What happens to parameter bias when multiple gravitational-wave signals arrive so close together that detectors can't fully resolve them?

# Parameter Bias in Overlapping Gravitational-Wave Signals

When multiple gravitational-wave signals arrive too close together for detectors to fully separate them, parameter estimation becomes systematically biased — meaning measurements of mass, spin, distance, and other properties systematically veer away from their true values [AG-2025.09-192].

How the Bias Arises

The fundamental problem is that detectors cannot cleanly isolate each signal. When you fit a waveform model to data containing two overlapping signals, the model "sees" a hybrid signal that doesn't match any single source [AG-2025.07-283]. The fitting algorithm then attributes features from the second signal to the first — for instance, miscalculating mass or mislocating the source on the sky. This is not random noise; it's a systematic distortion that pushes estimates in specific directions depending on the relative timing, strength, and orientation of the two events.

Which Parameters Suffer Most

The bias is not uniform across all parameters. Studies find that [AG-2025.09-192]:

• Chirp mass and symmetric mass ratio (which determine the intrinsic properties of the binary) are significantly biased
• Luminosity distance (how far away the event is) is biased
• Coalescence time (when the merger occurs) shows parameter-dependent shifts
• Sky position and source orientation suffer biases that depend strongly on detector network geometry and the source's location in the sky

Remarkably, the magnitude of bias depends not just on how close the signals are in time, but also on the source's position relative to the detector network and on the phase and orientation of each signal [AG-2025.09-192].

The Time-Frequency Window Matters

A useful heuristic is the time-frequency overlap method: two signals cause significant bias only if they overlap in both time *and* frequency [AG-2025.07-283]. If they are separated in frequency, or if the time gap is large enough that the waveforms don't interfere, bias can be much smaller. However, this is only accurate about 86% of the time in practical scenarios [AG-2025.07-283].

Workarounds for Next-Generation Detectors

Third-generation detectors like the Einstein Telescope and Cosmic Explorer will see roughly 10⁵ signals per year, with many overlapping [AG-2025.09-192]. The main mitigation strategy is joint parameter estimation — fitting all overlapping signals simultaneously rather than one at a time [AG-2025.07-283]. This is computationally expensive but can recover unbiased estimates if done correctly, including effects like precession and higher-order modes in the waveform.

A complementary finding: glitches (transient detector noise) that overlap with signals also bias parameters, but only if they are within the model's time-frequency box and have significant signal-to-noise ratio [AG-2026.04-211, AG-2025.06-690]. Being mindful of when glitch mitigation is truly necessary helps manage computational burden.

A Hidden Astrophysical Risk

Overlapping signals can mimic gravitational lensing — where light or gravity from an intervening mass bends the waves — potentially creating false discoveries of exotic effects [AG-2025.10-403]. This underscores why understanding and correcting overlap bias is essential for astrophysical interpretation.