[Paper Review] The Upper Initial Mass Function from Ultraviolet Spectral Lines

Claus Leitherer|arXiv (Cornell University)|Sep 1, 2010

Stellar, planetary, and galactic studies3 citations

TL;DR

This paper presents a new theoretical spectral library of massive stars based on radiation-hydrodynamical wind models to interpret ultraviolet spectral lines in integrated stellar populations. Using these models, the authors find no evidence for environmental dependence of the initial mass function (IMF), supporting a near-universal Salpeter-type IMF, while identifying stellar rotation and binary evolution as key missing factors in current models.

ABSTRACT

The space-ultraviolet wavelength region contains strong spectral lines from massive, hot stars. These features form in winds and are sensitive to luminosity and mass, and ultimately provide constraints on the initial mass function. New radiation-hydrodynamical models of stellar winds are used to construct a theoretical spectral library of massive stars for inclusion in population synthesis. The models are compared to observations of nearby star clusters, of starburst regions in local galaxies, and of distant star-forming galaxies. The data are consistent with a near-universal Salpeter-type initial mass function. We find no evidence of environmental effects on the initial mass function. Some model deficiencies are identified: stellar rotation and binary evolution are not accounted for and may become increasingly important in metal-poor systems.

Motivation & Objective

• To develop a theoretical spectral library of massive stars using radiation-hydrodynamical wind models for improved population synthesis.
• To use ultraviolet wind lines as direct tracers of massive star populations, avoiding uncertainties from nebular recombination and radiative transfer.
• To test whether the initial mass function (IMF) varies with environment by comparing models to observed UV spectra of star clusters, starbursts, and high-redshift galaxies.
• To identify limitations in current models, particularly the omission of stellar rotation and binary evolution effects.

Proposed method

• Employed the WM-Basic code for non-LTE, spherically extended, blanketed radiation-hydrodynamics calculations of hot star atmospheres and winds.
• Generated synthetic spectra from 900 to 3000 Å with 0.4 Å resolution, covering masses >5 M⊙ and metallicities from 0.05 to 2×Z⊙.
• Used P Cygni profiles in UV wind lines (e.g., C IV λ1550, Si IV λ1400, N V λ1240) as diagnostics of stellar luminosity and mass.
• Compared model spectra to observed UV data from nearby star clusters, local starbursts, and high-redshift Lyman break galaxies (LBGs).
• Treated dust reddening and chemical composition as external inputs, focusing on IMF and star-formation history as free parameters.
• Analyzed both cluster and field stellar populations to disentangle dynamical evolution effects from potential IMF variations.

Experimental results

Research questions

• RQ1 Does the initial mass function (IMF) of massive stars vary with environment, as suggested by some high-redshift galaxy observations?
• RQ2 Can ultraviolet wind lines in integrated spectra provide a robust, direct census of massive stars without relying on nebular emission proxies?
• RQ3 To what extent do current population synthesis models fail due to missing physics such as stellar rotation and binary evolution?
• RQ4 Why do some high-redshift Lyman break galaxies show anomalously strong He II λ1640 emission inconsistent with standard IMF and evolution models?
• RQ5 How do dynamical processes like cluster dissolution affect the observed IMF in field versus cluster environments?

Key findings

• The observed UV spectra of nearby star clusters, local starbursts, and high-redshift galaxies are consistent with a near-universal Salpeter-type initial mass function, with no significant environmental dependence.
• UV wind lines such as C IV λ1550, Si IV λ1400, and N V λ1240 are strong, stable features that trace stellar mass and luminosity, enabling direct IMF constraints.
• The observed deficit of strong wind lines in field stellar populations compared to clusters is better explained by age differences due to cluster dissolution than by a top-heavy IMF.
• The extremely strong He II λ1640 emission in high-redshift galaxy BX418 cannot be reproduced by standard models and suggests either an unreasonably young age, a top-heavy IMF, or missing physics such as rotation or binary evolution.
• Stellar rotation and binary evolution are identified as critical missing components in current models, as they significantly alter luminosity, temperature, and mass-loss rates in massive stars.
• The models show that wind lines are more reliable IMF diagnostics than nebular Hα because they form in the stellar wind itself, avoiding uncertainties from interstellar gas path lengths and radiative transfer.