Searching for Extra Dimensions with Gravitational Waves: Dark-Siren Constraints from GWTC-4

Using a hierarchical Bayesian analysis of 141 compact binary coalescences from GWTC-4 combined with galaxy data, this study constrains higher-dimensional gravitational wave propagation to a spacetime dimension of $D = 4.38^{+1.91}_{-1.01}$, finding results consistent with four-dimensional General Relativity while noting that the crossover scale remains poorly constrained.

The Gist

Imagine the universe as a giant, multi-story building. For over a century, physicists have been convinced that this building only has four dimensions: three of space (up/down, left/right, forward/backward) and one of time. This is the foundation of Einstein's General Relativity.

However, some theories suggest there might be hidden "basements" or "attics"—extra spatial dimensions we can't see. The big question is: Are there extra floors?

This paper is like a detective investigation trying to find out if gravity "leaks" into these hidden floors. Here is how the authors did it, explained simply:

The Detective's Tool: "Dark Sirens"

Usually, to measure how far away something is in space, astronomers look at a "standard candle"—a light source with a known brightness, like a specific type of star. If you know how bright it should be and see how dim it actually is, you can calculate the distance.

But many gravitational wave events (collisions of black holes or neutron stars) don't have a visible light partner. They are "dark." To find their distance, the authors used a clever trick called the "Dark Siren" method.

• The Analogy: Imagine hearing a siren in the distance but not seeing the ambulance. You can't see the ambulance, but you know the general neighborhood it's in. By looking at a map of all the galaxies in that direction (the "neighborhood"), the scientists can guess which galaxy the sound came from. Once they know the galaxy, they know its distance.
• The Data: They used a catalog of 141 cosmic collisions (the "sirens") detected by LIGO, Virgo, and KAGRA, and matched them against a massive map of 22 million galaxies.

The Test: Does Gravity Leak?

If extra dimensions exist, gravity might behave differently than light.

• Light is like a car stuck on a 4-lane highway (our 4D universe). It can only travel along those 4 lanes.
• Gravity (in these theories) is like a drone that can fly off the highway and into the 3D air above it (the extra dimensions).

If gravity leaks into these extra dimensions, the signal should get weaker (dampened) faster over long distances than light does. It's like if you shouted in a hallway (4D) versus shouting in a giant open field (higher dimensions); the sound would fade much quicker in the open field.

The Investigation

The authors took the 141 "dark siren" events and asked: "Do the gravitational waves fade away exactly as Einstein predicted for a 4D universe, or do they fade faster, suggesting extra dimensions?"

They used a super-computer method (Bayesian analysis) to compare the observed data against two possibilities:

1. The Standard Model: Gravity stays in 4 dimensions.
2. The Extra Dimension Model: Gravity leaks, and the universe has $D$ dimensions (where $D$ could be 5, 6, or more).

The Results: The Verdict

Here is what they found:

1. The Universe Looks 4D: When they ran the numbers, the data strongly suggested that the universe behaves exactly as if it has 4 dimensions. The calculated number of dimensions was roughly 4.38, but with a wide margin of error that comfortably includes the number 4.

   • Simple takeaway: There is no evidence of gravity leaking into extra dimensions. Einstein's 4D model still holds up.

2. The "Crossover" Scale Mystery: The theory involves a "crossover scale" ($R_c$). Think of this as a magic fence.

   • Inside the fence (close distances), gravity acts normal (4D).
   • Outside the fence (very far distances), gravity might start leaking into extra dimensions.
   • The data couldn't pin down where this fence is. The results kept piling up at the very edge of the range the scientists were willing to test. This means we simply don't have enough data yet to say how far away this "fence" is, or if it even exists.

3. The "Fence" Matters: The authors discovered that their answer depended heavily on where they decided to draw the line for the "fence" in their math.

   • If they assumed the fence was relatively close (within the range of the events they observed), they got a tighter answer pointing to 4 dimensions.
   • If they assumed the fence was incredibly far away (much further than any event they saw), the data became too weak to tell the difference between 4D and 5D.
   • Analogy: It's like trying to hear a whisper. If you assume the whisperer is right next to you, you can tell exactly what they said. If you assume they might be on the other side of the planet, you can't be sure if you heard a whisper or just the wind.

The Bottom Line

This paper is the first to use the latest batch of gravitational wave data (GWTC-4) to test for extra dimensions using this "Dark Siren" method.

The conclusion is clear: Based on current observations, gravity behaves exactly as if the universe has only four dimensions. There is no sign of it leaking into hidden extra spaces. However, the scientists admit that their ability to find these extra dimensions is currently limited by how far away the "fence" (the crossover scale) might be. As we detect more distant events and map more galaxies, we will be able to test this theory with even greater precision.

Technical Summary

Technical Summary: Searching for Extra Dimensions with Gravitational Waves: Dark-Siren Constraints from GWTC-4

Problem Statement
Higher-dimensional theories of gravity, such as braneworld scenarios and the Dvali-Gabadadze-Porrati (DGP) model, predict that gravitational waves (GWs) may propagate into extra spatial dimensions. This propagation leads to an anomalous damping of GW amplitude over cosmological distances compared to the predictions of four-dimensional General Relativity (GR). While previous studies have constrained these theories using individual "bright sirens" (e.g., GW170817) or limited subsets of binary black hole (BBH) mergers, this work addresses the need for a comprehensive constraint using the full population of compact binary coalescences (CBCs) from the Gravitational-Wave Transient Catalog 4.0 (GWTC-4). The primary challenge lies in distinguishing deviations in GW luminosity distance ($d_L^{GW}$) caused by extra dimensions from standard cosmological expansion and population evolution effects.

Methodology
The authors employ a hierarchical Bayesian framework using the "dark siren" method, which associates GW luminosity distances with galaxy redshifts without relying on electromagnetic counterparts for the GW events themselves.

1. Phenomenological Model: The study adopts a parameterization for the modified GW luminosity distance derived from DGP-like theories. The deviation from GR is characterized by the spacetime dimension number $D$ and a crossover scale $R_c$ (the scale at which gravity transitions from 4D to higher-dimensional behavior). The relationship is given by:
   $$d_L^{GW} = d_L^{EM} \left[ 1 + \left( \frac{d_L^{EM}}{R_c(1+z)} \right)^n \right]^{\frac{D-4}{2n}}$$
   where $d_L^{EM}$ is the electromagnetic luminosity distance calculated under the standard $\Lambda$CDM cosmology, and $n$ is a phenomenological parameter fixed to 1.

2. Data and Inference:

   • GW Events: The analysis utilizes 141 CBC events from GWTC-4 (137 BBHs and 4 events containing at least one neutron star) with a false alarm rate $< 0.25 \text{ yr}^{-1}$.
   • Galaxy Catalog: Line-of-sight (LOS) redshift information is drawn from the GLADE+ catalog, which contains $\sim$22 million galaxies. The analysis constructs LOS redshift priors using a pixelized approach (HEALPix $N_{side}=128$), incorporating both in-catalog galaxies (weighted by luminosity) and out-of-catalog contributions.
   • Population Models: Two mass distribution models are tested: the "FullPop-4.0" model (describing both neutron stars and black holes) and the "MultiPeak" model (BBHs only). The merger rate evolution follows a Madau-Dickinson-inspired power-law model.
   • Computational Framework: The inference is performed using the gwcosmo package, accelerated via GPU-based vectorized operations and the nessai nested sampling algorithm with normalizing flows.

3. Parameter Priors: The analysis varies the Hubble constant ($H_0$), the spacetime dimension ($D$), and the crossover scale ($\log(R_c/\text{Mpc})$). Two scenarios for $H_0$ are tested: a narrow prior $U(65, 77)$ km s$^{-1}$ Mpc$^{-1}$ (motivated by electromagnetic measurements) and a wide prior $U(10, 120)$. The dimension $D$ is sampled over $U(3, 12)$, and $\log(R_c/\text{Mpc})$ is sampled with a lower bound of 2.7 ($R_c \approx 500$ Mpc) and varying upper bounds.

Key Contributions

• First GWTC-4 Dark-Siren Constraints: This work provides the first constraints on higher-dimensional GW propagation using the full GWTC-4 dataset combined with the GLADE+ galaxy catalog.
• Computational Efficiency: The implementation of GPU-accelerated hierarchical Bayesian inference significantly reduces computational time, enabling the analysis of 141 events with complex population priors.
• Prior Sensitivity Analysis: The paper rigorously demonstrates that constraints on the spacetime dimension $D$ are highly sensitive to the assumed prior range of the crossover scale $R_c$, a factor often overlooked in previous studies.

Results

• Spacetime Dimension ($D$):
  • Under the narrow $H_0$ prior ($65-77$) and a crossover scale prior of $\log(R_c/\text{Mpc}) \in [2.7, 4.0]$, the analysis yields $D = 4.38^{+1.91}_{-1.01}$ (68% credible interval).
  • This result is consistent with the four-dimensional prediction of General Relativity ($D=4$).
  • When the upper bound of the $R_c$ prior is increased to 5.0 or 6.0, the constraint on $D$ weakens significantly, with the posterior distribution broadening.
• Crossover Scale ($R_c$): The posterior distribution for $R_c$ accumulates near the upper boundary of the prior range in all cases. This indicates that current observations cannot constrain the crossover scale from above; the data are consistent with $R_c$ being larger than the distances probed by the current GW population.
• Degeneracies: Strong degeneracies are observed between $H_0$ and $D$, as well as between $D$ and the merger rate spectral index $\gamma$. The narrow $H_0$ prior effectively breaks the $H_0$-$D$ degeneracy, tightening the constraint on $D$.
• Model Dependence:
  • The "FullPop-4.0" model (including neutron stars) yields slightly tighter constraints on $D$ than the "MultiPeak" model (BBHs only), likely due to the inclusion of nearby events with more precise distance measurements.
  • The choice of galaxy weighting scheme (luminosity-weighted vs. uniform) in the GLADE+ catalog has a negligible impact on the results, as the catalog is currently incomplete at the high redshifts where most GW events are detected.

Significance and Claims
The paper claims to provide the first GWTC-4 dark-siren constraints on higher-dimensional GW propagation. The authors assert that current observations remain consistent with four-dimensional General Relativity, finding no evidence for gravitational leakage into extra dimensions within the tested parameter space.

A central finding is the intrinsic limitation of current datasets: the ability to constrain deviations from GR depends critically on the assumed crossover scale $R_c$. If $R_c$ is significantly larger than the characteristic distances of observed GW events, the modified gravity effects are suppressed, rendering the data insensitive to $D$. Consequently, the paper concludes that while dark sirens are a powerful probe, their current constraining power is limited by the lack of high-redshift host galaxy information and the unknown scale of potential extra dimensions. Future improvements are expected from deeper galaxy surveys (e.g., DES, DESI, Euclid) and next-generation GW detectors (e.g., Einstein Telescope, Cosmic Explorer), which will access larger source populations and higher redshifts.