betaFeedback

WorksPhysics and Astronomy

Photometric Study of a Distant SU UMa-type Dwarf Nova KSP-OT-201803a

Abstract

We present the discovery and photometric study of a distant dwarf nova KSP-OT-201803a using BVI-band observation data obtained by using the KMTNet Supernova Program. One superoutburst (SO) and three normal outbursts are identified from two years of observations (2018-2019), which show that KSP-OT-201803a is an SU UMa-type dwarf nova (DN). Applying the phase dispersion minimization method to the superhump data during the plateau phase of the SO yields a superhump period of P_sh = 127.4 minutes (=2.123 hr =0.08847 days). By stacking ten images of the quiescent phase, we find the quiescent phase magnitude of V = 22.59 +/- 0.18 mag. The SO shows quiescent-to-peak amplitude of 4.97 +/- 0.18 mag and duration of 16.59 days which are typical for the SU UMa-type DNe. Using the superhump period-orbital period relation of SU UMa stars, we obtain the approximate distance of 3.7 +/- 0.3 kpc to KSP-OT-201803a and Galactocentric distance of 9.4 kpc and height from the Galactic plane of +1.2 kpc. Stars at these large distances may belong to the Galactic halo or thick disk, which contain old, metal-poor Population II stars; therefore, these systems may have undergone a different evolutionary process than Population I objects. The orbital period gap of DNe spans 2.45-3.18 hours. Assuming that the orbital period of KSP-OT-201803a is a few percent shorter than its superhump period, KSP-OT-201803a may have just emerged from the period gap and recently resumed mass transfer after crossing its lower boundary.

Introduction

Dwarf novae (DNe) are one of the cataclysmic variables (CVs) and close binary systems with a primary white dwarf (WD) and a Roche lobe-filling low mass dwarf companion ([46, 13]). The companion loses and transfers mass to the primary WD through an accretion disk around the primary. This causes evolution of the system toward a shorter orbital period (PorbP_{\mathrm{orb}}) and smaller mass ratio (q=M2/M1q = M_2/M_1). The thermal instability-induced outburst is observed as DNe with magnitude amplitude of 2 − 10 mag and the outburst durations of days to weeks ([27, 33]).

Corresponding author: Sang Chul Kim

DNe are classified into three types based on their optical light curves (LCs): U Geminorum (U Gem) or SS Cygni (SS Cyg)-type with relatively regular outbursts and longer orbital periods, Z Camelopardalis (Z Cam)-type with a standstill (intermediate brightness for some time), and SU Ursae Majoris (SU UMa)-type with long and brighter superoutbursts (SOs) and short and less bright normal outbursts (NOs) ([29]). SU UMa-type DNe are further classified into ER UMa ([24]), SU UMa ([43]), and WZ Sagittae (Sge)-types ([17]) in the decreasing order of mass transfer.

As the DNe of longer orbital period (Porb>3P_{\mathrm{orb}} > 3 hr) like Z Cam ([39]) and U Gem ([10]) evolve, the donor star has a sufficiently deep convective envelope and magnetic braking can efficiently remove orbital angular momentum from the binary through the interaction between the donor’s magnetic field and its stellar wind. This loss of the angular momentum causes the orbital period to decrease. When the secondary star becomes fully convective at an orbital period of Porb∼3P_{\rm orb} \sim3 hr, magnetic braking weakens, reducing the mass-transfer rate and causing the binary to become detached. This process is thought to produce the observed period gap in DNe between ∼2\sim2 and ∼3\sim3 hrs.

During the time the system stays in the period gap, gravitational radiation [30] becomes the dominant mechanism driving angular-momentum loss, causing the binary to shrink further. As the Roche lobe gradually shrinks, mass transfer resumes once the secondary fills its Roche lobe again near the lower edge of the period gap (∼2\sim2 hr). The system continues to evolve toward the minimum orbital period ([19, 37, 20, 38]), as the donor becomes increasingly degenerate. After reaching the period minimum, the orbital period begins to increase again, and such a system is termed a period-bouncer. There is, however, also an alternative idea of complexity of the surface magnetic field of the donor star for the explanation of the period gap ([9]).

While the lower (Pgap,−P_{\rm gap,-}) and upper (Pgap,+P_{\rm gap,+}) edge values of the period gap have some ranges: 2−2.452-2.45 hr and 3−3.183-3.18 hr, respectively, there seems to be no consensus on the exact values, especially for the lower edge value. As noted by Schreiber et al. [38], compared to the volume-limited samples, the SDSS samples contain fewer systems below the period gap because the low mass-transfer rates in these systems make them fainter. Consequently, low-accretion-rate CVs may be underrepresented in SDSS, which identifies CVs primarily through spectroscopy. This observational bias may reduce the sensitivity of SDSS to short-period systems with low mass-transfer rates, potentially affecting the value of Pgap,−P_{\rm gap,-}.

Following the approach of the KMTNet Supernova Program (KSP) ([23, 22, 24, 17]), to discovering and studying DNe using 24-hour photometric coverage with its 1.6 m telescopes, this paper presents the discovery, photometric study, and light curve analysis of a new SU UMa-type DN, KSP-OT-201803a. As shown in § 2, KSP-OT-201803a appears to be located at greater distance (∼3.7\sim3.7 kpc, Galactocentric distance of 9.4 kpc and +1.2 kpc from the Galactic plane) than the previously known population of DNe (≲1\lesssim1 kpc, [23]). At such large distances, these systems may belong to the Galactic halo or thick disk, which contain older, metal-poor Population II stars. Therefore, these systems may have undergone evolutionary processes different from those of Population I objects. By presenting the results for KSP-OT-201803a at large distance, we aim to identify and compile a sample of candidate Population II DNe for further investigation.

This paper is organized as follows. § 2 describes our observations and data reduction. § 3 presents the main results of our LC, color-evolution, and period analyses. § 4 discusses the period gap, Gaia observations, and the morphology of the outburst LC. Finally, a summary is provided in § 5.

Observations and Discovery

The coordinates of KSP-OT-201803a are αJ2000=09h58m3.25s\alpha_{\rm J2000}=09^{\rm h}58^{\rm m}3.25^{\rm s} (149.513530∘149.513530^\circ), δJ2000=−29∘35′40.3′′\delta_{\rm J2000}=-29^\circ35'40.3'' (−29.594519∘-29.594519^\circ) and (l,b)=(263.58∘,+19.73∘)(l,b)=(263.58^\circ,+19.73^\circ). In this paper, for the observations, we have used the three identical 1.6 m telescopes of the KMTNet ([18]) installed in Chile, Australia, and South Africa as part of the KMTNet Supernova Program (KSP) ([26]). KMTNet is equipped with wide-field (2∘×2∘2^\circ\times2^\circ at a pixel scale of 0.′′40.''4) CCD camera and an exposure time of 60 seconds are used for each of the BVIBVI filters. KSP-OT-201803a was first discovered at the KMTNet in the field of the lenticular galaxy NGC 3056 by the telescope in Chile in the BB-band (B=18.05±0.02B=18.05\pm0.02 mag) on 2018 Mar 29 00h15m50.4s00^{\rm h}15^{\rm m}50.4^{\rm s} (UT, MJD = 58206.011), and followed by VV and II bands: V=18.23±0.03V=18.23\pm0.03 mag at 00h17m16.8s00^{\rm h}17^{\rm m}16.8^{\rm s} (UT) and I=17.45±0.02I=17.45\pm0.02 mag at 00h20m09.6s00^{\rm h}20^{\rm m}09.6^{\rm s} (UT) on the same night. We performed monitoring observations for KSP-OT-201803a during MJD = 58150-58300 and 58520-58670 days (2018-2019 years). Figure 1 shows a VV-band image centered on KSP-OT-201803a, which is a stacked image produced by stacking ten individual 60-second exposures before the SO (MJD = 58170.1 to 58174.1 days).

V-band stacked image of KSP-OT-201803a

Figure 1. VV-band stacked image of ten individual exposures of 60 seconds each before the superoutburst during MJD = 58170.1 to 58174.1 days of KSP-OT-201803a. North is up, and east is to the left. Crosshair shows the position of KSP-OT-201803a and horizontal bar of 5″ is shown in the bottom-left corner.

We conducted point-spread function (PSF) photometry using the DAOPHOT-II package [42] and used the AAVSO Photometric All Sky Survey (APASS) database on the same target.2 For the standardization of the magnitudes, we performed a BB-band color correction of ΔB=B−B′≃0.27(B−V)+offset\Delta B = B - B^\prime\simeq0.27(B - V) + \text{offset}, where BB and B′B^\prime are color-corrected BB-band magnitudes in the standard Johnson system and uncorrected KNTNet BB-band magnitudes, respectively [31]. The AAVSO photometry is provided with the BViBV i-bands, therefore we applied another correction of I=i−0.4I = i - 0.4 mag [see 31].

We have used the Galactic dust extinction map of Green et al. [11] toward the position of KSP-OT-201803a. Using equation (7) of Mennickent et al. (1999) to convert the superhump period Psh=0.08847P_{\mathrm{sh}} = 0.08847 days of KSP-OT-201803a (see § 3.3 below) to the orbital period, we obtain Porb∼0.08444(93)P_{\mathrm{orb}} \sim0.08444(93) days3 (= 2.03 hr). The relation between the SO peak magnitude and the orbital period given by Patterson [32] (their equation (4)) yields the peak absolute magnitude of MV,peak∼4.55±0.18M_{\mathrm{V,peak}} \sim4.55 \pm0.18 mag for the SO peak of KSP-OT-201803a. This gives an extinction-uncorrected distance modulus of (V−MV)∼17.62(±0.02)−4.55(±0.18)≈13.07±0.18(V - M_V) \sim17.62(\pm0.02) - 4.55(\pm0.18) \approx13.07 \pm0.18 mag (see § 3.1 below for the peak magnitude of the SO) and an approximate distance of ∼4.1±0.4\sim4.1 \pm0.4 kpc. Using this rough distance estimate, the reddening is estimated to be E(B−V)≈0.07E(B - V) \approx0.07 mag and E(V−I)≈0.09E(V - I) \approx0.09 mag. Applying this reddening to the above distance estimate finally gives an extinction-corrected distance of d∼3.7±0.3d \sim3.7 \pm0.3 kpc.4 The magnitudes given in this paper are not extinction corrected values. Although we searched the Gaia DR3 catalog (Gaia Collaboration et al. [8]), we could not obtain a distance estimate for KSP-OT-201803a (see § 4.2).

Results

Light Curve

Figure 2 (a) shows the light curve (LC) of KSP-OT-201803a for all the observed duration of 2018 (before MJD = 58300 days) and 2019 (after MJD = 58500 days) with a seasonal gap between them and zoomed-in details of the superoutburst (SO) and three normal outbursts (NOs: N1, N2, and N3) are shown in panels, (b), (c), (d), and (e), respectively. Each of the four blue star symbols in panel (a) are the photometry results from 10 stacked images (each reaching the exposure time of 600 seconds). While we increased the tolerance range to include data with magnitude errors <0.2< 0.2 mag, some single-image photometry results still show values similar to those from the stacked-image photometry, especially at 2018 observations. KSP-OT-201803a can be faintly seen in some of the single exposure images, although not all images show this behavior. We have fitted the SO with order 8 polynomial function and shown as dashed line in panel (b). For the three NOs, data for N1 and N3 are relatively sparse compared to N2. For N2 we fitted a polynomial function with order 4 (panel (d)) and overplotted this function in panel (c) (N1) and panel (e) (N3) for comparison among the NOs. N1 appears to show a less luminous peak than N2, while N3 exhibits a LC shape quite similar to that of N2 based on the small number of data points for N3. However, these comparisons should be treated with caution because the numbers of data points for N1 and N3 are limited.

Multi-panel light curves and color evolution of KSP-OT-201803a

Figure 2. (a) VV-band light curve of KSP-OT-201803a from 2018 February to June (before MJD =58300= 58300 days) and 2019 February to July (after MJD =58500= 58500 days). Superoutburst (SO) is shown in detail in panel (b), and the three normal outbursts (NOs: N1, N2, and N3) are shown in panels (c), (d), and (e), respectively. Color evolution (lower panels) for each of the four outbursts is shown in panels (f), (g), (h), and (i) for the SO, N1, N2, and N3, respectively. Data with errors <0.2< 0.2 mag are used. Short green vertical lines at the bottom of panel (a) show the epochs when the VV-band observations were made. Four blue star symbols in panel (a) denote the quiescent phase magnitudes obtained from photometry on stacked images of 10 exposures (=600= 600 seconds). The mean of these quiescent phase magnitudes are denoted as blue horizontal dotted lines at V=22.59±0.18V = 22.59 \pm0.18 mag in panels (a) to (e). While the gray dashed line in panel (b) shows the polynomial fitting result for the SO with order 8, we tried another polynomial fitting for the N2 (order 4, panel (d)) and this fitting function is overplotted on N1 (panel (c)) and N3 (panel (e)) for comparison. The open circles in panels (f) to (i) are data points, the dashed lines in panel (f) are polynomial fittings and those in panel (g) to (i) are mean colors. The three star symbols on the left of panel (f) show the quiescent phase colors (see text for details).

From the 8th-order polynomial fit to the SO data, we find that the luminosity peak reached V=17.62±0.02V = 17.62 \pm0.02 mag at MJD = 58206.4605 days. After the peak, the plateau phase continued until MJD = 58218.7665 days, giving a plateau duration of ∼12.3\sim12.3 days. We obtained quiescent phase photometry at four epochs and are shown in Table 1 and in Figure 2 (a) as blue star symbols. The mean magnitudes of the quiescent phase are B=22.88±0.14B = 22.88 \pm0.14, V=22.59±0.18V = 22.59 \pm0.18, and I=22.18±0.14I = 22.18 \pm0.14 mag, and shown as blue horizontal dotted lines in panels (a) to (e) of Figure 2. Since we have adopted magnitude errors to be <0.2< 0.2 mag for the single image photometry, the quiescent phase photometry from 10 stacked images shows magnitudes similar to those measured from single images near MJD ∼58172\sim58172 and 58252 days, when the Moon was near its new phase and seeing conditions were good. This is the reason why the target can be seen faintly in some of the single images during these time spans. The magnitude amplitude of the SO between the peak and quiescent-phase is 4.97±0.184.97 \pm0.18 mag, rising and declining rates are 4.54 and 1.05 mag day−1^{-1}, and the duration is 16.59 days (from MJD = 58204.2568 to 58220.8514 days).

BandMJD MiddleMJD First imageMJD Last imageNumber of imagesmagerror
BB58172.203058170.202458174.20371023.050.12
VV58172.064358170.060458174.06821022.600.12
II58172.274958170.480258174.06961022.210.15
BB58251.532458250.071958252.99301022.710.08
VV58251.748658250.745058252.75221022.300.10
II58251.267358250.074858252.45991021.950.11
BB58573.081858571.104458575.05921022.990.13
VV58573.129858571.199058575.06061022.740.15
II58573.469658571.936258575.00311022.300.14
BB58626.901758626.032858627.77071022.770.09
VV58626.880258626.034158627.72621022.730.11
II58626.881658626.035558627.72771022.250.10

Table 1. Quiescent Phase Photometry of KSP-OT-201803a

The polynomial function fitting for the N2 gives the peak magnitude of V≈18.13V \approx18.13 mag at MJD = 58600.1942 days. The N2 amplitude is 4.46±0.184.46 \pm0.18 mag, rising and decline rates are 3.71 and 1.50 mag day−1^{-1}, and the duration is 4.18 days (from MJD = 58598.9930 to 58603.1772 days). Since we did not attempt to fit N1 and N3 due to insufficient data, the parameters of N2 are used to fit N1 and N3 solely to determine their peak MJDs, which are 58276.8942 and 58649.1942 days, respectively.

The SOs, NOs, and superhumps during the SO plateau indicate that KSP-OT-201803a is an SU UMa-type DN (e.g., Kato et al. [14]). Otulakowska-Hypka et al. [29] conducted a comprehensive statistical analysis of 653 DNe. Their Figure 13 shows a strong correlation between SO and NO amplitudes (ASOA_{\mathrm{SO}} vs. ANOA_{\mathrm{NO}}), placing KSP-OT-201803a well within the SU UMa-type DNe. Their Figures 15–17 present distributions of durations, amplitudes, and orbital periods for NOs and SOs. The corresponding parameters of KSP-OT-201803a fall well within the ranges observed in SU UMa-type DNe, regardless of the correlation strength.

Color Evolution

Figure 2 (f) to (i) (lower panels) shows the color evolutions of KSP-OT-201803a, where open circles are the observed data points, dashed lines in panel (f) are polynomial fittings, and those in panels (g) to (i) are mean colors. Figure 2 (f) shows that all the three colors remain constant (B−V=0.19±0.03B-V = 0.19\pm0.03, V−I=−0.31±0.03V-I = -0.31\pm0.03, and B−I=−0.12±0.03B-I = -0.12\pm0.03 mag) from MJD = 58205 to 58215 days during which the outburst brightness changes. As it gets closer to the end of the SO and to the quiescent phase magnitudes, all the colors get redder. This reddening near the end of the decline can be explained by the cooling of the accretion disk ([6, 5]). As a cooling front propagates inward, the II-band flux responds more sensitively than the shorter-wavelength fluxes, causing the colors containing the II-band get redder earlier than the B−VB-V color. This effect arises because the contribution from steady mass accretion, which is more prominent at longer wavelengths, becomes increasingly significant as soon as the outburst component, which is more evident at shorter wavelengths, begins to fade (Hameury et al. [12]).

We calculate the quiescent phase colors from the quiescent phase magnitudes obtained in §3.1 (average from Table 1) and show them as star symbols in Figure 2 (f). Just as in the case of KSP-OT-202104a (Kim et al. [17]), which is a WZ Sge-type DN belonging to the SU UMa-type DNe, the SO colors converge to the quiescent phase colors as the outburst fades away.

The mean colors of the three NOs generally remain constant and are shown in Table 2. The mean B−VB-V colors are 0.16, 0.17, and 0.24 mag, for N1, N2, and N3, respectively. As in the SO, the colors of N1 and N2 (panels (g) and (h)) show reddened values after the outbursts vanished and reached the quiescent phase. All the three color indices show systematically bluer in N1 and redder in N3 than in N2, although the offsets are small.

ColorSuperoutburst [mag]N1 [mag]N2 [mag]N3 [mag]
B−VB - V+0.19±0.03+0.19 \pm0.03+0.16±0.08+0.16 \pm0.08+0.17±0.05+0.17 \pm0.05+0.24±0.08+0.24 \pm0.08
V−IV - I−0.31±0.03-0.31 \pm0.03−0.34±0.04-0.34 \pm0.04−0.30±0.06-0.30 \pm0.06−0.27±0.07-0.27 \pm0.07
B−IB - I−0.12±0.03-0.12 \pm0.03−0.18±0.04-0.18 \pm0.04−0.12±0.06-0.12 \pm0.06−0.06±0.07-0.06 \pm0.07

Table 2. Mean Colors of the Outbursts of KSP-OT-201803a

For the SU UMa-type DN KSP-OT-201611a, Lee et al. [23] have shown that the color values are almost constant at B−V=0.23B-V = 0.23 and V−I=0.06V-I = 0.06 mag (extinction uncorrected) during the SO, while they show linear change during the NO (B−VB-V gets bluer and V−IV-I becomes redder). The nearly constant color values during the SOs of KSP-OT-201611a and our target are naturally explained by the constant viscosity model in the accretion disk, which predicts an approximately constant mass transfer rate ([23, 5]). Although the colors of KSP-OT-201803a N2 appear to remain nearly constant in Figure 2 (h), slight color variations are still evident. The B−VB-V color shows a slight tendency to become bluer after the peak, while the V−IV-I color reaches its bluest value before the peak and subsequently becomes progressively redder. These trends are consistent with the color evolution described by Lee et al. [23] for their NO. During the NO, thermal instability triggers a transition of the outer accretion disk from a cool state to a hot state. The heating front then propagates inward, eventually bringing the entire accretion disk into the hot state. At this stage, the system reaches its bluest color (Cannizzo et al. [6]). Subsequently, the development of UV emission in the inner disk slightly delays the occurrence of the bluest color in B−VB-V. In contrast, the bluest color is reached earlier in V−IV-I, followed by a more rapid evolution toward redder colors due to the supply of cool material to the outer regions of the accretion disk.

Period Analysis

During SOs of SU UMa-type DNe, small amplitude (up to 0.2 mag) magnitude oscillations known as ‘superhumps’ occur. These oscillations are caused by varying tidal dissipation in a precessing, eccentric and tidally distorted accretion disk (e.g., Kato et al. [14]). SU UMa-type DNe exhibit three stages during the superhump phase: ‘A’ is an early evolutionary phase with a longer superhump period, ‘B’ is an intermediate phase with systematically varying periods, and ‘C’ is a phase with shorter and almost constant period (Kato et al. [14]). To obtain the period of the superhump (PshP_{\mathrm{sh}}) during the plateau of the SO, we use the phase dispersion minimization (PDM: Stellingwerf [41]) method, which

finds periods from unevenly sampled time-series data [44, 17]. Due to the relatively low cadence of our observations, it is difficult to directly identify the superhump stages using an O-C diagram. However, empirical studies have shown that the transition from superhump stage A to stage B generally occurs a few days after the outburst peak [32, 14]. Since we measured the superhump period of KSP-OT-201803a using observations obtained at MJD 58210–58217, approximately 3.5 days after the VV-band peak, we consider the measured superhumps to be approximately in stage B.

Figure 3 shows the results of the PDM analysis. The black lines on the left panels are the median Θ\Theta and the shaded areas are the 90% confidence ranges from 1,000 bootstrap resamplings. All of the three (BB, VV, and II) bands show the same obvious peaks at Psh=127.4P_{\mathrm{sh}} = 127.4 minutes (= 2.123 hr = 0.08847 days). The right panels are phase-folded profiles drawn using this period with the sine curves fitted to the data shown in red solid lines. Considering the fact that the superhump periods of SU UMa-type DNe are usually a few percent longer than the orbital periods (“positive superhumps”; [34, 16, 17]), the orbital period of KSP-OT-201803a would be a few percent shorter than the superhump period derived here.

PDM analysis and phase-folded profiles of KSP-OT-201803a

Figure 3. Results of the PDM analysis (left) and phase-folded profiles (right) of KSP-OT-201803a. Panels (a) and (b) are for BB-band, (c) and (d) are for VV-band, and (e) and (f) are for II-band. In the left panels, the black solid lines are median Θ\Theta and the shaded areas show 90% confidence ranges from 1,000 bootstrap resamplings. All the BB, VV, and II-band data show the same superhump period of 127.4 minutes (= 2.123 hr = 0.08847 days). Red solid lines on the right panels are sine curves fitted to the data with the obtained period.

Discussion

Period Gap

The period gap is characterized by a deficit of CVs with orbital periods between 2 – 3 hours [35, 40], while there is a sharp cutoff at the lowest value known as the ‘period minimum’ of ∼76\sim76 minutes (70 – 80 min: Knigge [19]; [17]). To explain the period gap, the ‘disrupted magnetic braking’ model was proposed [35, 40]. The values of the upper and lower edges of the period gap are not known exactly: Knigge [19] and Knigge et al. [20] claim Pgap,−=2.15±0.03P_{\mathrm{gap,-}} = 2.15 \pm0.03 h and Pgap,+=3.18±0.04P_{\mathrm{gap,+}} = 3.18 \pm0.04 h, and Schreiber et al. [38] suggests Pgap,−=2.45P_{\mathrm{gap,-}} = 2.45 h and Pgap,+=3.18P_{\mathrm{gap,+}} = 3.18 h. It might also be possible that there would not be sharp boundaries for the period gap [38]. Considering the fact that usually the superhump period is a few percent longer than the orbital period (“positive superhumps”; [15, 17] (Osaki 2003)), it is probable that the orbital period of KSP-OT-201803a might be a bit shorter than Pgap,−P_{\mathrm{gap,-}}. This means that KSP-OT-201803a could be, if the magnetic braking model is correct, an object that just resumed the mass transfer and became observable to us. Better understanding in the future would be made on the changes taking place at this transition stage if enough observations are made for the objects near the edge.

Gaia Observations

We have searched the archival data of the American Association of Variable Star Observers, Catalina Real-Time Transient Survey [7], All-Sky Automated Survey for SuperNovae [21], and Zwicky Transient Facility (ZTF, [2]), to see if any of them contain signals from KSP-OT-201803a, but could not find any.

We searched the Gaia DR3 catalog [8] for distance information to KSP-OT-201803a but could not obtain a distance estimate. However, we found that KSP-OT-201803a had been observed six times by Gaia as shown in Table 3. The first three observations (mean G≃17.86G \simeq17.86 mag) correspond to ∼3\sim3 days after the peak of SO in Figure 2 (b). The detection in 2021 at G=17.58G = 17.58 mag appears to be made at another SO, while those in 2020 could be during a NO or another SO in rising or declining phases. Although the Gaia observations spanned a total of 3 years and 2 months, KSP-OT-201803a was detected only during outbursts because it was too faint during quiescence. Future observations with larger telescopes (e.g., Rubin/LSST) may enable us to investigate its recurrence cycle.

MJD (days)Date (UT)G (mag)
58209.165582018 Apr 1 03h 58m 26.1s03^{h}\ 58^{m}\ 26.1^{s}17.86
58209.415742018 Apr 1 09h 58m 39.9s09^{h}\ 58^{m}\ 39.9^{s}17.82
58209.489752018 Apr 1 11h 45m 14.4s11^{h}\ 45^{m}\ 14.4^{s}17.91
59026.632652020 Jun 26 15h 11m 01.0s15^{h}\ 11^{m}\ 01.0^{s}18.24
59026.706662020 Jun 26 16h 57m 35.4s16^{h}\ 57^{m}\ 35.4^{s}18.31
59372.198192021 Jun 7 04h 45m 23.6s04^{h}\ 45^{m}\ 23.6^{s}17.58

Table 3. Gaia Observations for KSP-OT-201803a (ID: Gaia18aus)

Morphology of the Outburst LCs

We overplotted the outburst LCs in Figure 4: SO VV-band data are shown in gray color (open circles and dashed line), three NOs in filled circles (N1 blue, N2 red, and N3 green) and red dashed lines. The three NOs are shown twice at the rising and declining parts of the SO. The three NOs match very well with each other and with the SO at the rising and declining parts, while there are some scattered points near the NO peaks implying a bit different peak brightnesses. Note that there are not enough data points at the rising part of the SO, therefore this part should not be regarded to be reliable.

Overlap of the outburst light curves

Figure 4. Overlap of the outburst light curves. VV-band SO is shown in gray open circles and gray dashed line. Three normal outbursts in VV-band (N1 blue, N2 red, and N3 green) are plotted as filled circles and red dashed lines twice at rising and declining parts.

For the 21 outbursts of the U Gem-type DN CH UMa during 1973−1998, Simon [45] showed that their peak brightnesses are bimodal: “bright” and “faint” being separated at 12 magvis_{\mathrm{vis}}. While the bright outbursts are classified into two types of fast and slow decays by the decay branches, the bright outbursts are symmetric and the faint outbursts are asymmetric (decay being slower than the rise). On the other hand, Bath & van Paradijs [1] classified the outburst LCs of SS Cygni from 1897 till 1979 into four types : A, B, C, and D from the steepest rise to the most shallow rise patterns. Outbursts propagating outward from the inner disk (‘inside-out’) produce symmetric LCs, whereas inward-propagating outbursts from the outer disk (‘outside-in’) yield asymmetric LCs with faster rises than decays [6, 5, 23]. The asymmetric LC in Figure 4, characterized by steeper rises than decays, supports the outside-in scenario as the more likely mechanism for the 2018−2019 outbursts of KSP-OT-201803a. The NOs in SU UMa systems are generally believed to be outside-in type [4], and outside-in outbursts appear to be more common than inside-out outbursts in numerous studies of individual SU UMa-type systems (see, e.g., Bruch et al. 2026).

Summary and Conclusion

In this work, from two years of observations by the KMTNet Supernova Program for the SU UMa-type DN KSP-OT-201803a, we have studied the LCs and obtained the results as summarized in the following.

  • One superoutburst (SO) and three normal outbursts (NOs) are identified during the observations of 2018 – 2019.

  • By stacking ten images of the quiescent phase, we estimate the quiescent phase magnitude of VV = 22.59 ± 0.18 mag.

  • The LC parameters of SO are : magnitude amplitude of 4.97 ± 0.18 mag, peak magnitude of VV = 17.62 ± 0.02 mag, and outburst duration of 16.59 days. The second NO (N2), among the three NOs, is most thoroughly observed and gives the following LC parameters : amplitude 4.46 ± 0.18

mag, peak magnitude V≈18.13V \approx18.13 mag, and duration of 4.18 days.

  • Applying the phase dispersion minimization method to the superhump data during the plateau phase of the SO yields a superhump period of Psh=127.4P_{\mathrm{sh}} = 127.4 minutes (= 2.123 hr = 0.08847 days). We assume the orbital period of KSP-OT-201803a would be a few percent shorter than this superhump period. Schreiber et al. [38] gives the lower and upper boundaries of the orbital period gap to be Pgap,−=2.45P_{\mathrm{gap,-}} = 2.45 hours and Pgap,+=3.18P_{\mathrm{gap,+}} = 3.18 hours, respectively. It might be possible that KSP-OT-201803a just crossed the lower boundary meaning the system recently resumed the mass transfer after the detached phase invisible to us for a while.

  • The asymmetric shapes of the outburst LCs of KSP-OT-201803a support the outside-in propagation scenario where the outburst instability started at the outer part of the accretion disk and moved inward.

We thank the referee for the very helpful comments that improved the manuscript. This research has made use of the KMTNet system operated by the Korea Astronomy and Space Science Institute (KASI) at three host sites of CTIO in Chile, SAAO in South Africa, and SSO in Australia. Data transfer from the host site to KASI was supported by the Korea Research Environment Open NETwork (KREONET). This research was supported by the Korea Astronomy and Space Science Institute under the R&D program (Project No. 2026-1-860-00) supervised by the Ministry of Science and ICT. This work has made use of data from the European Space Agency (ESA) mission Gaia (https://www.cosmos.esa.int/gaia), processed by the Gaia Data Processing and Analysis Consortium (DPAC, https://www.cosmos.esa.int/web/gaia/dpac/consortium). Funding for the DPAC has been provided by national institutions, in particular the institutions participating in the Gaia Multilateral Agreement.

References

  1. [1]Bath, G. T., & van Paradijs, J. 1983, Nature, 305, 33, doi: 10.1038/305033a0DOI
  2. [2]Bellm, E. C., Kulkarni, S. R., Graham, M. J., et al. 2019, PASP, 131, 018002, doi: 10.1088/1538-3873/aaecbeDOI
  3. [4]Buat-Ménard, V., Hameury, J.-M., & Lasota, J.-P. 2001, A&A, 366, 612, doi: 10.1051/0004-6361:20000107DOI
  4. [5]Cannizzo, J. K., & Kenyon, S. J. 1987, ApJ, 320, 319, doi: 10.1086/165545DOI
  5. [6]Cannizzo, J. K., Wheeler, J. C., & Polidan, R. S. 1986, ApJ, 301, 634, doi: 10.1086/163929DOI
  6. [7]Drake, A. J., Djorgovski, S. G., Mahabal, A., et al. 2009, ApJ, 696, 870, doi: 10.1088/0004-637X/696/1/870DOI
  7. [8]Gaia Collaboration, Vallenari, A., Brown, A. G. A., et al. 2023, A&A, 674, A1, doi: 10.1051/0004-6361/202243940
  8. [9]Garraffo, C., Drake, J. J., Alvarado-Gomez, J. D., Moschou, S. P., & Cohen, O. 2018, ApJ, 868, 60, doi: 10.3847/1538-4357/aae589DOI
  9. [10]Godon, P., Shara, M. M., Sion, E. M., & Zurek, D. 2017, ApJ, 850, 146, doi: 10.3847/1538-4357/aa9616DOI
  10. [11]Green, G. M., Schlafly, E., Zucker, C., Speagle, J. S., & Finkbeiner, D. P. 2019, ApJ, 887, 93, doi: 10.3847/1538-4357/ab5362DOI
  11. [12]Hameury, J.-M., Knigge, C., Lasota, J.-P., Hambsch, F.-J., & James, R. 2020, A&A, 636, A1, doi: 10.1051/0004-6361/202037631DOI
  12. [13]Inight, K., Gänsicke, B. T., Breedt, E., et al. 2023, MNRAS, 524, 4867, doi: 10.1093/mnras/stad2018DOI
  13. [14]Kato, T., Imada, A., Uemura, M., et al. 2009a, PASJ, 61, S395, doi: 10.1093/pasj/61.sp2.S395DOI
  14. [15]Kato, T., Pavlenko, E. P., Maehara, H., et al. 2009b, PASJ, 61, 601, doi: 10.1093/pasj/61.3.601DOI
  15. [16]Kato, T., Isogai, K., Hambsch, F.-J., et al. 2017, PASJ, 69, 75, doi: 10.1093/pasj/psx058DOI
  16. [17]Kim, S. C., Lee, Y., Moon, D.-S., et al. 2026, AJ, 172, 20, doi: 10.3847/1538-3881/ae6b7f
  17. [18]Kim, S.-L., Lee, C.-U., Park, B.-G., et al. 2016, Journal of Korean Astronomical Society, 49, 37, doi: 10.5303/JKAS.2016.49.1.37DOI
  18. [19]Knigge, C. 2006, MNRAS, 373, 484, doi: 10.1111/j.1365-2966.2006.11096.xDOI
  19. [20]Knigge, C., Barafe, I., & Patterson, J. 2011, ApJS, 194, 28, doi: 10.1088/0067-0049/194/2/28DOI
  20. [21]Kochanek, C. S., Shappee, B. J., Stanek, K. Z., et al. 2017, PASP, 129, 104502, doi: 10.1088/1538-3873/aa80d9DOI
  21. [22]Lee, Y., Kim, S. C., Moon, D.-S., et al. 2022, ApJL, 925, L22, doi: 10.3847/2041-8213/ac4c41DOI
  22. [23]Lee, Y., Moon, D.-S., Kim, S. C., et al. 2019, ApJ, 880, 109, doi: 10.3847/1538-4357/ab2985DOI
  23. [24]Lee, Y., Moon, D.-S., Kim, S. C., Park, H. S., & Ni, Y. Q. 2024, ApJ, 964, 186, doi: 10.3847/1538-4357/ad25ffDOI
  24. [26]Moon, D.-S., Kim, S. C., Lee, J.-J., et al. 2016, in Proc. SPIE, Vol. 9906, Ground-based and Airborne Telescopes VI, 99064I, doi: 10.1117/12.2233921DOI
  25. [27]Osaki, Y. 1996, PASP, 108, 39, doi: 10.1086/133689DOI
  26. [29]Otulakowska-Hypka, M., Olech, A., & Patterson, J. 2016, MNRAS, 460, 2526, doi: 10.1093/mnras/stw1120DOI
  27. [30]Paczynski, B., & Sienkiewicz, R. 1981, ApJL, 248, L27, doi: 10.1086/183616DOI
  28. [31]Park, H. S., Moon, D.-S., Zaritsky, D., et al. 2017, ApJ, 848, 19, doi: 10.3847/1538-4357/aa88abDOI
  29. [32]Patterson, J. 2011, MNRAS, 411, 2695, doi: 10.1111/j.1365-2966.2010.17881.xDOI
  30. [33]Patterson, J., Augusteijn, T., Harvey, D. A., et al. 1996, PASP, 108, 748, doi: 10.1086/133798DOI
  31. [34]Patterson, J., Kemp, J., Harvey, D. A., et al. 2005, PASP, 117, 1204, doi: 10.1086/447771DOI
  32. [35]Rappaport, S., Verbunt, F., & Joss, P. C. 1983, ApJ, 275, 713, doi: 10.1086/161569DOI
  33. [36]Reid, M. J., Menten, K. M., Brunthaler, A., et al. 2014, ApJ, 783, 130, doi: 10.1088/0004-637X/783/2/130DOI
  34. [37]Schmidtobreick, L., & Tappert, C. 2006, A&A, 455, 255, doi: 10.1051/0004-6361:20054625DOI
  35. [38]Schreiber, M. R., Belloni, D., & Schwope, A. D. 2024, A&A, 682, L7, doi: 10.1051/0004-6361/202348807DOI
  36. [39]Simonsen, M., Boyd, D., Goff, W., et al. 2014, JAAVSO, 42, 177, doi: 10.48550/arXiv.1402.0207DOI
  37. [40]Spruit, H. C., & Ritter, H. 1983, A&A, 124, 267
  38. [41]Stellingwerf, R. F. 1978, ApJ, 224, 953, doi: 10.1086/156444DOI
  39. [42]Stetson, P. B. 1987, PASP, 99, 191, doi: 10.1086/131977DOI
  40. [43]Tampo, Y., Kojiguchi, N., Kato, T., et al. 2025, PASJ, 77, 823, doi: 10.1093/pasj/psaf051DOI
  41. [44]Tampo, Y., Kojiguchi, N., Isogai, K., et al. 2026, MNRAS, 545, staf1964, doi: 10.1093/mnras/staf1964DOI
  42. [45]Šimon, V. 2000, A&A, 354, 103
  43. [46]Warner, B. 1995, Cambridge Astrophysics Series, 28

Paper details

vlnc:W_djtytqtu
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 10

downloads · 1

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents