Some progress on

1
Some progress on linking together models

• Nick Achilleos
• Lecturer, Department of Physics
• University College London
• With thanks to Patrick Guio and Dugan Witherick

2

• Report from UCL
• We have been pursuing activities which may
  produce some useful demonstrators and
  deliverables for the JRA3 Distributed Modelling
  Laboratory.
• Modelling auroral dynamics at Jupiter (PhD
  project by J. N. Yates)
• Coupling UCL Jovian magnetodisc model with a
  model of plasma flow (Achilleos, Guio)
• Other possibilities ?

3
Auroral Dynamics at Jupiter

• Model 1
• - The UCL global, axisymmetric model of Jupiters
  thermosphere and ionosphere (e.g. Smith and
  Aylward, Ann. Geo., 2009).
• Solves equations of fluid flow for the neutral
  thermosphere.
• Needs an algorithm for computing ionospheric
  current, esp. in the auroral region.
• Uses Model 2 the theoretical profile by Grodent
  and Gerard (JGR 2001) for auroral ionosphere
  (ion density versus altitude)

4
Model Linkage

• Architecture Auroral profile is built into a
  subroutine of UCL code

UCL model reads auroral profile as a template
for auroral ionosphere
The ion density profile is scaled at all
altitudes by the same factor, according to the
conductance properties calculated by the disc
dynamics routine - gives plasma rotation rate and
corresponding M-I currents.
5

• For more information, see paper submitted to PSS
  by Yates, Achilleos and Guio (2010, arxiv)
• Grodent / Gerard profile (ascii files?) may be
  added to JRA3 Catalogue
• We need more 1D theoretical profiles of auroral
  structure so that we may use the UCL Jovian and
  Kronian models as a testbed - this should be
  emphasised as a useful goal of the DNML.
• Note also the work of Nicholson et al (MNRAS
  2009) on precipitation at Mars, UCL are working
  on taking this further to a coupled model of
  Martian thermosphere and aurora.

6
Giant Planet Magnetospheres

• Example 1 UCL Magnetodisc model
• What is it ? A model which calculates
  self-consistent magnetic field structure and
  plasma distributions for a disc-like,
  axisymmetric, rotating magnetosphere. Used for
  studies of magnetospheric structure at Saturn and
  Jupiter (giant rapid rotators) e.g. Achilleos,
  Guio and Arridge (MNRAS, 2010), Achilleos et al
  (GRL, 2010)

7

• Present disc model assumes a fixed profile of
  plasma rotation (angular velocity ?M) versus
  radial distance.
• New configuration couples disc B-field model to
  an independent solver of plasma ?M

B Field Profile in Plasmadisc
?M Solver
UCL Magnetodisc Model
?M profile needed to update force balance,
re-compute B Field
8

• New configuration links two independent Matlab
  codes on different nodes in a local network
  through a simple Java class using IP Sockets.
• We could do a similar exercise for the
  magnetopause field model - at present we use
  average value from the formulae of Alexeev et al
  (Ann. Geo. 2005) (Catalogue entry ?)

B Field Profile in Plasmadisc
?M Solver
UCL Magnetodisc Model
?M profile needed to update force balance,
re-compute B Field
9

• A reminder of what JRA3 activities may foster in
  the long term
• My feeling is that these are beyond the scope of
  Europlanet (not enough resource) but they may
  indicate where we are headed if we follow some
  of the more simple ideas.
• Example A A time-dependent MHD model of Jovian
  or Kronian plasmadisc to couple to the
  appropriate atmospheric GCMs, to study transient
  auroral response.
• Example B A time- and space-dependent model of
  auroral precipitation associated with the
  Io-Jupiter current circuit, to investigate the
  nature of wind systems driven by such a localised
  source of auroral energy input.

10

• Progress on Catalogue-Linked Web Apps
• These are
• The H3 cooling function calculator (code from S.
  Miller) JRA3T3
• The VoiSe image segmentation tool (code from P.
  Guio) JRA3T4
• Web apps migrated to new webserver (beginning
  2011) and are now live at
• http//astroweb.projects.phys.ucl.ac.uk/europlanet
  jra3/
• Entries for these web apps have been update on
  the JRA3 catalogue.

11
Process diagram for each web app
Process Job
User
Results/Notification by Email
Input files?
Input parameters
Job details
Results?
Queue Job
Validate Parameters
Jobs
Job details
Job details
12

• Technologies used for Web Apps
• Web Interface and Parameter Validation
• Django web framework for Python
• Job Queuing System/Broker
• RabbitMQ Advanced Message Queuing Protocol (AMQP)
  compliant message queuing system
• Job Processing
• Django-Celery asynchronous task queue/job queue.
• Database for Job Detail Storage
• MySQL

13

• Upcoming Web Apps
• MarTIM
• Mars Thermosphere and Ionosphere Model
• 3D general circulation model of the Martian
  middle and upper atmosphere from 0.883 to
  9.910-8 Pa.
• Developed by Will Nicholson
• Possible coupling to TransMars
• UCL Magnetodisc Model
• Described earlier
• A library of disc model outputs for community use
  to be placed in the Catalog.

14

• Next steps?
• Action for all JRA3 Modellers Whose work is
  relevant to what you are trying to achieve ? Are
  they in the catalog - would they be willing to
  have an entry in the catalog ?
• The examples shown here illustrate the need in
  some areas of planetary science for progress by
  coupling models, even at a very basic level -
  e.g. GCMs as testbeds for auroral profiles
• There may also be a possibility that we could
  provide force-balance models for theoretical
  models of exoplanet discs (e.g. work by
  Khodachenko et al)
• Keep it simple, keep it realistic, make it a part
  of what you would naturally want to do in your
  own research programme
• From our meetings, it is clearer that IDIS
  ideally would provide a variety of databases
  which, in principle, could be queried from a
  system like the one described herein - keep this
  in mind, and keep informed of IDIS progress.
  (Note IDIS general meeting in Rome, June 7)