On the Precision of Full-Spectrum Fitting of Simple Stellar Populations. IV. A Systematic Comparison with Results from Colour-Magnitude Diagrams

TL;DR

This study systematically compares age, metallicity, and reddening determinations from resolved and unresolved stellar data, revealing good age agreement but discrepancies in metallicity and reddening, and assessing the detection of age spreads in clusters with eMSTO features.

Contribution

It provides a direct comparison of stellar population parameters derived from CMDs and integrated spectra using different isochrones, highlighting strengths and limitations of the full-spectrum fitting method.

Findings

Good agreement in age estimates between CMD and spectral methods.

Significant discrepancies in metallicity and reddening results.

eMSTO clusters often do not show detectable age spreads in integrated spectra.

Abstract

In this fourth paper of a series on the precision of ages of stellar populations obtained through the full-spectrum fitting technique, we present a first systematic analysis that compare the age, metallicity and reddening of star clusters obtained from resolved and unresolved data (namely colour-magnitude diagrams (CMDs) and integrated-light spectroscopy) using the same sets of isochrones. We investigate the results obtained with both Padova isochrones and MIST isochrones. We find that there generally is a good agreement between the ages derived from CMDs and integrated spectra. However, for metallicity and reddening, the agreement between results from analyses of CMD and integrated spectra is significantly worse. Our results also show that the ages derived with Padova isochrones match those derived using MIST isochrones, both with the full spectrum fitting technique and the CMD fitting method. However, the metallicity derived using Padova isochrones does not match that derived using MIST isochrones using the CMD method. We examine the ability of the full-spectrum fitting technique in detecting age spreads in clusters that feature the extended Main Sequence Turn Off (eMSTO) phenomenon using two-population fits. We find that 3 out of 5 eMSTO clusters in our sample are best fit with one single age, suggesting that eMSTOs do not necessarily translate to detectable age spreads in integrated-light studies.