Luis A. Gutiérrez Soto
Unveiling the Hidden Population of H-alpha Excess Sources in the Southern Sky: A Synergy of S-PLUS Photometry and Machine Learning
We present a novel method to identify and classify Hα-excess compact sources in the Southern Hemisphere using the 12-band photometry of the S-PLUS survey (7 narrow- and 5 broad-band filters). By employing color-color diagnostics, we isolate approximately 7,000 Hα candidates, including emission-line stars, young stellar objects, cataclysmic variables (CVs), planetary nebulae, QSOs, and compact galaxies with redshifted emission lines, as well as transients, X-ray binaries, and peculiar objects.
Combining S-PLUS colors with advanced clustering techniques (UMAP+HDBSCAN) and infrared data, we are able to distinguish between distinct classes within our Hα source list. A Random Forest model, trained on the HDBSCAN results, highlights the key color features that differentiate the various classes of Hα-excess sources, providing a robust framework for future studies, such as follow-up spectroscopy.
Event Timeslots (1)
Thursday
-
PDF
Unveiling the Hidden Population of H-alpha Excess Sources in the Southern Sky: A Synergy of S-PLUS Photometry and Machine Learning