The STAGGER-grid: A grid of 3D stellar atmosphere models

V. Synthetic stellar spectra and broad-band photometry^★

¹ Université Côte d’Azur, Observatoire de la Côte d’Azur, CNRS, Lagrange, CS 34229, 06300 Nice, France
e-mail: andrea.chiavassa@oca.eu
² Research School of Astronomy & Astrophysics, Australian National University, Cotter Road, Weston, ACT 2611, Australia
³ Stellar Astrophysics Centre, Department of Physics and Astronomy, Aarhus University, Ny Munkegade 120, 8000 Aarhus C, Denmark
⁴ Niels Bohr Institute, University of Copenhagen, Juliane Maries Vej 30, 2100 Copenhagen, Denmark
⁵ Centre for Star and Planet Formation, Natural History Museum of Denmark, University of Copenhagen, Øster Voldgade 5-7, 1350 Copenhagen, Denmark

Received: 20 October 2017
Accepted: 4 January 2018

Abstract

Context. The surface structures and dynamics of cool stars are characterised by the presence of convective motions and turbulent flows which shape the emergent spectrum.

Aims. We used realistic three-dimensional (3D) radiative hydrodynamical simulations from the STAGGER-grid to calculate synthetic spectra with the radiative transfer code OPTIM3D for stars with different stellar parameters to predict photometric colours and convective velocity shifts.

Methods. We calculated spectra from 1000 to 200 000 Å with a constant resolving power of λ∕Δλ = 20 000 and from 8470 and 8710 Å (Gaia Radial Velocity Spectrometer – RVS – spectral range) with a constant resolving power of λ∕Δλ = 300 000.

Results. We used synthetic spectra to compute theoretical colours in the Johnson-Cousins UBV (RI)_C, SDSS, 2MASS, Gaia, SkyMapper, Strömgren systems, and HST-WFC3. Our synthetic magnitudes are compared with those obtained using 1D hydrostatic models. We showed that 1D versus 3D differences are limited to a small percent except for the narrow filters that span the optical and UV region of the spectrum. In addition, we derived the effect of the convective velocity fields on selected Fe I lines. We found the overall convective shift for 3D simulations with respect to the reference 1D hydrostatic models, revealing line shifts of between −0.235 and +0.361 km s⁻¹. We showed a net correlation of the convective shifts with the effective temperature: lower effective temperatures denote redshifts and higher effective temperatures denote blueshifts. We conclude that the extraction of accurate radial velocities from RVS spectra need an appropriate wavelength correction from convection shifts.

Conclusions. The use of realistic 3D hydrodynamical stellar atmosphere simulations has a small but significant impact on the predicted photometry compared with classical 1D hydrostatic models for late-type stars. We make all the spectra publicly available for the community through the POLLUX database.