Balltracking: An highly efficient method for tracking flow fields

Abstract

We present a method for tracking solar photospheric flows that is highly efficient, and demonstrate it using high resolution MDI continuum images. The method involves making a surface from the photospheric granulation data, and allowing many small floating tracers or balls to be moved around by the evolving granulation pattern. The results are tested against synthesised granulation with known flow fields and compared to the results produced by Local Correlation tracking (LCT). The results from this new method have similar accuracy to those produced by LCT. We also investigate the maximum spatial and temporal resolution of the velocity field that it is possible to extract, based on the statistical properties of the granulation data. We conclude that both methods produce results that are close to the maximum resolution possible from granulation data. The code runs very significantly faster than our similarly optimised LCT code, making real time applications on large data sets possible. The tracking method is not limited to photospheric flows, and will also work on any velocity field where there are visible moving features of known scale length.