Globular clusters as cosmic clocks: New cosmological hints from their integrated light

¹ Dipartimento di Fisica e Astronomia “Augusto Righi”–Università di Bologna, Via Piero Gobetti 93/2, I-40129 Bologna, Italy
² INAF – Osservatorio di Astrofisica e Scienza dello Spazio di Bologna, Via Piero Gobetti 93/3, I-40129 Bologna, Italy
³ INAF – Osservatorio Astrofisico di Arcetri, Largo E. Fermi 5, I-50125 Firenze, Italy
⁴ ICC, University of Barcelona, Martí i Franquès, 1, E08028 Barcelona, Spain
⁵ ICREA, Pg. Lluis Companys 23, Barcelona 08010, Spain

^⋆ Corresponding author; elena.tomasetti2@unibo.it

Received: 30 October 2024
Accepted: 10 February 2025

Abstract

Aims. In this work, we explore the reliability and robustness in measuring the ages and main physical properties of a sample of old Milky Way globular clusters (GCs) from their integrated light. This approach sets the stage for using GCs as cosmic clocks at high redshift. Additionally, it enables us to establish an independent lower limit on the age of the Universe, and an upper limit on H₀.

Methods. We analysed a sample of 77 GCs from the WAGGS project, by first measuring their spectral features (Lick indices and spectroscopic breaks) with PyLick and then performing full spectral fitting with BAGPIPES. The analysis of Lick indices offers an initial estimate of the population’s age and metallicity, generally aligning well with values reported in the literature. However, it also highlights a subset of old clusters for which we estimate younger ages. This discrepancy is primarily attributed to the presence of horizontal branches (HBs) with complex morphologies, which are not accounted for in the stellar population models. With full spectral fitting we measured the GCs’ ages, metallicities, and masses, testing how removing the cosmological prior on the ages affects the final results.

Results. Compared to isochrone fitting estimates, ages are best recovered when the cosmological prior is removed, with a 20% increase in the number of GCs showing ages compatible with literature values within ±1.5 Gyr. The derived metallicity and mass are consistently in good agreement with the reference values, regardless of HB morphology, [Z/H], or the fit settings. The average discrepancies across the entire sample are ⟨Δ[Z/H]⟩ = −0.02 ± 0.24 dex for metallicity and ⟨Δlog(M_⋆/M_⊙)⟩ = 0.04 ± 0.28 dex for mass. Metal-rich GCs ([Z/H] ≥ −0.4) showing a red HB (with morphological parameter HBR > 0) are the sub-group in which ages are best recovered. In this group, 70% of the results align with literature values within ±1.5 Gyr. Identifying the tail of the oldest cosmology-independent ages with a Gaussian mixture model, we obtained a sample of 24 objects with ⟨age⟩ = 13.4 ± 1.1 Gyr.

Conclusions. Being a natural lower limit on the age of the Universe, we used the age of the oldest GCs to constrain the Hubble constant, obtaining H₀ = 70.5_−6.3^+7.7 km s⁻¹ Mpc⁻¹ (stat+syst) when a flat ΛCDM with Ω_m = 0.30 ± 0.02 (based on low-z measurements) was assumed. Validating the analysis of GCs based on their integrated light lays the foundation for extending this type of study to high redshift, where GCs have begun to appear in lensed fields, thanks to JWST.