The hierarchical fragmentation of filaments and the role of sub-filaments

TL;DR

This study uses hydrodynamical simulations to explore how sub-filaments within larger filaments influence core formation, revealing diverse core environments and challenging previous theories of characteristic fragmentation scales.

Contribution

It demonstrates that core formation occurs in two distinct environments within sub-filaments and shows no characteristic fragmentation length-scale, aligning better with observations than earlier models.

Findings

Cores form in isolated or hub environments within sub-filaments.

Hub cores are more massive and have wider mass distributions.

Fragmentation signatures are similar across multiple scales.

Abstract

Recent observations have revealed the presence of small fibres or sub-filaments within larger filaments. We present a numerical fragmentation study of fibrous filaments investigating the link between cores and sub-filaments using hydrodynamical simulations performed with the moving-mesh code Arepo. Our study suggests that cores form in two environments: (i) as isolated cores, or small chains of cores, on a single sub-filament, or (ii) as an ensemble of cores located at the junction of sub-filaments. We term these isolated and hub cores respectively. We show that these core populations are statistically different from each other. Hub cores have a greater mean mass than isolated cores, and the mass distribution of hub cores is significantly wider than isolated cores. This fragmentation is reminiscent of parsec-scale hub-filament systems, showing that the combination of turbulence and gravity leads to similar fragmentation signatures on multiple scales, even within filaments. Moreover, the fact that fragmentation proceeds through sub-filaments suggests that there exists no characteristic fragmentation length-scale between cores. This is in opposition to earlier theoretical works studying fibre-less filaments which suggest a strong tendency towards the formation of quasi-periodically spaced cores, but in better agreement with observations. We also show tentative signs that global collapse of filaments preferentially form cores at both filament ends, which are more massive and dense than other cores.