Inversions for Average Supergranular Flows Using Finite-frequency Kernels

TL;DR

This study uses finite-frequency kernels and time-distance inversion to analyze average supergranular flows, revealing a significant vertical velocity peak beneath the surface, consistent with recent hypotheses about deep flows.

Contribution

It introduces a detailed inversion method for supergranular flows using finite-frequency kernels and statistical averaging over thousands of supergranules.

Findings

Peak vertical velocity of 117±2 m/s at 1.2 Mm depth

Root-mean-square vertical velocity of 21 m/s

Large-amplitude vertical flows exist beneath supergranules' surface

Abstract

I analyse the maps recording the travel-time shifts caused by averaged plasma anomalies under an "average supergranule", constructed by means of statistical averaging over 5582 individual supergranules with large divergence signals detected in two months of HMI Dopplergrams. By utilising a three-dimensional validated time-distance inversion code, I measure the peak vertical velocity of 117+/-2 m/s in depths around 1.2 Mm in the centre of the supergranule and root-mean-square vertical velocity of 21 m/s over the area of the supergranule. A discrepancy between this measurement and the measured surface vertical velocity (a few m/s) can be explained by the existence of the large-amplitude vertical flow under the surface of supergranules with large divergence signals, recently suggested by Duvall & Hanasoge (2012).