A statistical investigation into coupled magnetospheric-ionospheric dynamics

via multi-scale, multi-instrument, data assimilation


A. Grocott, O. Amm, J.B.H Baker, M.P. Freeman, S.E. Haaland, B. Hubert, G. Lu, F. Pitout, I.J. Rae, and T.K. Yeoman

 

Objectives and expected output: The objectives of our proposal are two-fold.  Initially, we will investigate data assimilation methods that would be suitable for application to space physics, and establish techniques for the integration of existing geophysical data products.  Once in place, these techniques will be available to the wider community for future exploitation.  We then intend to implement these new techniques, with the specific aim of delivering a more complete characterisation of the nature of magnetospheric-ionospheric dynamics.  This output will take the form of publications in refereed scientific journals, with both elements of our work providing a framework of understanding in which data from future missions can be analysed and interpreted.  It is our intention that the development of new analysis techniques will itself suggest the more specific science goals we are able to address and we therefore do not want to be overly prescriptive on what they might be at this stage.  However, here we describe the data-sets and existing techniques which will form the basis of our programme, and give some examples of the science questions we intend to address.


1. Data and assimilation methods:  Clearly, in order to address the issues outlined above we will need access to a wide variety of data-sets and scientific capabilities.  In order to produce statistically valid characterisations these data-sets will need to cover a sufficient interval of time (preferably of the order of one half solar cycle, minimum).  The primary data-sets we will use are therefore SuperDARN ionospheric radars (data available from 1995-present), the IMAGE FUV auroral imager (2000-2005), the Cluster satellites (2001-present), ACE upstream interplanetary data (1997-present), and ground-based magnetometer networks, such as IMAGE (1982-present), Greenland (1981-present), CARISMA (2003-present, although earlier CANOPUS data is also available), Antarctic LPM (2001-present).  Ancillary data-sets such as WIND, Geotail, and Polar are also publicly available via the internet.  Existing assimilation methods which will form the basis of our work include the SuperDARN Map Potential technique, ground-magnetometer inversion methods, auroral boundary identification techniques and precipitation/conductivity measurement methods.  We intend to investigate both the combination of existing techniques, and the extension of techniques to statistical applications.  We also intend to perform multi-scale analysis of various data-sets, via the integration or mapping of observations across multiple spatial scales.  This has two specific advantages.  Firstly, where multi-scale analysis is made on a single data-set (e.g. small-scale radar electric field measurements versus global ionospheric convection maps) continuity of data at both spatial scales is guaranteed.  Secondly, where measurements from different instruments at different spatial scales are integrated (e.g. the magnetic mapping of spacecraft data into the ionosphere) direct comparison of otherwise incompatible observations is possible.  Using these techniques, we intend to incorporate the influence, or contribution, of small-scale phenomena into the same conceptual framework of larger-scale dynamics.


2. Science Goals: We aim to pursue a full characterisation of the average M-I system response to a range of controlling parameters such as the IMF strength and orientation, the solar wind speed and density, and different levels of geomagnetic activity.  Primary features we will be looking to characterise are the convection, magnetic field topology, particle distributions and auroral morphology.  Secondary characteristics which may be possible to study include the auroral conductivity, field-aligned currents and plasma energisation mechanisms.  Specifically we hope to address problems such as the following:

  1. BulletWhat controls whether the magnetosphere is driven into the substorm cycle or into some alternate mode of convection, and what exactly characterises the different modes?

  2. BulletHow does the magnetosphere respond to, and what are the governing factors for, substorms that initiate at different latitudes and local times, and how does this relate to substorm magnitude and magnetic storms?

  3. BulletCan we reconcile the differences between the global dynamics inferred from observations using different instrumentation, e.g. satellite versus radar observations of convection, or convection versus geomagnetic quantifications of energy transfer into the magnetosphere?