The data confirm a long-established model – much to the team’s surprise.

The Kilo-Degree Survey (KiDS) has been observing large parts of the southern sky over a period of eight years in order to gain valuable new insights into the distribution of matter in the universe. On March 18, 2025, the final data set “KiDS-Legacy” was published under the leadership of researchers from Bochum, Leiden, Edinburgh, Newcastle, and London.

Previous KiDS analyses had cast doubt on the standard model of cosmology: The data had suggested a more uniform distribution of matter than the standard model predicts based on measurements from the Planck satellite. However, after analyzing the now complete KiDS data set with improved methods and calibration data, the results are consistent with the standard model, which describes how the Universe and structures within evolved over cosmic time.

“We took great care to optimize all parts of our analysis, which was a time-consuming process,” says Dr Angus Wright from Ruhr University Bochum. “The fact that the result now deviates so much from our previous analyses came as a surprise – but we were able to identify the reasons behind these changes.” The final evaluation is described in five publications that have been published or submitted for publication in the journal “Astronomy & Astrophysics”. All papers are accessible via the document server arXiv since March 26, 2025.

Determining the matter distribution with gravitational lensing

There are various methods for determining the density and structure of matter. The KiDS team used gravitational lensing in this instance: Massive objects deflect the light from distant galaxies so that these galaxies appear in a distorted shape and in a different place than they actually are when viewed from Earth. Cosmologists can use these distortions to estimate the mass of the deflecting objects and, ultimately, the total mass of the Universe. “The major advantage over other methods is that gravitational lensing can also be used to detect the dominant dark matter and make it visible, so to speak,” explains Dr. Robert Reischke from the University of Bonn.

To this end, the researchers need to know quantities such as the distances between the light source, the deflecting object and the observer. The researchers make use of the redshift to calculate these factors; redshift describes an effect where light shifts more and more towards longer wavelengths as it travels from more distant galaxies through the expanding Universe before it reaches Earth.

Images of 41 million galaxies taken with the Very Large Telescope Survey Telescope were included in the analysis. The KiDS data covers an area of 1,347 square degrees of the sky, i.e., almost ten percent of the sky where we can see past our own Galaxy.

Calculating the distance of galaxies based on redshift

In order to determine the redshift of such a large number of galaxies, the team used the photometric method: The researchers took nine images of each galaxy at different wavelengths and determined the brightness of the galaxies in each image; from this they were able to infer the redshift. The redshift can be measured more precisely by spectroscopy, but it would be too time-consuming to apply that method to millions of faint galaxies.

Still, for some galaxies both spectroscopic and photometric data are available, so that the KiDS team can calibrate its photometric measurements of redshifts with these precise spectroscopic data. While the previous analysis KiDS-1000 used spectroscopic data from approximately 25,000 galaxies for the calibration, data from as many as 126,000 galaxies were available for KiDS-Legacy. In addition, the researchers used optimized methods and new computer simulations for the analysis to reduce systematic uncertainties in the final data set.

Following the optimized evaluation, the team was able to include more distant galaxies in the final analysis than in the previous one. While KiDS-1000 was limited to galaxies with a maximum distance of 8.5 billion light years, KiDS-Legacy can now observe galaxies 10.4 billion light years away.

Blind analysis to ensure unbiased results

In the field of cosmology, it is common practice to evaluate data sets blindly to avoid any bias due to previous analyses or personal hypotheses. Before starting the analysis, researchers send the catalog of all galaxies to a third party, who changes a certain parameter for each galaxy, resulting in three variants of the data set: one with the real measured values and two with slightly different values. The researchers analyzing the data set don’t know which is the real data. They carry out their analysis with all data sets and only then learn which is the correct result. Once this step is completed, the analysis method is no longer changed.

KiDS team taken by surprise

According to the KiDS-Legacy data, the matter in Space is distributed somewhat more unevenly than KiDS-1000 had revealed. “Many tests of the internal consistency of the data show that this final analysis is significantly more robust than previous studies,” points out Dr. Benjamin Stölzner from Ruhr University Bochum. The team also compared the new results with those of other surveys. Earlier KiDS analyses had indicated a discrepancy with the Planck Survey, which estimates the matter density based on the cosmic microwave background, a radiation that was emitted shortly after the Big Bang and can still be measured today.

“The discrepancies in our KiDS data set have caused quite a stir in the research community in recent years,” explains Dr Marika Asgari, Lecturer in Data Science, Newcastle University

School of Mathematics, Statistics and Physics, “Ironically, we are now resolving this discrepancy ourselves. To our surprise, the KiDS-Legacy data contain no evidence to suggest errors in the standard model of cosmology.” The researchers explain how the differences in the various KiDS analyses come about in the current publications.

Cooperation partners

The Kilo-Degree Survey was headed by Professor Hendrik Hildebrandt from the Chair of Observational Cosmology at Ruhr University Bochum (Germany), Professor Koen Kuijken from Leiden Observatory at Leiden University (The Netherlands), Professor Catherine Heymans from the Institute for Astronomy at University of Edinburgh (UK), Dr. Marika Asgari from Newcastle University (UK), and Professor Benjamin Joachimi from University College London (UK).

Funding

The research was funded by, among others, the German Research Foundation (Hi 1495/5-1, SFB 1491), the European Research Council (ERC Consolidator Grant number 770935), the DLR project 50QE2305, as well as the Max Planck Society and the Alexander von Humboldt Foundation in the framework of the Max Planck-Humboldt Research Award endowed by the German Federal Ministry of Education and Research.

Original publications

Angus Wright et al.: The Fifth Data Release of the Kilo Degree Survey: Multi-Epoch Optical/NIR Imaging Covering Wide and Legacy-Calibration Fields, in: Astronomy & Astrophysics, 2024,  DOI: 10.1051/0004-6361/202346730, https://www.aanda.org/articles/aa/full_html/2024/06/aa46730-23/aa46730-23.html, paper download on arXiv: https://arxiv.org/abs/2503.19439

Robert Reischke et al.: KiDS-Legacy: Covariance Validation and the Unified OneCovariance Framework for Projected Large-Scale Structure Observables, paper download on arXiv: https://arxiv.org/abs/2410.06962

Benjamin Stölzner et al.: KiDS-Legacy: Consistency of Cosmic Shear Measurements and
Joint Cosmological Constraints with External Probes, paper download on arXiv: https://arxiv.org/abs/2503.19442

Angus Wright et al.: KiDS-Legacy: Redshift Distributions and Their Calibration, paper download on arXiv: https://arxiv.org/abs/2503.19440

Angus Wright et al.: KiDS-Legacy: Cosmological Constraints from Cosmic Shear with

the Complete Kilo-Degree Survey, paper download on arXiv: https://arxiv.org/abs/2503.19441

Image caption: The image shows the same matter distribution but here the KiDS footprint (the two main strips) are put closer to each other on a flat surface. The circle shows a zoom in region which is contrasted to the size of the moon in the sky. 

 Press release with thanks from Ruhr University Bochum