A grid of synthetic spectra for the study of multiple populations on globular clusters Vinicius Branco 1 , Paula Coelho 1 , and Ariane Lan¸ con 2 . Instituto de Astronomia, Geof´ ısica e Ciˆ encias Atmosf´ ericas, Sao Paulo, Brazil. 1 Observatoire astronomique de Strasbourg, Strasbourg, France. 2 Email: vbranco@usp.br Summary Globular Clusters are challenging and fascinating astronomy objects. Once understood to be simple stellar populations, it is now known that stars in massive GCs present chemical variations not expected by the current theoretical predictions. In this work, we have calculated stellar spectra to study the integrated properties of GCs, considering three iron metallicities and two generation of stars — a primordial one following the pattern of galactic halo stars and a second one with anti-correlated CNONa abundances. We show that the modelled stellar population spectra present signatures of the CNONa anti-correlations, thus affecting our interpretation of the integrated spectrum of a GC and the Lick indices. Before applying our models to extragalactic environments, where only the integrated spectra of GCs can be accessed, we are calibrating our stellar spectra grid to a sample of metal-poor empirical spectra drawn from the X-Shooter Library. This will allow us to construct a more reliable grid of synthetic spectra to study the multiple populations on GCs. Introduction Globular Clusters (GCs) are invaluable to advance our knowledge of both galactic and extragalactic astronomy. The last decade is seeing a change of paradigm regarding these systems: once understood to be simple stellar populations, now evidence accumulates that nearly all massive Galactic GCs harbour multiple populations (MPs) of stars. Yet the stellar populations models to date often still study these systems on the basis of simple stellar populations. Our work aims at improving stellar population models of integrated spectra by incorporating what is already known in terms of the chemical mix- tures of the multiple populations. For that, we build on our previous work and in synergy with the POPSYCLE collaboration [1]. We believe this will allow us to better exploit the data of extragalactic GCs, expected to be observed in the coming years. Methodology Our recent work Branco (2020, [2], [3], “B20 ”) and Branco et al. (in prep ), followed up on the first efforts by Coelho et al. (2011, “C11 ”, [4]) in understanding the effect of MPs in the integrated light of GCs. To calculate the models and the synthetic spec- tra, we used ATLAS12 and SYNTHE codes [5], with the Sun to calibrate our atomic and molecular ingredients. We produced two stellar population synthesis (SPS) models, covering three metallicities. For each [Fe/H] value, we calculated α-enhanced models representing a first generation of stars and a α- and helium-enhanced second generation including the observed chemical abundance variations on C, N, O and Na on stars of Galactic GCs. Figure 1 shows how the SPS integrated spectra contrast to each other at various metallicities, as well as the residual flux difference between first and second generation spectra. Figure 1. SPS integrated spectra representing the first (purple) and second (yellow) generation of stars for three metallicities (top panel) and their residual flux difference (i.e., first generation flux minus second generation flux, bottom panel). Results For a metallicity of [Fe/H] = -0.7, C11 found that the Lick indices appreciably affected by the chemical abundance anticorrelations, as expected by the presence of MPs, were CN 1 , CN 2 , Ca4227,G4300 and NaD. Later on, B20 produced a grid of stellar spectra expanding the metallicity coverage of C11 towards less and more metallic systems. The effects on the Balmer lines were confirmed in B20, and for the first time the influence of the MPs on those Lick indices were found in different metallicity regimes. Conclusion and Perspectives We have confirmed the results by C11 and expanded the metallicity coverage of that study in B20. Now, as part of the POPSYCLE collaboration, we are computing models mimicking the stellar parameters of a sub-sample of the X-Shooter Library (XSL) (Figure 2) [6]. We will calibrate the atomic opacities to this empirical sample, aiming at building a reliable synthetic-XSL grid. With improved stellar population models based on this new grid, we expect to be able to study the multiple populations in the integrated light of extragalactic GCs. Figure 2. CMD of the XSL-DR2 (gray dots) and the selected observations for stellar modeling (purple dots). Size and hue indicate the metallicity of the star. References [1] POPSYCLE, “Population Synthesis for Clusters and Galaxies (supported by ANR-19-CE31-0022, France),” [2] V. Branco and P. Coelho, BAAA, vol. 61C, pp. 85–85, 2020. [3] V. Branco, M.S. thesis, IAG/USP, doi:10.11606/D.14.2020.tde-15122020-164651, 2020. [4] P. Coelho, S. M. Percival, and M. Salaris, ApJ, vol. 734, no. 1, p. 72, 2011. [5] R. Kurucz, “ATLAS12, SYNTHE, ATLAS9, WIDTH9, et cetera,” MSAIt, 2005. [6] A. Gonneau et al., A&A, vol. 634, A133, 2020.