Scientists are now leveraging the faint trails of disrupted stars, known as stellar streams, to map the distribution of dark matter surrounding galaxies. David Chemaly, Elisabeth Sola, and Vasily Belokurov, all from the Institute of Astronomy, Cambridge, alongside Sergey Kosposov, GyuChul Meyong, and HanYuan Zhang, detail a novel hierarchical Bayesian framework for inferring the shapes of dark matter halos using only two-dimensional images of these streams. This research is significant because it unlocks the potential to study dark matter in external galaxies without requiring difficult-to-obtain kinematic data , a major hurdle in current astrophysical research. By combining individual stream analyses with a powerful new modelling tool called StreaMAX, the team demonstrates that population-level constraints on halo morphology, distinguishing between flattened, spherical, and elongated shapes, are achievable, paving the way for large-scale dark matter studies with upcoming surveys like Euclid and Rubin/LSST.
The research, published in MNRAS, presents a hierarchical Bayesian framework that infers the population distribution of halo flattening solely from projected stream tracks, essentially, the visible paths of these disrupted stellar structures. This breakthrough circumvents the need for detailed kinematic data, which is often unavailable for streams outside our own Milky Way, offering a powerful new tool for cosmological studies. The team achieved this by developing StreaMAX, a new JAX-accelerated particle-spray package that generates stream models orders of magnitude faster than conventional methods, enabling efficient exploration of parameter space and robust statistical inference.
The study meticulously forward-models streams using StreaMAX, fitting each stream to an axisymmetric dark matter halo model to obtain a posterior probability distribution for its flattening. These individual posteriors are then combined using hierarchical reweighting, a sophisticated statistical technique that accounts for uncertainties and biases inherent in the projections and individual fits. Experiments utilising mock data reveal that while individual stream fits exhibit modest precision and projection-induced complexities, aggregating these fits yields remarkably accurate and confident constraints on the overall population distribution of dark matter halo morphologies. This allows scientists to confidently distinguish between oblate, spherical, and prolate halo shapes at a population level, providing crucial insights into galaxy formation and the nature of dark matter.
This innovative approach demonstrates that ensembles of purely photometric streams, those observed only in two dimensions, contain sufficient information to constrain dark matter halo shapes in external galaxies. Crucially, the computational cost of this method scales linearly with sample size, making it highly practical for application to the vast datasets expected from forthcoming surveys like Euclid and Rubin/LSST. The research establishes a pathway to population-level inferences of halo morphology without relying on difficult-to-obtain kinematic measurements, opening up new avenues for understanding the distribution and properties of dark matter in the Universe. By leveraging the abundance of stellar streams, astronomers can now probe the shapes of dark matter haloes across a wider range of galaxies and cosmic environments.
The work unveils a significant advancement in our ability to map the invisible dark matter scaffolding that governs galaxy formation. While six-dimensional phase-space data provides the most precise modelling for streams within the Milky Way, this new technique extends these capabilities to external galaxies where such detailed data is often lacking. The team’s success hinges on the efficiency of StreaMAX, which dramatically reduces the computational burden of stream modelling, and the power of hierarchical Bayesian inference, which effectively combines information from multiple streams to overcome the limitations of individual observations. This combination allows for robust statistical analysis and accurate determination of the underlying population distribution of dark matter halo shapes, offering a powerful new tool for cosmological research and furthering our understanding of the Universe’s hidden components.
Halo flattening inference from stellar stream modelling is
Scientists developed a hierarchical Bayesian framework to infer the population distribution of dark matter halo flattening using only projected stellar stream tracks. This work addresses the limitation that streams around external galaxies are often observed as two-dimensional photometric data, lacking the kinematic information needed for precise dark matter modelling. The research team pioneered a novel approach, forward-modelling streams with StreaMAX, a new JAX-accelerated particle-spray package that achieves stream generation speeds orders of magnitude faster than traditional methods. For each stream analysed, they fitted an axisymmetric dark matter halo model, obtaining a posterior probability distribution representing the flattening of the halo.
Subsequently, these individual posteriors were combined using hierarchical reweighting, a statistical technique that allows for robust constraints on the overall population distribution of halo morphologies. To validate their methodology, the team conducted experiments using mock data, demonstrating that individual flattening fits, while exhibiting modest precision and projection-induced multi-modalities, accurately recover the correct values. Crucially, aggregating these individual fits yielded accurate and confident constraints on the underlying population distribution, successfully distinguishing between oblate, spherical, and prolate halo shapes. The computational efficiency of this approach is noteworthy, as the total cost scales linearly with the sample size, enabling analysis of large datasets.
The study harnessed the power of StreaMAX by implementing a particle-spray technique, where numerous particles are launched to simulate the stream’s trajectory and brightness distribution. This innovative method allows for rapid generation of stream models, facilitating efficient parameter estimation. Researchers then employed a Bayesian inference procedure, utilising Markov Chain Monte Carlo (MCMC) sampling to explore the parameter space and obtain posterior distributions for the halo flattening. The hierarchical reweighting scheme effectively combines information from multiple streams, mitigating the impact of individual stream uncertainties and improving the overall robustness of the results. This approach demonstrates that ensembles of purely photometric streams contain sufficient information to constrain dark matter halo shapes at the population level, offering a practical path for future studies. With forthcoming large-scale surveys like Euclid and Rubin/LSST poised to deliver unprecedented samples of stellar streams, this methodology promises to unlock new insights into the nature of dark matter and the formation of galaxies without relying on difficult-to-obtain kinematic measurements.
Halo flattening inferred from stellar stream tracks supports
Scientists have developed a new hierarchical Bayesian framework to infer the population distribution of dark matter halo flattening using only projected stellar stream tracks. The team forward-modelled streams in StreaMAX, a JAX-accelerated particle-spray package achieving orders of magnitude faster stream generation compared to traditional methods. Experiments revealed that individual stream fits recover the correct flattening with modest precision, but exhibit projection-induced multi-modalities, a common challenge in astronomical modelling. Nevertheless, aggregating these individual fits yields accurate and confident constraints on the underlying population distribution of dark matter halo morphologies, clearly distinguishing between oblate, spherical, and prolate populations.
Data shows the total computational cost scales linearly with sample size, making this approach highly efficient for large datasets. Measurements confirm that ensembles of purely photometric streams carry sufficient information to constrain dark matter halo shapes in external galaxies at the population level, a significant breakthrough in galactic halo studies. Results demonstrate that the framework accurately recovers halo flattening, despite the complexities introduced by projection effects. The study utilized mock data to validate the methodology, ensuring the reliability of the inferred population distributions.
Tests prove that the hierarchical reweighting technique effectively combines information from multiple streams, mitigating the impact of individual fit uncertainties. The breakthrough delivers a practical path to population-level inferences of halo morphology without requiring any kinematic measurements, a substantial advantage for observing distant galaxies. Scientists recorded that this approach will be particularly powerful with forthcoming data from the Euclid and Rubin/LSST surveys, promising a wealth of photometric stream data. The research establishes a new method for probing dark matter halo shapes, relying solely on observable stream tracks and advanced computational techniques. Measurements confirm the potential to unlock insights into galaxy formation physics and test alternative dark matter models through population-level analyses of halo morphology. This work offers a novel pathway to understanding the distribution of dark matter in the universe.
👉 More information
🗞 Hierarchical bayesian inference: constraining population distribution of dark matter halo shapes via stellar streams
🧠ArXiv: https://arxiv.org/abs/2601.15373