Ages and masses of 0.64 million red giant branch stars from LAMOST Survey

Dr. Ya-Qian Wu of National Astronomical Observatories of Chinese Academy of Sciences (NAOC) presents a catalog of stellar age and mass estimated for a sample of 640986 red giant branch (RGB) stars of the Galactic disk from the LAMOST Galactic Spectroscopic Survey (DR4).


This work has been published online in the Monthly Notices of the Royal Astronomical Society on February 16th, 2019 at


Understanding the formation and evolution of the galactic disk is one of the most vital problems in the field of galaxy formation and evolution. As our host galaxy, the Milky Way is a typical disk galaxy and is also the only galactic disk whose individual stars can be fully resolved. A detailed characterizing for the structure and chemo-dynamic properties of the Galactic disk is thus of vital importance to understand the formation and evolution of galactic disk(s). In recent years, a series of space high-precision photometry surveys and ground large-scale spectral surveys provides significant opportunities for studying the chemo-dynamic properties of the Galactic disk. In the fourth data release (DR4), the LAMOST Galactic Survey released more than 6.5 million spectra observed with LAMOST by June 2016, including 1 million giant stellar spectra. Delivering reliable age estimates for a large sample of red giant stars is timely for the study of Galactic structure and archaeology.


In the Teff–log g diagram, RGB and RC stars have overlaps due to the metallicity effect. Since the metallicity determinations from the spectra have considerable errors, a simple cut in the Teff–log g diagram to distinguish RC and RGB stars may lead to considerable contamination or undesired patterns. In this work, we distinguish RGB and RC utilizing period spacing derived from the spectra with a machine-learning method based on kernel principal component analysis (KPCA). Cross-validation suggests our method is capable of distinguishing RC from RGB stars with an only 2 per cent contamination rate for stars with signal-to-noise ratio (SNR) higher than 50. The ages and masses of these RGB stars are determined from their LAMOST spectra with the KPCA method by taking the LAMOST-Kepler giant stars having asteroseismic parameters and the LAMOST-TGAS sub-giant stars based on isochrones as training sets. Examination suggests that the age and mass estimates of our RGB sample stars with SNR > 30 have a median error of 30 per cent and 10 per cent, respectively. Stellar ages are found to exhibit positive vertical and negative radial gradients across the disk.


Dr. Maosheng Xiang from Max-Planck Institute for Astronomy and Prof. Gang Zhao at NAOC have contributed to this work as key co-authors. This work was supported by Joint Research Fund in Astronomy of NSFC and Chinese Academy of Sciences.



Fig. Colour-coded distribution of the median age for stars of different spatial bins in the R–Z plane. The adopted bin size is 0.25 kpc in the Rdirection and 0.1 kpc in the Z-direction.

Related accessories
Related documents
Copyright © National Astronomical Observatories, Chinese Academy of Sciences
Address: 20A Datun Road, Chaoyang District, Beijing, China code: 100012
Tel: 010-64888708 E-mail: