Solar Physics

The group has a long tradition of solar physics research including the Wilson effect in sunspots, Eddington-Sweet circulation in stars, and Sweet-Parker magnetic reconnection theory. Recent and current work spans theoretical and observational solar physics, the latter utilising data from numerous solar space missions, and is concentrated in the following areas:
   

High energy processes in flares

(Allan, Brown, Fletcher, MacKinnon, Haydock, MacDonald, Toner, Vogt)
Main collaborations: Huntsville, Lockheed Martin, Montana, NASA Goddard, UC Berkeley, Meudon, Zurich, ISAS.

Solar flares are dramatically evident in the high energy radiation they produce - for example hard X-rays (20-100keV) and gamma-rays (1MeV and above) generated by fast electrons and protons accelerated during the flare. Work in the group concentrates on the theory of the production of these radiations in solar plasma, the theory of acceleration and propagation of the particles responsible for them, and diagnostic (inversion) methods for quantitative interpretation of observed photon distributions (such as will be obtained with the HESSI mission) to recover useful diagnostic information about the particle spectra and evolution (due to e.g., scattering, wave generation), and atmospheric conditions. This has much in common with diagnostics in laboratory plasmas

• Observational Studies

  The primary diagnostics of energetic particles in solar flares are hard X-rays, currently observed with the Hard X-ray Telescope onboard the UK/US/Japanese Yohkoh satellite, which is able to image in the range 20-93 keV, and make crude spectra. Data from HXT have been used in multi-wavelength flare campaigns studying individual flares, to work out the relationships between photospheric magnetic field to that of the flare at X-ray and UV wavelengths, with a view to identifying the parts of the magnetic field in which the principle acceleration event takes place, and comparing the evolution with models. This also involves UV/EUV data from the Transition Region and Coronal Explorer (TRACE) satellite, magnetic data from the Michelson Doppler Imager instrument onboard SoHO, and soft X-rays observed with the SXT instrument onboard the Yohkoh satellite.

The positions of the footpoints of EUV flare ribbons evolving in time (LH) superimposed on the underlying magnetic field in the region.

HXT data will, to some extent, be superseded by that from the High Energy Solar Spectroscopic Imager (HESSI), on which Glasgow is a Co-I Institute. This will provide higher resolution spectra, continuous from the `firm' end at 5keV (never before done) up to gamma-rays, with the ability to resolve strong gamma-ray diagnostic lines. It will also provide higher resolution images, and the first ever spatially resolved gamma-ray flare observations.

Energetic, neutral emissions from the Sun (gamma-rays and neutrons) tell us about the extremes of solar particle acceleration. With colleagues from Meudon and Toulouse (data from the Franco-Soviet GRANAT spacecraft), and from New Hampshire and Riverside, California (data from the Compton Telescope - Comptel - on NASA's Compton Gamma-Ray Observatory), we are involved in a range of data analysis and modelling projects. We bring Glasgow expertise in instrument deconvolution to the Comptel neutron data, and in modelling pion decay gamma-rays to the GRANAT data. Such data, when combined with complete instrumental response treatments and modelling calculations, can constrain tightly the flare primary ion distribution at energies above 300 MeV/nucleon.

The distribution of flare protons with energies of a few hundred keV and below is still, largely speaking, an unknown quantity, as very few diagnostics exist for it. Linear polarization arising from impact excitation of atoms by protons in this wavelength range has been proposed as a viable diagnostic, and we are working with linear polarization observations from the THEMIS instrument, complemented by theoretical calculations, to exploit this diagnostic potential.

• Theoretical Studies

  Theoretical work on high energy processes concentrates on the acceleration and subsequent transport of particles in flare plasmas, and the behaviour of electric and magnetic fields in the vicinity of three-dimensional magnetic nulls:

  The acceleration of fast particles near the null points in two- and three- dimensional collisionless reconnection regions is being modelled using test-particle simulations to solve exactly the particle equations of motion in this non-adiabatic regime. In general, following acceleration, particles are transported throughout the solar corona, where they radiate and are scattered; transport problems are treated analytically and numerically. For example, we are developing parametric methods to solve the problem of the generation of waves by streaming electrons in a collisional plasma.

  In cases in which analytic solutions are not possible, we have developed efficient test-particle simulations, stochastic simulations, for modelling the evolution of the particle distribution function. Stochastic simulations are based on the stepwise integration of the differential equations describing particle motion, which in general contain a stochastic scattering (diffusion) term, due to collisions with waves and particles. The fact that this is essentially a test-particle technique allows physically realistic background field and plasma distributions to be modelled, along with the effect of multiple influences on the particle distributions.

  Discussions of magnetic reconnection usually assume what happens in the dissipation region can be described by the currents and fields of a larger-scale MHD calculation, with a scalar resistivity. A previous Glasgow PhD demonstrated that a self-consistent description of 2D reconnection cannot be obtained in this form. In our current work we search for a self-consistent description of steady, 3D reconnection near a neutral point. Existing, cold plasma magnetofluid descriptions form the starting point for this work, but our aim is to iterate towards a state in which diffusion region dynamics dictate self-consistent electric and magnetic fields, getting away from the getting away from the limitations of a calculation invoking (even 'inertial') scalar resistivity.

Atmospheric Structure and Dynamics

(Diver, Fletcher, Potts)
Main collaborations: Boulder HAO, Buenos Aires, Lockheed Martin, Cambridge, Meudon

The plasma environment of the solar corona is structured on all observable size- and time-scales by the magnetic field which permeates it. To understand the complexity of the observations it is necessary to combine models of the magnetic field and its evolution with the observed data. This is being done for flares and also for non-flaring active regions, where multi-thermal observations, magnetic data and plasma diagnostics assist in understanding the three-dimensional structure and evolution of the magnetised coronal plasma.

Three-dimensional magnetic field reconstructions (courtesy of Christina Mandrini) and the S-shaped active region which we are modelling.

The Lattice-Boltzmann method (see Cosmic and Laboratory Plasmas and 1.3 below) will be used to model the solar photospheric flows past flux-tube footpoints, in order to calculate the pressure distribution around them. The effect of the flow, in terms of net forces and circulation will be accommodated in an accompanying MHD numerical simulation of the flux-tube evolution, driven by the LB results. This will allow coupling of the gas-dynamical energy of the solar surface flow, into magnetic energy release in the the upper solar atmosphere. In particular, such numerical simulations, driven by data from TRACE and SoHO (among other sources), will allow the intrinsic toroidal twist (safety factor) to be inferred for erupting flux tubes, thus giving vital diagnostic information on subsurface solar plasmas as a modelling framework for Solar-B.

Simulations of Magnetised Plasma Processes

(Diver, MacKinnon, Potts)
Main collaborations: Open University, Thessaloniki, Univ. Central Lancashire.

• Lattice Boltzmann Simulations

  The interaction of photospheric flows with magnetic flux tubes is being modelled using the Lattice Boltzmann technique, which returns to a kinetic treatment of fluid flow, but with a highly restricted number of velocity components. The Boltzmann equation is solved by streaming notional 'particles' on a lattice, allowing collisions. The collision rules encapsulate the physical laws of the system (conservation laws, continuity, etc). This technique is particularly suited to solving solar problems due to its ability to deal with complex and moving boundaries such as that of an irregular flux tube being distorted by flow, and its ability to include the energy equation, MHD effects and compression.
  

• Avalanche Models and Self-Organised Criticality

  Statistical models for solar flares are being developed, motivated by the various sorts of observational evidence pointing to a highly fragmented character for the solar flare energy release process. We model this supposing that many elementary energy release events self-organise, by nearest-neighbour interactions, to give much larger events, with the relative frequency of occurrence of flares of different sizes as a consequence. Success in such a project will have many implications for overall understanding of flare physics.

  Comparison of observed hard X-ray light curves (left) with temporal development of a 3-D percolation avalanche model (right) suggests these models may aid in understanding flare fragmented energy release.
  

We are investigating this idea in an illustrative way, using models of elementary, discrete events on a grid. The simplest possible class of models are formally identical with those studied in percolation theory. We have demonstrated the following: power-law event size distributions, as observed, arise naturally in such models; the presence of a phase transition in the models is essential to the occurrence of a power-law - but they need not be in a state of criticality; observed temporal behaviour supports models in which nearest neighbours interact in a correlated way, e.g. as if governed by a local conservation law. Statements of the last type are of particular interest in astrophysics, potentially narrowing the range of possibilities when unknown physics is involved. Additionally, we are studying the effect of non-local communication in 'sandpile' type, self-organised critical (SOC) models, demonstrating the destruction of the SOC state and a transition to limit cycle behaviour.

Solar Wind and IPS

(Woan)

(See radio astronomy). An important feature of low frequency radio astronomy is the effect of the interstellar and interplanetary media on wave propagation. By combining scintillation ('twinkling') measurements of distant quasars and radiogalaxies with models of plasma dynamics we can infer much about the nature of both. The Astronomy and Astrophysics group holds the primary data from the New Cambridge IPS Survey (1990 to 1994). This unique data set comprises daily measurements of the scintillation of approximately 800 extragalagtic radio sources distributed over the sky as seen from the Mullard Radio Astronomy Observatory, Cambridge, UK. The antenna itself is now no longer maintained and is unable to make further observations, so this database represents a unique view of the inner heliosphere throughout the early 1990s. Recent analysis of these data by Glasgow workers has identified a new single-station velocity mapping technique and has compared IPS-based parameters of the solar wind derived over this period with in situ measurements (from IMP-8, SAMPEX and GOES) and from other ground-based facilities. This also demonstrated the usefulness of large-scale heliospheric density mapping for geomagnetic storm prediction.

Observational Solar Polarimetry

(Clarke)

Observational solar polarimetry is undertaken locally. One novel experiment has involved precision measurements of the global (total) polarization from the solar disk using medium passbands and also within H beta, with a view to making comparisons of polarimetric variability with solar-type stars -- making solar-stellar connections by polarimetry. We have detected a polarization of ~ 0.005% during the passage of a large sunspot complex across the solar disk.

Solar limb polarimetry has also been explored at H alpha with a view to investigating possible differences between the equator and poles, this being an issue with respect to claims by other researchers that photometric measurements reveal that the solar chromosphere is prolate. However, our measurements, with a precision of p~ +/- 0.02 provided a null result.
 

A sample of data of the centre-to-limb solar polar and equatorial q polarization as a function of the heliocentric angle as recorded on 1997 July 9. Error bars  are indicated for the polarization measurements; the uncertainties on position angle are not indicated.