Detectability of Gravitational-Wave Memory with LISA: A Bayesian Approach

This paper employs state-of-the-art LISA simulations and Bayesian analysis to evaluate the observatory's capability to detect and characterize the gravitational-wave displacement memory effect from individual massive black hole binary mergers, thereby establishing detection criteria, reconstruction precision, and expected event rates for testing General Relativity.

The Gist

Imagine the universe as a giant, invisible trampoline made of space and time. When massive objects, like two black holes, dance around each other and crash together, they create ripples on this trampoline. We call these ripples gravitational waves.

For years, scientists have been listening to the "chirp" of these waves—the sound of the black holes spiraling in and merging. But according to Einstein's theory of General Relativity, there is a second, stranger effect that happens when the black holes merge. It's called the gravitational wave memory effect.

The "Permanent Dent" Analogy

Think of the memory effect like a permanent dent in a car's fender after a crash.

• The Chirp (Oscillatory Wave): This is the shaking and rattling of the car during the crash. It vibrates back and forth, but eventually, the shaking stops, and the car settles.
• The Memory (Displacement): This is the dent itself. After the shaking stops, the metal doesn't return to its original flat shape. It stays slightly bent. In space, this means that after the gravitational waves pass, the distance between two points in space is permanently stretched or squeezed. It's a "scar" left on the universe.

The Mission: LISA

Currently, our detectors (like LIGO) are like ears tuned to hear high-pitched screams. They are great at hearing the "chirp" of smaller black holes, but the "dent" (memory) is a very low-frequency, slow-moving signal. It's too quiet and too slow for current ground-based detectors to hear clearly.

Enter LISA (Laser Interferometer Space Antenna). LISA is a future space-based detector, essentially a giant triangle of satellites floating in space. It is designed to listen to the deep, low-frequency rumble of massive black holes. The authors of this paper asked: "Can LISA actually hear this permanent dent?"

How They Tested It

The researchers didn't wait for LISA to launch. Instead, they built a virtual laboratory using supercomputers.

1. Creating the Sound: They simulated thousands of massive black hole collisions. They created two versions of the sound for each collision:
   • Version A: Just the normal "chirp" (no dent).
   • Version B: The "chirp" plus the permanent "dent" (memory).
2. Adding Static: They added "static noise" to simulate the background hiss of the universe and the instrument's own limitations, making it realistic.
3. The Detective Work (Bayesian Analysis): They used a statistical method called Bayesian analysis. Imagine you are a detective trying to solve a mystery. You have a suspect (the memory effect) and an alibi (no memory). You look at the evidence (the noisy data) and ask: "Is it more likely that the suspect was there, or that they weren't?"
   • They calculated a score called the Bayes Factor. If the score is high enough, it means the evidence strongly supports the idea that the "dent" is real.

The Findings

The paper presents three main discoveries, explained simply:

1. The "Volume" Threshold
The researchers found that to hear the memory effect, the "dent" needs to be loud enough. They calculated that the memory signal needs a specific volume level (called a Signal-to-Noise Ratio, or SNR) of about 3 to be detectable, and 5 to be detected with high confidence.

• Analogy: It's like trying to hear a whisper in a noisy room. If the whisper is too quiet, you can't tell if it's there. But if it's loud enough (above the threshold), you can be sure it's a whisper and not just random noise.

2. The "Helper" Effect
Sometimes, the "dent" helps us understand the crash better.

• Analogy: Imagine trying to guess how heavy a box is by shaking it. If the box is very light and the shaking is messy, it's hard to tell. But if the box leaves a permanent dent on the floor, that dent gives you extra clues about how heavy it was.
• The paper found that for smaller or quieter black hole collisions, including the memory effect in the math helps scientists figure out the black holes' properties (like their mass and spin) more accurately. For the loudest, biggest collisions, the "chirp" is already so clear that the "dent" doesn't add much new information.

3. The "Cosmic Lottery"
Finally, they looked at the "Cosmic Lottery." They simulated a universe full of black holes (using population models) to see how many times LISA might win the prize of detecting a memory effect.

• The Result: It depends on how the black holes formed.
  • If black holes form from "heavy seeds" (giant gas clouds collapsing early in the universe), LISA has a very good chance of detecting this memory effect in individual events.
  • If they form from "light seeds" (remnants of the first stars), it's harder, but there is still a chance, especially if we wait a long time (10 years) and listen to many events.

The Bottom Line

This paper is a "proof of concept" for the future. It tells us that:

• The "permanent dent" in space (memory effect) is real and calculable.
• LISA is the right tool to find it.
• We have a clear rule for when we can say, "Yes, we found it!" (When the signal is loud enough).
• Depending on how the universe built its black holes, we might be able to see this effect in our first few years of listening, opening a new window to test Einstein's theories in a way we never could before.

The authors did not claim this would cure diseases or change daily life; they simply mapped out the path to hearing a new, fundamental sound of the universe.

Technical Summary

Technical Summary: Detectability of Gravitational-Wave Memory with LISA: A Bayesian Approach

Problem Statement
Gravitational wave (GW) astronomy offers a new window into the universe, with future space-based observatories like the Laser Interferometer Space Antenna (LISA) expected to detect signals from massive black hole binaries (MBHBs) with high signal-to-noise ratios (SNR). These loud signals provide an opportunity to test fundamental physics effects predicted by General Relativity (GR), specifically the displacement memory effect. This effect corresponds to a permanent, non-oscillatory deformation of spacetime caused by the passage of gravitational radiation. While current ground-based detectors (LVK) and Pulsar Timing Arrays (PTAs) have not yet definitively observed this effect—largely due to frequency band limitations and the need for statistical stacking of many events—LISA's sensitivity in the 0.1 mHz to 1 Hz range suggests individual events might be detectable. The challenge lies in distinguishing the memory signal, which is a low-frequency, monotonic offset, from the dominant oscillatory GW signal and instrumental noise.

Methodology
The authors employ a rigorous Bayesian framework to assess the detectability of the memory effect using simulated LISA data. The methodology proceeds through several key stages:

1. Waveform Modeling and Memory Implementation:

   • Two waveform models are utilized: NRHybSur3dq8 CCE (a numerical relativity surrogate with Cauchy Characteristic Extraction that naturally includes memory) and SEOBNRv5HM (an effective-one-body model without inherent memory).
   • For the SEOBNRv5HM model and the dominant (2,2) mode of the CCE model, the memory contribution is computed separately using a simplified integral formula derived from the energy flux of the high-frequency oscillatory modes (specifically the (2,0) mode).
   • The waveforms are transformed into the detector frame, rotated by polarization angles, and processed through the LISA response and Time-Delay Interferometry (TDI) pipeline to generate mock A, E, and T channels.

2. Bayesian Analysis and Detectability Criteria:

   • The study uses dynamical nested sampling (Dynesty) to compute the Bayesian evidence for two competing models: one including the memory effect (o + mem) and one excluding it (o).
   • The Bayes Factor ($B_{mem}$) is calculated as the ratio of evidences. A detection is claimed if the log-Bayes factor ($\log_{10} B_{mem}$) exceeds a threshold of 2 (corresponding to "decisive" evidence on the Jeffreys scale).
   • To overcome the computational cost of full evidence calculations across the entire parameter space, the authors establish a correlation between the Bayes factor and the memory-specific SNR ($SNR_{mem}$). They utilize a "peaked likelihood" approximation to rapidly estimate $\log_{10} B$ for various noise realizations, characterizing the statistical dispersion caused by noise.

3. Parameter Estimation:

   • The study performs parameter estimation to determine if including memory in the waveform model improves the reconstruction of binary parameters (mass, spin, distance, inclination) and to assess the precision with which the memory amplitude can be measured.
   • A geometric parameter $\gamma$ is introduced to scale the memory amplitude ($h_{mem} = \gamma \times h_{GR}^{mem}$), allowing for tests of GR predictions.

4. Astrophysical Population Analysis:

   • The detectability criteria are applied to MBHB population catalogues from Barausse et al., covering various formation scenarios (light vs. heavy seeds, delay times, supernovae feedback).
   • Simulations of 4-year and 10-year LISA missions are conducted to estimate the number of events where the memory effect is detectable.

Key Contributions and Results

• Detection Threshold: The analysis establishes a quantitative link between the Bayes factor and the memory SNR. The authors derive a detection threshold of $SNR_{mem} \approx 3$ for a "favorable" detection, while an $SNR_{mem} \gtrsim 5$ is required to ensure confident detection (minimizing the risk of noise-induced false negatives) across different total masses and noise realizations.
• Noise Variability: The study characterizes the dispersion of the Bayes factor due to instrumental noise. It finds that while the mean Bayes factor depends primarily on $SNR_{mem}$, the variance is mass-dependent. Higher total masses shift the signal frequency closer to the low-frequency noise floor, increasing the uncertainty in detection.
• Parameter Estimation Impact:
  • For high-SNR events ($SNR_{tot} \gtrsim 500$), the inclusion of memory has a negligible impact on the reconstruction of standard binary parameters, as the oscillatory component already tightly constrains them.
  • For lower-SNR events ($SNR_{tot} \sim 100$), the memory effect can help lift degeneracies between parameters (e.g., inclination and luminosity distance), particularly for lighter, equal-mass binaries where the merger signal is less resolved.
  • The study highlights that neglecting the oscillatory component of the (2,0) mode (distinct from the memory component) can introduce biases, suggesting that higher-order modes are critical for accurate parameter estimation.
• Amplitude Reconstruction: The precision of the memory amplitude measurement ($\delta\gamma$) scales with $SNR_{mem}$ according to a power law ($\delta\gamma \propto SNR_{mem}^{-1.1}$). The authors provide estimates for the $SNR_{mem}$ required to achieve specific precision levels (e.g., 10% precision requires $SNR_{mem} \approx 18.8$).
• Astrophysical Rates: Applying these criteria to population models, the study finds that LISA is expected to detect the memory effect in individual events, particularly in "heavy-seed" formation scenarios. "Light-seed" scenarios show greater variability but still yield favorable events. The number of detectable events depends heavily on the assumed delay between galaxy mergers and black hole coalescence.

Significance and Claims
The paper claims to provide the first comprehensive Bayesian assessment of the memory effect's detectability for individual MBHB events with LISA. By moving beyond simple SNR thresholds to a full model-comparison framework, the authors establish robust criteria for claiming detection. The work demonstrates that LISA is capable of not only detecting the memory effect but also reconstructing its amplitude with sufficient precision to test GR and constrain alternative theories of gravity. The authors modestly note that their results rely on specific waveform assumptions (e.g., non-precessing binaries) and that future work should address precessing systems and more complex memory mode decompositions. The study concludes that while the memory effect is a subtle signal, the high SNR of MBHBs in the LISA band makes it a viable target for fundamental physics tests within the mission's planned lifetime.