Exploring Coronal Heating Using Unsupervised Machine-Learning

TL;DR

This paper introduces an unsupervised machine learning technique to analyze impulsive radio emissions from the quiet Sun, aiming to understand coronal heating by characterizing their morphology in solar images.

Contribution

It presents a novel unsupervised learning method for morphological analysis of solar radio emissions, aiding in solving the coronal heating mystery.

Findings

Identified approximately 34,500 features as 2D elliptical Gaussians in solar radio images.

Analyzed over 8000 images spanning 70 minutes of solar data.

Demonstrated the technique's robustness in characterizing impulsive emissions.

Abstract

The perplexing mystery of what maintains the solar coronal temperature at about a million K, while the visible disc of the Sun is only at 5800 K, has been a long standing problem in solar physics. A recent study by Mondal(2020) has provided the first evidence for the presence of numerous ubiquitous impulsive emissions at low radio frequencies from the quiet sun regions, which could hold the key to solving this mystery. These features occur at rates of about five hundred events per minute, and their strength is only a few percent of the background steady emission. One of the next steps for exploring the feasibility of this resolution to the coronal heating problem is to understand the morphology of these emissions. To meet this objective we have developed a technique based on an unsupervised machine learning approach for characterising the morphology of these impulsive emissions. Here we present the results of application of this technique to over 8000 images spanning 70 minutes of data in which about 34,500 features could robustly be characterised as 2D elliptical Gaussians.