Solar horizontal flow evaluation using neural network and numerical simulation with snapshot data

TL;DR

This paper presents a neural network-based method to evaluate the horizontal velocity in the solar photosphere using snapshot data, improving upon traditional techniques like LCT by requiring only a single intensity image and achieving high correlation with simulated velocities.

Contribution

The authors develop a neural network approach that estimates horizontal solar velocities from a single intensity snapshot, outperforming traditional methods and applicable to observed data.

Findings

Achieved a correlation coefficient of 0.83 using only intensity data.

Improved correlation to 0.90 when including LoS velocity as input.

Successfully applied the method to observed solar data.

Abstract

We suggest a method that evaluates the horizontal velocity in the solar photosphere with easily observable values using a combination of neural network and radiative magnetohydrodynamics simulations. All three-component velocities of thermal convection on the solar surface have important roles in generating waves in the upper atmosphere. However, the velocity perpendicular to the line of sight (LoS) is difficult to observe. To deal with this problem, the local correlation tracking (LCT) method, which employs the difference between two images, has been widely used, but LCT has several disadvantages. We develop a method that evaluates the horizontal velocity from a snapshot of the intensity and the LoS velocity with a neural network. We use data from numerical simulations for training the neural network. While two consecutive intensity images are required for LCT, our network needs just one intensity image at only a specific moment for input. From these input array, our network outputs a same-size array of two-component velocity field. With only the intensity data, the network achieves a high correlation coefficient between the simulated and evaluated velocities of 0.83. In addition, the network performance can be improved when we add LoS velocity for input, enabling achieving a correlation coefficient of 0.90. Our method is also applied to observed data.