🔭 astrophysics

Identifying galactic binary systems of neutron stars and black holes with LISA

This paper demonstrates that while LISA will detect approximately 100 galactic binary systems containing neutron stars and black holes, it will only be able to accurately infer their component masses and distinguish them from double white dwarfs for roughly 10% to 50% of these systems by measuring orbital eccentricity and frequency derivatives, leaving the nature of the remaining binaries uncertain.

Original authors: Hannah Middleton, Panagiota Kolitsidou, Antoine Klein, Connor Adam, Rowan Chalmers, Alberto Vecchio

Published 2026-08-17

📖 7 min read🧠 Deep dive

CC BY 4.0

Original authors: Hannah Middleton, Panagiota Kolitsidou, Antoine Klein, Connor Adam, Rowan Chalmers, Alberto Vecchio

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 as a giant, cosmic orchestra. For decades, we've been listening to its music using telescopes that catch light, radio waves, and X-rays. But there's a whole other layer of sound we've only just started to hear: gravitational waves. Think of these not as sound waves traveling through air, but as ripples in the fabric of spacetime itself, created when massive objects dance or crash together. It's like feeling a vibration in the floor when a heavy truck drives by, rather than hearing the engine.

In this cosmic orchestra, there are two main types of "instruments" that make the loudest ripples: black holes and neutron stars. Black holes are like cosmic vacuum cleaners, so dense that not even light can escape them. Neutron stars are the ultra-dense, city-sized corpses of exploded stars, packed so tightly that a teaspoon of their material would weigh a billion tons. For a long time, we've mostly found these objects by looking at the light they emit or by spotting how they pull on their neighbors. But many of them are quiet, dark, or hidden behind thick clouds of dust, making them invisible to our traditional telescopes.

Enter LISA, the Laser Interferometer Space Antenna. You can think of LISA as a giant, floating ear in space, designed to listen to the specific "hum" of binary systems—pairs of these heavy objects orbiting each other. While ground-based detectors are great at hearing the loud "crash" when two black holes smash together, LISA is tuned to hear the steady, high-pitched "whine" of pairs that are still dancing in a wide circle, long before they collide. The big question is: if LISA hears a whine, can it tell us what is making the sound? Is it a pair of black holes, a pair of neutron stars, or perhaps a mix? Or is it just a pair of white dwarfs (the smaller, cooler cousins of neutron stars) that are far more common? This is the puzzle the paper tackles.

The Great Cosmic Mix-Up

Imagine you are at a crowded party, and you hear a hum coming from a specific corner. You know there are about 10,000 people in the room, and 99% of them are wearing white shirts (white dwarfs), while only a tiny handful are wearing black suits (black holes) or neon vests (neutron stars). If you hear a hum, how do you know if it's coming from a black suit or a white shirt? Usually, you'd look at the person to see what they're wearing. But in space, these objects are often too dim or too far away to see with a telescope.

This is exactly the challenge facing the Laser Interferometer Space Antenna (LISA). The mission is expected to detect about 100 pairs of black holes and neutron stars dancing around each other in our galaxy. However, it will also detect about 10,000 pairs of white dwarfs. The problem is that for many of these pairs, the "whine" they make sounds almost identical. They are so far from crashing together that their orbit doesn't change much during the mission, making it hard to tell them apart just by listening.

The Detective Work in the Data

The authors of this paper decided to play detective using a computer simulation. They didn't just guess; they built a fake universe inside their computers. They used a realistic model of how stars are born, live, and die to create a "population" of 300 binary systems: 100 pairs of black holes, 100 pairs of a black hole and a neutron star, and 100 pairs of neutron stars. They then simulated what LISA would "hear" from these systems.

Their goal was to see if LISA could measure the specific "fingerprints" left on the sound wave that would reveal the mass of the objects. To do this, they looked for three specific clues hidden in the data:

How fast the pitch changes: Even a slow change in the frequency of the sound (the "chirp") tells us about the combined mass of the pair.
The wobble of the orbit: If the orbit isn't a perfect circle, the point of closest approach (the periapse) will slowly rotate. The speed of this rotation tells us the total mass.
The shape of the orbit: The amount of "squishiness" or eccentricity in the orbit affects how the sound evolves.

By measuring these three things together, the team could mathematically work backward to figure out the individual masses of the two dancing objects. If they knew the masses, they could say, "Aha! This object is 1.4 times the mass of our Sun, so it must be a neutron star," or "This one is 10 times the mass of the Sun, so it's definitely a black hole."

What They Found

The results were a mix of good news and "it depends." The simulation showed that LISA will be a fantastic detective for about half of the neutron star pairs. For roughly 50% of the binary neutron star systems, LISA will be able to measure their masses accurately enough to confirm they are indeed neutron stars.

However, the job is harder for the black holes. For binary black holes and mixed pairs (one black hole, one neutron star), LISA will only be able to identify the masses for about 10% of the systems. Why the difference? It turns out that the neutron star pairs in the simulation tend to be closer to us or orbiting in a way that makes their "whine" louder and clearer, allowing LISA to pick up those subtle clues about their mass. The black hole pairs are often further away or have orbits that make the clues harder to hear.

For the systems where LISA can measure the masses, the accuracy is impressive. The team found that the error in the mass measurement could be as low as 1%, but for some trickier systems, the error could be as high as 100%. Even with a 100% error, the measurement is often good enough to rule out the possibility that the objects are white dwarfs, which is a huge step forward.

The Bonus: A Map to the Stars

There's a second superpower here. When LISA measures these specific clues, it doesn't just learn the mass; it also learns exactly where the pair is located in our galaxy. The team found that for the systems where they could measure the masses, they could also pinpoint the location in the Milky Way with a precision that would allow other telescopes to look at that specific spot.

This is like finding a lost phone not just by hearing its ring, but by triangulating its position so precisely that you can walk right up to it. Once we know where these binary systems are, astronomers can point optical or radio telescopes at them to look for any leftover light or signs of their environment. This could help solve the mystery of how these pairs formed in the first place.

The Bottom Line

So, what does this all mean? The paper suggests that while LISA won't be able to identify every single black hole or neutron star pair it hears, it will definitely be able to solve the mystery for a significant number of them—specifically, about 10% of the black hole pairs and 50% of the neutron star pairs.

For the rest, the nature of the objects might remain a bit of a mystery, hidden behind the limits of the data. But even for those, LISA will provide a map to their location, inviting other scientists to come and take a closer look. It's a bit like finding a few clear footprints in the snow that tell you exactly what kind of animal passed by, while the rest of the tracks are just a blur. But those clear footprints are enough to start a whole new chapter in our understanding of how these heavy cosmic objects are born and evolve. The authors are careful to note that these numbers come from simulations based on our current best guesses of how stars behave, so the real numbers might vary, but the potential for discovery is definitely there.

Technical Summary: Identifying Galactic Binary Systems of Neutron Stars and Black Holes with LISA

Problem Statement
The Laser Interferometer Space Antenna (LISA) is expected to detect approximately $10^4$ double white dwarf (DWD) systems within the Milky Way, alongside a much smaller population of double compact objects (DCOs) comprising binary black holes (BHBHs), binary neutron stars (NSNSs), and black hole-neutron star (BHNS) systems. The primary challenge lies in distinguishing these rare DCOs (estimated at $\sim$ 100 systems) from the overwhelming background of DWDs. While component masses are the primary classifier for these objects, short-period galactic binaries ( $\sim 10^7$ – $10^3$ years from coalescence) produce quasi-monochromatic signals. In the absence of additional information, the standard gravitational wave (GW) phase evolution for these systems carries insufficient information to constrain individual component masses, making classification difficult.

Methodology
The authors conduct an end-to-end analysis using synthetic LISA data sets generated from an astrophysically realistic population of galactic binaries. The population model follows the isolated binary evolution channel, specifically utilizing the fiducial model from Wagg et al. (2022, W22), which predicts detection rates of 74 BHBHs, 42 BHNSs, and 8 NSNSs over a 4.5-year mission.

The analysis employs a Bayesian inference pipeline to explore the accuracy of mass measurements. The study generates synthetic Time-Delay-Interferometry (TDI) data streams (channels A, E, T) containing GW signals from 100 simulated sources for each DCO type (300 total). The signal model accounts for orbital eccentricity, periapse precession, and frequency derivatives, modeled as a Taylor expansion of the phase evolution.

Key methodological steps include:

Signal Generation: Drawing source parameters (masses, semi-major axis, eccentricity, 3D location) from the W22 distribution and assigning random orientation parameters.
Inference: Computing posterior probability density functions (PDFs) for 11 observable parameters (including amplitude, inclination, frequency $f_{GW}$ , frequency derivative $\dot{f}_{GW}$ , periapse precession frequency $f_{PP}$ , and eccentricity $e$ ) using the nested sampling algorithm nessai.
Measurability Criteria: A parameter is deemed "measured" if the ratio of its median value to the 16th–50th percentile width ( $S_k = \text{med}(\theta_k) / 3\sigma_{\theta_k}$ ) exceeds 1.
Mass Derivation: If $f_{GW}$ , $\dot{f}_{GW}$ , $f_{PP}$ , and $e$ are measured, the total mass ( $M$ ) and chirp mass ( $\mathcal{M}$ ) can be derived, allowing for the calculation of individual component masses ( $m_1, m_2$ ).

Key Contributions and Results
The study quantifies LISA's capability to measure individual component masses and classify DCOs based on an isolated binary evolution population:

Measurement Success Rates: LISA is projected to measure the individual component masses for approximately 10% of detected BHBHs and BHNSs and 50% of detected NSNSs. The higher success rate for NSNSs is attributed to their observation at higher frequencies and signal-to-noise ratios (SNR).
Precision: For systems where masses are measurable, typical fractional mass errors range from $\approx 1\%$ to $100\%$ , depending on specific source parameters (mass, period, eccentricity, distance).
Parameter Correlations: The study finds that measuring the total mass (via $f_{PP}$ and $e$ ) is generally more feasible than measuring the chirp mass (via $\dot{f}_{GW}$ and $e$ ). However, if the chirp mass is measurable, the total mass is also measurable.
Classification Capability:
- BHBHs: For systems with measurable masses, the primary component is confidently identified as a black hole. The secondary component's classification is less certain if its mass distribution overlaps with the neutron star range ( $\approx 2.0$ – $2.7 M_\odot$ ).
- NSNSs: Mass measurements are generally precise enough to exclude black holes. However, distinguishing NSNSs from massive white dwarfs (WDs) remains challenging due to the overlap in the Chandrasekhar limit mass range.
- Eccentricity as a Discriminator: The authors note that while most DWDs are circularized by the time they enter the LISA band, a small fraction may retain eccentricity. Conversely, DCOs formed via isolated evolution retain measurable eccentricity ( $e \gtrsim 10^{-3}$ ), which can help distinguish them from circular DWDs, though eccentric DWDs in dense environments (e.g., globular clusters) present a potential confusion source.
Sky Localization: For the subset of binaries where masses are measurable, LISA can also determine their 3-dimensional position in the Milky Way. Typical fractional distance errors are $\approx 1\%$ – $100\%$ , with sky localization areas ( $90\%$ credible regions) ranging from $10^{-2}$ to $10$ deg $^2$ .

Significance and Limitations
The paper claims that these detections will provide a "small but precious sample" of $\sim 10$ GW-selected galactic DCOs. This sample will complement existing radio, X-ray, astrometric, and microlensing surveys, offering new insights into the formation pathways of compact objects, the Galactic population, and the Milky Way's star formation history. The ability to localize these sources enables targeted follow-up observations to identify optical counterparts and characterize their environments, potentially resolving classification ambiguities (e.g., NS vs. WD).

The authors maintain a modest tone regarding the scope of their findings:

The results are based on a specific theoretical population (isolated binary evolution) and may vary with different formation channels (e.g., dynamical formation in dense clusters).
The analysis assumes known, Gaussian, stationary noise and does not account for data gaps or noise transients, which would increase statistical uncertainties.
The study does not address the computational challenges of the "global fit" required to disentangle the "forest of lines" from eccentric sources in the actual LISA data stream.
The distinction between NSs and WDs based solely on mass measurements is acknowledged as an open question requiring further investigation, potentially aided by the environmental context provided by LISA's localization capabilities.

The Great Cosmic Mix-Up

The Detective Work in the Data

What They Found

The Bonus: A Map to the Stars

The Bottom Line

More like this