Search

The dividing line between gamma-ray bursts (GRBs) and ordinary stripped-envelope core-collapse supernovae (SNe) is yet to be fully understood. Observationally mapping the variety of ejecta outcomes (ultra-relativistic, mildly-relativistic or non-relativistic) in SNe of Type Ic with broad lines (Ic-BL) can provide a key test to stellar explosion models. However, this requires large samples of the rare Ic-BL events with follow-up observations in the radio, where fast ejecta can be probed largely free of geometry and viewing angle effects. Here, we present the results of a radio (and X-ray) follow-up campaign of 16 SNe Ic-BL detected by the Zwicky Transient Facility (ZTF). Our radio campaign resulted in 4 counterpart detections and 12 deep upper limits. None of the events in our sample is as relativistic as SN 1998bw and we constrain the fraction of SN 1998bw-like explosions to $< 19\%$ (3$\sigma$ Gaussian equivalent), a factor of $\approx 2$ smaller than previously established. We exclude relativistic ejecta with radio luminosity densities in between $\approx 5\times10^{27}$ erg s$^{-1}$ Hz$^{-1}$ and $\approx 10^{29}$ erg s$^{-1}$ Hz$^{-1}$ at $t\gtrsim 20$ d since explosion for $\approx 60\%$ of the events in our sample. This shows that SNe Ic-BL similar to the GRB-associated SN 1998bw, SN 2003lw, SN 2010dh, or to the relativistic SN 2009bb and iPTF17cw, are rare. Our results also exclude an association of the SNe Ic-BL in our sample with largely off-axis GRBs with energies $E\gtrsim 10^{50}$ erg. The parameter space of SN2006aj-like events (faint and fast-peaking radio emission) is, on the other hand, left largely unconstrained and systematically exploring it represents a promising line of future research., Comment: 28 pages, 11 figures, submitted to AAS journal

We present SNIascore, a deep-learning based method for spectroscopic classification of thermonuclear supernovae (SNe Ia) based on very low-resolution (R $\sim100$) data. The goal of SNIascore is fully automated classification of SNe Ia with a very low false-positive rate (FPR) so that human intervention can be greatly reduced in large-scale SN classification efforts, such as that undertaken by the public Zwicky Transient Facility (ZTF) Bright Transient Survey (BTS). We utilize a recurrent neural network (RNN) architecture with a combination of bidirectional long short-term memory and gated recurrent unit layers. SNIascore achieves a $<0.6\%$ FPR while classifying up to $90\%$ of the low-resolution SN Ia spectra obtained by the BTS. SNIascore simultaneously performs binary classification and predicts the redshifts of secure SNe Ia via regression (with a typical uncertainty of $<0.005$ in the range from $z = 0.01$ to $z = 0.12$). For the magnitude-limited ZTF BTS survey ($\approx70\%$ SNe Ia), deploying SNIascore reduces the amount of spectra in need of human classification or confirmation by $\approx60\%$. Furthermore, SNIascore allows SN Ia classifications to be automatically announced in real-time to the public immediately following a finished observation during the night., Comment: 12 pages, 5 figures, 2 tables, accepted for publication in ApJL

We present a public catalog of transients from the Zwicky Transient Facility (ZTF) Bright Transient Survey (BTS), a magnitude-limited (m<19 mag in either the g or r filter) survey for extragalactic transients in the ZTF public stream. We introduce cuts on survey coverage, sky visibility around peak light, and other properties unconnected to the nature of the transient, and show that the resulting statistical sample is spectroscopically 97% complete at <18 mag, 93% complete at <18.5 mag, and 75% complete at <19 mag. We summarize the fundamental properties of this population, identifying distinct duration-luminosity correlations in a variety of supernova (SN) classes and associating the majority of fast optical transients with well-established spectroscopic SN types (primarily SN Ibn and II/IIb). We measure the Type Ia SN and core-collapse (CC) SN rates and luminosity functions, which show good consistency with recent work. About 7% of CC SNe explode in very low-luminosity galaxies (M_i > -16 mag), 10% in red-sequence galaxies, and 1% in massive ellipticals. We find no significant difference in the luminosity or color distributions between the host galaxies of Type II and Type Ib/c supernovae, suggesting that line-driven wind stripping does not play a major role in the loss of the hydrogen envelope from their progenitors. Future large-scale classification efforts with ZTF and other wide-area surveys will provide high-quality measurements of the rates, properties, and environments of all known types of optical transients and limits on the existence of theoretically predicted but as of yet unobserved explosions., Comment: Accepted to ApJ. Live BTS statistics are available online at https://sites.astro.caltech.edu/ztf/bts/bts.php and an interactive catalog of our sample is available at https://sites.astro.caltech.edu/ztf/bts/explorer.php