Kepler Presearch Data Conditioning I - Architecture and Algorithms for Error Correction in Kepler Light Curves

Abstract: Kepler provides light curves of 156,000 stars with unprecedented precision. However, the raw data as they come from the spacecraft contain significant systematic and stochastic errors. These errors, which include discontinuities, systematic trends, and outliers, obscure the astrophysical signals in the light curves. To correct these errors is the task of the Presearch Data Conditioning (PDC) module of the Kepler data analysis pipeline. The original version of PDC in Kepler did not meet the extremely high performance requirements for the detection of miniscule planet transits or highly accurate analysis of stellar activity and rotation. One particular deficiency was that astrophysical features were often removed as a side-effect to removal of errors. In this paper we introduce the completely new and significantly improved version of PDC which was implemented in Kepler SOC 8.0. This new PDC version, which utilizes a Bayesian approach for removal of systematics, reliably corrects errors in the light curves while at the same time preserving planet transits and other astrophysically interesting signals. We describe the architecture and the algorithms of this new PDC module, show typical errors encountered in Kepler data, and illustrate the corrections using real light curve examples.

• The paper presents a Bayesian MAP method that refines error correction in Kepler light curves to preserve key astrophysical signals.
• It replaces the traditional least-squares approach with a strategy that leverages correlations within the data to manage systematic errors.
• The new architecture includes advanced algorithms for handling SPSDs and outliers, supported by a quantitative Goodness Metric for performance.

Analyzing Error Correction in Kepler Light Curves: Architectural and Algorithmic Advances

The paper "Kepler Presearch Data Conditioning I -- Architecture and Algorithms for Error Correction in Kepler Light Curves" by M. C. Stumpe et al. details significant advancements in the data analysis processes utilized within the Kepler Mission, a cornerstone project in the search for exoplanets through transit photometry. The primary focus is the enhancement of the Kepler Presearch Data Conditioning (PDC) module, specifically to address systematic and stochastic errors that obscure the astrophysical signals in the observed light curves.

The Kepler Mission and Data Challenges

The Kepler Mission's objective is the detection of Earth-size exoplanets in the habitable zones of Sun-like stars by monitoring the light curves of about 156,000 stars. This endeavor demands unprecedented photometric precision, necessitating stringent error correction mechanisms to accurately detect the faint signals of transiting exoplanets. The raw light curves produced by Kepler are marred by various systemic errors including discontinuities, systematic trends, and noise, which the PDC module is tasked with correcting.

Key Improvements in the PDC Module

The paper elaborates on a comprehensive rewrite of the PDC module, implemented in Kepler SOC 8.0, harnessing a Bayesian Maximum A Posteriori (MAP) approach to light curve cotrending. This represents a notable advancement over the former method which employed a least squares fitting approach using ancillary engineering data (PDC-LS). The Bayesian methodology prioritizes error removal while preserving significant astrophysical features, addressing the previous shortcomings where essential signals were inadvertently removed.

Specific Architectural and Algorithmic Changes

1. Bayesian MAP Approach: Unlike PDC-LS, which corrected errors via least squares-fitting against engineering data, PDC-MAP uses a Bayesian approach which draws upon correlations within the light curves themselves. This method is more adept at distinguishing and preserving signals such as stellar variability, which the least squares-based method often misidentified as errors.
2. Handling Systematic Variations: Systematic errors are corrected by fitting the light curves against basis vectors derived from correlated targets within the same channel. The Bayesian approach includes priors from correlated observations which constrain the fits and prevent overfitting.
3. Discontinuity and Outlier Management: The introduction of sophisticated algorithms for correcting Sudden Pixel Sensitivity Dropoffs (SPSD) and outliers ensures that such local errors do not disrupt the accurate detection of transits.
4. Performance Metrics and Future Improvements: The implementation includes a Goodness Metric to quantitatively assess the quality of error correction, focusing on systematic error removal, noise introduction, and preservation of stellar variability. Future enhancements may include adaptive selection of parameters and improved handling of systematic errors at multiple scales.

Implications and Speculation on Future Developments

The advancements delineated in this work have significant implications for both the practical detection of exoplanets and theoretical modeling of stellar activities. By more finely tuning error correction mechanisms, the Kepler science team can ensure the robustness of their results, fostering higher confidence in subsequent analyses of exoplanetary data. As observational astronomy continues to advance, it is plausible that similar Bayesian techniques will become standard in the processing pipelines of future missions, enhancing the precision and reliability of data analysis across diverse astronomical domains.

This paper stands as a testament to the continual refinement in data processing required to unlock the full potential of space-based observatories, underscoring the significance of sophisticated algorithmic interventions in the pursuit of distant worlds.