. Estimates for (HEEQ) longitude are +/- 6o
8
The new tool, EEGGL (Eruptive Event Generator using Gibson-Low
configuration – see Splash page
http://ccmc.gsfc.nasa.gov/analysis/EEGGLInfo/EEGGL.html
and the tool itself: http://ccmc.gsfc.nasa.gov/analysis/EEGGL/) has
been recently developed at the CCMC (Goddard Space Flight
Center) in collaboration with the University of Michigan.
Based on the magnetogram and evaluation of the CME initial
location, speed, and start time the user may
choose the active region from which the CME originates;
then the EEGGL tools provides the parameters of the Gibson-Low
magnetic configuration to parameterize the CME;
the recommended parameters may be used then to drive the CME
propagation from the low solar corona to 1 AU using the global code
for simulating the solar corona and inner heliosphere. To achieve this,
the EEGGL has a link to the run submission web page, which helps
the user to fill in the request form for a simulation run.
Newly Developed EEGGL tool
EEGGL tool (historic events): chose AR
9
For a start time, 2012-07-12.13:51 calculate CR number 2125 and
Carrington longitude 83. Find AR in the synoptic magnetic map for
CR2125 near the point with longitude 83+/-6o and latitude -200
CME origin point as found from StereoCAT
Find Parameters for GL configuration
10
Choose and mark bipolar configuration of solar spots in this AR
1.Mark positive and negative spots
2.Click “Recommended parameters”
Fill in Form to Request Simulation Run
11
With the found parameters for GL configuration request a run.
1. Parameters are found
2. Request SWMF run
Submit Your Run and Wait
12
SWMF Run-On-Request with EEGGL
13
Simulates Solar Corona (SC) in spherical coordinates (about 3
million cells) and Inner Heliosphere (IH) in Cartesian
coordinates, on AMR grid (about 35 million cells) with an
improved resolution within the cone in which the CME
propagates.
Superimpose the Gibson-Low configuration with the
observationally constrained parameters, to simulate CME
Steady-state simulation of the state prior to eruption takes
approximately 17 hours at 120 CPUs with the CCMC cluster hilo.
Simulation of 4-10 hours of the CME evolution in the SC and
then about 3 days of its evolution in the IH takes approximately
16 hours at 120 CPUs.
The only relevant slide.
14
Transitioning the EEGGL to the CCMC. Problems:
EEGGL in the CCMC is the web application and the UofM team
cannot handle the web application at the CCMC by multiple
reasons (classification, network security)
EEGGL in the CCMC is the web application while the code
development in the UofM is done in languages (f95, C++, IDL)
not suited for web application at all.
This results in an excessive labor needs when the CCMC has to
reshape new versions of the code to get them work at the CCMC
Solution of the problems:
Algorithms are implemented in Python. At the CCMC this part of
the code one-to-one converts to the server side (“back-end”) of
the EEGGL
The emulator is implemented in Python/IDL/Fortran-executable.
This emulator both allows us (UofM) to keep using/developing the
EEGGL as standing-alone code and describes, how the browser
side (“front-end”) of the EEGGL at the CCMC may look like.
Future Work
15
Acknowledgement
The collaboration between the CCMC and University of Michigan is
supported by the NSF SHINE grant 1257519 (PI Aleksandre
Taktakishvili). "The work performed at the University of Michigan
was partially supported by National Science Foundation grants
AGS-1322543 and PHY-1513379, NASA grant NNX13AG25G, the
European Union’s Horizon 2020 research and innovation program
under grant agreement No 637302 PROGRESS. We would also like
to acknowledge high-performance computing support from: (1)
Yellowstone (ark:/85065/d7wd3xhc) provided by NCAR's
Computational and Information Systems Laboratory, sponsored by
the National Science Foundation, and (2) Pleiades operated by
NASA's Advanced Supercomputing Division
We will add a capability to simulate real-time CMEs based on
the existing automated real-time simulation system .
We will add a capability to superimpose the Titov-Demouline
flux rope.
Today’s Experience. 1. Start the EEGGL
script with the today’s noon magnetogram
16
Today’s Experience. #
2. Choose am active region
17
Today’s Experience. 3. Enjoy result (zoom)
18
Today’s Experience.#
4. CME parameters.
19
#CME
T UseCme
T DoAddFluxRope
342.50 LongitudeCme
10.50 LatitudeCme
0.87 OrientationCme
GL TypeCme
0.60 Stretch
1.80 Distance
0.49 Radius
-10.15 BStrength
0.0 Density
1.0 ModulationRho
1.0 ModulationP