Continuing the project described by Kato et al. (2009, arXiv:0905.1757), we collected times of superhump maxima for 102 SU UMa-type dwarf novae observed mainly during the 2014-2015 season and characterized these objects. Our project has greatly impro
ved the statistics of the distribution of orbital periods, which is a good approximation of the distribution of cataclysmic variables at the terminal evolutionary stage, and confirmed the presence of a period minimum at a period of 0.053 d and a period spike just above this period. The number density monotonically decreased toward the longer period and there was no strong indication of a period gap. We detected possible negative superhumps in Z Cha. It is possible that normal outbursts are also suppressed by the presence of a disk tilt in this system. There was no indication of enhanced orbital humps just preceding the superoutburst, and this result favors the thermal-tidal disk instability as the origin of superoutbursts. We detected superhumps in three AM CVn-type dwarf novae. Our observations and recent other detections suggest that 8% of objects showing dwarf nova-type outbursts are AM CVn-type objects. AM CVn-type objects and EI Psc-type object may be more abundant than previously recognized. OT J213806, a WZ Sge-type object, exhibited a remarkably different feature between the 2010 and 2014 superoutbursts. Although the 2014 superoutburst was much fainter the plateau phase was shorter than the 2010 one, the course of the rebrightening phase was similar. This object indicates that the O-C diagrams of superhumps can be indeed variable at least in WZ Sge-type objects. Four deeply eclipsing SU UMa-type dwarf novae (ASASSN-13cx, ASASSN-14ag, ASASSN-15bu, NSV 4618) were identified. We studied long-term trends in supercycles in MM Hya and CY UMa and found systematic variations of supercycles of ~20%.
We report on a superoutburst of the AM CVn-type object SDSS J090221.35+381941.9 [J0902; orbital period 0.03355(6) d] in 2014 March-April. The entire outburst consisted of a precursor outburst and the main superoutburst, followed by a short rebrighten
ing. During the rising branch of the main superoutburst, we detected growing superhumps (stage A superhumps) with a period of 0.03409(1) d. During the plateau phase of the superoutburst, superhumps with a shorter period (stage B superhumps) were observed. Using the orbital period and the period of the stage A superhumps, we were able to measure the dynamical precession rate of the accretion disk at the 3:1 resonance, and obtained a mass ratio (q) of 0.041(7). This is the first successful measurement of the mass ratio in an AM CVn-type object using the recently developed stage A superhump method. The value is generally in good agreement with the theoretical evolutionary model. The orbital period of J0902 is the longest among the outbursting AM CVn-type objects, and the borderline between the outbursting systems and systems with stable cool disks appears to be longer than had been supposed.
Continuing the project described by Kato et al. (2009, PASJ, 61, S395, arXiv:0905.1757), we collected times of superhump maxima for 56 SU UMa-type dwarf novae mainly observed during the 2013-2014 season and characterized these objects. We detected ne
gative superhumps in VW Hyi and indicated that the low number of normal outbursts in some supercycle can be interpreted as a result of the disk tilt. This finding, combined with the Kepler observation of V1504 Cyg and V344 Lyr, suggests that the disk tilt is responsible for modulating the outburst pattern in SU UMa-type dwarf novae. We also studied the deeply eclipsing WZ Sge-type dwarf nova MASTER OT J005740.99+443101.5 and found evidence of a sharp eclipse during the phase of early superhumps. The profile can be reproduced by a combination of the eclipse of the axisymmetric disk and the uneclipsed light source of early superhumps. This finding confirms the lack of evince of a greatly enhanced hot spot during the early stage of WZ Sge-type outburst. We detected growing (stage A) superhumps in MN Dra and give a suggestion that some of SU UMa-type dwarf novae situated near the critical condition of tidal instability may show long-lasting stage A superhumps. The large negative period derivatives reported in such systems can be understood a result of the combination of stage A and B superhumps. The WZ Sge-type dwarf novae AL Com and ASASSN-13ck showed a long-lasting (plateau-type) rebrightening. In the early phase of the rebrightening, both objects showed a precursor-like outburst, suggesting that the long-lasting rebrightening is triggered by a precursor outburst.
We studied the background dwarf nova of KIC 11412044 in the Kepler public data and identified it with GALEX J194419.33+491257.0. This object turned out to be a very active SU UMa-type dwarf nova having a mean supercycle of about 150 d and frequent no
rmal outbursts having intervals of 4-10 d. The object showed strong persistent signal of the orbital variation with a period of 0.0528164(4) d (76.06 min) and superhumps with a typical period of 0.0548 d during superoutbursts. Most of the superoutbursts were accompanied by a precursor outburst. All these features are unusual for this very short orbital period. We succeeded in detecting the evolving stage of superhumps (stage A superhumps) and obtained a mass ratio of 0.141(2), which is unusually high for this orbital period. We suggest that the unusual outburst properties are a result of this high mass ratio. We suspect that this object is a member of the recently recognized class of cataclysmic variables (CVs) with a stripped core evolved secondary which are evolving toward AM CVn-type CVs. The present determination of the mass ratio using stage A superhumps makes the first case in such systems.
Continuing the project described by Kato et al. (2009a, arXiv:0905.1757), we collected times of superhump maxima for SU UMa-type dwarf novae mainly observed during the 2012-2013 season. We found three objects (V444 Peg, CSS J203937 and MASTER J212624
) having strongly positive period derivatives despite the long orbital period (Porb). By using the period of growing stage (stage A) superhumps, we obtained mass ratios for six objects. We characterized nine new WZ Sge-type dwarf novae. We made a pilot survey of the decline rate of slowly fading part of SU UMa-type and WZ Sge-type outbursts. The decline time scale was found to generally follow the expected Porb^(1/4) dependence and WZ Sge-type outbursts also generally follow this trend. There are some objects which show slower decline rates, and we consider these objects good candidates for period bouncers. We also studied unusual behavior in some objects, including BK Lyn which made a transition from an ER UMa-type state to the novalike (standstill) state in 2013 and unusually frequent occurrence of superoutbursts in NY Ser and CR Boo. We applied least absolute shrinkage and selection operator (Lasso) power spectral analysis, which has been proven to be very effective in analyzing the Kepler data, to ground-based photometry of BK Lyn and detected the dramatic disappearance of the signal of negative superhumps in 2013. We suggested that the mass-transfer rates did not vary strongly between the ER UMa-type state and novalike state in BK Lyn, and this transition was less likely caused by a systematic variation of the mass-transfer rate.
We made a supplemental study of the superoutbursts and superhumps in SU UMa stars by using the recently released Kepler public data of V1504 Cyg and V344 Lyr. One of the superoutbursts in V1504 Cyg was preceded by a precursor normal outburst which wa
s well separated from the main superoutburst. The superhump first appeared during the descending branch of the precursor normal outburst and it continued into quiescence (the deep dip between the precursor and the main superoutburst), and it began to grow in amplitude with the growth of the main superoutburst after quiescence ended. A similar phenomenon was also observed in V344 Lyr. This observation demonstrates very clearly that the superoutburst was triggered by the superhump (i.e., by the tidal instability), supporting the thermal-tidal instability model. Smak (2013, Acta Astron., 63, 109, arXiv:1301.0187) criticized our previous paper (Osaki and Kato, 2013, PASJ, 65, 50, arXiv:1212.1516) and challenged our main conclusion that various observational lines of evidence of V1504 Cyg support the thermal-tidal instability model for the superoutbursts of SU UMa stars. We present our detailed accounts to all of his criticisms by offering clear explanations. We conclude that the thermal-tidal instability model is after all only the viable model for the superoutbursts and superhumps in SU UMa stars.
We analyzed the Kepler long cadence data of KIC 7524178 (=KIS J192254.92+430905.4), and found that it is an SU UMa-type dwarf nova with frequent normal outbursts. The signal of the negative superhump was always the dominant one even during the supero
utburst, in contrast to our common knowledge about superhumps in dwarf novae. The signal of the positive superhump was only transiently seen during the superoutburst, and it quickly decayed after the superoutburst. The frequency variation of the negative superhump was similar to the two previously studied dwarf novae in the Kepler field, V1504 Cyg and V344 Lyr. This is the first object in which the negative superhumps dominate throughout the supercycle. Nevertheless, the superoutburst was faithfully accompanied by the positive superhump, indicating that the tidal eccentric instability is essential for triggering a superoutburst. All the pieces of evidence strengthen the thermal-tidal instability as the origin of the superoutburst and supercycle, making this object the third such example in the Kepler field. This object had unusually small (~1.0 mag) outburst amplitude and we discussed that the object has a high mass-transfer rate close to the thermal stability limit of the accretion disk. The periods of the negative and positive superhumps, and that of the candidate orbital period were 0.07288 d (average, variable in the range 0.0723-0.0731 d), 0.0785 d (average, variable in the range 0.0772-0.0788 d) and 0.074606(1) d, respectively.
We observed the 2012-2013 superoutburst of the newly identified transient SSS J122221.7-311523 and found that this object showed successive two superoutbursts. Superhumps grew in amplitude during the second superoutburst and showed a characteristic p
attern of period change reflecting the growth of the superhump. Assuming that the periods of superhumps during the growing stage [0.07721(1) d] and post-superoutburst stage [0.07673(3) d], represent the dynamical precession rates at the radius of the 3:1 resonance and the radius immediately after the superoutburst, respectively, we found that this object has a very small mass ratio q=M2/M1 < 0.05. The possible orbital period from quiescent data suggests q=0.045, one of the smallest among hydrogen-rich cataclysmic variables. The long orbital period and low q make this object a perfect candidate for a period bouncer. We suggest that the peculiar pattern of double superoutburst is a result of a low q and may be characteristic to period bouncers.
We propose a new dynamical method to estimate binary mass ratios by using the period of superhumps in SU UMa-type dwarf novae during the growing stage (the stage A superhumps). This method is based on a working hypothesis in which the period of the s
uperhumps at the growing stage is determined by the dynamical precession rate at the 3:1 resonance radius, a picture suggested in our new interpretation of the superhump period evolution during the superoutburst (Osaki, Kato 2013, arXiv:1305.5877). By comparison with the objects with known mass ratios, we show that our method can provide sufficiently accurate mass ratios comparable to those obtained by quiescent eclipse observations. This method is very advantageous in that it requires neither eclipses, nor an experimental calibration. It is particularly suited for exploring the low mass-ratio end of the evolution of cataclysmic variables, where the secondary is undetectable by conventional methods. Our analysis suggests that previous estimates of mass ratios using superhump periods during superoutburst were systematically underestimated for low mass-ratio systems and we provided a new calibration. It suggests that most of WZ Sge-type dwarf novae have secondaries close to the border of the lower main-sequence and brown dwarfs, and most of the objects have not yet reached the evolutionary stage of period bouncers. Our result is not in contradiction with an assumption that the observed minimum period (~77 min) of ordinary hydrogen-rich cataclysmic variables is indeed the period minimum. We highlight the importance of early observation of stage A superhumps and propose a future desirable strategy of observation.
We have studied the short-cadence Kepler public light curves of SU UMa stars, V344 Lyr and V1504 Cyg extending over a period of more than two years by using power spectral analysis. We determined the orbital period of V344 Lyr to be Porb=0.087903(1)
d. We also reanalyzed the frequency variation of the negative superhump in a complete supercycle of V1504 Cyg with additional data of the O-C diagram, confirming that its characteristic variation is in accordance with the thermal-tidal instability model. We present a new two-dimensional period analysis based on a new method of a least absolute shrinkage and selection operator (Lasso). The new method gives very sharp peaks in the power spectra, and it is very useful for studying of the frequency variation in cataclysmic variable stars. We also analyzed simultaneous frequency variations of the positive and negative superhumps. If they are appropriately converted, it is found that they vary in unison, indicating that they represent a disk-radius variation. We have also studied the frequency (or period) variations of positive superhumps during superoutbursts. These variations can be understood in a qualitative way by combining of the disk radius variation and the variation of pressure effects during a superoutburst. A sudden excitation of oscillation with a frequency range near to the negative superhump (which we call impulsive negative superhump) was observed in the descending branch of several outbursts of V344 Lyr. These events seem to have occurred just prior to the next superoutburst, and to act as a lead of the impending superoutburst.
