Item

Abstract

The exact formation mechanism of massive galaxy in the universe still become an open question in modern astrophysics. Radio emission from active galactic nucleus (AGN) is known to be suppressing stellar growth in the most massive galaxies, thus becoming significant ingredient in modeling galaxy formation process. Radio luminosity function across cosmic redshifts (z) is an important tool to constrain the co- evolutionary models of radio AGN and its host galaxy. Therefore, we aim to calculate radio luminosity function at 1.4 GHz frequency for radio AGN in the range of 0.1 ≤ z ≤ 1.3 in this work. The radio data was taken from deep VLA observation of Stripe 82 field at angular resolution of 1.8” and 52 μJy sensitivity. On the other hand, the optical/near- infrared data was taken from Dark Energy Survey DR1 observation in g, r, i, z, Y bands with co-added catalog limiting magnitude of i = 23.44. We estimated the photometric redshift (photo-z) of each sources by using combined two machine learning algorithms: neural networks and boosted regression trees. We successfully performed accurate photo-z measurement with average bias 〈δ〉 = −3.5 × 10⁻³, scatter 〈σ〉 = 0.15 and outlier fraction 〈f(3σ)〉 = 0.06. By using 1/V _max method, we calculated the luminosity function, then constrained their evolution with pure density evolution (PDE) or pure luminosity evolution (PLE) model. At median z = {0.31, 0.59, 0.88, 1.10}, we found the power-law index of PDE is α_D = {1.29, 1.43, 1.73, 0.94} while for PLE is α_L = {2.28, 2.59, 3.19, 1.63}. Our result is consistent with previous studies and gives better constraint to radio AGN luminosity and density evolution power-law indexes due to larger number of sources (6900) and wider covered sky fraction (92 deg²).