[Paper Review] Probing the mass-loss history of AGB and red supergiant stars from CO rotational line profiles I. Theoretical model -- Mass-loss history unravelled in VY CMa

L. Decin, S. Hony+4 more|UvA-DARE (University of Amsterdam)|Jun 13, 2006

Astrophysics and Star Formation Studies66 references99 citations

TL;DR

This paper presents a non-LTE radiative transfer code that models CO rotational line profiles to reconstruct the mass-loss history of AGB and red supergiant stars. Applied to VY CMa, it reveals a high-mass-loss episode of ~3.2×10⁻⁴ M⊙ yr⁻¹ lasting ~100 years ~1000 years ago, preceded by 800 years of low mass loss (~1×10⁻⁶ M⊙ yr⁻¹), with current mass loss at ~8×10⁻⁵ M⊙ yr⁻¹.

ABSTRACT

Context: Mass loss plays a dominant role in the evolution of low mass stars while they are on the Asymptotic Giant Branch (AGB). The gas and dust ejected during this phase are a major source in the mass budget of the interstellar medium. Recent studies have pointed towards the importance of variations in the mass-loss history of such objects. Aims: By modelling the full line profile of low excitation CO lines emitted in the circumstellar envelope, we can study the mass-loss history of AGB stars. Methods: We have developed a non-LTE radiative transfer code, which calculates the velocity structure and gas kinetic temperature of the envelope in a self-consistent way. The resulting structure of the envelope provides the input for the molecular line radiative calculations which are evaluated in the comoving frame. The code allows for the implementation of modulations in the mass-loss rate. This code has been benchmarked against other radiative transfer codes and is shown to perform well and efficiently. Results: We illustrate the effects of varying mass-loss rates in case of a superwind phase. The model is applied to the well-studied case of VY CMa. We show that both the observed integrated line strengths as the spectral structure present in the observed line profiles, unambiguously demonstrate that this source underwent a phase of high mass loss (~ 3.2E-4 Msun/yr) some 1000 yr ago. This phase took place for some 100 yr, and was preceded by a low mass-loss phase (~ 1E-6 Msun/yr) taking some 800 yr. The current mass-loss rate is estimated to be in the order of 8E-5 Msun/yr. Conclusions: In this paper, we demonstrate that both the relative strength of the CO rotational line profiles and the (non)-occurrence of spectral structure in the profile offer strong diagnostics to pinpoint the mass-loss history.

Motivation & Objective

• To develop a non-LTE radiative transfer code that self-consistently models velocity structure, kinetic temperature, and molecular line emission in circumstellar envelopes.
• To use CO rotational line profiles as diagnostics for time-variable mass-loss histories in AGB and red supergiant stars.
• To test the model on VY CMa, a well-studied red supergiant with complex nebular structures.
• To determine whether spectral line shapes and strengths can uniquely constrain past mass-loss rates, including episodic events.
• To benchmark the code against existing models and validate its accuracy and efficiency.

Proposed method

• The code solves the energy and momentum equations for gas dynamics in the circumstellar envelope, including heating and cooling processes.
• It computes the kinetic temperature structure self-consistently, accounting for dust-gas coupling, radiative heating, and cooling by CO rotational transitions.
• Molecular line emission is calculated in the comoving frame using non-LTE level populations, with CO rotational transitions up to J=7–6.
• The model allows for time-dependent mass-loss rates, enabling simulation of episodic or modulated mass loss.
• The code is benchmarked against established radiative transfer codes to validate its performance and accuracy.
• The model is applied to observed CO line profiles of VY CMa across multiple transitions (J=1–0 to J=7–6).

Experimental results

Research questions

• RQ1 Can CO rotational line profiles uniquely constrain the time-dependent mass-loss history of AGB and red supergiant stars?
• RQ2 What physical conditions (temperature, velocity structure) are required to reproduce the observed CO line shapes in VY CMa?
• RQ3 Did VY CMa experience a high-mass-loss episode in the past, and if so, what were its duration and rate?
• RQ4 How do variations in mass-loss rate affect the relative strengths and spectral structure of CO lines?
• RQ5 Is the observed nebular morphology in VY CMa consistent with the inferred mass-loss history from line profile analysis?

Key findings

• VY CMa experienced a high-mass-loss episode ~1000 years ago with a rate of ~3.2×10⁻⁴ M⊙ yr⁻¹, lasting approximately 100 years.
• This high-mass-loss phase was preceded by 800 years of low mass loss at a rate of ~1×10⁻⁶ M⊙ yr⁻¹.
• The current mass-loss rate is estimated at ~8×10⁻⁵ M⊙ yr⁻¹, significantly lower than the peak episode.
• The model shows that both the relative strengths of CO lines and the presence of spectral structure (e.g., double-peaked features) are strong diagnostics of past mass-loss variations.
• The derived mass-loss history is consistent with Hubble and near-IR observations showing nebular arcs and condensations at 1′′–7′′ from the star.
• The results demonstrate that assuming constant mass loss leads to a significantly incomplete picture of the star’s evolutionary history.