Shedding light on the darkest galaxiesstructure and substructures in sextans and other milky way dwarf spheroidals

Resumen

This thesis is focused on studying the structure and substructures in dwarf spheroidal galaxies (dSphs) satellites of the Milky Way (MW), mainly on the spatial distribution, but also on the internal kinematic and metallicity properties of their stellar component, from the analysis of large samples of individual stars. Due to the proximity of these small galaxies to a much larger system such as the MW, we also searched for possible tidal tails caused by their mutual interaction, which was one of the main objectives of this thesis; the presence of tidal tails would indicate departure from dynamical equilibrium, invalidating one of the main assumptions on which determinations of the dark matter content and distribution of these galaxies rely on. The dSphs studied in this thesis are Sextans, Ursa Minor (UMi) and the recently discovered Eridanus II (Eri II). Sextans was chosen because of its very low central surface brightness and much larger extent compared to other similarly luminous MW dSphs, which make it a good candidate for being tidally disturbed by the MW, but at the same time hard to study in detail, due to the difficulties of mapping the large angular extent of its stellar body and of separating its stars from the very numerous contaminants (e.g. foreground MW stars). Its study was the most extensive one. As such, it was split in two parts. In the first part we obtained in visitor mode (PI: B. McMonigal; Observer: L. Cicuéndez; 6 nights) very spatially extended CTIO/DECam g and r deep photometry (covering ~ 20 deg^2 and reaching out to ~ 2 magnitudes below the oldest main-sequence turn-off) of the galaxy and explored its structural properties, searching for possible tidal tails/debris as well. We obtained the structural parameters of both the overall stellar population and its different evolutionary phases by making use of a Bayesian Markov chain Monte Carlo (MCMC) method. By making some improvements to a literature matched filter analysis of the colour-magnitude diagram, we also decontaminated their corresponding surface density maps and looked for possible departures from axisymmetry. The outcome was that Sextans is significantly more concentrated and less extended than previous studies suggested, with no signs of tidal disturbance found down to a surface brightness of ~ 31.8 mag/arcsec^2 in V-band. In agreement with previous findings, the old and metal poor stellar components such as the Blue Horizontal Branch stars were found to be considerably more extended than the rest of evolutionary phases, while bright Blue Stragglers (BSs) are more concentrated than faint ones. Nonetheless, this different spatial distributions between bright and faint BSs is compatible with the general age and metallicity gradients found in the overall stellar population of the galaxy, and therefore BSs could have formed just by evolution of binaries and not necessarily by the disruption of a globular cluster, as suggested in the literature. In addition, we updated the membership probabilities of two literature spectroscopic catalogues by making use of an improved form of a literature "expectation maximization" technique, making them publicly available together with the photometric catalogue of point-sources used in the rest of the study. The second part of the study dealt with the subsequent discovery of clear observational signs of past accretion/merger events in Sextans. Using the same photometric catalogue and the high probable spectroscopic members from the first part of the study, we found that the spatial distribution of stars varies as a function of the colour/metallicity: while the blue (metal-poor) red giant branch and main-sequence turn-off stars have a rather round and regular distribution, the spatial distribution of the red ones (metal-rich) is much more elliptical and irregularly shaped, with the presence of a distinct "shell-like" overdensity in the northeast side. We also detected a "ring-like" feature on the line-of-sight (LOS) velocities, also visible when using indicators sensitive to the metallicity, which displays a considerably larger systemic velocity and lower mean metallicity than the rest of stars. All these features make Sextans, with a stellar mass of just ~ 5 x 10^5 solar masses, the smallest galaxy with clear observational signs of accretion to date. This fact is particularly important as it lowers the faint end of the galaxy mass function at which some predictions of the hierarchical theory of galaxy formation have been qualitatively verified. For the study of UMi we used archive CFHT/MegaCam g and r photometry instead, which were also recently published and analyzed in Muñoz et al. (2018). This photometric dataset is the most deep and spatially extended one available for this system so far (covering ~ 4 deg^2 and reaching out to ~ 2.5 magnitudes below the oldest main-sequence turn-off). UMi is another very interesting target because it has a much higher ellipticity than most of the other dSphs, lacks a radial metallicity gradient and, in addition, there have been reports of extra-tidal stars that suggest the possible presence of tidal tails, which would indicate departure from tidal equilibrium. Hence, we performed for UMi a similar analysis as in the first part of the study of Sextans. Besides, as in the case of Sextans, UMi is hard to study in detail, due to the difficulties of mapping its large angular extent and of separating its stars from the numerous contaminants (e.g. foreground MW stars). Thus, by taking advantage of the tools developed during the study of Sextans, we obtained the structural parameters of the overall stellar population of UMi and statistically decontaminated its surface density map through two different methods, searching for deviations from axisymmetry as well. In this way the obtained structural parameters were in overall agreement with previous determinations from the literature, and apart from a foreground stellar cluster already discovered in this same dataset, we also detected other internal substructures. One of these was a previously reported cold secondary peak in the stellar density that could result from a disrupted stellar cluster as based on numerical simulations, as in principle the rest of stellar over-dense clumps here detected that make the inner regions of UMi highly structured; this might complicate the analysis of the internal properties of this galaxy. We further detect one stellar overdensity close to the galactic centre that could also result from a central black hole whose possible presence was often investigated in the literature, with reported detections in both X-rays and radio. Regarding signs of tidal disturbances, they were not found down to a surface brightness limit of ~ 32 mag/arcsec^2 in V-band. Finally, we also calculated the membership probabilities of a literature spectroscopic catalogue and analyzed the metallicity and LOS velocity distributions of the probable UMi's members, not finding any radial gradient either on the individual metallicities or in the velocity dispersion. Finally, we studied the case of Eri II and its candidate star cluster recently reported in the literature, because the existence of such cluster would have important implications: it would favour a cored dark matter halo for the galaxy over a cusped one, it would constrain the existence of massive compact halo objects with masses larger than ~ 5 solar masses, it would extend the empirical relationship between the number of hosted clusters and the mass of the host galaxy about two orders of magnitude below the previous limit in galactic stellar mass, and it would imply that some MW star clusters could be hosted in undetected halos with extremely faint stellar counterparts. We studied this system by making use of Magellan Clay/MegaCam and very deep ACS/WFC photometry, in order to better characterize the structural parameters of the putative cluster. By fitting its structural parameters with the MCMC method used for UMi and Sextans, they turned out to be all compatible with the ones of Eri II, which led to the hypothesis that this feature could be just the central cusp of the galactic surface density profile. Besides, the available observational evidence and a fitted ellipticity equal to 0.4, which would make this cluster the most elliptical one known to date with all the rest having ellipticities lower than 0.3, suggest it would be a very unusual star cluster in case it would not be just the central cusp of the stellar density profile of Eri II.