[Paper Review] The VLT-FLAMES Tarantula Survey - XI. A census of the hot luminous stars and their feedback in 30 Doradus

E. Doran, P. A. Crowther+17 more|UvA-DARE (University of Amsterdam)|Aug 15, 2013

Stellar, planetary, and galactic studiesPhysics and Astronomy115 references72 citations

TL;DR

This study presents the first comprehensive census of hot luminous stars in the 30 Doradus star-forming region using multi-epoch spectroscopy from the VLT-FLAMES Tarantula Survey. It reveals that massive stars—particularly those with initial masses >100 M⊙ and W-R/Of/WN types—contribute disproportionately to ionizing radiation and stellar wind feedback, with ~6% of ionizing photons escaping the region, and shows that standard models like Starburst99 underestimate feedback by factors of 2–9.

ABSTRACT

We compile the first comprehensive census of hot luminous stars in the 30 Doradus (30 Dor) star forming region of the LMC. The census extends to a radius of 10arcmin (150pc) from the central cluster, R136. Stars were selected photometrically and combined with the latest spectral types. 1145 candidate hot luminous stars were identified of which >700 were considered genuine early type stars that contribute to feedback. We assess the spectroscopic completeness to be 85% in outer regions (>5pc) but fall to 35% in the vicinity of R136, giving a total of 500 hot luminous stars with spectroscopy. Stellar calibrations and models were used to obtain their physical parameters before integrated values were compared to global observations and the population synthesis code, Starburst99. The 31 W-R and Of/WN stars made large contributions to the total ionising and wind luminosities of ~40% and ~50%, respectively. Stars with Minit>100Msun also showed high contributions to the global feedback, ~25% in both cases. Such massive stars are not accounted for by the current Starburst99 code, which underestimated the ionising and wind luminosities of R136 by factors of ~2 and ~9, respectively. The census inferred a SFR of 0.073+/-0.04Msun/yr for 30 Dor, typically higher than results from popular SFR calibrations. However, it remained consistent with a far-UV luminosity tracer and a combined Halpha and mid-infrared tracer, but only after correcting for Halpha extinction. The global ionising output exceeded measurements from the associated gas and dust, suggesting ~6(+55/-6)% of ionising photons escape the region. When studying very luminous star forming regions, it is therefore essential to include the most massive stars to ensure a reliable energy budget. If 30 Dor is typical of other large star forming regions, estimates of the SFR will be underpredicted if this escape fraction is not accounted for.(abridged)

Motivation & Objective

• To compile a complete census of hot luminous stars in the 30 Doradus region, focusing on early-type and massive stars.
• To assess spectroscopic completeness and identify the most massive and luminous stars, including W-R and Of/WN types.
• To measure integrated ionizing luminosity and stellar wind luminosity to re-evaluate the star formation rate (SFR) and ionizing photon escape fraction.
• To test the accuracy of population synthesis models like Starburst99 in reproducing observed feedback in extreme H II regions.
• To determine the contribution of the most massive stars (>100 M⊙) to the total energy budget of the region.

Proposed method

• Combined photometric selection with multi-epoch spectroscopy from the VLT-FLAMES Tarantula Survey to identify hot luminous stars.
• Applied stellar atmosphere models and calibrations (e.g., Martins & Pleijel 2006) to derive physical parameters from spectral types and photometry.
• Used extinction laws (R_V = 3.5–4.2) derived from K-band photometry to correct for dust reddening and estimate intrinsic magnitudes.
• Calculated integrated ionizing luminosity and wind luminosity from individual stellar contributions and compared to global UV to FIR observations.
• Validated results against population synthesis models (Starburst99) and observational tracers (far-UV, Hα, mid-IR).
• Estimated the ionizing photon escape fraction by comparing total stellar output to gas-phase ionization measurements.

Experimental results

Research questions

• RQ1 What is the complete census of hot luminous stars in the 30 Doradus region, particularly in the dense R136 cluster?
• RQ2 How complete is the spectroscopic sampling of early-type stars, especially in the crowded central region?
• RQ3 What fraction of the total ionizing luminosity and wind luminosity is contributed by the most massive stars (>100 M⊙) and W-R/Of/WN stars?
• RQ4 To what extent do standard population synthesis models like Starburst99 underestimate the feedback from massive stars in extreme H II regions?
• RQ5 What is the global ionizing photon escape fraction from 30 Doradus, and how does it affect SFR estimates?

Key findings

• The census identified 1,145 candidate hot luminous stars, with 500 confirmed via spectroscopy, of which >700 are genuine early-type stars.
• Spectroscopic completeness drops to 35% near R136 but reaches 85% at >5 pc, indicating significant incompleteness in the cluster core.
• W-R and Of/WN stars, though only 31 in number, contribute ~40% to the total ionizing luminosity and ~50% to the wind luminosity.
• Stars with initial masses >100 M⊙ (mostly H-rich WN stars) contribute ~25% to both ionizing luminosity and wind luminosity.
• Starburst99 underestimates the ionizing luminosity of R136 by a factor of ~2 and the wind luminosity by a factor of ~9, due to missing very massive stars.
• The inferred star formation rate is 0.073 ± 0.04 M⊙ yr⁻¹, consistent with far-UV and combined Hα + mid-IR tracers only after correcting for Hα extinction, and implies a photon escape fraction of ~6% (6⁺⁵⁵₋₆%), indicating significant ionizing output leakage.