2013A&A...554A..55H

Core condensation is a critical step in the star-formation process, but it is still poorly characterized observationally. We have studied the 10 pc-long L1495/B213 complex in Taurus to investigate how dense cores have condensed out of the lower density cloud material. We observed L1495/B213 in C¹⁸O(1-0), N₂H⁺(1-0), and SO(J_N=3₂-2₁) with the 14m FCRAO telescope, and complemented the data with dust continuum observations using APEX (870µm) and IRAM 30m (1200µm). From the N₂H⁺ emission, we identify 19 dense cores, some starless and some protostellar. They are not distributed uniformly, but tend to cluster with relative separations on the order of 0.25pc. From the C¹⁸O emission, we identify multiple velocity components in the gas. We have characterized them by fitting Gaussians to the spectra and by studying the distribution of the fits in position-position-velocity space. In this space, the C¹⁸O components appear as velocity-coherent structures, and we identify them automatically using a dedicated algorithm (FIVE: Friends In VElocity). Using this algorithm, we identify 35 filamentary components with typical lengths of 0.5pc, sonic internal velocity dispersions, and mass-per-unit length close to the stability threshold of isothermal cylinders at 10K. Core formation seems to have occurred inside the filamentary components via fragmentation, with few fertile components with higher mass-per-unit length being responsible for most cores in the cloud. On large scales, the filamentary components appear grouped into families, which we refer to as bundles. Core formation in L1495/B213 has proceeded by hierarchical fragmentation. The cloud fragmented first into several pc-scale regions. Each of these regions later fragmented into velocity-coherent filaments of about 0.5pc in length. Finally, a small number of these filaments fragmented quasi-statically and produced the individual dense cores we see today.