SDSS-IV MaNGA: Unveiling Galaxy Interaction by Merger Stages with Machine Learning

TL;DR

This study employs machine learning to classify galaxy merger stages using SDSS-IV MaNGA data, revealing physical processes in galaxy interactions with high accuracy and identifying key features from images and velocity maps.

Contribution

The paper introduces a machine learning approach to classify galaxy merger stages with high precision, utilizing imaging and velocity data, applicable to large surveys.

Findings

High precision (over 90%) in binary merger classification.

Effective multi-stage classification with over 85% precision.

Images and velocity maps are key features for identifying interaction stages.

Abstract

We use machine learning techniques to classify galaxy merger stages, which can unveil physical processes that drive the star formation and active galactic nucleus (AGN) activities during galaxy interaction. The sample contains 4,690 galaxies from the integral field spectroscopy survey SDSS-IV MaNGA, and can be separated to 1,060 merging galaxies and 3630 non-merging or unclassified galaxies. For the merger sample, there are 468, 125, 293, and 174 galaxies in (1) incoming pair phase, (2) first pericentric passage phase, (3) aproaching or just passing the apocenter, and (4) final coalescence phase or post-mergers. With the information of projected separation, line-of-sight velocity difference, SDSS gri images, and MaNGA Ha velocity map, we are able to classify the mergers and their stages with good precision, which is the most important score to identify interacting galaxies. For the 2-phase classification (binary; non-merger and merger), the performance can be high (precision>0.90) with LGBMClassifier. We find that sample size can be increased by rotation, so the 5-phase classification (non-merger, 1, 2, 3, and 4 merger stages) can be also good (precision>0.85). The most important features come from SDSS gri images. The contribution from MaNGA Ha velocity map, projected separation, and line-of-sight velocity difference can further improve the performance by 0-20%. In other words, the image and the velocity information are sufficient to capture important features of galaxy interactions, and our results can apply to the entire MaNGA data as well as future all-sky surveys.