We present a study of the M 83 cluster population, covering the disc of the galaxy between radii of 0.45 and 4.5 kpc. We aim to probe the properties of the cluster population as a function of distance from the galactic centre. We observe a net decline in cluster formation efficiency (Γ, i.e. the amount of star formation happening in bound clusters) from about 26% in the inner region to 8% in the outer part of the galaxy. The recovered Γ values within different regions of M 83 follow the same Γ versus star formation rate density relation observed for entire galaxies. We also probe the initial cluster mass function (ICMF) as a function of galactocentric distance. We observe a significant steepening of the ICMF in the outer regions (from −1.90 ± 0.11 to −2.70 ± 0.14) and for the whole galactic cluster population (slope of −2.18 ± 0.07) of M 83. We show that this change of slope reflects a more fundamental change of the 'truncation mass' at the high-mass end of the distribution. This can be modelled as a Schechter function of slope −2 with an exponential cut-off mass (M c ) that decreases significantly from the inner to the outer regions (from 4.00 to 0.25 × 10 5 M ⊙ ) while the galactic M c is ≈ 1.60 × 10 5 M ⊙ . The trends in Γ and ICMF are consistent with the observed radial decrease of the Σ(H 2 ), hence in gas pressure. As gas pressure declines cluster formation becomes less efficient. We conclude that the host galaxy environment appears to regulate 1) the fraction of stars locked in clusters; 2) the upper mass limit of the ICMF, consistently described by a near-universal slope −2 truncated at the high-mass end.

We study the stellar cluster population in two adjacent fields in the nearby, face‐on spiral galaxy M83 using multiwavelength Wide Field Camera 3/Hubble Space Telescope imaging. After automatic detection procedures, the clusters are selected through visual inspection to be centrally concentrated, symmetric, and resolved on the images, which allows us to differentiate between clusters and likely unbound associations. We compare our sample with previous studies and show that the differences between the catalogues are largely due to the inclusion of a large numbers of diffuse associations within previous catalogues as well as the inclusion of the central starburst region, where the completeness limit is significantly worse than in the surrounding regions. We derive the size distribution of the clusters, which is well described by a lognormal distribution with a peak at ∼2.5 pc, and find evidence for an expansion in the half‐light radius of clusters with age. The luminosity function of the clusters is well approximated by a power law with an index of −2 over most of the observed range; however, a steepening is seen at MV=−9.3 and −8.8 in the inner and outer fields, respectively. Additionally, we show that the cluster population is inconsistent with a pure power‐law mass distribution, but instead exhibits a truncation at the high‐mass end. If described as a Schechter function, the characteristic mass is 1.6 × 105 and 0.5 × 105 M⊙ for the inner and outer fields, respectively, in agreement with previous estimates of other cluster populations in spiral galaxies. Comparing the predictions of the mass‐independent disruption (MID) and mass‐dependent disruption (MDD) scenarios with the observed distributions, we find that both models can accurately fit the data. However, for the MID case, the fraction of clusters destroyed (or mass lost) per decade in age is dependent on the environment; hence, the age and mass distributions of clusters are not universal. In the MDD case, the disruption time‐scale scales with galactocentric distance (being longer in the outer regions of the galaxy) in agreement with analytic and numerical predictions. Finally, we discuss the implications of our results on other extragalactic surveys, focusing on the fraction of stars that form in clusters and the need (or lack thereof) for infant mortality.

The study of young massive clusters can provide key information for the formation of globular clusters, as they are often considered analogues. A currently unanswered question in this field is how long these massive clusters remain embedded in their natal gas, with important implications for the formation of multiple populations that have been used to explain phenomena observed in globular clusters. We present an analysis of ages and masses of the young massive cluster population of M83. Through visual inspection of the clusters, and comparison of their spectral energy distributions (SEDs) and position in colour-colour space, the clusters are all exposed (no longer embedded) by <4 Myr, most likely less, indicating that current proposed age spreads within older clusters are unlikely. We also present several methods of constraining the ages of very young massive clusters. This can often be difficult using SED fitting due to a lack of information to disentangle age-extinction degeneracies and possible inaccurate assumptions in the models used for the fitting. The individual morphology of the Hα around each cluster has a significant effect on the measured fluxes, which contributes to inaccuracies in the age estimates for clusters younger than 10 Myr using SED fitting. This is due to model uncertainties and aperture effects. Our methods to help constrain ages of young clusters include using the near-infrared and spectral features, such as Wolf-Rayet stars.

Context. Recent studies have started to cast doubt on the assumption that most stars are formed in clusters. Observational studies of field stars and star cluster systems in nearby galaxies can lead to better constraints on the fraction of stars forming in clusters. Ultimately this may lead to a better understanding of star formation in galaxies, and galaxy evolution in general. Aims. We aim to constrain the amount of star formation happening in long-lived clusters for four galaxies through the homogeneous, simultaneous study of field stars and star clusters. Methods. Using HST/ACS and HST/WFPC2 images of the galaxies NGC 45, NGC 1313, NGC 5236, and NGC 7793, we estimate star formation histories by means of the synthetic CMD method. Masses and ages of star clusters are estimated using simple stellar population model fitting. Comparing observed and modeled luminosity functions, we estimate cluster formation rates. By randomly sampling the stellar initial mass function (SIMF), we construct artificial star clusters and quantify how stochastic effects influence cluster detection, integrated colors, and age estimates. Results. Star formation rates appear to be constant over the past 10 7 −10 8 years within the fields covered by our observations. The number of clusters identified per galaxy varies, with a few detected massive clusters (M ≥ 10 5 M ) and a few older than 1 Gyr. Among our sample of galaxies, NGC 5236 and NGC 1313 show high star and cluster formation rates, while NGC 7793 and NGC 45 show lower values. We find that stochastic sampling of the SIMF has a strong impact on the estimation of ages, colors, and completeness for clusters with masses ≤10 3 −10 4 M , while the effect is less pronounced for high masses. Stochasticity also makes size measurements highly uncertain at young ages (τ 10 8 yr), making it difficult to distinguish between clusters and stars based on sizes. Conclusions. The ratio of star formation happening in clusters (Γ) compared to the global star formation appears to vary for different galaxies. We find similar values to previous studies (Γ ≈ 2%-10%), but we find no obvious relation between Γ and the star formation rate density (Σ SFR ) within the range probed here (Σ SFR ∼ 10 −3 −10 −2 M yr −1 kpc −2 ). The Γ values do, however, appear to correlate with the specific U-band luminosity (T L (U), the fraction of total light coming from clusters compared to the total U-band light of the galaxy).

Recent discoveries have put the picture of stellar clusters being simple stellar populations into question. In particular, the colormagnitude diagrams of intermediate age (1-2 Gyr) massive clusters in the Large Magellanic Cloud (LMC) show features that could be interpreted as age spreads of 100-500 Myr. If multiple generations of stars are present in these clusters then, as a consequence, young (<1 Gyr) clusters with similar properties should have age spreads of the same order. In this paper we use archival Hubble Space Telescope (HST) data of eight young massive LMC clusters (NGC 1831, NGC 1847, NGC 1850, NGC 2004, NGC 2100, NGC 2136, NGC 2157 and NGC 2249 to test this hypothesis. We analyzed the color-magnitude diagrams of these clusters and fitted their star formation history to derive upper limits of potential age spreads. We find that none of the clusters analyzed in this work shows evidence for an extended star formation history that would be consistent with the age spreads proposed for intermediate age LMC clusters. Tests with artificial single age clusters show that the fitted age dispersion of the youngest clusters is consistent with spreads that are purely induced by photometric errors. As an additional result we determined a new age of NGC 1850 of ∼100 Myr, significantly higher than the commonly used value of about 30 Myr, although consistent with early HST estimates.