English

Unsupervised Spectral Unmixing For Telluric Correction Using A Neural Network Autoencoder

Instrumentation and Methods for Astrophysics 2021-11-18 v1 Machine Learning

Abstract

The absorption of light by molecules in the atmosphere of Earth is a complication for ground-based observations of astrophysical objects. Comprehensive information on various molecular species is required to correct for this so called telluric absorption. We present a neural network autoencoder approach for extracting a telluric transmission spectrum from a large set of high-precision observed solar spectra from the HARPS-N radial velocity spectrograph. We accomplish this by reducing the data into a compressed representation, which allows us to unveil the underlying solar spectrum and simultaneously uncover the different modes of variation in the observed spectra relating to the absorption of H2O\mathrm{H_2O} and O2\mathrm{O_2} in the atmosphere of Earth. We demonstrate how the extracted components can be used to remove H2O\mathrm{H_2O} and O2\mathrm{O_2} tellurics in a validation observation with similar accuracy and at less computational expense than a synthetic approach with molecfit.

Keywords

Cite

@article{arxiv.2111.09081,
  title  = {Unsupervised Spectral Unmixing For Telluric Correction Using A Neural Network Autoencoder},
  author = {Rune D. Kjærsgaard and Aaron Bello-Arufe and Alexander D. Rathcke and Lars A. Buchhave and Line K. H. Clemmensen},
  journal= {arXiv preprint arXiv:2111.09081},
  year   = {2021}
}

Comments

Presented at Workshop on Machine Learning and the Physical Sciences (NeurIPS 2021)

R2 v1 2026-06-24T07:42:03.676Z