The PIC Code Comparison Project emerged from discussions during the 2009 KITP workshop Particle Acceleration in Astrophysical Plasmas. The idea is to agree on a standard set of test setups, and try to run as many codes as possible with them. We will all learn from the exercise, and understand what makes some codes better or worse at certain tests.
There now exist many excellent Particle-In-Cell codes in the astrophysical community, but different groups have implemented different field solvers, filtering of currents, interpolation methods, and, fundamentally, different numerical methods. There already exist a vast and classic literature on PIC codes, but most of these tests are concerned about non-relativistic plasma dynamics dynamics, and we therefore feel that there is a need to redo some of this work, with a focus on tests, which are relevant for the astrophysical community.
By running the same tests with many different codes we hope to
* Get an understanding of what is needed to improve on the stability of the codes for longtime runs; this may to some extend be tested with synthetic benchmarks. Both to check for numerical heating of the particles, and generation of EM-waves with high wave numbers ('Cherenkov radiation')
* Understand how different codes compare in real world setups. What is for example the minimal number of skin depths per cell, and particles per cell, which are needed to get a collisionless shock, with the correct jump conditions ? Can useful physics be extracted from such a minimal experiment ? How much better does the resolution have to be, for tracing synthetic radiation, and looking at heating / acceleration in the shock structure ?
* Leave a legacy of tests that people who are developing PIC codes can use as the standard benchmark to test against.
Before the workshop begins most of the groups who are attending should have run some of the tests, so that we have material to work with. We will focus on getting work done, with only one or two talks per day, and office space for everybody.
The workshop is organised by the Computational Astrophysics Group and the Niels Bohr International Academy, and will be held at NBIA at Niels Bohr Institute in Copenhagen.
Results from the tests can be uploaded to a wiki, which is hosted here
The deadline for registration is March 1
Participation in the workshop is free, and we can offer some financial support for some of the participants. Affordable accommodation possibilities near the Niels Boher Institute can be found on this page