Danger Zone: Establishing Buffers for Enhanced Classification in BPT Diagrams

TL;DR

This paper applies unsupervised machine learning to classify optical spectra from galaxies, identifying ambiguous regions and establishing buffer zones to improve classification accuracy in diagnostic diagrams.

Contribution

It introduces a novel application of UMAP to classify galaxy spectra and delineate boundary zones where traditional methods are ambiguous.

Findings

UMAP effectively classifies spectra with ambiguous classifications.

Boundary zones are identified where spectra frequently change classes.

Physically interesting subsets are found within ambiguous spectra.

Abstract

This study utilizes unsupervised machine learning, specifically the uniform manifold approximation and projection (UMAP) algorithm, to classify optical spectra originating from star-forming regions, Seyferts, and low-ionization (nuclear) emission-line regions (LI(N)ERs) based on their line ratios. Typically, the ionization source of a region is determined from intensity ratio of different combinations of pairs of spectral lines. However, using current boundary definitions, $\sim10$\% of spectra change classes between diagnostic diagrams. We apply the machine learning technique to $\sim$1.3 million optical spectra from 6,439 galaxies observed in the MaNGA survey. By training UMAP on consistently classified data, we can classify these ``ambiguous'' spectra, and delineate boundary zones where such ambiguities arise. Furthermore, we identify physically interesting subsets within the ambiguous spectra. Future work will incorporate additional parameters, such as alternative emission line ratios and velocity dispersions, to enhance classification accuracy.