The Carnegie Supernova Project-I. Spectroscopic analysis of stripped-envelope supernovae

TL;DR

This study analyzes 170 spectra of 35 stripped-envelope supernovae, identifying spectral features, sub-types, and correlations, and demonstrates a PCA-based method for robust supernova classification during the photospheric phase.

Contribution

It provides detailed spectral analysis, identifies sub-types and spectral feature origins, and introduces a PCA-based classification approach for SE supernovae.

Findings

Identification of silicon contribution in SNe IIb spectral features

Detection of hydrogen signatures in some SNe Ib

Successful PCA-based separation of SE SN sub-types with ~80% accuracy

Abstract

An analysis leveraging 170 optical spectra of 35 stripped-envelope (SE) core-collapse supernovae observed by the Carnegie Supernova Project-I and published in a companion paper is presented. Mean template spectra are constructed for the SNe IIb, Ib and Ic sub-types and parent ions associated with designated spectral features are identified with the aid of the spectral synthesis code SYNAPPS. Our modeled mean spectra suggest the ~6150~\AA\ feature in SNe~IIb may have an underlying contribution due to silicon, while the same feature in some SNe Ib may have an underlying contribution due to hydrogen. Standard spectral line diagnostics consisting of pseudo-equivalent widths (pEW) and blue-shifted Doppler velocity are measured for each of the spectral features. Correlation matrices and rolling mean values of both spectral diagnostics are constructed. A Principle Component Analysis (PCA) is applied to various wavelength ranges of the entire data set and suggests clear separation among the different SE SN sub-types, which follows from trends previously identified in the literature. In addition, our finds reveal the presence of two SNe IIb sub-types, a handful of SNe Ib displaying signatures of weak, high-velocity hydrogen, and a single SN~Ic with evidence of weak helium features. Our PCA results can be leveraged to obtain robust sub-typing of SE SN based on a single spectrum taken during the so-called photospheric phase, separating SNe IIb from SNe Ib with ~80 percent completion.