Friday, 10 October 2014

Flare prediction by sunspot dynamics

Today we had the first Solar Physics seminar for this academic year we were presented with an exciting piece of research investigating the prediction of solar flare events through studies of sunspot dynamics.

Solar flares occur due to the sudden release of energy stored in active-region magnetic fields. To date, the precursors to flaring are still not fully understood, although there is evidence that flaring is related to changes in the topology or complexity of an active-region's magnetic field. Understanding of these events thus require a study of reconnection phenomena. The seminar was introduced with a description of reconnection phenomena with descriptions of the Sweet-Parker model, Petschek model and collisionless reconnection.

The study of Mason et al showed that Flare occurrence is statistically associated with changes in several characteristics of the line-of-sight magnetic field in solar active regions (ARs). They calculated magnetic measures throughout the disk passage of 1075 ARs spanning solar cycle 23 and found a statistical relationship between the solar magnetic field and flares. This study of over 71,000 magnetograms and 6000 flares uses superposed epoch (SPE) analysis to investigate changes in several magnetic measures surrounding flares and ARs completely lacking associated flares. 

Virtually all X-class flares produce a coronal mass ejection (CME), and each CME carries magnetic helicity into the heliosphere. Using magnetograms from the Michelson Doppler Imager on the Solar and Heliospheric Observatory, La Bonte surveyed magnetic helicity injection into 48 X-flare-producing active regions recorded by the MDI between 1996 July and 2005 July. Their survey revealed that a necessary condition for the occurrence of an X-flare is that the peak helicity flux has a magnitude >6×1036 Mx2 s-1. X-flaring regions also consistently had a higher net helicity change during the ~6 day measurement intervals than nonflaring regions.

Murray  examined the evolution of the magnetic field in active region NOAA 10953 the study used Hinode/SOT-SP data over a period of 12 hours leading up to and after a GOES B1.0 flare.

The seminar presented a study of dynamic phenomena in flaring active regions and found diagnostically useful features for the assessment of risk, intensity and the time of imminent flares. Targeted features are sunspots and their data were taken from the SOHO/MDI-Debrecen  Sunspot Data Catalogue (SDD). The method described employed an appropriately defined proxy measure of the non-potentiality at the  photosphere, i.e. the horizontal gradient of the magnetic field between two subgroups of spots with opposite polarities at the polarity inversion line of the active region. The value and temporal variation of this proxy contains information about the expected time and intensity of imminent flares. Additionally, it allows an investigator to track the free energy content of the active region.

The curve fitted to the measured points can be described by the following formula:
where |K1| = 265 gauss and |K2| = 1067 gauss. This function relates a Bmean magnetic field to an A umbral area.  The proxy measure used to represent the magnetic field gradient between two spots of opposite polarities having areas A1 and A2 and at a distance d, is defined as:
The seminar picked out the results for a number of events e.g. NOAA AR9393  on March 28 2001 at 06:23 UT, NOAA AR9661 on October 17 2001 at 16:37  and NOAA AR10226 on December 19 2002, at 16:54 UT. Images and results from these are shown below.

The figure above shows images of AR NOAA 9393: continuum image, reconstruction from SDD, magnetogram. The diagrams of the right column: variations of GM in areas 1 and 3; diagrams of area 2: variations of GM, distances, flux amounts. The two lower axes show the longitudinal distance from the central meridian (LCM) and the radial coordinate (r) of area 2.
Results for the active region NOAA 9661.
Results for the active region NOAA 10226 within the selected area 1 between 2002 December 16 and 21

The results show a quick growth of the GM magnetic flux gradient defined by equation (2), this is not surprising, since this is the process of building up the non-potential component of the active region magnetic field. The rise takes about 1.5–2 days; the steepness of the strengthening is only weakly indicative of the flare intensity.

The high maximum of GM at about 3 × 106 Wb m−1 is also obvious because this is the measure of non-potentiality in the given region. This maximum value shows the only unambiguous relationship with the intensity of the released flare. The results enable the derivation of a relation which can be used as a tool to estimate the intensity of the expected flare from the measured maximum of the flux gradient.

There is a decrease of GM after maximum and until the flare onset. This might be the most surprising pre-flare signature because it would be expected that the state of strongest gradient is the most efficient trigger of the eruption. The after-maximum weakening prior to the flare may imply that the condition for the reconnection is not only the compression of the oppositely oriented flux ropes but also a subsequent loosening.

There is significant fluctuation during the decrease, this is also a characteristic feature of the pre-flare dynamics, but it is not related to the flare intensity. Its measured values are between (3 × 105 and 6 × 105 Wb m−1 during the decrease prior to the X-flares, whereas it is 7.5 × 104 Wb m−1 in the quiet domain No. 1 of active region NOAA 9393.

The obtained characteristics are mostly signatures for major flares. The flare class M5 as a lower threshold may seem to be arbitrary, but it can be justified. It was shown that for M4-class flares the maximum of GM is about 1 × 106 Wb m−1, while other results show that the fluctuation of GM may be as high as 0.7 × 106 Wb m−1; thus, the temporal variation of GM may be covered by the fluctuation in the case of weak flares. The recognizability of the presented behavior decreases toward weaker flares at about GM ~ 1 × 106 Wb m−1 and M4 class; it can be identified for flares more intense than M5.

There were questions regarding the rules for assessing the sunspot grouping. For example, what is the criteria for selection of an area or spot group fluxes opposite polarity  prior to flare indication of reconnection event? Future work will compare results generated  from analysis in the photosphere to analysis in the Chromosphere.

Pre-flare Dynamics of Sunspot Groups (ApJ)

Pre-Flare Dynamics of Sunspot Groups (arXiv)

M. B. Korsós, T. Baranyi, and A. Ludmány

Papers on flare forecasting
Mason et al,2010 Testing Automated Solar Flare Forecasting with 13 Years of Michelson Doppler Imager Magnetograms  (abstract) (pdf)

LaBonte et al 2007 Survey of Magnetic Helicity Injection in Regions Producing X-Class Flares (abstract) (pdf) 

Wang 2007, Structure and Evolution of Magnetic Fields Leading to Solar Flares (abstract) (pdf)

Murray 2012 The Evolution of Sunspot Magnetic Fields Associated with a Solar Flare (abstract) (pdf)

Studies of reconnection in laboratory plasmas
Yamada 1999, Recent Results from Magnetic Reconnection Experiment (MRX) at PPPL (abstract) (pdf)

M. Yamada, H. Ji et al., Phys. Rev. Lett. 78, 3117(1997)
H. Ji, M. Yamada et al., Phys. Rev. Lett. 80, 3256 (1998).

Tuesday, 13 May 2014

The Solar Global Eigenmodes of Oscillation

This post presents work we have undertaken to understand the dynamics generated in the solar atmosphere by the solar global oscillating eigenmode. We describe a model for studying dynamics in the solar atmosphere and describe simulations performed using this model. The results provide evidence of induced dynamics in the solar atmosphere.

The objectives of this work are to;
  • Demonstrate code validity by repeating earlier hydrodynamic 2D models of Malins and Erdelyi,
  • demonstrate cavity modes in the Chromosphere and surface modes in the transition region and
  • understand the dynamics allowing mode tunneling and energy leakage into the corona.
The Earliest observations of dynamical motions of the solar surface dates back to around 1918, (St.John C.E. et al, from Doppler Shift when studying solar rotation). In 1960 Leighton, reported observed Vertical motions of 300-400m/s, these were explained by Ulrich 1970. The solar p modes are generated by global resonant oscillations and turbulent motions just beneath the photosphere.  Modes which trapped below the photosphere are refracted by a sharp change in density. There is  a power peak at 5 – minutes these are p mode oscillations.  The picture below shows an example of a power spectrum obtained from 144 days the MDI Medium-l data for the modes averaged over the azimuthal order m. The power concentrates in ridges corresponding to solar acoustic (p) modes. The lowest weak ridge corresponds to the fundamental (f) mode.

It is found that  modes are evanescent around the photosphere, propagation into the corona is inhibited. However, modes can tunnel through and propagate into the solar atmosphere. Earlier work of Malins and Erdelyi using 2D hydrodynamical models to study dynamics with point drivers demonstrated cavity modes in the Chromosphere. There are also surface modes at the transition region. It is interesting to understand dynamics which enable tunneling.  Local acoustic cut-off is a natural period, disturbances at the cut off can cause dynamic responses. Propagation occurs below the cutoff, there is evanescence above the cut-off.  As demonstrated by Schmitz, 1998, the Cutoff can be calculated in different ways and for an Isothermal atmosphere or a highly stratified atmosphere. For the simulations undertaken here we represent oscillation modes as a vibrating membrane driver located at temperature minimum. We ensure that the simulation is constructed such that the driver delivers the same amount of energy for different modes.

The plots below display the computed cut-off for atmosphere models  

In the context of oscillating systems, the eigenmodes (or normal modes) of that system refers to the sinusoidal motion with which the  system oscillates. In particular all parts of the system move with the same frequency and phase relation. This motion of the normal mode is termed a resonance. Consider a solution for an oscillating membrane

Since the solution satisfies the wave equation we have the following relation for the modes of oscillation

The total membrane energy is obtained by integrating over its surface, this may be expressed as

Here we have made the wave speed the same as the speed of sound and we use the expression for an adiabatic gas

If we assume that the total membrane energy for all modes is the same as the fundamental 00 mode i.e.
then evaluation of the surface integral above the relationship between the mode amplitudes

Expressing the membrane solution using the wave number

The wave number can be expressed as
Here we have used
Using c=8.4km/s (speed of sound at the photosphere) this can be calculated from the VALIIIc data

We can calculate the frequency of oscillation using
For each mode of oscillation we keep the ratio
we also keep the energy fixed by using the relationship for Anm above

A visualisation of such a membrane with oscillations described by the above relations is shown in the animation below illustrating the 1,1 mode

This was obtained using the scilab script membrane_modes.sce which is available on github.

Simulations were performed using the stratified MHD code enabled for GPUs (SMAUG).  For the computational Model we used a Model of the stratified solar atmosphere using data sets from VALIIIC and McWhirter.  The model height is a 6Mm through the atmosphere with a cross section of 4Mmx4Mm. The computational domain was divided into 128x128x128 equal computational elements. Simulations were run with membrane drivers, with periods of 30s,180s,300s these were run for the 00,01 and 02 modes.

To study the energy leakage and the evolution of the membrane oscillations we present simulation results showing a view through the simulation box for the vertical component of the velocity. We also present plots showing the integral of the total energy for each time step through the solar atmosphere. The integration of the total energy is performed over the model cross section and at each height of the computational domain. Since most of the energy is trapped near the photosphere, we plot the energies for the transition region and into the Corona.

Results for the 30s driver with 0,1 mode are shown below

Results for the 180s driver with 0,1 mode are shown below

Results for the 300s driver with 0,1 mode are shown below

The results above show that the 180s and 300s modes provide energy leakage  with period the same as that of the driver, the 30s driver is the least efficient at coronal  energy leakage. The results also provide evidence for non linear behaviour. Using the formula

we compute the frequencies for the normal modes, for the 00, 01 and 03 mode. For the 00 mode a 670s driver was used, for the 01 mode a 430s driver was used and for the 03 mode a driver with a period of 230s was used. For these normal modes of oscillation we obtain.

The image above shows distance time Plots for 0,1 mode from left to right 30s, 180s, 300s for the vertical component of the velocity. It is clearly seen that the 30s mode does not support oscillations in the corona. The 180s and 300s modes support standing modes in the transition region and the corona. It is informative to summarise this collection of results by presenting plots which show the energy deposited into the corona and transition zone. The energies have been integrated over these regions of height and also over time, an average was taken over the times. For computing the integration the chromosphere is from the base of the simulation box to a height of 1.78Mm,  the transition region is between 1.78 and 2.16Mm and the corona is the remainder of the box above 2.16Mm. For the integrated energies we compute the difference between the total energy and the background energy, i.e. the perturbed energy,  these integrated energies are shown in the plots below for the different modes and different driver frequencies.

The 180s fundamental is effective at leaking energy into the transition region but energy leaks from the atmosphere  and back to the transition region. The low values for the 30s driver are related to the cutoff? For the fundamental mode the 300s driver appears to be the most effective at enabling energy leakage into the atmosphere. Not all wave energy of the driver will go to the corona. The drivers can excite surface waves in the TR region, waves can be reflected from the TR region.


  • The GPU code performs well 
  • The results provide evidence for energy leakage into the corona
  • The fundamental mode 5 min driver (300s) is effective for energy supply to the corona region and the 3 min driver is effective for TR
  • 01 mode  5 min nothing for corona and 3 min again effective for TR
  • 02 mode 5 min nothing for corona and 3 min effective for corona
  • The distance time plots illustrate cavity modes in the chromosphere
  • For the 300s and 180s drivers there is a clear indication  of induced dynamics in the corona
  • There are unexplained resonances e.g. for 30s driver probably resulting from non-linear behaviour
  • Further characterisation of the normal modes and run models with a greater range of modes of oscillation
Further work will consider genuine MHD examples for a vertical B field, a horizontal B field and for flux tubes.


We also consider magnetic effects referiing to 

Magnetohydrodynamic waves driven by p-modes Elena Khomenko, Irantzu Calvo Santamaria

 Simulations of the Dynamics Generated in the Solar Atmosphere by Solar Global Oscillating Eigenmodes (slides for NAM2014, June 2014, Portsmouth Univesity)

F.Schmitz and B. Fleck Astronomy and Astrophysics, v.337, p.487-494 (1998)

Course Notes on Oscillations and Waves

Friday, 4 April 2014

Modelling Magnetic Reconnection and Tearing

Today we saw the return of the Solar Physics seminars at The University of Sheffield. The seminar entitled "Magnetic reconnection and tearing in a 3D current sheet about a solar coronal null" was given by David Pontin from The University of Dundee. We heard about a study of magnetic structure in the solar atmosphere and attempts to explain energetic events such as flares and other explosive events. These events resulting in reorganisation and restructuring of the magnetic fields result from reconnection phenomena. To understand reconnection we have to look at the exciting territory of resistive plasma dynamics. We are reminded of the contrasting situations of ideal fluid dynamics with infinite Reynolds number and the case of the viscous fluids with finite Reynolds numbers. In dissipative dynamics we therefore have  boundary layers (e.g. boundaries between regions of differing viscosity).  Such dissipation effects arise when we introduce resistive boundaries in MHD. The resistive boundary layers give new types of plasma instability caused by a loss of conservation of magnetic flux and the reconnection of magnetic field lines.

The animation above illustrates such reconnection occuring at so called x-points where flux conservation is broken and the reconnection occurs. Magnetic null points are regions where the magnetic field goes to zero, such points arise during explosive reconnection events in the active solar atmosphere. Three-dimensional magnetic null points are ubiquitous in the solar corona and in any generic mixed-polarity magnetic field. We discuss the nature of flux transfer during reconnection an isolated coronal null point, that occurs across the fan plane when a current sheet forms about the null. We then go on to discuss the breakup of the current sheet via a non-linear tearing-type instability and show that the instability threshold corresponds to a Lundquist number comparable to the 2D case. We also discuss the resulting topology of the magnetic field, which involves a layer in which open and closed magnetic fields are effectively mixed, with implications for particle transport.
We heard about the spine-fan field structure which may enclose null points and can lead to reconnection events, such a structure is shown above. Magnetic field lines outlining the spine and fan structures associated with a magnetic null, located in a separatrix dome above a parasitic polarity. The shading on the lower surface represents the normal component of B and the dashed line marks the polarity inversion line. Using  resistive MHD to model such a null point located in the centre of a simulation box a stream function applied at the boundaries is used to cause advection of the spines in opposite direction, this is described by the advection coefficient, Ad.
The figure above illustrates the initial magnetic field configuration. The spine (in blue) lies along the x-axis and the fan (in red) lies in the x = 0 plane. Coloured arrows indicated the field direction. The black arrows show the direction of shear applied by the driver at the boundaries. Bottom: the driving profile applied to the x-boundaries with Ad = 80.

The change in connectivity with two simultaneous pulses (top) one pulse (bottom) and when v0 = 0.04, Ad = 80 and η = 5 × 10-4. The red and blue flux tubes (traced from the driving boundaries) are initially connected to the gold field lines anchored on the side boundaries. Taken at times t = 1 (left), 3 (middle) and 5 (right). These results were obtained using the 3D resistive Copenhagen code of Nordlund and Galsgaard (1997).

 The figures above show representative field lines traced from fixed footpoints (marked by spheres) located close to the spine axes, at times t = 0, 0.33, 0.38, 0.5, 0.62, 0.8. The footpoints are located in the negative sources, so the motions of the field lines exhibit the flux velocity wout. The location of the null is marked by the red diamond.

 Frames showing the magnetic field components (arrows) and current density (shading) in the y = 0 plane over x ∈ [ − 0.8, 0.8], z ∈ [0, 0.8] for the MHD simulation. From top to bottom, times t = 1.0, 2.4, 4.6, and 6.0.

(a) Representative magnetic field lines at t = 0 in a subsection of the domain for the MHD simulation. The shading on the z = 0 plane represents the vertical magnetic field strength on that plane. (b) Pattern of the driving flow close to the parasitic polarity. Also shown are the normal component (Bz) of the magnetic field at the photosphere, z = 0, at (c) t = 0 and (d) t = 3.0.

Asymmetric Reconnection at 3D Magnetic Null Points from UK Solar Physics on Vimeo.

The following images show individual frames from the asymmetric reconnection event
 Above, the shaded contours show |J| (scaled to the maximum of each snapshot), while the arrows depict the plasma flow. For the case of v0 = 0.01 with only one pulse.
 Above, the shaded contours show |J| (scaled to the maximum of each snapshot), while the arrows depict the plasma flow. For the case of v0 = 0.01 and tlag = 1.8.

Sample magnetic field lines in the y = 0 plane showing the null point structure at (a) t = 0 and (b) t = 0.8. The vectors show magnetic field orientation and the background shading is proportional to Ey. The field lines in (a) and (b) are traced from the same fixed footpoint locations in the positive polarities near the fan.

Field lines of the simple model magnetic field (described in Section 3) outlining the spine and fan of the magnetic null at t = 0. The shading on the lower surface represents the normal component of the magnetic field there. The circle on the lower surface shows the approximate location of the footprint of the separatrix dome (i.e., fan separatrix surface). The numbers refer to source numbering discussed in the text.

It was interesting to see some of the movies illustrating the propagation of plasmoids through the current street.We heard about the dynamics of flux ropes and how these reconnection events introduce torsional wave motions. Much of this seminar was summarised in the excellent UK  Solar Physics nugget entitled "Asymmetric Reconnection at 3D Magnetic Null Points".


  • Pontin, D. I., Priest, E. R. and Galsgaard, K. On the nature of reconnection at a solar coronal null point above a separatrix dome, Astrophys. J., 774, 154 (2013).  arXiv preprint journal webpage
  • Wyper, P. F. and Pontin, D. I. Kelvin-Helmholtz instability in a current-vortex sheet at a 3D magnetic null, Phys. Plasmas, 20, 032117 (2013). arXiv preprint
  • Wyper, P. F., Jain, R. and Pontin, D. I. Spine-Fan reconnection. The influence of temporal and spatial variation in the driver, A&A, 545, A78 (2012). journal webpage
  • Pontin, D. I. Theory of magnetic reconnection in solar and astrophysical plasmas, Phil. Trans R. Soc. A, 370, 3169-3192 (2012).  pdf

Tuesday, 11 March 2014

Our Wobbling Star!

During a recent discussion I presented a fairly simple application providing an animation of spherical harmonics. The purpose of this post is to address the questions which arose during that discussion. I've already alluded to some of the work we are undertaking in the earlier posting "Dynamics of the Solar Atmosphere Generated by the Eigen Modes of Solar Global Oscillations". The modelling we have described requires us to provide a mathematical description of excitation phenomena which may arise in different solar structures. The resulting formulation is used as an input to our computational models. In this post I'd like to begin to understand the global solar phenomena  which may provide an important input into our models of solar excitation's. The driver for our model is spatially structured and is extended in a sinusoidal profile across the base of the computational model.  Our objective is to run computational models which  recreate the atmospheric motions generated by global resonant oscillations.

As commented by Christensen-Daalsgard, the earliest observations of motions of the solar surface probably date back to around 1918, the motions which were detected arose from measurements of the changes in the Doppler shift that were observed during attempts to measure the rate of rotation of the sun (St.John C.E. et al ). Since then, observations using a variety of space and ground based solar telescopes clearly demonstrate the diverse range of wave phenomena which are observed in the solar atmosphere. One of the earliest to be discovered was the five minute oscillation, the p-mode. The solar p modes are generated by global resonant oscillations and turbulent motions just beneath the photosphere. The restoring force for these oscillations is the pressure within the different solar structures, variation of the pressure and subsequently the speed of sound determines the dynamics of these modes. The resulting propagation of this wave energy into the solar atmosphere may be used as a diagnostic tool to predict some of the physical characteristics of solar atmospheric structures. 

In 1960 Leighton presented the first direct evidence for motion tangential to the solar surface this was in addition to the vertical motions which were found with velocities in the range 0.3-0.4km/s, these were observed within a region of scale 3Mm. The vertical motions showed a strong oscillatory character with a period of 296s there were 2-4 oscillations before the vibrations died out. In 1970 Ulrich provided one of the first analyses of these so called 5 minute oscillations, a study of the sub photospheric layers was undertaken. Standing acoustic waves may be trapped in a layer below the photosphere. It was suggested that the power output from these oscillations above the temperature minumum  is a significant contribution to the dissipation of energy from the chromosphere to the corona.

A study of these oscillations and their interactions enables a determination of the properties of the solar interior. This subject known as Helioseismology has been discussed by many authors two well known papers are as follows. 
Lecture Notes on Stellar Oscillations
Helioseismology  - J. Christensen-Dalsgaard
Observations indicate that oscillations are damped within a few periods. Global helioseismology involves the study of standing modes which are the result of the constructive interference of the localised modes. The constant wobbling of our nearest star is a result of this superposition of normal modes. We can start to understand the nature of these oscillations by considering the hydrodynamic wave equation for a gravitating slab. This is derived by Goedbloed and Poedts in Principles of Megnetohydrodynamics, see also Goedbloed's presentation.  A similar derivation is provided in the texts by Christense-Dalsgaard, above. To summarise, the derivation uses the equations of gas dynamics and neglects disturbances of the gravitational field (the Cowling approximation). Assuming a static background  (e.g. hydrodynamic equlibrium) the hydrodynamical equations are perturbed and the equations are expressed using the lagrangian displacement. If v1 is the perturbed fluid velocity the lagrangian displacement is defined as

This approach leads to the following wave equation

Using Cartesian co-ordinates the normal mode solutions can be written as

The Cartesian approach is useful for slab geometries and for model geometries where the curvature of the solar surface can be neglected. For helioseismology it is necessary to treat the problem in full spherical geometry. With spherical solutions the wave number, k, can be quantized in terms of the solar radius as follows
Considering the radial component of the normal mode expansion in spherical coordinates, the displacement and perturbed pressure may be expressed as

The spherical harmonics are defined as 
Plm is the legendre polynomial and clm is defined such that integral of the modulus of the spherical harmonic evaluates to unity. l and m are integers which must satisfy -l<=m<=l. The indices l and m are the mode numbers for the oscillation. The mode index n is determined from the radial solutions. Some of these modes are shown below.

The images illustrate clearly that for higher values of l there is more localisation of the amplitudes. l increases localisation in the theta direction whilst increasing m, increases the number of periods in the phi direction around the rotational axis, i.e. m is the number of modes found around the equator. For m=+/-l and with large values of m the modes are concentrated near the equatorial region. A simple routine used to generate the above plots using the open source matlab clone, scilab is available at

Harmonic oscillator Example Scilab routine

The p modes with high l numbers  are more localized to the exterior of the sun. The so called f modes are gravitational oscillations of the solar surface. Gravitational modes generated in the interior are challenging to detect, these very low frequency modes whose restoring force arises from adiabatic expansion deep in the solar interior are evanescent, their amplitude near the photosphere is significantly reduced making them difficult to detect. The plot below shows the observed frequency power spectrum of a 5-day segment of BiSON Sun-as-a-star data. The dark solid line is the smoothed spectrum, while the dashed line is an estimate of the background (from Chaplin et al 2008). 

The picture above show the observed mode velocity amplitudes Vobs (thick grey line) and theoretically computed amplitudes V (dotted line) of the Sun. The plotted amplitudes are the mean amplitudes from analyzing 950 independent 5-day segments of BiSON Sun-as-a-star data (from Chaplin et al 2008). Information from the whole surface of the Sun is collected and is then interpreted to find information about the standing waves in the Sun's interior causing it to oscillate. Because some of these modes are formed by sound waves which penetrate into the deep interior of the Sun we are provided with an unprecedented opportunity to study the core regions. Here we have briefly reviewed models enabling us to gain insight into the nature of the wobbling motions of the sun with this knowledge we may begin to understand how the dynamics of the solar atmosphere are influenced by the global eigenmodes. In a further posting we consider the justification for modelling the drivers in our computational study of the solar eigenmodes of oscillation. So far we have used a hydrodynamical model for this study which must be extended to include the complexities introduced by the solar magnetic structures.