The Apache Point Observatory extra-Galactic Evolution Experiment (APOeGEE): Chemical Abundance Trends for Seven Dwarf Spheroidal Galaxies in the APOGEE Survey

TL;DR

This study harnesses APOGEE DR17 data to derive homogeneous chemical abundances for seven MW dwarf spheroidal galaxies, introducing a robust upper-limit framework to address weak, metal-poor lines in low-S/N spectra. By combining APOGEE radial velocities with Gaia EDR3 proper motions, the authors assemble a 6-sigma membership sample of 731 kinematic candidates, of which 518 giants yield reliable [Fe/H] and [Mg/Fe] measurements across multiple elements after applying carefully calibrated upper limits. The resulting abundance patterns reveal two broad chemical evolution pathways—episodic bursts and continuous star formation—with alpha-element knees occurring at very low metallicities, and they show that many dSphs contribute little to the MW halo at [Fe/H] > -2.0. Comparisons with high-resolution literature, after systematic offsets, largely validate the APOGEE trends while highlighting bursty histories in Carina and Fornax and more quiescent histories in Sculptor and Draco. Overall, the work demonstrates how combining homogeneous APOGEE data with upper-limit treatments enables detailed inferences about the SFHs and environment-driven evolution of dwarf galaxies, with implications for understanding the assembly of the Milky Way halo.

Abstract

In addition to comprehensive surveys of the Milky Way bulge, disk, and halo, the Apache Point Galactic Evolution Experiment (APOGEE) project observed seven dwarf spheroidal satellites (dSphs) of the Milky Way: Carina, Sextans, Sculptor, Draco, Ursa Minor, Bootes 1, and Fornax. APOGEE radial velocities, stellar parameters, and Gaia EDR3 proper motions are used to identify member stars in the vicinity of each dwarf. To properly analyze the abundance patterns of these galaxies, a novel procedure was developed to determine the measurable upper limits of the APOGEE chemical abundances as a function of the effective temperature and the spectral signal-to-noise ratio. In general, the APOGEE abundance patterns of these galaxies (limited to [Fe/H] $>$ -2.5) agree with those found in high-resolution optical studies after abundance offsets are applied. Most of the galaxies studied have abundance patterns that are distinctly different from the majority of stars found in the MW halo, suggesting that these galaxies contributed little to the MW halo above [Fe/H] $>$ -2.0. From these abundance patterns, we find that these dSphs tend to follow two types of chemical evolution paths: episodic and continuous star formation, a result that is consistent with previous photometric studies of their star formation histories. We explore whether mass and/or environment have an impact on whether a galaxy has an episodic or continuous star formation history, finding evidence that, in addition to the galaxy's mass, proximity to a larger galaxy and the cessation of star formation may drive the overall shape of the chemical evolution.

Figures (15)

• Figure 1: Four panel targeting and membership demonstrations for all seven dSphs, in the following order: Bootes 1, Carina, Draco, Fornax, Sculptor, Sextans, and Ursa Minor. In all panels, dSph members are filled circles and non-members are open circles, stars targeted as literature members are in orange and candidates in blue, and targets with overall too low spectral quality to be evaluated confidently are indicated with a black "x." For each dSph the panels are as follows: (a) Plate-centric sky distribution with the plate FOV in red, the central coordinates of the dSph are shown with the solid grey lines, and the 3 and 6 times the half-light radius is indicated with dashed lines McConnachie_2012. (b) Proper motion distribution of targets, the central PM is given by the solid lines, and the circles represent 3 and 6 sigma from the central value. The inset panel expands to a much larger PM range. (c) Radial velocity versus the radial distance from the dwarf center with the central velocity from APOGEE DR17 RVs as solid lines and 3 and 6 sigma given as dashed lines. The error bars plotted in this panel are VSCATTER that measures RV variability. The inset panel expands to a much larger RV range. (d) Kiel diagram for stars with calibrated ASPCAP results. Asterisks in this panel flag those stars with generally unreliable abundances. The complete figure set (7 images) is available in the online journal.
• Figure 2: Gaia absolute color-magnitude diagrams for each dSph. Points are colored by [Fe/H], as indicated by the color bar at right. In each panel, a fit to Ursa Minor is reproduced (black dashed line).
• Figure 3: Continuum normalized APOGEE spectra of dSph stars as a function of metallicity (top row), or S/N for stars observed in the south (middle row) and in the north (bottom row). Relatively strong lines are identified in a spectral window around the strong Mg I lines (left panel) and Al I lines (right panel). Lines marked as O indicate OH features. Pixels with uncertainties $> 25\%$ are masked as transparent for clarity. In the top row, stars with S/N$\sim 70$ with T$_{\rm eff}$$\sim 4550$ and $\log g$$\sim 0.5$ are compared from [M/H] from -1.2 to -2.4 from top to bottom. In the middle and bottom panel, stars with T$_{\rm eff}$$\sim 4550$ and $\log g$$\sim 0.5$ are compared with S/N 100 to 30 from top to bottom.
• Figure 4: The T$_{\rm eff}$ and $\log g$ errors from ASPCAP are a complicated function of spectral S/N (as measured by snrev) and the overall metallicity ([M/H]). We exclude stars with errors greater than by removing those with errors greater than 200 K and 0.3 in T$_{\rm eff}$ and $\log g$, respectively (indicated by dashed lines), because these stars have parameters that are too uncertain to provide usable information.
• Figure 5: The APOGEE [C/Fe], [N/Fe], [(C+N)/Fe], and [C/N] are shown as a function of metallicity for stars with good abundances in each dSph galaxy (i.e., not identified as upper limits and having uncertainties less than 0.3 dex). The left panels show the abundances of individual stars (Fornax - dark red pentagons, Sculptor - red circles, Sextans - orange squares, Carina - yellow triangles, Ursa Minor - cyan diamonds, Draco - blue downward triangles, Bootes I - purple crosses) in comparison to LOWESS smoothed distributions of four MW comparison samples (high-$\alpha$ disk - black line, low-$\alpha$ disk - gray, the bulge-like sample - brown, low-$\alpha$ halo - light brown). In the right panels we show the equivalent LOWESS smoothed abundance distributions for each of the dSphs (dashed lines with the same color-coding), and are transparent at metallicities below which more than half of the abundance measurements are considered bad for that galaxy. Potential carbon stars are also identified (points surrounded by a black circle).