Exploring Neutron Star Merger Spectral Features with Dimensionality Reduction

Nicole Ford ( Université McGill )

Kilonovae are likely a key site of heavy r-process element production in the universe, and their optical/infrared spectra contain insights into both the properties of the ejecta and the conditions of the r-process. However, event GW170817/AT2017gfo is the only kilonova so far with well-observed spectra. To understand the diversity of absorption features that might be observed in future kilonovae spectra, we use the TARDIS Monte Carlo radiative transfer code to simulate a suite of optical kilonova spectra that spans a wide range of ejecta properties, spectral features, and abundance patterns. To identify the strongest and most common absorption lines, we perform dimensionality reduction using an autoencoder (AE), and apply a Bayesian Gaussian Mixture Model to find clusters in the latent space. We show that future kilonovae will likely display strong absorption by Sr (strontium) II, Y (yttrium) I - II, and Zr (zirconium) I - II, with strong lanthanide contributions at low electron fractions. When a new candidate kilonova spectrum is observed, our results will enable a rapid assessment of the dominant absorption lines and key ejecta properties, providing context on where this event falls within the larger `zoo' of kilonova possibilities.