[Paper Review] Radiative hydrodynamics simulations of red supergiant stars: I. interpretation of interferometric observations

A. Chiavassa, B. Plez+2 more|arXiv (Cornell University)|Jul 10, 2009

Stellar, planetary, and galactic studiesPhysics and Astronomy31 references93 citations

TL;DR

This study uses 3D radiative-hydrodynamics simulations with CO 5 BOLD to model convection in red supergiant stars and interpret interferometric observations of Betelgeuse (α Ori). It demonstrates that granulation patterns from these simulations produce visibility curves and closure phases that match observed data far better than limb-darkened disk models, confirming the presence of large convective cells on the star’s surface.

ABSTRACT

Context. It has been suggested that convection in Red Supergiant (RSG) stars gives rise to large-scale granules causing observable surface inhomogeneities. This convection is also extremely vigorous, and suspected to be one of the causes of mass-loss in RSGs. It must thus be understood in details. Evidence has been accumulated that there are asymmetries in the photospheres of RSGs, but detailedstudies of granulation are still lacking. Interferometric observations offer an exciting possibility to tackle this question, but they are still often interpreted using smooth symmetrical limb-darkened intensity distributions, or very simple spotted ad hoc models. Aims. We explore the impact of the granulation on visibility curves and closure phases using the radiative transfer code OPTIM3D. We simultaneously assess how 3D simulations of convection in RSG with CO5BOLD can be tested against these observations. Methods. We use 3D radiative-hydrodynamics (RHD) simulations of convection to compute intensity maps at various wavelengths and time, from which we derive interferometric visibility amplitudes and phases. We study their behaviour with time, position angle, and wavelength, and compare them to observations of the RSG alpha Ori Results. We provide average limb-darkening coefficients for RSGs. We detail the prospects for the detection and characterization of granulation (contrast, size) on RSGs. We demonstrate that our RHD simulations provide an excellent fit to existing interferometric observation of alpha Ori, contrary to limb darkened disks. This confirms the existence of large convective cells on the surface of Betelgeuse.

Motivation & Objective

• To investigate how surface granulation in red supergiant stars affects interferometric visibility curves and closure phases.
• To test whether 3D radiative-hydrodynamics simulations can reproduce observed interferometric data of α Ori (Betelgeuse) better than traditional limb-darkened disk models.
• To assess the detectability of granulation features (size, contrast, timescale) using current and future interferometric instruments.
• To evaluate the limitations of current simulations, particularly the use of grey radiative transfer and limited spatial resolution, and their impact on observable signatures.
• To provide a framework for using high-spatial-frequency interferometric measurements to characterize stellar convection in red supergiants.

Proposed method

• 3D radiative-hydrodynamics (RHD) simulations of a 12 M⊙ red supergiant were performed using the CO 5 BOLD code with a star-in-a-box configuration and open boundary conditions.
• The simulations include radiative transfer in LTE using a grey Rosseland mean opacity, with luminosity and gravity derived from the model’s average structure over time.
• Intensity maps were generated from the RHD simulations at various wavelengths and temporal snapshots to compute interferometric visibility amplitudes and closure phases.
• The visibility curves and closure phases were compared to actual interferometric observations of α Ori, particularly from the VLTI and CHARA arrays.
• The simulations were analyzed across different spatial frequencies, position angles, and wavelengths to assess detectability of granulation features.
• The impact of non-grey opacities and higher resolution was evaluated by comparing results to previous grey simulations, noting that non-grey effects would reduce temperature and intensity contrasts.

Experimental results

Research questions

• RQ1 Can 3D RHD simulations of red supergiant convection reproduce the observed visibility curves and closure phases of α Ori better than limb-darkened disk models?
• RQ2 What is the detectable signature of large-scale granulation in interferometric data, and at what spatial frequencies and wavelengths is it most prominent?
• RQ3 How do visibility variations with time, position angle, and wavelength reflect the properties of convective cells (size, contrast, timescale) on red supergiant surfaces?
• RQ4 To what extent do the approximations in current RHD simulations—particularly grey radiative transfer and limited resolution—affect the predicted interferometric observables?
• RQ5 What observational strategies (e.g., multi-epoch, multi-configuration, high-resolution spectral measurements) are most effective for detecting and characterizing granulation in red supergiants?

Key findings

• The 3D RHD simulations provide an excellent fit to the observed visibility amplitudes and closure phases of α Ori, significantly outperforming limb-darkened disk models.
• The simulations confirm the existence of large convective cells on Betelgeuse’s surface, with characteristic sizes on the order of 100–200 R⊙, consistent with observed angular scales.
• Average limb-darkening coefficients for red supergiants were derived from the simulations, providing improved atmospheric models for interferometric interpretation.
• Visibility variations with position angle and time are detectable with current interferometers (e.g., VLTI, CHARA) if measurement errors are kept below 10%, especially in the second, third, and fourth visibility lobes.
• High spectral resolution observations (1% error) of visibility as a function of wavelength can detect correlations with spectral features, revealing changes in radius, limb-darkening, or granulation pattern.
• The use of grey radiative transfer in simulations leads to higher temperature and intensity contrasts than in non-grey models; thus, non-grey effects would reduce visibility fluctuations, implying that current simulations may overestimate the amplitude of granulation signatures.