Pulsar Research

Introduction

Welcome to the pulsar research page.

Members

• Prof. Andy Shearer
• Eoin O'Connor
• Albaset Dhannoon

Some Past Members

• Dr. John McDonald
• Susan Collins
• Dr. Diarmad De Búrca

Collaborating Institutions

• NUI, Galway: Computational Astrophysics Laboratory [Schools of Physics and Mathematics]

Pulsar Modelling

An understanding the physical processes at work within the pulsar magnetosphere has remained elusive despite nearly forty years of observations and theory. There probably can not be an analytical solution to this and the consensus is that only numerical solutions will work. Under CosmoGrid 1 we developed DYMPNA3D, a next generation 3D multi-purpose plasma simulation code based on the Particle-In-Cell, (PIC), methodology. The beauty of this methodological approach is that it is free from the constraints of initial assumptions, from which many of the alternative fluid techniques suffer. The PIC methodology involves the imposition of a spatial mesh on the plasma filled region of interest, the plasma then interacts with this spatial mesh, on which the plasma attributes are deposited. This reduces the N squared problem of the Coulomb interactions to an Nlog(N) problem in which the plasma interacts indirectly through the imposed spatial mesh. The N squared order of a Coulomb problem makes it computationally infeasible for the simulation of large-scale plasma's. By reducing the order of the problem to Nlog(N), the Particle-In-Cell approach to modelling plasma's essentially yields a first principle's approach to modelling plasma's in which, no simplifying assumptions are required. This approach deals with the plasma itself on a fundamental level, depositing charge and current densities on the spatial mesh, solving the electromagnetic fields from Maxwell's equations, calculating the forces on the plasma particles and moving them accordingly.

The code is now entering its final development stages having completed its first, GenOne, and second, GenTwo, test phases and has been successfully applied in the regime of pulsar magnetospheric plasma distributions. The initial results correlate very well with theoretical predictions and with results of previous numerical simulations in this area. The primary design remit for the code was that it was to be constructed in an extremely modular fashion in order to facilitate ease of modification to investigate a plethora of kinetic phenomena in a diversity of astrophysical objects of interest ranging from objects as obscure as Pulsars and Brown Dwarfs to objects as familiar as the Earth, Jupiter and the Sun. The code is fully relativistic so that it is applicable to kinetic plasma phenomena in all areas of interest in which such phenomena are known to play a fundamental role in the atmospheric dynamics and emission mechanisms.

The code is fully parallel enabling it to make efficient use of modern large scale computing resources. This is a critical aspect of the codes development as most of the kinetic phenomena of interest in terms of its application are well below the best resolvable spatial scales of today's modern telescopes and thus large scale numerical simulation is our only means of probing such phenomena. The large computational requirements of such simulations and the computational methodology of the PIC approach advocate the need for the utilisation of modern high-end computing facilities to make the simulation runs temporally efficient. Another extremely important aspect of the developed code is that is has an immense amount of post-processing built in to it. One of the quintessential problems with today's large scale simulation codes and simulation runs is the vast amounts of data produced. The analysis of the resulting data can be expensive and time consuming as virtually of of the publicly available plasma simulation codes leave any and all post-processing to the user. DYMPNA3D allows for very accurate control of the simulation output with a plethora of inbuilt options for the output. The plasma data can be output to standard data files to allow for arbitrary post-processing, but the output can also be pre-formatted in VTK files which allow the plasma data to be readily visualised in any of the freely available front ends for VTK, like Paraview. All of the vector data from the simulation runs can be output in vectorised VTK files or in terms of their Cartesian components at either the end of the simulation run or at the end of each iteration of the code to allow study of the temporal evolution of the plasma. This makes the investigation of arbitrary kinetic plasma phenomena a much more viable proposition as the ease of modification of the code, (due to its modular construction), the flexibility and scalability of the code, the 3D nature of the developed code and the inbuilt post-processing provide a unique cutting-edge platform for such investigations.

As mentioned, the code has already been successfully applied to the investigation of plasma distributions in the pulsar magnetosphere confirming the postulated stable charge-separated non-neutral Dome-Torus plasma configuration with imminent publication of the obtained results. Future applications of interest include the simulation of the Brown Dwarf emission mechanism simulation of a proposed mechanism for the emission of Giant Radio pulses from pulsars, (GRP's) i.e. turbulent plasma wave packet collapse, and comparison and verification of the most popular pulsar high energy emission models through high-end simulation. Also, investigation of the possible mechanisms which would allow pulsar magnetospheric plasma to propagate beyond the identified trapped Dome-Torus distributions to generate the observed jets and equatorial outflows seen with the Chandra X-ray observatory would be included. Exemplar studies of the Earth's magnetosphere and Jupiter's emission mechanisms are also planned.