Application of the Trend Filtering Algorithm on the MACHO Database

TL;DR

This paper demonstrates the successful application of the Trend Filtering Algorithm (TFA) to the MACHO database, significantly improving the detection of variable stars by removing systematics and noise from photometric data.

Contribution

The study adapts TFA for variable star detection in the MACHO database, revealing many variables previously undetected due to systematics and noise.

Findings

TFA reduces trend-dominated detections from 60% to nearly zero.

387 variables identified, 183 of which are new detections.

Improved variable star detection in crowded fields using TFA.

Abstract

Due to the strong effect of systematics/trends in variable star observations, we employ the Trend Filtering Algorithm (TFA) on a subset of the MACHO database and search for variable stars. TFA has been applied successfully in planetary transit searches, where weak, short-lasting periodic dimmings are sought in the presence of noise and various systematics (due to, e.g., imperfect flat fielding, crowding, etc). These latter effects introduce colored noise in the photometric time series that can lead to a complete miss of the signal. By using a large number of available photometric time series of a given field, TFA utilizes the fact that the same types of systematics appear in several/many time series of the same field. As a result, we fit each target time series by a (least-square-sense) optimum linear combination of templates and frequency-analyze the residuals. Once a signal is found, we reconstruct the signal by employing the full model, including the signal, systematics and noise. We apply TFA on the brightest ~5300 objects from subsets of each of the MACHO Large Magellanic Cloud fields #1 and #79. We find that the Fourier frequency analysis performed on the original data detect some 60% of the objects as trend-dominated. This figure decreases essentially to zero after using TFA. Altogether, We detect 387 variables in the two fields, 183 of which would have remained undetected without using TFA. Where possible, we give preliminary classification of the variables found.