Evolution of Neutron Star Environment in the Galactic Halo: Implications for Dark Matter Accretion

Neutron stars (NS) are one of the indirect detection probes for dark matter (DM). The presence of DM is known to affect the observable properties of NS. Theoretical calculations for various equations of state of a DM admixed NS lead to estimates of the DM mass in a NS of the order of 10−2 M⊙. On the other hand, simplistic estimates of the amount of DM that is accreted on to the NS in the Solar neighborhood, over its age, suggest that this number is of the order 10−14 M⊙. Various studies have addressed this non-agreement theoretically by explaining the mechanisms leading to higher fraction than expected from smooth spherically symmetric accretion. In this work, we attempt to assess the role of the dynamic DM environmental density in the Galactic halo to explain possible enhancement in DM accretion. We consider a high resolution N-body simulation and method of Voronoi tessellation to calculate local DM density around putative NS locations in a statistically representative sample of Milky Way analogues. We infer that the dynamics of substructure may enhance the DM mass in NS by about a factor 2 as compared to the baseline, spherically symmetric expectation. Environmental effects therefore cannot explain the orders of magnitude discrepancy between equation of state and accretion based estimates of DM admixed in NS.

Improvements to AI systems:

1. Dynamic Density Field Estimation for Astroparticle Accretion Models

• Improvement: Replace static, spherically symmetric DM density assumptions in NS accretion codes with a time-resolved, spatially inhomogeneous density field derived from N-body simulations (via Voronoi tessellation).

• What the improved AI can do: Predict DM accretion rates onto NSs in realistic Galactic environments, accounting for substructure clumps and orbital motion, and quantify the enhancement factor (e.g., 2×) over baseline models.

2. Uncertainty-Aware Multi-Scale Simulation Coupling

• Improvement: Build an AI framework that automatically couples high-resolution N-body halo simulations (e.g., MW analogues) with NS equation-of-state and accretion physics, propagating statistical uncertainties from halo-to-halo variance and tessellation resolution.

• What the improved AI can do: Output probability distributions for DM mass in NSs, rather than single values, enabling robust comparisons between observational constraints and theoretical predictions.

3. Anomaly Detection for Discrepancy Attribution

• Improvement: Train a classifier to distinguish between environmental (density-driven) and intrinsic (equation-of-state) contributions to DM mass in NSs, using features like local density variance, accretion history, and NS age.

• What the improved AI can do: Automatically flag NSs where environmental effects are insufficient (as in this paper) and identify which physical mechanisms (e.g., capture from DM spikes, self-interaction) are needed to resolve order-of-magnitude gaps.

4. Generative Surrogate for Halo Substructure

• Improvement: Use the Voronoi-based density maps as training data to create a generative model (e.g., GAN or normalizing flow) that synthesizes realistic DM density fluctuations around arbitrary NS positions in MW-like galaxies.

• What the improved AI can do: Rapidly generate thousands of plausible accretion environments without running expensive N-body simulations, enabling large-scale Monte Carlo studies of DM admixed NS populations.

5. Bayesian Inference for NS-DM Parameter Estimation

• Improvement: Integrate the simulated density enhancement factors into a Bayesian hierarchical model that jointly infers DM particle properties (mass, cross-section) and NS equation of state from observed NS thermal or rotational data.

• What the improved AI can do: Provide posterior constraints that explicitly account for environmental uncertainty, preventing biased conclusions about DM microphysics from ignoring halo dynamics.

Neutron stars (NS) are one of the indirect detection probes for dark matter (DM). The presence of DM is known to affect the observable properties of NS. Theoretical calculations for various equations of state of a DM admixed NS lead to estimates of the DM mass in a NS of the order of 10-2 M. On the other hand, simplistic estimates of the amount of DM that is accreted on to the NS in the Solar neighborhood, over its age, suggest that this number is of the order 10-14 M. Various studies have addressed this non-agreement theoretically by explaining the mechanisms leading to higher fraction than expected from smooth spherically symmetric accretion. In this work, we attempt to assess the role of the dynamic DM environmental density in the Galactic halo to explain possible enhancement in DM accretion. We consider a high resolution N-body simulation and method of Voronoi tessellation to calculate local DM density around putative NS locations in a statistically representative sample of Milky Way analogues. We infer that the dynamics of substructure may enhance the DM mass in NS by about a factor 2 as compared to the baseline, spherically symmetric expectation. Environmental effects therefore cannot explain the orders of magnitude discrepancy between equation of state and accretion based estimates of DM admixed in NS.