We present new computations of the equilibrium and non-equilibrium cooling efficiencies and ionization states for low-density radiatively cooling gas containing cosmic abundances of the elements H, He, C, N, O, Ne, Mg, Si, S, and Fe. We present results for gas temperatures between 10 4 and 10 8 K, assuming dust-free and optically thin conditions, and no external radiation. For non-equilibrium cooling we solve the coupled time-dependent ionization and energy loss equations for a radiating gas cooling from an initially hot, 5 × 10 6 K, equilibrium state, down to 10 4 K. We present results for heavy element compositions ranging from 10 −3 to 2 times the elemental abundances in the Sun. We consider gas cooling at constant density (isochoric) and at constant pressure (isobaric). We calculate the critical column densities and temperatures at which radiatively cooling clouds make the dynamical transition from isobaric to isochoric evolution. We construct ion ratio diagnostics for the temperature and metallicity in radiatively cooling gas. We provide numerical estimates for the maximal cloud column densities for which the gas remains optically thin to the cooling radiation. We present our computational results in convenient on-line figures and tables (see http://wise-obs.tau.ac.il/∼orlyg/cooling/).

(Fig. 1). The first BB temperature estimate was obtained via careful extrapolation of the UVOT UV M2 and P60 g + i light curves back to the first P48 (detection) point (see inset of Fig. 1 and text for details). The earlytime BB temperature estimates, within the first half day after explosion, are also in agreement with our temperature estimates from the modeling of the early Keck spectra (Fig. 4), showing the highly ionised emission lines at temperatures > ∼ 50 kK. The shaded region and the solid black line (a running mean of the region) denote bolometric luminosity estimates based on the multiband photometry ( Fig. 1) according to three methods used to calculate the total flux from the SED: interpolation, order-4 polynomial fit, and BB fits. The top end of the shaded region can be regarded as our best lower limit on the real bolometric luminosity, based on the photometric observations. The red triangles denote a (more conservative) lower limit on the bolometric luminosity obtained from our spectra (Fig. 2, Fig. 3), beginning with the early set of 4 Keck spectra at < ∼ 10 hr after explosion and ending with our latest spectrum at 57.2days. The blue triangles show the luminosity as obtained by our best BB temperature and radius estimates (Fig. 4), L = 4πR 2 σT 4 ; the luminosity in the first point, at ∼ 3.8 hr after explosion, exceeds 10 44 erg s −1 .

There is a consensus that type Ia supernovae (SNe Ia) arise from the thermonuclear explosion of white dwarf stars that accrete matter from a binary companion. However, direct observation of SN Ia progenitors is lacking, and the precise nature of the binary companion remains uncertain. A temporal series of high-resolution optical spectra of the SN Ia PTF 11kx reveals a complex circumstellar environment that provides an unprecedentedly detailed view of the progenitor system. Multiple shells of circumstellar material are detected, and the SN ejecta are seen to interact with circumstellar material starting 59 days after the explosion. These features are best described by a symbiotic nova progenitor, similar to RS Ophiuchi.

Recent observations have revealed that some Type Ia supernovae exhibit narrow, time-variable Na I D absorption features. The origin of the absorbing material is controversial, but it may suggest the presence of circumstellar gas in the progenitor system prior to the explosion, with significant implications for the nature of the supernova progenitors. We present the third detection of such variable absorption, based on six epochs of high-resolution spectroscopy of the Type Ia supernova SN 2007le from the Keck I telescope and the Hobby-Eberly Telescope. The data span a time frame of approximately three months, from 5 days before maximum light to 90 days after maximum. We find that one component of the Na I D absorption lines strengthened significantly with time, indicating a total column density increase of ∼ 2.5 × 10 12 cm −2 . The data limit the typical timescale for the variability to be more than 2 days but less than 10 days. The changes appear to be most prominent after maximum light rather than at earlier times when the ultraviolet flux from the supernova peaks. As with SN 2006X, we detect no change in the Ca II H and K absorption lines over the same time period, rendering line-of-sight effects improbable and suggesting a circumstellar origin for the absorbing material. Unlike the previous two supernovae exhibiting variable absorption, SN 2007le is not highly reddened (E B−V = 0.27 mag), also pointing toward circumstellar rather than interstellar absorption. Photoionization calculations show that the data are consistent with a dense (10 7 cm −3 ) cloud or clouds of gas located ∼ 0.1 pc (3 × 10 17 cm) from the explosion. These results broadly support the single-degenerate scenario previously proposed to explain the variable absorption, with mass loss from a nondegenerate companion star responsible for providing the circumstellar gas. We also present possible evidence for narrow Hα emission associated with the supernova, which will require deep imaging and spectroscopy at late times to confirm.