No Arabic abstract
Recurrent novae are binary stars in which a white dwarf accretes matter from a less evolved companion, either a red giant or a main-sequence star. They have dramatic optical brightenings of around 5-6 mag in V in less than a day, several times a century. These occur at variable and unpredictable intervals, and are followed by an optical decline over several weeks, and activity from the X-ray to the radio. The unpredictability of recurrent novae and related stellar types can hamper systematic study of their outbursts. Here we analyse the long-term lightcurve of RS Ophiuchi, a recurrent nova with six confirmed outbursts, most recently in 2006 February. We confirm the previously suspected 1945 outburst, largely obscured in a seasonal gap. We also find a signal via wavelet analysis that can be used to predict an incipient outburst up to a few hundred days before hand. This has never before been possible. In addition this may suggest that the preferred thermonuclear runaway mechanism for the outbursts will have to be modified, as no pre-outburst signal is anticipated in that case. If our result indeed points to gaps in our understanding of how outbursts are driven, we will need to study such objects carefully to determine if the white dwarf is growing in mass, an essential factor if these systems are to become Type Ia Supernovae. Determining the likelihood of recurrent novae being an important source population will have implications for stellar and galaxy evolution.
Optical spectra of the 2006 outburst of RS Ophiuchi beginning one day after discovery to over a year after the outburst are presented here. The spectral evolution is found to be similar to that in previous outbursts. The early phase spectra are dominated by hydrogen and helium (I & II) lines. Coronal and nebular lines appear in the later phases. Emission line widths are found to narrow with time, which is interpreted as a shock expanding into the red giant wind. Using the photoionisation code CLOUDY, spectra at nine epochs spanning 14 months after the outburst peak, thus covering a broad range of ionisation and excitation levels in the ejecta, are modelled. The best-fit model parameters indicate the presence of a hot white dwarf source with a roughly constant luminosity of 1.26 x 10^{37} erg/s. During first three months, the abundances (by number) of He, N, O, Ne, Ar, Fe, Ca, S and Ni are found above solar abundances; abundances of these elements decreased in the later phase. Also presented are spectra obtained during quiescence. Photoionisation model of the quiescence spectrum indicates the presence of a low luminosity accretion disk. The helium abundance is found to be subsolar at quiescence.
We report observations of the flickering variability of the symbiotic recurrent nova RS~Oph at quiescence in five bands ($UBVRI$). We find evidence of a correlation between the peak-to-peak flickering amplitude ($Delta F$) and the average flux of the hot component ($F_{rm av}$). The correlation is highly significant, with a correlation coefficient of 0.85 and a $p$-value of~$sim 10^{-20}$. Combining the data from all wavebands, we find a dependence of the type $Delta F propto F^k_{rm av}$, with power-law index $k = 1.02 pm 0.04$ for the $UBVRI$ flickering of RS~Oph. Thus, the relationship between the amplitude of variability and the average flux of the hot component is consistent with linearity. The rms amplitude of flickering is on average 8 per cent ($pm2$ per cent) of $F_{rm av}$. The detected correlation is similar to that found in accreting black holes/neutron stars and cataclysmic variables. The possible reasons are briefly discussed. The data are available upon request from the authors.
Near-infrared spectra are presented for the recent 2006 outburst of the recurrent nova RS Ophiuchi (RS Oph).We report the rare detection of an infrared shock wave as the nova ejecta plows into the pre-existing wind of the secondary in the RS Oph system consisting of a white dwarf (WD) primary and a red giant secondary. The evolution of the shock is traced through a free expansion stage to a decelerative phase. The behavior of the shock velocity with time is found to be broadly consistent with current shock models. The present observations also imply that the WD in the RS Oph system has a high mass indicating that it could be a potential SNIa candidate. We also discuss the results from a recent study showing that the near-IR continuum from the recent RS Oph eruption does not originate in an expanding fireball. However, the present work shows that the IR line emission does have an origin in an expanding shock wave.
We present infrared spectroscopy of the recurrent nova RS Ophiuchi, obtained 11.81, 20.75 and 55.71 days following its 2006 eruption. The spectra are dominated by hydrogen recombination lines, together with HeI, OI and OII lines; the electron temperature of ~10^4 K implied by the recombination spectrum suggests that we are seeing primarily the wind of the red giant, ionized by the ultraviolet flash when RS Oph erupted. However, strong coronal emission lines (i.e. emission from fine structure transitions in ions having high ionization potential) are present in the last spectrum. These imply a temperature of 930000K for the coronal gas; this is in line with x-ray observations of the 2006 eruption. The emission line widths decrease with time in a way that is consistent with the shock model for the x-ray emission.
Following the Swift X-ray observations of the 2006 outburst of the recurrent nova RS Ophiuchi, we developed hydrodynamical models of mass ejection from which the forward shock velocities were used to estimate the ejecta mass and velocity. In order to further constrain our model parameters, here we present synthetic X-ray spectra from our hydrodynamical calculations which we compare to the Swift data. An extensive set of simulations was carried out to find a model which best fits the spectra up to 100 days after outburst. We find a good fit at high energies but require additional absorption to match the low energy emission. We estimate the ejecta mass to be in the range (2-5) x 10^{-7} solar masses and the ejection velocity to be greater than 6000 km/s (and probably closer to 10,000 km/s). We also find that estimates of shock velocity derived from gas temperatures via standard model fits to the X-ray spectra are much lower than the true shock velocities.