⚛️ general relativity

Wave-optics imprints of dark matter subhalos on strongly lensed gravitational waves. II. Saddle images and detectability

This paper extends the analysis of wave-optics imprints from dark matter subhalos on strongly lensed gravitational waves to saddle-point images, demonstrating that LISA can detect these frequency-dependent distortions with high significance to probe substructure masses between $10^4$ and $10^7\,M_\odot$ that are inaccessible to electromagnetic observations.

Original authors: Shin'ichiro Ando

Published 2026-06-23

📖 4 min read🧠 Deep dive

CC BY 4.0

Original authors: Shin'ichiro Ando

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine the universe is filled with invisible "ghosts" called dark matter. While we know big clumps of these ghosts form galaxies, physicists suspect there are also tiny, invisible clumps (subhalos) floating around, too small to hold any stars. Because they have no light, we can't see them with telescopes.

This paper proposes a new way to "see" these invisible ghosts using gravitational waves—ripples in space-time caused by crashing black holes.

Here is the story of the paper, broken down into simple concepts:

1. The Cosmic Lens and the "Ghost" Shadows

Imagine a massive galaxy sitting between us and a crashing pair of black holes. This galaxy acts like a giant cosmic magnifying glass (a lens). Usually, this lens creates two images of the same crash:

The "Minimum" Image: A bright, clear picture.
The "Saddle" Image: A slightly distorted, often fainter picture.

Now, imagine a tiny, invisible dark matter ghost (a subhalo) is floating right in front of one of these images. Because the gravitational waves are so fast and rhythmic, they don't just get magnified; they get scrambled. The ghost leaves a tiny, frequency-dependent "fingerprint" or distortion on the wave, like a subtle ripple on a pond caused by a pebble.

2. The New Challenge: The "Saddle" Image

In a previous paper, the authors figured out how to read these fingerprints on the bright "Minimum" image. But in this paper, they tackled the harder job: the "Saddle" image.

Think of the "Minimum" image like a bowl. If you roll a marble (the wave) into it, it settles at the bottom. The math is stable and easy to calculate.
The "Saddle" image is like a mountain pass (the shape of a horse saddle). If you roll a marble there, it can roll off in two different directions. The math for this shape is notoriously tricky because the "ripples" don't close up neatly; they stretch out infinitely.

The authors had to invent a special time-traveling math trick (a time-domain method) to solve this. They had to subtract the huge, messy background noise of the mountain pass to find the tiny, subtle signal left by the dark matter ghost. It's like trying to hear a whisper in a hurricane by perfectly canceling out the sound of the wind.

3. The Big Discovery: The "Parity Split"

When they finally solved the math, they found something fascinating about how the dark matter ghosts affect the two images differently:

The Fluctuations: The "jitter" or random shaking caused by the ghosts is about the same size for both images.
The Average Effect: However, the average effect is opposite.
- The Minimum image gets slightly brighter (magnified) on average.
- The Saddle image gets slightly dimmer (demagnified) on average.

It's as if the dark matter ghosts have a preference: they tend to boost the "good" image and dim the "bad" one. This confirms that the "Saddle" image is indeed the most fragile and sensitive to these invisible ghosts.

4. Can We Actually Detect This?

The authors ran a massive simulation (a "Monte Carlo" ensemble) with 1,000 different random arrangements of dark matter ghosts to see if the LISA space telescope (a future mission designed to listen to gravitational waves) could hear this.

The Setup: They imagined a loud crash of black holes (about a million times the mass of our Sun) happening behind a galaxy.
The Result: If the crash happens very close to the edge of the galaxy's "magnifying glass" (the caustic), the combined signal from both images is strong enough to be detected with high confidence (over 99% certainty) in about 62% of the cases.
The Catch: This only works if the source is very close to the edge of the lens. If it's further away, the signal fades too quickly to be heard.

5. The Bottom Line

This paper proves that strongly lensed gravitational waves are a powerful new tool. They can detect invisible dark matter clumps that are too small for any telescope to see.

By listening to the "whispers" left on both the bright and the saddle images, and using their new math to handle the tricky saddle shape, scientists could potentially find 10 to 20 of these dark matter ghosts during the lifetime of the LISA mission. It's a way to weigh the invisible universe by listening to the echoes of colliding black holes.

Technical Summary: Wave-optics imprints of dark matter subhalos on strongly lensed gravitational waves. II. Saddle images and detectability

Problem Statement
The abundance and density structure of low-mass dark matter subhalos ( $10^4$ – $10^7 M_\odot$ ) serve as a primary discriminant between Cold Dark Matter (CDM) and alternative models. However, the absence of baryons in these subhalos renders them invisible to electromagnetic observations. While gravitational waves (GWs) offer a wave-optics (WO) probe of substructure via frequency-dependent distortions in the amplification factor, previous work (Paper I) was limited to the "minimum" image parity in strongly lensed systems. This paper addresses two open questions: (1) How do subhalos imprint on the "saddle" image parity, which is known to be fragile in geometric optics (GO) flux-ratio anomalies? (2) Is the combined signal from both image parities detectable by the Laser Interferometer Space Antenna (LISA) when marginalizing over unknown source and lens parameters?

Methodology
The authors extend the WO formalism to the saddle-point image for a galaxy-scale lens (NFW halo + singular isothermal sphere) at $z_L=0.5$ lensing a massive black hole binary (MBHB) at $z_S=1.5$ .

Saddle-Point Calculation: Evaluating the amplification factor at a saddle is numerically delicate because the equal-arrival-time contours are open (hyperbolic) rather than closed, and the subhalo signal is a small residual atop a divergent background. The authors develop a dedicated time-domain method:
- They separate the signal into an analytic macro GO template (the bare quadratic saddle) and a numerically transformed WO subhalo residual.
- The residual is computed using a Filon quadrature for highly oscillatory integrals on a non-uniform grid clustered near the saddle delay.
- An analytic far-field tail is added to handle the non-cancelling logarithmic potential of the truncated NFW subhalos at large delays.
- This method is validated against an independent contour-free area-method arbiter.
Monte Carlo Ensemble: The authors generate 1,000 realizations of CDM subhalo populations (masses $10^2$ – $10^9 M_\odot$ ) around both minimum and saddle images. They compute the per-image amplification factors $F(w)$ and the complex modulation $\eta(w) = F(w)/F_{GO}(w) - 1$ .
Detectability Framework: Since LISA observes the total lensed waveform (the coherent sum of minimum and saddle), the authors construct a detection statistic that:
- Demodulates the macro-image interference (separating the two images in the time domain or cepstral domain) to recover per-image transfer functions.
- Projects out degenerate parameters (amplitude, Morse phase, arrival time) using a matched-filter inner product. This isolates the non-absorbable, frequency-dependent structure of the subhalo signal.
- Combines the significance of the minimum and saddle images in quadrature.

Key Results

Parity Asymmetry in Mean Demagnification: While the fluctuation amplitude of the subhalo imprint is parity-blind (affecting both images at a similar percent level), the mean modulation splits by parity. The minimum image is net magnified ( $dc \approx +1.3\%$ ), while the saddle image is net demagnified ( $dc \approx -0.8\%$ ). This systematic split correlates with the local tidal shear with opposite signs for the two parities, confirming that the geometric-optics "saddle fragility" persists in the wave-optics regime as a mean (de)magnification effect.
Modulation Characteristics: Subhalos induce percent-level amplitude modulations and $\sim 10^{-2}$ rad phase shifts across the LISA band ( $10^{-4}$ – $10^{-1}$ Hz). The modulation is dominated by subhalos in the $10^4$ – $10^7 M_\odot$ range.
Detection Significance: For a fiducial MBHB source of total mass $\sim 10^6 M_\odot$ located near the lens caustic ( $y_{src} \lesssim 0.1$ ), the combined detection significance exceeds $5\sigma$ in 62% of the Monte Carlo realizations. The minimum image contributes more to the significance than the saddle, but including the saddle raises the combined detection rate significantly (from 40% for the minimum alone to 62% combined for the $10^6 M_\odot$ case).
Dependence on Source Position: The detection significance scales steeply with the source impact parameter ( $\rho_{sub} \propto y_{src}^{-1.6}$ ). Robust per-event detection requires the source to lie very close to the caustic ( $y_{src} \lesssim 0.1$ ).

Significance and Claims
The paper claims that strongly lensed gravitational waves provide a sensitive, complementary probe of dark matter substructure at scales ( $10^4$ – $10^7 M_\odot$ ) inaccessible to electromagnetic observations.

Complementarity: Unlike flux-ratio anomalies in quasars which rely on the brightness of the saddle, this method detects the frequency-dependent distortion of the entire waveform, allowing the saddle's fragility to be measured even when the signal is not anomalously faint in a static sense.
Population Forecasts: Folding the per-event significances with optimistic strong-lensing rate forecasts suggests 10–20 substructure detections over the LISA mission. However, the authors note that conservative source and lens populations could reduce this yield to well below one event.
Diagnostic Power: The signal is enhanced for denser substructures (e.g., primordial black holes or core-collapsed self-interacting dark matter), offering a route to constrain not just the abundance but also the internal density of dark substructure.

The authors conclude that while the absolute number of detections depends heavily on uncertain astrophysical rates, the per-event sensitivity is robust: even a single loud, strongly lensed event near a critical curve can serve as a powerful diagnostic of subgalactic dark matter structure.

1. The Cosmic Lens and the "Ghost" Shadows

2. The New Challenge: The "Saddle" Image

3. The Big Discovery: The "Parity Split"

4. Can We Actually Detect This?

5. The Bottom Line

More like this