Here we report that the most rapidly recurring nova, M31N 2008-12a, which erupts annually, is surrounded by a nova super-remnant which demonstrates that M31N 2008-12a has erupted with high frequency for millions of years.
We present a comprehensive review of all observations of the eclipsing recurrent Nova LMC 1968 in the Large Magellanic Cloud which was previously observed in eruption in 1968, 1990, 2002, 2010, and most recently in 2016. We derive a probable recurren
ce time of $6.2 pm 1.2$ years and provide the ephemerides of the eclipse. In the ultraviolet-optical-IR photometry the light curve shows high variability right from the first observation around two days after eruption. Therefore no colour changes can be substantiated. Outburst spectra from 2016 and 1990 are very similar and are dominated by H and He lines longward of 2000 Angstrom. Interstellar reddening is found to be E(B-V) = $0.07pm0.01$. The super soft X-ray luminosity is lower than the Eddington luminosity and the X-ray spectra suggest the mass of the WD is larger than 1.3 M$_odot$. Eclipses in the light curve suggest that the system is at high orbital inclination. On day four after the eruption a recombination wave was observed in Fe II ultraviolet absorption lines. Narrow line components are seen after day 6 and explained as being due to reionisation of ejecta from a previous eruption. The UV spectrum varies with orbital phase, in particular a component of the He II 1640 Angstrom emission line, which leads us to propose that early-on the inner WD Roche lobe might be filled with a bound opaque medium prior to the re-formation of an accretion disk. Both this medium and the ejecta can cause the delay in the appearance of the soft X-ray source.
Nova LMC 2009a is confirmed as a Recurrent Nova (RN) from positional coincidence with nova LMC 1971b. The observational data set is one of the most comprehensive for any Galactic or extragalactic RN: optical and near-IR photometry from outburst until
over 6 years later; optical spectra for the first 6 months, and Swift satellite Ultraviolet and X-ray observations from 9 days to almost 1 year post-outburst. We find $M_V = -8.4pm0.8_{mathrm{r}}pm0.7_{mathrm{s}}$ and expansion velocities between 1000 and 4000 km s$^{-1}$. Coronal line emission before day 9 indicates shocks in the ejecta. Strengthening of He II $lambda$4686 preceded the emergence of the Super-Soft Source (SSS) in X-rays at $sim63-70$ days, which was initially very variable. Periodic modulations, $P=1.2$ days, most probably orbital in nature, were evident in the UV and optical from day 43. Subsequently, the SSS shows an oscillation with the same period but with a delay of 0.28P. The progenitor system has been identified; the secondary is most likely a sub-giant feeding a luminous accretion disk. Properties of the SSS infer a white dwarf (WD) mass $1.1 mathrm{M}_odot lesssim M_{rm WD} lesssim 1.3 mathrm{M}_odot$. If the accretion occurs at constant rate, $dot{it{M}}_{rm acc} simeq 3.6^{+4.7}_{-2.5} times 10^{-7} mathrm{M}_odot$ yr$^{-1}$ is needed, consistent with nova models for an inter-eruption interval of 38 years, low outburst amplitude, progenitor position in the color-magnitude diagram, and spectral energy distribution at quiescence. We note striking similarities between LMC 2009a and the Galactic nova KT Eri, suggesting that KT Eri is a candidate RN.
The Andromeda Galaxy recurrent nova M31N 2008-12a has been caught in eruption nine times. Six observed eruptions in the seven years from 2008 to 2014 suggested a duty cycle of ~1 year, which makes this the most rapidly recurring system known and the
leading single-degenerate Type Ia Supernova progenitor candidate; but no 2010 eruption has been found so far. Here we present evidence supporting the recovery of the 2010 eruption, based on archival images taken at and around the time. We detect the 2010 eruption in a pair of images at 2010 Nov 20.52 UT, with a magnitude of m_R = 17.84 +/- 0.19. The sequence of seven eruptions shows significant indications of a duty cycle slightly shorter than one year, which makes successive eruptions occur progressively earlier in the year. We compared three archival X-ray detections with the well observed multi-wavelength light curve of the 2014 eruption to accurately constrain the time of their optical peaks. The results imply that M31N 2008-12a might have in fact a recurrence period of ~6 months (175 +/- 11 days), making it even more exceptional. If this is the case, then we predict that soon two eruptions per year will be observable. Furthermore, we predict the next eruption will occur around late Sep 2015. We encourage additional observations.
V745 Sco is a recurrent nova, with the most recent eruption occurring in February 2014. V745 Sco was first observed by Swift a mere 3.7 hr after the announcement of the optical discovery, with the super-soft X-ray emission being detected around four
days later and lasting for only ~two days, making it both the fastest follow-up of a nova by Swift and the earliest switch-on of super-soft emission yet detected. Such an early switch-on time suggests a combination of a very high velocity outflow and low ejected mass and, together with the high effective temperature reached by the super-soft emission, a high mass white dwarf (>1.3 M_sun). The X-ray spectral evolution was followed from an early epoch where shocked emission was evident, through the entirety of the super-soft phase, showing evolving column density, emission lines, absorption edges and thermal continuum temperature. UV grism data were also obtained throughout the super-soft interval, with the spectra showing mainly emission lines from lower ionization transitions and the Balmer continuum in emission. V745 Sco is compared with both V2491 Cyg (another nova with a very short super-soft phase) and M31N 2008-12a (the most rapidly recurring nova yet discovered). The longer recurrence time compared to M31N 2008-12a could be due to a lower mass accretion rate, although inclination of the system may also play a part. Nova V745 Sco (2014) revealed the fastest evolving super-soft source phase yet discovered, providing a detailed and informative dataset for study.
The Andromeda Galaxy recurrent nova M31N 2008-12a had been observed in eruption ten times, including yearly eruptions from 2008-2014. With a measured recurrence period of $P_mathrm{rec}=351pm13$ days (we believe the true value to be half of this) and
a white dwarf very close to the Chandrasekhar limit, M31N 2008-12a has become the leading pre-explosion supernova type Ia progenitor candidate. Following multi-wavelength follow-up observations of the 2013 and 2014 eruptions, we initiated a campaign to ensure early detection of the predicted 2015 eruption, which triggered ambitious ground and space-based follow-up programs. In this paper we present the 2015 detection; visible to near-infrared photometry and visible spectroscopy; and ultraviolet and X-ray observations from the Swift observatory. The LCOGT 2m (Hawaii) discovered the 2015 eruption, estimated to have commenced at Aug. $28.28pm0.12$ UT. The 2013-2015 eruptions are remarkably similar at all wavelengths. New early spectroscopic observations reveal short-lived emission from material with velocities $sim13000$ km s$^{-1}$, possibly collimated outflows. Photometric and spectroscopic observations of the eruption provide strong evidence supporting a red giant donor. An apparently stochastic variability during the early super-soft X-ray phase was comparable in amplitude and duration to past eruptions, but the 2013 and 2015 eruptions show evidence of a brief flux dip during this phase. The multi-eruption Swift/XRT spectra show tentative evidence of high-ionization emission lines above a high-temperature continuum. Following Henze et al. (2015a), the updated recurrence period based on all known eruptions is $P_mathrm{rec}=174pm10$ d, and we expect the next eruption of M31N 2008-12a to occur around mid-Sep. 2016.