Wide Area VISTA Extra-galactic Survey (WAVES): Unsupervised star-galaxy separation on the WAVES-Wide photometric input catalogue using UMAP and ${\rm{\scriptsize HDBSCAN}}$
Authors:
Todd L. Cook,
Behnood Bandi,
Sam Philipsborn,
Jon Loveday,
Sabine Bellstedt,
Simon P. Driver,
Aaron S. G. Robotham,
Maciej Bilicki,
Gursharanjit Kaur,
Elmo Tempel,
Ivan Baldry,
Daniel Gruen,
Marcella Longhetti,
Angela Iovino,
Benne W. Holwerda,
Ricardo Demarco
Abstract:
Star-galaxy separation is a crucial step in creating target catalogues for extragalactic spectroscopic surveys. A classifier biased towards inclusivity risks including spurious stars, wasting fibre hours, while a more conservative classifier might overlook galaxies, compromising completeness and hence survey objectives. To avoid bias introduced by a training set in supervised methods, we employ an…
▽ More
Star-galaxy separation is a crucial step in creating target catalogues for extragalactic spectroscopic surveys. A classifier biased towards inclusivity risks including spurious stars, wasting fibre hours, while a more conservative classifier might overlook galaxies, compromising completeness and hence survey objectives. To avoid bias introduced by a training set in supervised methods, we employ an unsupervised machine learning approach. Using photometry from the Wide Area VISTA Extragalactic Survey (WAVES)-Wide catalogue comprising 9-band $u-K_s$ data, we create a feature space with colours, fluxes, and apparent size information extracted by ${\rm P{\scriptsize RO} F{\scriptsize OUND}}$. We apply the non-linear dimensionality reduction method UMAP (Uniform Manifold Approximation and Projection) combined with the classifier ${\rm{\scriptsize HDBSCAN}}$ to classify stars and galaxies. Our method is verified against a baseline colour and morphological method using a truth catalogue from Gaia, SDSS, GAMA, and DESI. We correctly identify 99.72% of galaxies within the AB magnitude limit of $Z = 21.2$, with an F1 score of 0.9970 across the entire ground truth sample, compared to 0.9871 from the baseline method. Our method's higher purity (0.9966) compared to the baseline (0.9780) increases efficiency, identifying 11% fewer galaxy or ambiguous sources, saving approximately 70,000 fibre hours on the 4MOST instrument. We achieve reliable classification statistics for challenging sources including quasars, compact galaxies, and low surface brightness galaxies, retrieving 95.1%, 84.6%, and 99.5% of them respectively. Angular clustering analysis validates our classifications, showing consistency with expected galaxy clustering, regardless of the baseline classification.
△ Less
Submitted 17 June, 2024;
originally announced June 2024.
COWS all tHE way Down (COWSHED) I: Could cow based planetoids support methane atmospheres?
Authors:
William J. Roper,
Todd L. Cook,
Violetta Korbina,
Jussi K. Kuusisto,
Roisin O'Connor,
Stephen D. Riggs,
David J. Turner,
Reese Wilkinson
Abstract:
More often than not a lunch time conversation will veer off into bizarre and uncharted territories. In rare instances these frontiers of conversation can lead to deep insights about the Universe we inhabit. This paper details the fruits of one such conversation. In this paper we will answer the question: How many cows do you need to form a planetoid entirely comprised of cows, which will support a…
▽ More
More often than not a lunch time conversation will veer off into bizarre and uncharted territories. In rare instances these frontiers of conversation can lead to deep insights about the Universe we inhabit. This paper details the fruits of one such conversation. In this paper we will answer the question: How many cows do you need to form a planetoid entirely comprised of cows, which will support a methane atmoosphere produced by the planetary herd? We will not only present the necessary assumptions and theory underpinning the cow-culations, but also present a thorough (and rather robust) discussion of the viability of, and implications for accomplishing, such a feat.
△ Less
Submitted 30 March, 2022;
originally announced March 2022.
