Collisional dark matter and the structure of dark halos

TL;DR

The paper assesses how a large elastic cross-section for dark matter particle collisions affects nonlinear halo structure by comparing collisionless CDM to a fluid-like limit using high-resolution cluster resimulations with SPH. Using the GADGET code, they quantify differences in cluster shape, central density cusps, circular velocity profiles, and subhalo content within $R_{200}$. They find that the fluid limit yields nearly spherical halos with steeper central cusps and only modest suppression of substructure, while collisionless halos remain elongated with less concentrated cores. The results argue against large self-interaction cross-sections as a solution to CDM halo problems and suggest that intermediate cross-sections require further investigation, as halo structure could behave nontrivially between the two extremes.

Abstract

We study how the internal structure of dark halos is affected if Cold Dark Matter particles are assumed to have a large cross-section for elastic collisions. We identify a cluster halo in a large cosmological N-body simulation and resimulate its formation with progressively increasing resolution. We compare the structure found in the two cases where dark matter is treated as collisionless or as a fluid. For the collisionless case our results agree with those of other workers. Collisional dark matter results in a cluster which is more nearly spherical and has a more singular central density profile. Substructure within the cluster is only weakly suppressed relative to the collisionless case. The observed structure of dwarf galaxies argues against self-interacting dark matter if, as seems likely, intermediate cross-sections produce structure lying between the extremes we have simulated.