Tiger Y-Y Hsiao

We present a distant T-type brown dwarf candidate at ≈2.55 kpc discovered in the Cosmic Evolution Early Release Science (CEERS) fields by James Webb Space Telescope (JWST) NIRCam. In addition to the superb sensitivity, we utilised 7 filters from JWST in near-IR and thus is advantageous in finding faint, previously unseen brown dwarfs. From the model spectra in new JWST/NIRCam filter wavelengths, the selection criteria of F115W-F277W<-0.8 and F277W-F444W>1.1 were chosen to target the spectrum features of brown dwarfs having temperatures from 500 K to 1300 K. Searching through the data from Early Release Observations (ERO) and Early Release Science (ERS), we find 1 promising candidate in the CEERS field. The result of SED fitting suggested an early T spectral type with a low effective temperature of Teff ≈1300 K, the surface gravity of log g ≈ 5.25 cms−2, and an eddy diffusion parameter of logKzz ≈ 7 cm2s−1, which indicates an age of ≈1.8 Gyr and a mass of ≈0.05 M⊙. In contrast to typically found T dwarf within several hundred parsecs, the estimated distance of the source is ≈2.55 kpc, showing the JWST’s power to extend the search to a much larger distance.

Fast radio bursts (FRBs) are astronomical transients with millisecond time-scales. Although most of the FRBs are not observed to repeat, a few of them are detected to repeat more than hundreds of times. There exist a large variety of physical properties among these bursts, suggesting heterogeneous mechanisms of FRBs. In this paper, we conduct a categorization on the extremely frequently repeating FRB 20201124A with the assistance of machine learning, as such techniques have the potential to use subtle differences and correlations that humans are unaware of to better classify bursts. The research is carried out by applying the unsupervised Uniform Manifold Approximation and Projection (umap) model on the FRB 20201124A data provided by Five-hundred-meter Aperture Spherical radio Telescope. The algorithm eventually categorizes the bursts into three clusters. In addition to the two categories in previous work based on waiting time, a new way for categorization has been found. The three clusters are either high energy, high frequency, or low frequency, reflecting the distribution of FRB energy and frequency. Importantly, a similar machine learning result is found in another frequently repeating FRB 20121102A, implying a common mechanism among this kind of FRB. This work is one of the first steps towards the systematical categorization of the extremely frequently repeating FRBs.