🔭 astrophysics

First application of weak lensing peak steepness statistics to HSC Y1 data: effectively probing halo density profiles

This paper presents the first application of weak lensing peak steepness statistics to HSC Y1 data, demonstrating its effectiveness in constraining halo density profiles and providing evidence for strong baryonic feedback effects at cluster scales.

Original authors: Ziwei Li, Xiangkun Liu, Tianyu Zhang, An Zhao, Chuzhong Pan, Shuo Yuan, Qiao Wang, Zuhui Fan

Published 2026-08-26

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Ziwei Li, Xiangkun Liu, Tianyu Zhang, An Zhao, Chuzhong Pan, Shuo Yuan, Qiao Wang, Zuhui Fan

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

The universe is not a smooth, empty void. It is a vast, web-like structure where invisible clumps of dark matter hold together the galaxies we can see. To understand how this cosmic web formed and how it behaves, astronomers need a way to weigh these invisible clumps and map their density. One of the most powerful tools for this job is weak gravitational lensing. As light travels across billions of light-years from distant galaxies to our telescopes, the gravity of the dark matter clumps it passes bends the light slightly, distorting the shapes of those background galaxies. By measuring these tiny distortions across the sky, scientists can reconstruct a map of where the mass is located, even though they cannot see the mass itself.

For years, researchers have looked at the height of the peaks in these mass maps—the tallest peaks usually correspond to the most massive clusters of galaxies. However, a new approach is emerging that looks not just at how tall these peaks are, but at how steep their sides are. This "steepness" tells a different story. While the height of a peak reveals the total amount of mass, the steepness of its slope reveals how that mass is packed together inside. If the matter is tightly concentrated in the center, the peak will be steep; if it is spread out more loosely, the peak will be gentler. This distinction is crucial because the way matter packs itself is influenced by the complex physics of normal matter, such as gas and stars, which can push against dark matter and change its distribution. Understanding this interaction helps scientists refine their models of the universe and test the nature of dark matter itself.

A team of researchers has now taken this new method of measuring peak steepness and applied it to real data for the first time. Using observations from the Hyper Suprime-Cam on the Subaru Telescope in Hawaii, they analyzed a vast patch of the sky containing millions of galaxies. The team focused on the highest peaks in their mass maps, which are dominated by massive galaxy clusters. They combined the traditional measurement of peak height with their new measurement of peak steepness to create a more complete picture of the universe's structure. By comparing their observations against thousands of computer simulations, they were able to test how well their theoretical models matched reality.

The results of this analysis point toward a universe where the centers of massive galaxy clusters are less dense than previously thought in simple models. When the researchers measured the concentration of matter in these clusters, they found a value significantly lower than what is seen in simulations that include only dark matter. In those dark-matter-only simulations, the matter is packed very tightly. The real data, however, suggests that the matter is more spread out. This difference aligns with the idea that powerful feedback from normal matter—such as the energy released by supermassive black holes or the explosion of stars—has pushed the dark matter outward, smoothing out the core of these clusters. The team found that their combined analysis of height and steepness provided a much tighter constraint on this concentration than looking at height alone, proving that the steepness statistic is a sensitive and powerful tool for probing the inner workings of these cosmic giants.

The study also carefully checked for potential errors that could skew the results. The researchers accounted for the fact that some galaxies in the background might actually be members of the foreground clusters, which would dilute the lensing signal. They also tested whether the natural alignment of galaxy shapes, a phenomenon known as intrinsic alignment, could mimic the signal they were looking for. Their tests showed that these effects were negligible for their specific analysis. Furthermore, they checked for "B-modes," a type of signal that should not exist in a perfect gravitational lensing map and would indicate errors in the data processing. The absence of these signals confirmed that their data was clean and reliable.

By applying this new statistical tool to real observations, the team has demonstrated that the steepness of weak lensing peaks is a viable and effective way to study the density profiles of dark matter halos. Their findings suggest that the influence of baryonic feedback—the physical processes involving normal matter—is strong enough to significantly alter the structure of massive clusters. This result supports the view that the universe is shaped not just by the invisible gravity of dark matter, but also by the dynamic, energetic interactions of the visible matter within it. As future surveys gather even more data, this method of looking at the slope of the cosmic peaks promises to reveal even more details about the hidden architecture of our universe.

Technical Summary: First Application of Weak Lensing Peak Steepness Statistics to HSC Y1 Data

Problem and Motivation
Weak gravitational lensing (WL) has become a primary probe for mapping the large-scale mass distribution of the Universe. While traditional analyses rely heavily on two-point correlation functions, the non-Gaussian nature of the cosmic matter distribution necessitates higher-order statistics to extract maximal cosmological information. Among these, WL peak statistics have long been recognized as efficient probes. However, the majority of existing peak analyses focus solely on peak height. Recent theoretical and machine-learning studies suggest that the "steepness" of a peak profile—essentially the second derivative of the convergence field—contains additional cosmological information, particularly regarding the inner density profiles of dark matter halos. These profiles encode critical information about baryonic feedback and dark matter properties. Despite this potential, peak steepness had not yet been applied to real observational data. This paper addresses that gap by presenting the first application of WL peak steepness statistics to the Hyper Suprime-Cam (HSC) Year 1 (Y1) data.

Methodology
The authors utilize the HSC Y1 shear catalog (S16A), selecting 52 regular square fields covering a total effective area of approximately 57.4 deg² after masking and boundary corrections. The analysis pipeline involves:

Convergence Reconstruction: Shear fields are smoothed with a Gaussian kernel ( $\theta_G = 1.5$ arcmin) and reconstructed into convergence maps ( $\kappa_N$ ) using the non-linear Kaiser-Squires inversion.
Peak Identification: High peaks are defined as pixels with $\kappa_N$ values higher than their eight neighbors and a signal-to-noise ratio $\nu \geq 4$ .
Steepness Calculation: For each identified peak, the steepness ( $S$ ) is calculated using a discrete operator: $S = 4\kappa_N(i,j) - \sum \kappa_N(\text{nearest 4 pixels})$ . This corresponds to the negative Laplacian of the convergence field. The signal-to-noise ratio for steepness is defined as $x = S/\sigma_{N,2}$ .
Theoretical Modeling: The authors employ a halo-based theoretical model valid for high peaks dominated by massive clusters. The model incorporates:
- HMcode2016: Used to model baryonic feedback phenomenologically by reducing the halo concentration parameter. The mass-concentration relation is parameterized as $c(M, z) = A(1+z_f)/(1+z)$ , where $A$ is the amplitude parameter.
- Halo Contributions: The convergence field is decomposed into contributions from massive halos ( $K_H$ ), large-scale structure projections ( $K_{LSS}$ ), and shape noise ( $N$ ).
- Systematics Modeling: The model explicitly accounts for cluster member contamination (dilution of WL signals and modification of noise profiles) using single-halo simulations and member excess fractions from cluster catalogs. Intrinsic alignment (IA) effects and B-mode tests are also evaluated.
Validation and Fitting: The pipeline is validated using 1,000 bootstrap simulation mocks based on N-body simulations. Markov Chain Monte Carlo (MCMC) fitting is performed on the joint peak height and steepness counts to constrain cosmological parameters ( $\Omega_m, \sigma_8$ ) and the concentration amplitude $A$ .

Key Contributions

First Observational Application: This work marks the first time peak steepness statistics have been applied to real observational WL data (HSC Y1), demonstrating the feasibility of the method.
Probing Halo Density Profiles: The study demonstrates that peak steepness is significantly more sensitive to the halo concentration parameter $A$ than peak height alone. This allows for statistical constraints on the inner mass distribution of halos.
Systematic Handling: The authors develop and implement a refined treatment for cluster member contamination, modeling its impact on both the halo profile derivatives and the noise field, rather than applying a simple global dilution factor.
Validation: The theoretical model is rigorously validated against simulation mocks, showing excellent agreement in both peak height and steepness distributions.

Results

Cosmological Constraints: Combining peak height and steepness statistics, the authors derive $S_8 = 0.76^{+0.08}_{-0.07}$ (with a Maximum-A-Posteriori, MAP, of 0.79). This result is consistent with other HSC WL studies.
Concentration Parameter: The analysis yields a concentration amplitude of $A = 1.93^{+1.33}_{-1.16}$ (MAP = 1.70).
Comparison with Simulations: This measured value of $A$ is approximately $1\sigma$ lower than the value derived from dark-matter-only simulated mocks ( $A = 3.34^{+1.52}_{-1.74}$ , MAP=3.31) and is more consistent with HMcode2016 fits to simulations including strong Active Galactic Nuclei (AGN) feedback (e.g., OWLS simulations).
Systematics Checks: B-mode tests using the imaginary part of the reconstructed convergence field show negligible B-mode signals, consistent with pure shape noise. Intrinsic alignment (IA) effects were found to be minimal within the statistical uncertainties of the survey area.

Significance
The paper claims that the peak steepness statistic is a powerful new tool for WL cosmology. Its primary significance lies in its enhanced sensitivity to the density profiles of massive halos compared to peak height statistics. The results tentatively support the presence of strong baryonic feedback effects at cluster scales, as indicated by the lower concentration parameter $A$ compared to dark-matter-only expectations. The authors conclude that this method offers a promising avenue for statistically constraining baryonic feedback and dark matter properties, complementing traditional two-point correlation analyses. They note that while the current framework uses HMcode2016, future work will aim to incorporate more sophisticated baryonic models (like HMcode2020 or BCM) to derive more physically linked information.

More like this