Data Analysis
transit-least-squares avatar

transit-least-squares

Transit Least Squares (TLS) algorithm for detecting exoplanet transits in light curve data, offering higher sensitivity than Lomb-Scargle for transit-shaped signals.

Introduction

The Transit Least Squares (TLS) skill is an advanced analytical tool designed for astronomers, data scientists, and researchers working with exoplanetary transit data. Unlike the standard Lomb-Scargle periodogram, which is optimized for sine-like variations, TLS specifically models the geometric shape of planetary transits, making it significantly more effective at detecting shallow dips in stellar brightness caused by orbiting planets. This skill provides a complete workflow for processing light curve data, from cleaning outliers and flattening trends to searching for periodic signals and masking detected candidates to identify multi-planet systems.

  • Performs non-sinusoidal transit model fitting to detect planetary signals with high precision.

  • Handles raw light curve data including time, flux, and critical flux uncertainties.

  • Integrates with the Lightkurve package for efficient data handling and preprocessing.

  • Includes robust features for period refinement, phase-folding, and visual model validation.

  • Facilitates iterative discovery of multiple planets via systematic transit masking and secondary signal searches.

  • Calculates standard astrophysical metrics such as Signal Detection Efficiency (SDE), Signal-to-Noise Ratio (SNR), transit depth, and epoch-based transit timings.

  • Ensure input data includes flux_err (flux uncertainties) to prevent inaccurate detection results.

  • Use a two-stage approach: conduct a broad global search first, followed by a narrow grid refinement around identified candidate periods to achieve higher precision.

  • Interpret SDE values greater than 6 as potential candidates and greater than 9 as strong detections; evaluate SNR scores against a threshold of 7.

  • Be aware of potential period aliasing in data with gaps, particularly if the algorithm suggests testing multiples of the discovered period.

  • Recommended preprocessing pipeline: remove sigma-clipping outliers, flatten the light curve to remove instrumental trends, then apply the TLS power search.

  • Ideal for processing high-cadence photometric observations, identifying T0 (transit epoch) and calculating planetary transit parameters for further characterization.

Repository Stats

Stars
1,086
Forks
271
Open Issues
39
Language
PDDL
Default Branch
main
Sync Status
Idle
Last Synced
Apr 30, 2026, 04:03 PM
View on GitHub