SPORK That Spectrum: Increasing Detection Significances from High-Resolution Exoplanet Spectroscopy with Novel Smoothing Algorithms

TL;DR

This paper introduces SPORK, a novel smoothing algorithm that significantly enhances detection significance in high-resolution exoplanet spectroscopy by minimizing systematic noise, demonstrated on real data from HD 209458 b and HD 179949 b.

Contribution

The paper presents SPORK, a new normalization and smoothing technique that improves detection significance in exoplanet atmospheric studies beyond existing methods.

Findings

Detection significance increased from 5.78 to 9.71 sigma for HD 209458 b.

Detection significance increased from 4.19 to 5.90 sigma for HD 179949 b.

Applicable to various spectroscopic data sets, including archival and simulated data.

Abstract

Spectroscopic studies of planets outside of our own solar system provide some of the most crucial information about their formation, evolution, and atmospheric properties. In ground-based spectroscopy, the process of extracting the planet's signal from the stellar and telluric signal has proven to be the most difficult barrier to accurate atmospheric information. However, with novel normalization and smoothing methods, this barrier can be minimized and the detection significance dramatically increased over existing methods. In this paper, we take two examples of CRIRES emission spectroscopy taken of HD 209458 b and HD 179949 b and apply SPORK (SPectral cOntinuum Refinement for telluriKs) and iterative smoothing to boost the detection significance from 5.78 to 9.71 sigma and from 4.19 sigma to 5.90 sigma, respectively. These methods, which largely address systematic quirks introduced by imperfect detectors or reduction pipelines, can be employed in a wide variety of scenarios, from archival data sets to simulations of future spectrographs.