Cluster-formation in the Rosette molecular cloud at the junctions of filaments (original) (raw)
Related papers
The Progression of Star Formation in the Rosette Molecular Cloud
The Astrophysical Journal, 2013
Using Spitzer Space Telescope and Chandra X-ray Observatory data, we identify YSOs in the Rosette Molecular Cloud (RMC). By being able to select cluster members and classify them into YSO types, we are able to track the progression of star formation locally within the cluster environments and globally within the cloud. We employ nearest neighbor method (NNM) analysis to explore the density structure of the clusters and YSO ratio mapping to study age progressions in the cloud. We find a relationship between the YSO ratios and extinction which suggests star formation occurs preferentially in the densest parts of the cloud and that the column density of gas rapidly decreases as the region evolves. This suggests rapid removal of gas may account for the low star formation efficiencies observed in molecular clouds. We find that the overall age spread across the RMC is small. Our analysis suggests that star formation started throughout the complex around the same time. Age gradients in the cloud appear to be localized and any effect the H II region has on the star formation history is secondary to that of the primordial collapse of the cloud.
Understanding star formation in molecular clouds
Astronomy and Astrophysics, 2022
Probability distribution functions of the total hydrogen column density (N-PDFs) are a valuable tool for distinguishing between the various processes (turbulence, gravity, radiative feedback, magnetic fields) governing the morphological and dynamical structure of the interstellar medium. We present N-PDFs of 29 Galactic regions obtained from Herschel imaging at high angular resolution (18), covering diffuse and quiescent clouds, and those showing low-, intermediate-, and high-mass star formation (SF), and characterize the cloud structure using the ∆-variance tool. The N-PDFs show a large variety of morphologies. They are all double-log-normal at low column densities, and display one or two power law tails (PLTs) at higher column densities. For diffuse, quiescent, and low-mass SF clouds, we propose that the two log-normals arise from the atomic and molecular phase, respectively. For massive clouds, we suggest that the first log-normal is built up by turbulently mixed H 2 and the second one by compressed (via stellar feedback) molecular gas. Nearly all clouds have two PLTs with slopes consistent with self-gravity, where the second one can be flatter or steeper than the first one. A flatter PLT could be caused by stellar feedback or other physical processes that slow down collapse and reduce the flow of mass toward higher densities. The steeper slope could arise if the magnetic field is oriented perpendicular to the LOS column density distribution. The first deviation point (DP), where the N-PDF turns from log-normal into a PLT, shows a clustering around values of a visual extinction of A V (DP1) ∼ 2-5. The second DP, which defines the break between the two PLTs, varies strongly. In contrast, the width of the N-PDFs is the most stable parameter, with values of σ between ∼0.5 and 0.6. Using the ∆-variance tool, we observe that the A V value, where the slope changes between the first and second PLT, increases with the characteristic size scale in the ∆-variance spectrum. We conclude that at low column densities, atomic and molecular gas is turbulently mixed, while at high column densities, the gas is fully molecular and dominated by self-gravity. The best fitting model N-PDFs of molecular clouds is thus one with log-normal low column density distributions, followed by one or two PLTs.
Monthly Notices of the Royal Astronomical Society, 2014
We present a model for the radiative output of star clusters in the process of star formation suitable for use in hydrodynamical simulations of radiative feedback. Gas in a clump, defined as a region whose density exceeds 10 4 cm −3 , is converted to stars via the random sampling of the Chabrier IMF. A star formation efficiency controls the rate of star formation. We have completed a suite of simulations which follow the evolution of accretion-fed clumps with initial masses ranging from 0 to 10 5 M ⊙ and accretion rates ranging from 10 −5 to 10 −1 M ⊙ yr −1. The stellar content is tracked over time which allows the aggregate luminosity, ionizing photon rate, number of stars, and star formation rate (SFR) to be determined. For a fiducial clump of 10 4 M ⊙ , the luminosity is ∼4×10 6 L ⊙ with a SFR of roughly 3×10 −3 M ⊙ yr −1. We identify two regimes in our model. The accretion-dominated regime obtains the majority of its gas through accretion and is characterized by an increasing SFR while the reservoirdominated regime has the majority of its mass present in the initial clump with a decreasing SFR. We show that our model can reproduce the expected number of O stars, which dominate the radiative output of the cluster. We find a nearly linear relationship between SFR and mass as seen in observations. We conclude that our model is an accurate and straightforward way to represent the output of clusters in hydrodynamical simulations with radiative feedback.
A universal route for the formation of massive star clusters in giant molecular clouds
Nature Astronomy, 2018
Young massive star clusters (YMCs, with M ≥10 4 M ʘ) are proposed modernday analogues of the globular clusters (GCs) that were products of extreme star formation in the early universe 14. The exact conditions and mechanisms under which YMCs form remain unknown 4,5 − a fact further complicated by the extreme radiation fields produced by their numerous massive young stars 69. Here we show that GCsized clusters are naturally produced in radiationhydrodynamic simulations of isolated 10 7 M ʘ Giant Molecular Clouds (GMCs) with properties typical of the local universe, even under the influence of radiative feedback. In all cases, these massive clusters grow to GClevel masses within 5 Myr via a roughly equal combination of filamentary gas accretion and mergers with several less massive clusters. Lowering the heavyelement abundance of the GMC by a factor of 10 reduces the opacity of the gas to radiation and better represents the highredshift formation conditions of GCs 10,11. This results in higher gas accretion leading to a mass increase of the largest cluster by a factor of ~4. When combined with simulations of less massive GMCs 12 (10 46 M ʘ), a clear relation emerges between the maximum YMC mass and the mass of the host GMC. Our results demonstrate that YMCs, and potentially GCs, are a simple extension of local cluster formation to more massive clouds and do not require suggested exotic formation scenarios 1416. Star clusters are the cradles of star formation and grow within Giant Molecular Clouds (GMCs) − large collections of turbulent molecular gas and dust with masses of 10 47 M ʘ and scale sizes typically 10-200 parsecs 17,18. Most star clusters forming in lower mass clouds (10 46 M ʘ) in the local Universe are relatively low mass (<10 4 M ʘ) 19,20 but, while rare, there are much highermass examples nearby (called Young Massive Clusters or YMCs, >10 4 M ʘ). The infrequency of local YMCs, in combination with their short < 10 Myr timescale of formation, makes their formation conditions and mechanisms uncertain. In particular, it is debated whether YMCs form from the hierarchical collapse of a single high density region of molecular gas, or whether they are assembled via the conglomeration of several distinct subclusters 2,5. Interpretations are further complicated by the extreme radiation fields produced by the many hot O stars in YMCs, which heat and ionize the gas surrounding the cluster and act to suppress star formation 6,9,12. The associated radiation pressure on dust grains can help remove the natal gas 7,8. These effects, deemed "radiative feedback", decrease the star formation efficiency within a GMC. Most previous studies of radiative feedback, however, focus on lowmass cluster formation 6,9 , so the degree to which feedback shapes the formation of YMCs is not well known. At the highmass end of star clusters but at the opposite end of the age scale are the globular clusters (GCs) − relics from an epoch of extreme star formation in the early universe (redshifts z > 2). GCs have presentday masses ~10 47 M ʘ and cover a wide range of metallicities (heavyelement abundances) that are typically subsolar (2.5 < log Z/Z ʘ < 0) 21,22. Direct observation of GC formation at high redshift is only now becoming possible 2325. This, in combination with evidence for multiple stellar populations in GCs, has resulted in numerous scenarios that invoke special conditions in the early universe to explain their formation 1416. YMCs are presentday analogs of GCs 3 , so understanding YMC formation may provide deep insights into the origin of GCs. We study the formation and assembly of YMCs during the first crucial ~5 Myr of their evolution through radiationhydrodynamic (RHD) simulations of isolated, turbulent 10 7 M ʘ GMCs with physical properties typical of those found in our Galaxy. Clouds of this mass are indeed present in the local universe, but are thought to be more abundant during the much more gasrich early universe when GCs formed 1. The imposed turbulence produces a system of dynamically evolving filaments, in which clusters form (Supplementary Videos 14). We perform two simulations: one at Solar metallicity (Z ʘ), and one at a tenth Solar metallicity (0.1 Z ʘ) that is closer to matching the early universe. Star clusters are represented via localized sink particles and their stellar content is prescribed by a subgrid model. One main parameter of the subgrid model is the density threshold for cluster formation which we take to be 10 4 cm 3 based on the observational divide between starless and starforming clumps in the local Milky Way 19. We have tested the effects of varying this parameter up to 10 6 cm 3 and find that the final total mass of the most massive cluster is little affected (see Methods and Extended Data Fig. 1). Clusters grow both by accreting gas from their host filaments and by merging with other clusters. A raytracing radiativetransfer scheme is used for heating, solving the ionization state of the gas, and inducing radiation pressure. This work represents the first time the detailed evolutionary history of massive clusters in a 10 7 M ʘ GMC with the inclusion of ionization feedback and radiation pressure has been studied. We do not include the effects of stellar winds which have been shown to reduce the star formation in young star forming regions 26. We also do not supernovae (SNe) feedback in our simulation but have verified that SNe would not significantly alter our results over the timescales considered (see Methods). Here we focus on the formation and evolution of the most massive cluster in each simulation, referred to as the YMC hereafter. The simulations show (Supplementary Videos 14) that clusters are born within filaments and move with the filamentary gas flows. The YMCs form in high column density filaments at 1.54 and 0.84 Myr for the Z ʘ and 0.1 Z ʘ GMCs (Fig. 1a and 1b). Both YMCs continue to gain mass by gas inflow from their host filaments, but they also grow by several merging events. The smaller clusters that they capture originate as distant as 21 pc (Z ʘ) and 41 pc (0.1 Z ʘ) away from the formation location of the YMC, though the average separations are 15 pc and 19 pc. The entire GMC environment, therefore, needs to be considered when tracing YMC formation. There are a total of 229 and 146 cluster particles at the end of the simulations, resembling the subclustered regions and starforming clumps in 30 Doradus 27. Examining the YMC histories reveals key details about their growth (Fig. 2). The Z ʘ cluster undergoes 9 mergers that are roughly equally spaced throughout its lifetime. Conversely, the 0.1 Z ʘ cluster participates in 23 mergers with most occurring after 3 Myr. Most mergers are with smaller clusters; the average mass ratio between the captured cluster and the YMC at time of merger is 8.2% and 9.9%. Mergers occur primarily between clusters moving within the filaments. The small clusters also grow by gas accretion, but rarely undergo mergers themselves before being absorbed by the YMC. The final total masses of the YMCs at the end of the 5 Myr run are 2.84x10 5 M ʘ (for Z ʘ) and 1.54x10 6 M ʘ (for 0.1 Z ʘ) including both their stellar and gaseous components (see Methods). Of this mass, 50% and 46% was obtained through mergers. Their final stellar masses excluding gas are 2.12x10 5 M ʘ and 8.69x10 5 M ʘ , placing them well within the range of observed YMCs and GCs. The stellar mass quoted here represents the total mass of the stars formed in the YMC. The bound
Monthly Notices of the Royal Astronomical Society, 2009
We present a new large-scale survey of the J = 3-2 12 CO emission covering 4.8 deg 2 around the Rosette Nebula. The results reveal the complex dynamics of the molecular gas in this region. We identify about 2000 compact gas clumps having a mass distribution given by dN/dM ∼ M −1.8 , with no dependence of the power-law index on distance from the central O stars. A detailed study of a number of the clumps in the inner region shows that most exhibit velocity gradients in the range 1-3 km s −1 pc −1 , generally directed away from the exciting nebula. The magnitude of the velocity gradient decreases with distance from the central O stars, and we compare the apparent clump acceleration with a photoionized gas acceleration model. For most clumps outside the central nebula, the model predicts lifetimes of a few 10 5 yr. In one of the most extended of these clumps, however, a near-constant velocity gradient can be measured over 1.7 pc, which is difficult to explain with radiatively driven models of clump acceleration. As well as the individual accelerated clumps, an unresolved limb-brightened rim lies at the interface between the central nebular cavity and the Rosette Molecular Cloud. Extending over 4 pc along the edge of the nebula, this region is thought to be at an earlier phase of disruption than the accelerating compact globules. Blueshifted gas clumps around the nebula are in all cases associated with dark absorbing optical globules, indicating that this material lies in front of the nebula and has been accelerated towards us. Redshifted gas shows little evidence of associated line-of-sight dark clouds, indicating that the dominant bulk molecular gas motion throughout the region is expansion away from the O stars. In addition, we find evidence that many of the clumps lie in a molecular ring, having an expansion velocity of 30 km s −1 and radius 11 pc. The dynamical timescale derived for this structure (∼10 6 yr) is similar to the age of the nebula as a whole (2 × 10 6 yr). The J = 3-2/1-0 12 CO line ratio in the clumps decreases with radial distance from the exciting O stars, from 1.6 at ∼8 pc distance to 0.8 at 20 pc. This can be explained by a gradient in the surface temperature of the clumps with distance, and we compare the results with a simple model of surface heating by the central luminous stars. We identify seven high-velocity molecular flows in the region, with a close correspondence between these flows and embedded young clusters or known young luminous stars. These
The Herschel view of star formation in the Rosette molecular cloud under the influence of NGC 2244
Astronomy & Astrophysics, 2010
Context. The Rosette molecular cloud is promoted as the archetype of a triggered star-formation site. This is mainly due to its morphology, because the central OB cluster NGC 2244 has blown a circular-shaped cavity into the cloud and the expanding H II-region now interacts with the cloud. Aims. Studying the spatial distribution of the different evolutionary states of all star-forming sites in Rosette and investigating possible gradients of the dust temperature will help to test the 'triggered star-formation' scenario in Rosette. Methods. We use continuum data obtained with the PACS (70 and 160 µm) and SPIRE instruments (250, 350, 500 µm) of the Herschel telescope during the Science Demonstration Phase of HOBYS. Results. Three-color images of Rosette impressively show how the molecular gas is heated by the radiative impact of the NGC 2244 cluster. A clear negative temperature gradient and a positive density gradient (running from the H II-region/molecular cloud interface into the cloud) are detected. Studying the spatial distribution of the most massive dense cores (size scale 0.05 to 0.3 pc), we find an age-sequence (from more evolved to younger) with increasing distance to the cluster NGC 2244. No clear gradient is found for the clump (size-scale up to 1 pc) distribution.
On the indeterministic nature of star formation on the cloud scale
Monthly Notices of the Royal Astronomical Society, 2018
Molecular clouds are turbulent structures whose star formation efficiency (SFE) is strongly affected by internal stellar feedback processes. In this paper, we determine how sensitive the SFE of molecular clouds is to randomized inputs in the star formation feedback loop, and to what extent relationships between emergent cloud properties and the SFE can be recovered. We introduce the YULE suite of 26 radiative magnetohydrodynamic simulations of a 10 000 solar mass cloud similar to those in the solar neighbourhood. We use the same initial global properties in every simulation but vary the initial mass function sampling and initial cloud velocity structure. The final SFE lies between 6 and 23 per cent when either of these parameters are changed. We use Bayesian mixed-effects models to uncover trends in the SFE. The number of photons emitted early in the cluster's life and the length of the cloud provide the strongest predictors of the SFE. The HII regions evolve following an analytic model of expansion into a roughly isothermal density field. The more efficient feedback is at evaporating the cloud, the less the star cluster is dispersed. We argue that this is because if the gas is evaporated slowly, the stars are dragged outwards towards surviving gas clumps due to the gravitational attraction between the stars and gas. While star formation and feedback efficiencies are dependent on non-linear processes, statistical models describing cloud-scale processes can be constructed.
2005
To complement the optical absorption-line survey of diffuse molecular gas in Paper I, we obtained and analyzed far ultraviolet H_2 and CO data on lines of sight toward stars in Cep OB2 and Cep OB3. Possible correlations between column densities of different species for individual velocity components, not total columns along a line of sight as in the past, were examined and were interpreted in terms of cloud structure. The analysis reveals that there are two kinds of CH in diffuse molecular gas: CN-like CH and CH^+-like CH. Evidence is provided that CO is also associated with CN in diffuse molecular clouds. Different species are distributed according to gas density in the diffuse molecular gas. Both calcium and potassium may be depleted onto grains in high density gas, but with different dependences on local gas density. Gas densities for components where CN was detected were inferred from a chemical model. Analysis of cloud structure indicates that our data are generally consistent ...
Star formation induced by cloud–cloud collisions and galactic giant molecular cloud evolution
Publications of the Astronomical Society of Japan, 2018
Recent radio observations towards nearby galaxies started to map the whole disk and to identify giant molecular clouds (GMCs) even in the regions between galactic spiral structures. Observed variations of GMC mass functions in different galactic environment indicates that massive GMCs preferentially reside along galactic spiral structures whereas inter-arm regions have many small GMCs. Based on the phase transition dynamics from magnetized warm neutral medium to molecular clouds, Kobayashi et al. 2017 proposes a semi-analytical evolutionary description for GMC mass functions including cloud-cloud collision (CCC) process. Their results show that CCC is less dominant in shaping the mass function of GMCs compared with the accretion of dense HI gas driven by the propagation of supersonic shock waves. However, their formulation does not take into account the possible enhancement of star formation by CCC. Radio observations within the Milky Way indicate the importance of CCC for the formation of star clusters and massive stars. In this article, we reformulate the time evolution equation largely modified from Kobayashi et al. 2017 so that we additionally compute star formation subsequently taking place in CCC clouds. Our results suggest that, although CCC events between smaller clouds are more frequent than the ones between massive GMCs, CCC-driven star formation is mostly driven by massive GMCs > ∼ 10 5.5 M ⊙ (where M ⊙ is the solar mass). The resultant cumulative CCC-driven star formation may amount to a few 10 per cent of the total star formation in the Milky Way and nearby galaxies.