The Carnegie Supernova Project II

Abstract

We present ultra-violet (UV) to mid-infrared (MIR) observations of the long-lasting Type IIn supernova (SN) 2013L obtained by the Carnegie Supernova Project II (CSP-II) beginning two days after discovery and extending until +887 days (d). The SN reached a peak r-band absolute magnitude of ≈ −19 mag and an even brighter UV peak, and its light curve evolution resembles that of SN 1988Z. The spectra of SN 2013L are dominated by hydrogen emission features, characterized by three components attributed to different emission regions. A unique feature of this Type IIn SN is that, apart from the first epochs, the blue shifted line profile is dominated by the macroscopic velocity of the expanding shock wave of the SN. We are therefore able to trace the evolution of the shock velocity in the dense and partially opaque circumstellar medium (CSM), from ∼ 4800 km s −1 at +48 d, decreasing as t −0.23 to ∼ 2700 km s −1 after a year. We performed spectral modeling of both the broad-and intermediate-velocity components of the Hα line profile. The high-velocity component is consistent with emission from a radially thin, spherical shell located behind the expanding shock with emission wings broadened by electron scattering. We propose that the intermediate component originates from preionized gas from the unshocked dense CSM with the same velocity as the narrow component, ∼ 100 km s −1 , but also that it is broadened by electron scattering. These features provide direct information about the shock structure, which is consistent with model calculations. The spectra exhibit broad O i and [O i] lines that emerge at +144 d and broad Ca ii features. The spectral continua and the spectral energy distributions (SEDs) of SN 2013L after +132 d are well reproduced by a two-component black-body (BB) model; one component represents emitting material with a temperature between 5×10 3 to 1.5×10 4 K (hot component) and the second component is characterized by a temperature around 1-1.5×10 3 K (warm component). The warm component dominates the emission at very late epochs ( +400 d), as is evident from both the last near infrared (NIR) spectrum and MIR observations obtained with the Spitzer Space Telescope. Using the BB fit to the SEDs, we constructed a bolometric light curve that was modeled together with the unshocked CSM velocity and the shock velocity derived from the Hα line modeling. The circumstellar-interaction model of the bolometric light curve reveals a mass-loss rate history with large values ( 1.7 × 10 −2 − 0.15 M yr −1 ) over the ∼25 -40 years before explosion, depending on the radiative efficiency and anisotropies in the CSM. The drop in the light curve at ∼ 350 days and the presence of electron scattering wings at late epochs indicate an anisotropic CSM. The mass-loss rate values and the unshocked-CSM velocity are consistent with the characteristics of a massive star, such as a luminous blue variable (LBV) undergoing strong eruptions, similar to η Carina. Our analysis also suggests a scenario where pre-existing dust grains have a dis...

The slow demise of the long-lived SN 2005ip

“…Fig. 6 of Smith et al (2009b) shows that this line has a 'shoulder' at ∼5300 km s −1 on the blue side in the spectra at ∼100 d. As discussed in detail by Taddia et al (2020), this shoulder may be caused by the macroscopic shock velocity, in contrast to the wings caused by electron scattering. Our late-time Hα spectra (Section 3.2.4) indicate a velocity of ∼2700 km s −1 at ∼1000 d, but the line profile still suggests contributions from electron scattering.…”

“…While the blue side can be well fitted with an electron-scattering wing, the line centroid is shifted to the blue, with a 'shoulder' at ∼−800 km s −1 . As shown by Taddia et al (2020) and Dessart et al (2015), this type of profile can be explained by a combination of emission from the shock wave and emission from the pre-shock CSM. The emission from the shock is responsible for the blueshifted shoulder, while the central component is coming from pre-ionized gas in the CSM.…”

“…As seen from equations ( 4) and ( 6), besides the value of v shock , an important uncertainty in these estimates is the efficiency parameter, , which depends on the importance of shock-wave instabilities and other multidimensional effects (Taddia et al 2020), and may be in the range ∼0.1-1. For these reasons, the total radiated energy and total mass should be taken as lower limits.…”

Spitzer’s Last Look at Extragalactic Explosions: Long-term Evolution of Interacting Supernovae

“…Another group of SNe IIn (SNe 2005ip, 2006jd, and 2013L) seem to be a bit fainter during the first ∼2000 days (note, however, that in the first two cases there are no mid-IR data before ∼950 days and ∼1150 days, respectively), but decline more slowly thereafter. In these SNe, the most probable scenario is a large, pre-existing dust shell that is continuously heated by energetic photons generated by ongoing CSM interaction (see, e.g., Fox et al 2010Fox et al , 2011Fox et al , 2020Stritzinger et al 2012;Andrews et al 2017;Taddia et al 2020). Based on this picture, the decline rates of these SNe may hint that the level of CSM interaction continuously decreases in the last years (except in the case of SN 2010mc, which seems to be in a very long "plateau" phase even at ∼3000 days).…”

A Diversity of Wave-driven Presupernova Outbursts

“…These SNe extend across virtually all spectroscopic classes of SNe (including Type II-P, II-L, IIn, Ib, Ic, Ibn, Ic-BL, superluminous Ic, etc.). There are many other SNe that show narrow emission lines and late-time interaction, such as SN 2004dk (Mauerhan et al 2018;Pooley et al 2019), SN 2010bt (Elias-Rosa et al 2018b), SN 2012ab (Bilinski et al 2018), SN hunt151 (Elias-Rosa et al 2018a), SN 2013L (Taddia et al 2020), ASASSN-15no (Benetti et al 2018), iPTF 16eh (Lunnan et al 2018), SN 2017dio (Kuncarayakti et al 2018), and SN 2017ens (Chen et al 2018), indicative of extreme pre-SN mass loss occurring approximately decades before the SN.…”