We present a neural network autoencoder structure that is able to extract essential latent spectral features from observed spectra and then reconstruct a spectrum from those features. Because of the training with a set of unpeculiar spectra, the network is able to reproduce a spectrum of high signal-to-noise ratio that does not show any spectral peculiarities even if they are present in an observed spectrum. Spectra generated in this manner were used to identify various emission features among spectra acquired by multiple surveys using the HERMES spectrograph at the Anglo-Australian telescope. Emission features were identified by a direct comparison of the observed and generated spectra. Using the described comparison procedure, we discovered 10,364 candidate spectra with a varying degree of H$\alpha$/H$\beta$ emission component produced by different physical mechanisms. A fraction of those spectra belongs to the repeated observation that shows temporal variability in their emission profile. Among emission spectra, we find objects that feature contributions of a nearby rarefied gas (identified through the emission of [NII] and [SII] lines) that was identified in 4004 spectra, which were not all identified as having H$\alpha$ emission. Positions of identified emission-line objects coincide with multiple known regions that harbour young stars. Similarly, detected nebular emission spectra coincide with visually-prominent nebular clouds observable in the red all-sky photographic composites.

Binary stellar systems form a large fraction of the Galaxy's stars. They are useful as laboratories for studying the physical processes taking place within stars, and must be correctly taken into account when observations of stars are used to study the structure and evolution of the Galaxy. We present a sample of 12760 well-characterised double-lined spectroscopic binaries that are appropriate for statistical studies of the binary populations. They were detected as SB2s using a t-distributed stochastic neighbour embedding (t-SNE) classification and a cross-correlation analysis of GALAH spectra. This sample consists mostly of dwarfs, with a significant fraction of evolved stars and several dozen members of the giant branch. To compute parameters of the primary and secondary star ($T_{\rm eff[1,2]}$, $\log g_{[1,2]}$, [Fe/H], $V_{r[1,2]}$, $v_{\rm mic[1,2]}$, $v_{\rm broad[1,2]}$, $R_{[1,2]}$, and $E(B-V)$), we used a Bayesian approach that includes a parallax prior from Gaia DR2, spectra from GALAH, and apparent magnitudes from APASS, Gaia DR2, 2MASS, and WISE. The derived stellar properties and their distributions show trends that are expected for a population of close binaries (a $<$ 10 AU) with mass ratios $0.5 \leq q \leq 1$. The derived metallicity of these binary stars is statistically lower than that of single dwarf stars from the same magnitude-limited sample.

We present isochrone ages and initial bulk metallicities ($\rm [Fe/H]_{bulk}$, by accounting for diffusion) of 163,722 stars from the GALAH Data Release 2, mainly composed of main sequence turn-off stars and subgiants ($\rm 7000 K>T_{eff}>4000 K$ and $\rm log g>3$ dex). The local age-metallicity relationship (AMR) is nearly flat but with significant scatter at all ages; the scatter is even higher when considering the observed surface abundances. After correcting for selection effects, the AMR appear to have intrinsic structures indicative of two star formation events, which we speculate are connected to the thin and thick disks in the solar neighborhood. We also present abundance ratio trends for 16 elements as a function of age, across different $\rm [Fe/H]_{bulk}$ bins. In general, we find the trends in terms of [X/Fe] vs age from our far larger sample to be compatible with studies based on small ($\sim$ 100 stars) samples of solar twins but we now extend it to both sub- and super-solar metallicities. The $\alpha$-elements show differing behaviour: the hydrostatic $\alpha$-elements O and Mg show a steady decline with time for all metallicities while the explosive $\alpha$-elements Si, Ca and Ti are nearly constant during the thin disk epoch (ages $\lessapprox $ 12 Gyr). The s-process elements Y and Ba show increasing [X/Fe] with time while the r-process element Eu have the opposite trend, thus favouring a primary production from sources with a short time-delay such as core-collapse supernovae over long-delay events such as neutron star mergers.

Using kinematics from Gaia and the large elemental abundance space of the second data release of the GALAH survey, we identify two new members of the Fimbulthul stellar stream, and chemically tag them to massive, multi-metallic globular cluster $\omega$ Centauri. Recent analysis of the second data release of Gaia had revealed the Fimbulthul stellar stream in the halo of the Milky Way. It had been proposed that the stream is associated with the $\omega$ Centauri, but this proposition relied exclusively upon the kinematics and metallicities of the stars to make the association. In this work, we find our two new members of the stream to be metal-poor stars that are enhanced in sodium and aluminium, typical of second population globular cluster stars, but not otherwise seen in field stars. Furthermore, the stars share the s-process abundance pattern seen in $\omega$ Centauri, which is rare in field stars. Apart from one star within 1.5 deg of $\omega$ Centauri, we find no other stars observed by GALAH spatially near $\omega$ Centauri or the Fimbulthul stream that could be kinematically and chemically linked to the cluster. Chemically tagging stars in the Fimbulthul stream to $\omega$ Centauri confirms the earlier work, and further links this tidal feature in the Milky Way halo to $\omega$ Centauri.

The Orion complex is arguably the most studied star-forming region in the Galaxy. While stars are still being born in the Orion nebula, the oldest part was believed to be no more than 13 Myr old. In order to study the full hierarchy of star formation across the Orion complex, we perform a clustering analysis of the Ori OB1a region using new stellar surveys and derive robust ages for each identified stellar aggregate. We use Gaia DR2 parameters supplemented with radial velocities from the GALAH and APOGEE surveys to perform clustering of the Ori OB1a association. Five overdensities are resolved in a six-dimensional parameter space (positions, distance, proper motions, and radial velocity). Most correspond to previously known structures (ASCC 16, 25 Orionis, ASCC 20, ASCC 21). We use Gaia DR2, Pan-STARRS1 and 2MASS photometry to fit isochrones to the colour-magnitude diagrams of the identified clusters. The ages of the clusters can thus be measured with ~10% precision. While four of the clusters have ages between 11 and 13 Myr, the ASCC 20 cluster stands out at an age of 21 $\pm$ 3 Myr. This is significantly greater than the age of any previously known component of the Orion complex. To some degree, all clusters overlap in at least one of the six phase-space dimensions. We argue that the formation history of the Orion complex, and its relation to the Gould belt, must be reconsidered. A significant challenge in reconstructing the history of the Ori OB1a association is to understand the impact of the newly discovered 21 Myr old population on the younger parts of the complex, including their formation.

The latest Gaia data release enables us to accurately identify stars that are more luminous than would be expected on the basis of their spectral type and distance. During an investigation of the 329 best Solar twin candidates uncovered among the spectra acquired by the GALAH survey, we identified 64 such over-luminous stars. In order to investigate their exact composition, we developed a data-driven methodology that can generate a synthetic photometric signature and spectrum of a single star. By combining multiple such synthetic stars into an unresolved binary or triple system and comparing the results to the actual photometric and spectroscopic observations, we uncovered 6 definitive triple stellar system candidates and an additional 14 potential candidates whose combined spectrum mimics the Solar spectrum. Considering the volume correction factor for a magnitude limited survey, the fraction of probable unresolved triple stars with long orbital periods is ~2 %. Possible orbital configurations of the candidates were investigated using the selection and observational limits. To validate the discovered multiplicity fraction, the same procedure was used to evaluate the multiplicity fraction of other stellar types.

Swan bands - characteristic molecular absorption features of the C$_2$ molecule - are a spectroscopic signature of carbon-enhanced stars. They can also be used to identify carbon-enhanced metal-poor (CEMP) stars. The GALAH (GALactic Archaeology with Hermes) is a magnitude-limited survey of stars producing high-resolution, high signal-to-noise spectra. We used 627,708 GALAH spectra to search for carbon-enhanced stars with a supervised and unsupervised classification algorithm, relying on the imprint of the Swan bands. We identified 918 carbon-enhanced stars, including 12 already described in the literature. An unbiased selection function of the GALAH survey allows us to perform a population study of carbon-enhanced stars. Most of them are giants, out of which we find 28 CEMP candidates. A large fraction of our carbon-enhanced stars with repeated observations show variation in radial velocity, hinting that there is a large fraction of variables among them. 32 of the detected stars also show strong Lithium enhancement in their spectra.

Identifying and characterizing young stars in the Solar neighbourhood is essential to find and describe planets in the early stages of their evolution. This work seeks to identify nearby young stars showing a Lithium 6707.78$\,\AA$ absorption line in the GALAH survey. A robust, data-driven approach is used to search for corresponding templates in the pool of 434,215 measured dwarf spectra in the survey. It enables a model-free search for best-matching spectral templates for all stars, including M dwarfs with strong molecular absorption bands. 3147 stars have been found to have measurable Lithium: 1408 G and 892 K0-K5 dwarfs (EW(Li)$>$0.1$\,$\AA), 335 K5-K9 ($>$0.07$\,$\AA) and 512 M0$\sim$M4 dwarfs ($>$0.05$\,$\AA). Stars with such Lithium features are used to investigate the possibility of searching for young stars above the main sequence based merely on their parallaxes and broad-band photometry. Selection of young stars above the main sequence is highly effective for M dwarfs, moderately effective for K dwarfs and ineffective for G dwarfs. Using a combination of the Lithium information and the complete 6D kinematics from Gaia and GALAH, 305 new candidate moving group members have been found, 123 of which belong to the Scorpius-Centaurus association, 36 to the Pleiades and 25 to the Hyades clusters.

GALAH is a large-scale magnitude-limited southern stellar spectroscopic survey. Its second data release (GALAH DR2) provides values of stellar parameters and abundances of 23 elements for 342,682 stars (Buder et al.). Here we add a description of the public release of radial velocities with a typical accuracy of 0.1 km/s for 336,215 of these stars, achievable due to the large wavelength coverage, high resolving power and good signal to noise ratio of the observed spectra, but also because convective motions in stellar atmosphere and gravitational redshift from the star to the observer are taken into account. In the process we derive medians of observed spectra which are nearly noiseless, as they are obtained from between 100 and 1116 observed spectra belonging to the same bin with a width of 50 K in temperature, 0.2 dex in gravity, and 0.1 dex in metallicity. Publicly released 1181 median spectra have a resolving power of 28,000 and trace the well-populated stellar types with metallicities between -0.6 and +0.3. Note that radial velocities from GALAH are an excellent match to the accuracy of velocity components along the sky plane derived by Gaia for the same stars. The level of accuracy achieved here is adequate for studies of dynamics within stellar clusters, associations and streams in the Galaxy. So it may be relevant for studies of the distribution of dark matter.

Sparse open clusters can be found at high galactic latitudes where loosely populated clusters are more easily detected against the lower stellar background. Because most star formation takes place in the thin disk, the observed population of clusters far from the Galactic plane is hard to explain. We combined spectral parameters from the GALAH survey with the Gaia DR2 catalogue to study the dynamics and chemistry of five old sparse high-latitude clusters in more detail. We find that four of them (NGC 1252, NGC 6994, NGC 7772, NGC 7826) - originally classified in 1888 - are not clusters but are instead chance projections on the sky. Member stars quoted in the literature for these four clusters are unrelated in our multi-dimensional physical parameter space; the published cluster properties are therefore irrelevant. We confirm the existence of morphologically similar NGC 1901 for which we provide a probabilistic membership analysis. An overdensity in three spatial dimensions proves to be enough to reliably detect sparse clusters, but the whole 6-dimensional space must be used to identify members with high confidence, as demonstrated in the case of NGC 1901.

We present a study using the second data release of the GALAH survey of stellar parameters and elemental abundances of 15 pairs of stars identified by Oh et al 2017. They identified these pairs as potentially co-moving pairs using proper motions and parallaxes from Gaia DR1. We find that 11 very wide (>1.7 pc) pairs of stars do in fact have similar Galactic orbits, while a further four claimed co-moving pairs are not truly co-orbiting. Eight of the 11 co-orbiting pairs have reliable stellar parameters and abundances, and we find that three of those are quite similar in their abundance patterns, while five have significant [Fe/H] differences. For the latter, this indicates that they could be co-orbiting because of the general dynamical coldness of the thin disc, or perhaps resonances induced by the Galaxy, rather than a shared formation site. Stars such as these, wide binaries, debris of past star formation episodes, and coincidental co-orbiters, are crucial for exploring the limits of chemical tagging in the Milky Way.