🔭 astrophysics

CMB lensing imprints of cosmic voids in DESI Legacy Survey DR9 LRGs with photometric redshift calibration

By utilizing a meticulously calibrated DESI Legacy Survey DR9 LRG sample and Planck 2018 lensing data, this study achieves a record 17σ detection of CMB lensing imprints from cosmic voids that fully aligns with $\Lambda$ CDM predictions, thereby resolving the previously reported "lensing-is-low" tension through rigorous systematic control.

Original authors: Simone Sartori

Published 2026-05-20

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Simone Sartori

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, three-dimensional sponge. Most of the sponge is made of "stuff" (galaxies and dark matter), but there are also huge, empty holes inside it called cosmic voids.

This paper is about a scientific detective story involving these empty holes and the oldest light in the universe: the Cosmic Microwave Background (CMB). Think of the CMB as a giant, ancient wallpaper that covers the entire sky.

Here is the story of what the researchers found, explained simply:

The Mystery: The "Too Weak" Signal

For a while, scientists have been trying to see how these cosmic voids affect the ancient wallpaper. According to our best theory of the universe (called ΛCDM), the empty voids should act like a diverging lens. If you look through a void, the wallpaper behind it should look slightly stretched out or "demagnified."

However, recent studies hit a snag. When they looked at real data, the stretching effect was much weaker than the theory predicted. It was like looking at a funhouse mirror that wasn't distorting the image as much as the instructions said it should. Scientists called this the "lensing-is-low" tension. They were worried: Is our theory of the universe wrong? Or is something else messing up the measurements?

The Investigation: A Better Map and a Better Mirror

The author, Simone Sartori, decided to solve this mystery using a massive new dataset from the DESI Legacy Survey. Imagine this as having a super-high-definition map of 10 million red galaxies (the "tracer" galaxies that help us find the voids).

To solve the mystery, the team didn't just look at the real data; they built a perfectly matched simulation (a "mock" universe) to compare against it.

Here is the clever part:

The Calibration: Previous studies might have compared real data to simulations that were slightly "off" in how they handled galaxy positions and distances. It's like trying to compare a photo taken with a wide-angle lens to a photo taken with a telephoto lens and expecting them to look identical.
The Fix: This team used over one million real spectroscopic measurements (extremely precise distance readings) to "tune" their simulation. They adjusted the simulation so that the "holes" in the fake universe had the exact same size, shape, and distribution errors as the holes in the real universe. They made the simulation a perfect twin of reality.

The Experiment: Stacking the Signals

Because a single void is too small to see clearly (it's like trying to hear a whisper in a hurricane), the team used a technique called stacking.

Imagine you have 140,000 whispers (voids). You can't hear one, but if you line them all up perfectly and play them at the same time, you can hear the combined sound clearly.
They lined up all 140,000 voids from their map and looked at the CMB wallpaper behind them.

They also sorted the voids into two groups based on their "personality":

Void-in-Voids: Big, lonely empty spaces surrounded by other empty spaces.
Void-in-Clouds: Smaller empty spaces trapped inside dense clusters of galaxies.

The Result: The Mystery is Solved

When they compared their real data to their perfectly tuned simulation, the "lensing-is-low" tension disappeared.

The Match: The stretching effect they saw in the real universe matched the simulation almost perfectly.
The Score: They measured an "amplitude" of 1.016. Since the theory predicts exactly 1.0, this is a near-perfect match.
The Confidence: They were so sure of this result that they achieved a statistical significance of 17 sigma. In the world of science, 5 sigma is usually enough to claim a discovery; 17 sigma is like winning the lottery every day for a year. It is an incredibly strong signal.

The Takeaway

The paper concludes that the universe is behaving exactly as our standard theory (ΛCDM) predicts. The previous "weak signal" problem wasn't because the theory was wrong; it was because the simulations used for comparison weren't calibrated correctly.

By using a massive amount of real data to fix the simulation, the researchers showed that the "noise" was actually just a mismatch in the tools they were using. Once the tools were aligned, the universe's behavior was perfectly normal.

In short: The cosmic voids are doing exactly what they are supposed to do. The "glitch" was in the comparison, not in the universe.

Technical Summary: CMB Lensing Imprints of Cosmic Voids in DESI Legacy Survey DR9 LRGs

Problem Statement
The lensing imprint of cosmic voids on the Cosmic Microwave Background (CMB) serves as a critical probe for testing the $\Lambda$ CDM cosmological model. Voids induce a demagnification effect, creating a negative signal in lensing convergence ( $\kappa$ ) maps. While stacking techniques have historically yielded high-significance detections, recent literature has reported a persistent "lensing-is-low" tension, where observed stacked signals deviate from $\Lambda$ CDM mock predictions by up to $\sim3\sigma$ . This discrepancy is highly sensitive to void identification strategies and the properties of tracer galaxy samples. The central hypothesis addressed in this work is that uncontrolled mismatches between observational data and simulated mocks—specifically regarding sparseness and photometric redshift ( $z_{phot}$ ) error distributions—may act as hidden systematics that mimic a breakdown of the standard model.

Methodology
To investigate this tension, the authors analyzed 3D cosmic voids identified in the Luminous Red Galaxy (LRG) sample from the DESI Legacy Survey Data Release 9 (DR9) and cross-correlated them with the Planck 2018 CMB lensing convergence map.

Data Sets: The study utilized approximately 10.4 million LRGs from the DESI Legacy Survey covering $\sim19,500$ deg $^2$ in the redshift range $0.35 < z < 0.95$ . This dataset was supplemented by over one million spectroscopic redshifts from early DESI survey stages for calibration purposes. The CMB lensing observable was a reconstructed $\kappa$ map from Planck 2018, smoothed with a Gaussian filter (FWHM = $0.5^\circ$ ) to suppress small-scale noise.
Mock Calibration: A critical component of the methodology was the rigorous calibration of the Buzzard mock catalogs. The authors employed a two-step procedure to ensure structural consistency between observed and simulated tracer populations:
1. Redshift Error Correction: The mock photometric redshift error distributions were corrected by sampling the $z_{spec} - z_{phot}$ distribution derived from the million DESI spectra.
2. Sparseness Matching: The mocks were downsampled in redshift bins ( $\Delta z = 0.01$ ) to align their number density with the observed LRGs.
  This process ensured that simulated voids were subject to the same selection effects and density profiles as the observed data.
Void Identification and Classification: Voids were identified using the REVOLVER algorithm, which employs Voronoi tessellation to define underdense basins. The sample was restricted to $0.35 < z < 0.95$ $0.35 < z < 0.95$ , yielding 140,712 voids. To disentangle signals from different large-scale environments, voids were categorized by the parameter $\lambda_v$ $λ_{v}$ (a proxy for central gravitational potential):
- Void-in-voids: $\lambda_v < -5$ (highly underdense)
- Transition sample: $-5 \le \lambda_v < 5$
- Void-in-clouds: $\lambda_v \ge 5$ (slightly underdense)
  Tomographic analysis was performed by dividing the sample into four redshift bins of $\Delta z = 0.15$ .
Analysis: The study measured the stacked radial profile of CMB convergence patches (side length $5R_v$ ) for each void. The amplitude parameter $A_\kappa$ was fitted via $\chi^2$ minimization against the calibrated templates, utilizing an inverse covariance matrix corrected by the Anderson-Hartlap factor.

Key Contributions and Results
The primary contribution of this work is the resolution of the "lensing-is-low" tension through precise systematic control rather than new physics.

Detection Significance: The analysis achieved record detection significances for this class of study. The full sample yielded a signal-to-noise ratio (S/N) of 14.06. When categorized by environment, the S/N reached $\sim17\sigma$ for both void-in-voids ($16.94$) and void-in-clouds ($17.02$).
Amplitude Measurement: The study measured an amplitude parameter of $A_\kappa = 1.016 \pm 0.054$ for the full sample. This value indicates full agreement between observations and the simulated $\Lambda$ CDM templates.
Redshift Evolution: The tomographic analysis confirmed that the amplitude parameter remains stable and close to unity ( $A_\kappa \sim 1$ ) across the entire redshift range ( $0.35 < z < 0.95$ ), consistent with standard theoretical expectations.
Systematic Resolution: By matching sparseness and $z_{phot}$ error distributions using over one million DESI spectra, the authors demonstrated that the previously reported tensions were likely artifacts of mismatches between mocks and observations.

Significance
The paper concludes that the "lensing-is-low" tension is not an indication of new physics but rather a consequence of uncontrolled systematics in previous analyses. The work highlights the necessity of accurate systematic control, particularly in the calibration of mock catalogs to match the sparseness and photometric redshift errors of real data. Furthermore, it establishes that distinguishing void populations based on their gravitational potential is essential for boosting detection significance and revealing the distinct physical signatures of voids in different large-scale environments. The results presented here, summarized in Sartori et al. (2025), provide a robust validation of the $\Lambda$ CDM model using CMB lensing of cosmic voids.