Automatic Recognition and Characterisation of Supergranular Cells from Photospheric Velocity

AbstractWe have developed an exceptionally noise resistant method for accurate and automatic identification of supergranular cell boundaries from velocity mea-surements. Due to its high noise tolerance the algorithm can produce reliable cell patterns with only very small amounts of smoothing of the source data in com-parison to conventional methods. In this paper we describe the method and test it with simulated data. We then apply it to the analysis of velocity fields derived from high-resolution continuum data from MDI (Michelson Doppler Imager) on SOHO. From this, we can identify certain basic properties of supergranulation cells, such as their characteristic sizes, the flow speeds within cells and their dependence on cell areas at high resolution. The effect of the noise and smoothing on the derived cell boundaries is investigated and quantified using simulated data. We show in detail the evolution of supergranular cells over their lifetime, including observations of emerging, splitting, and coalescing cells. A key result of our analysis of cell internal velocities is that there is a simple linear relation between cell size and cell internal velocity, rather than the power law usually suggested