Abstract |
One of the most important and difficult problems in modern Astrophysics is the
classification of galaxies based on their activity. A lot of progress has been done over the
years with numerous diagnostics to have been developed; optical and infrared methods being
the most successful and popular among them. In the recent years with the advent of the allsky surveys from space telescopes, infrared diagnostics for AGN selection methods have
become popular. Unfortunately, we find that some of them are neither complete or reliable in
galaxies located in the local Universe. In addition, the class of passive galaxies is absent from
these diagnostics. For these reasons, we embarked in the development of a new threedimensional activity diagnostic based on machine-learning methods and WISE infrared
photometry. In this project, we consider the classes of star-forming, AGN, composite and
passive galaxies. We find that a diagnostic based on three features derived from the three
WISE bands (1, 2 and 3): absolute magnitude on the band 2, band 1 – band 2 color and band
2 – band 3 color, offers precision above 90% for star-forming and passive galaxies. In
addition, using the new diagnostic, we classify 68.7% of the galaxies found in the HECATE
catalog. Finally, we estimate the activity demographics in the local Universe based on the
results from the classification of the full HECATE catalog.
|