Is the Binary Black Hole Population Inference from Gravitational-Wave Data Robust?

TL;DR

Waveform modeling uncertainties can significantly distort inferred features in the binary black hole mass distribution from gravitational-wave data, potentially overshadowing statistical uncertainties and affecting astrophysical conclusions.

Contribution

This work demonstrates the impact of waveform systematics on BBH population inference and discusses methods to mitigate these effects for more robust astrophysical insights.

Findings

Waveform uncertainties can distort mass distribution features more than statistical errors.

Systematic errors can affect the inferred slope of the power-law distribution.

Mitigation strategies are discussed to improve future population analyses.

Abstract

Gravitational-wave observations are playing an instrumental role in understanding the population of binary compact objects in the Universe. These observations have begun to hint at the mass distribution of binary black holes (BBHs), with tentative evidence for features in the mass distribution beyond a simple power-law. Such features, hence, can be connected with different formation scenarios of BBHs and lead to important astrophysical conclusions. However, it is crucial to understand whether these features are truly astrophysical or connected with any unknown systematics. We show in this work that waveform modelling uncertainties can significantly distort inferred features in the BBH mass distribution, which can be more pronounced than the statistical uncertainty, even with the current generation detectors, which can peak close to the lower edge of the pair instability supernovae (PISN) mass gap, and also can impact the slope of the power-law distribution. So, in order to have a confirmed detection of astrophysical features in the BBH mass distribution and connecting them with BBH formation channels, it is important to consider waveform systematics in the astrophysical population analysis. We show the typical scaling of the systematic error and discuss a few avenues to mitigate this effect for robust measurements in the future.