A principal component analysis approach to the morphology of Planetary Nebulae

TL;DR

This paper uses Principal Component Analysis to analyze optical data of Planetary Nebulae, revealing correlations between morphological types and chemical abundances, and providing insights into their physical properties.

Contribution

It introduces a PCA-based method to distinguish morphological differences in PNe and links these to chemical and physical parameters, offering new insights into their structure.

Findings

PC2 correlates with PNe morphology and chemical abundances.

Bipolar PNe have higher nitrogen and electron density than elliptical and round PNe.

PC2 shows a minimum at 0.89, related to initial mass of 2.6 solar masses.

Abstract

Principal Component Analysis (PCA) is a well-known technique used to decorrelate a set of vectors. It has been applied to explore the star formation history of galaxies or to determine distances of mass-lossing stars. Here we apply PCA to the optical data of Planetary Nebulae (PNe) with the aim of extracting information about their morphological differences. Preliminary analysis of a sample of 55 PNe with known abundances and morphology shows that the second component (PC2), which results from a relation produced by the parameters log(N/O), initial and final mass of PNe, is depending on the morphology of PNe. It has been found that when log(N/O) < -0.18 the PNe's nitrogen is low independently on the oxygen abundance for either Bipolar (B), Elliptical (E) or Round (R) PNe. An interesting result is that both E and R PNe have log(N/O) < 0 while only B PNe show negative and positive values. Consequently, B PNe are expected to have higher nitrogen values than the E and R PNe. Following that and a second sample of 35 PNe, n_{e} is also found to be higher in B PNe. Also, in all PNe morphologies PC2 appears to have a minimum at 0.89 and PNe's initial mass at 2.6 solar masses. 5-D diagrams between PCAs components and physical parameters are also presented. More results will follow while simple models will be applied in order to try to give a physical meaning to the components.