High-resolution spectroscopic monitoring observations of FU Orionis-type object, V960 Mon
Sunkyung Park, Jeong-Eun Lee, Tae-Soo Pyo, Daniel T. Jaffe, Gregory N. Mace, Hyun-Il Sung, Sang-Gak Lee, Wonseok Kang, Hyung-Il Oh, Tae Seog Yoon, Sung-Yong Yoon, Joel D. Green
aa r X i v : . [ a s t r o - ph . S R ] J u l Draft version July 8, 2020
Typeset using L A TEX default style in AASTeX62
High-resolution spectroscopic monitoring observations of FU Orionis-type object, V960 Mon
Sunkyung Park, Jeong-Eun Lee, Tae-Soo Pyo, Daniel T. Jaffe, Gregory N. Mace, Hyun-Il Sung, Sang-Gak Lee, Wonseok Kang, Hyung-Il Oh, Tae Seog Yoon, Sung-Yong Yoon, and Joel D. Green School of Space Research, Kyung Hee University1732, Deogyeong-daero, Giheung-gu, Yongin-si, Gyeonggi-do, 17104, Republic of [email protected], [email protected] Subaru Telescope, National Astronomical Observatory of Japan650 North Aohoku Place, Hilo, HI 96720, USA Department of Astronomy, University of Texas at Austin2515 Speedway, Austin, TX, USA Korea Astronomy and Space Science Institute776, Daedeok-daero, Yuseong-gu, Daejeon, 34055, Republic of Korea Seoul National University1 Gwanak-ro, Gwanak-gu, Seoul 08826, Republic of Korea National Youth Space Center200, Deokheungyangjjok-gil, Dongil-myeon, Goheung-gun, Jeollanam-do, 59567, Republic of Korea Department of Astronomy and Atmospheric Sciences, Kyungpook National UniversityDaegu 41566, Republic of Korea Space Telescope Science InstituteBaltimore, MD 21218, USA
ABSTRACTWe present the results of high-resolution (R ≥ ± − and 36.3 ± − , respectively. We also report the detection of[S II ] and H emission lines, which are jet/outflow tracers and rarely found in FUors. Keywords:
Stars: formation — Stars: protostars — Stars: individual: V960 Mon, 2MASS J06593168-0405277 — Techniques: spectroscopic INTRODUCTIONLow-mass stars form by gravitational collapse in dense molecular clouds. The material is transferred from an infallingenvelope to a disk and disk material is channeled into the central protostar along the magnetic field (Hartmann et al.1998), growing the mass of the protostar. However, the accretion mechanism from the disk to the central star isstill poorly understood. A steady accretion rate of ∼ × − M ⊙ yr − has been adopted in the standard accretionmodel (Shu 1977). However, the observed luminosities of young stellar objects (YSOs) are often lower compared tothe standard accretion model with the constant accretion rate, which is known as luminosity problem (Kenyon et al.1990; Dunham et al. 2010). A promising explanation of the luminosity problem is an episodic accretion model, whereprotostars spend most of their time in low accretion rates, and thus, with low luminosities, and occasional, relativelybrief bursts of accretion dominate the time-averaged flow of material onto the central star and produce temporarilyhigh luminosities and observable phenomena (Dunham et al. 2010; Audard et al. 2014). Park et al.
FU Orionis-type objects (hereafter, FUors) are an observable evidence of episodic accretion; they are low-mass YSOsshowing large-amplitude outbursts in the optical (∆V ≥ − to a few 10 − M ⊙ yr − ; Hartmann & Kenyon 1996; Herbig et al. 2003; Hartmann 2009; Audard et al.2014). In this context, FUors are ideal testbeds to study episodic accretion. During the outburst, the disk is about100-1000 times brighter than the protostar, so the continuum source of the spectrum observed in optical and near-infrared (NIR) is the disk midplane (Hartmann & Kenyon 1996). The major heating source of disk midplane is viscousheating caused by an accretion process (Armitage 2011).Originally, FUors were identified by their large brightness increase (∆V ≥ > α and Na I D lines, (8) strong CO absorption features, (9) strong water absorption bands at theedges of the H band, (10) a strong blue-shifted He I absorption line profile, and (11) wavelength-dependent spectraltypes; optical and infrared spectra are consistent with F-G and K-M supergiant or giant, respectively.There are only about 30 known FUors and FUor-like objects. Low-resolution optical and NIR spectra are a powerfultool with which to identify FUors (Hartmann & Kenyon 1996). With the addition of high-resolution data, we canstudy the physical and kinematic structure of the inner disk. To date, HBC 722 is the only FUor that has been studiedwith high-resolution (R ≡ λ /∆ λ ≥ ±
163 pc. The optical and NIR spectra show characteristics of FUors (Hillenbrand 2014; Reipurth & Connelley2015; Pyo et al. 2015; K´osp´al et al. 2015; Caratti o Garatti et al. 2015; Takagi et al. 2018). While there are inten-tionally taken pre-outburst data for only two FUors: V1057 Cyg (Herbig 1977) and HBC 722 (Cohen & Kuhi 1979),V960 Mon lies close to the Galactic plane so Galactic plane surveys can furnish information about the pre-outburststage (K´osp´al et al. 2015). Therefore, the V960 Mon dataset provides the opportunity to investigate the entire FUorphenomenon from its pre-outburst to its post-outburst phase.K´osp´al et al. (2015) studied the properties of the pre-outburst stage of V960 Mon. They estimated the centralprotostellar temperature as about 4000 K and central protostellar mass as about 0.75 M ⊙ . From the estimatedproperties, V960 Mon was classified as Class II before the outburst. Hackstein et al. (2015) suggested an oscillatingperiod of 17 days from the post-outburst light curve. Caratti o Garatti et al. (2015) revealed the existence of anextended disk-like structure, a very low-mass companion, and a bump which is thought to be a closer companion.They suggest that the target is a triple system, and the observed NIR spectral features of V960 Mon are similarto those of HBC 722 (Miller et al. 2011). Jurdana-ˇSepi´c & Munari (2016) re-constructed the historical light curvefrom 1899 to 1989 but did not find any brightness change similar to the outburst that occurred in 2014. Recently,Takagi et al. (2018) presented observations of spectroscopic variations of H α and nearby atomic lines and suggestedthat these variations are caused by the decreasing mass accretion rate. pectra of V960 Mon OBSERVATIONS AND DATA REDUCTION2.1.
Optical Observations
We observed V960 Mon using BOES from 2015 February 11 to 2018 December 19 with a spectral resolution of 30,000using a 300 µ m fiber. BOES is an echelle spectrograph (Kim et al. 2002) attached to the 1.8 m optical telescope atBohyunsan Optical Astronomy Observatory (BOAO) in Korea, and it covers the full optical wavelength from 3,900 ˚Ato 9,900 ˚A. To increase the signal-to-noise ratio (S/N), we binned the spectra over 2 × echelle package. For each image, biassubtraction was conducted, and each aperture from the spectral images was extracted using a master flat-field image.As part of the flat-fielding process, we corrected the interference fringes and pixel-to-pixel variations of the spectrumimages. A ThAr lamp spectrum was used for wavelength calibration. Continuum fitting was performed by the continuum task. Finally, heliocentric velocity correction was applied by using the rvcorrect task and the publishedradial velocity of V960 Mon (38.1 ± − ; Takagi et al. 2018).2.2. Near-infrared Observations
We observed V960 Mon with IGRINS installed on the 2.7 m Harlan J. Smith Telescope (HJST) at McDonaldObservatory and on the 4.3 m Discovery Channel Telescope (DCT) at Lowell Observatory from 2014 December 25 to2017 November 26. IGRINS provides high-resolution (R ∼ µ m) andK (1.96-2.46 µ m) bands with a single exposure (Yuk et al. 2010; Park et al. 2014). Table 1 lists the observing log forIGRINS and gray dashed lines in Fig. 1 indicate the dates of IGRINS observations.We reduced the H and K spectra using the IGRINS pipeline (Lee & Gullikson 2017) for flat-fielding, sky subtraction,correcting the distortion of the dispersion direction, wavelength calibration, and combining the spectra. Telluricstandard stars (A0 V) were observed immediately after or before each observation of V960 Mon for telluric correction.Continuum fitting and telluric correction were performed using custom IDL routines. We applied the same method forthe entire data reduction as described in Park et al. (2018). We report a S/N for each spectrum based on the medianvalue in the order that covers from 2.21 to 2.24 µ m. The S/N is typically 190, and ranges from 93 to 289. Finally, aheliocentric velocity correction was applied using the same method as used for the optical spectra. RESULTS AND ANALYSIS3.1.
Wind Features
The mass loss rate for Class II YSOs is about 10% of the mass accretion rate (Hartmann & Kenyon 1996; Hartmann2009; Ellerbroek et al. 2013; Bally 2016). FUors have more powerful winds than other YSOs because their mass accre-tion rates ( ∼ − to 10 − M ⊙ yr − ) are about three orders of magnitudes greater than other Class II YSOs ( ∼ − to 10 − M ⊙ yr − ; Hartmann & Kenyon 1996; Herbig et al. 2003; Hartmann 2009; Audard et al. 2014; Hartmann et al.2016). During the outburst, high-velocity winds (several hundred km s − ) can be present (Hartmann & Kenyon 1996).Calvet et al. (1993) and Hartmann & Kenyon (1996) showed that wind features can arise from the accreting disk. Thestronger wind lines are formed at the vertically outer part of the disk atmosphere, which indicate the largest expansionvelocities and show strongly blue-shifted absorption profiles.The optical spectra of the V960 Mon show several wind features in lines of H β I D doublet (5889 ˚Aand 5895 ˚A), and H α α and H β lineprofiles, in particular, show clear changes.H α has a P Cygni profile with a strong and broad blue-shifted absorption component extending to about -400 km s − ,which is produced by an outflowing wind (Hartmann & Kenyon 1996; Hartmann 2009; Herbig 2009; Reipurth & Aspin2010; Lee et al. 2011). A red-shifted emission component is also present. The variation of the blue-shifted absorptioncomponent with time is significant: the depth was the most profound at the first observation (2015 February, shortlyafter the outburst) and became shallower until 2015 October. About one year after the outburst, the absorption Park et al. component disappeared and the blue-shifted side of the line was in emission after 2015 December (Fig. 2). At thesame time, the width of the blue-shifted absorption features of H β and Na I D doublets in Fig. 2 became narrowerand shallower with time. Fig. 3 shows line variations of H β (left panel) and H α (right panel) with time. As shownin Figs. 2 and 3, the depth variation of the broad absorption component of H α and the width variation of H β occursimultaneously. These changes in wind features imply that the blue-shifted component is continually weakening sinceour observation, and the high-velocity component of the wind became too weak to be detectable around 2015 December.The changes in the blue-shifted component of wind features can be explained by the decreasing mass accretion rate.3.2. Disk Features
A disk in Keplerian rotation can produce double-peaked line profiles whose peak separations decrease with increasingwavelength (Hartmann & Kenyon 1996; Hartmann 2009; Zhu et al. 2007, 2009). In addition to the double-peaked lineprofile, a Keplerian rotational disk can produce a boxy profile with a flat bottom and steep wings (Petrov & Herbig2008).Several atomic metal lines were detected with double-peaked or boxy profiles in optical and NIR spectra (Fig. 4and Fig. 5). The double-peaked lines are relatively clear at the first observations (2015 February for BOES and 2014December for IGRINS), consistent with wind features (Section 3.1). The S/N is the best at the first observation datesfor each observation of BOES and IGRINS because the source was at its brightest immediately after the outburst andbecame fainter with time (Fig. 1 and Hackstein et al. 2015). The optical and NIR spectra of FUors during outburstoriginate from the disk rather than the central star because of the significant mass accretion rate (Petrov & Herbig1992; Hartmann & Kenyon 1996). The dimming results from the decreasing mass accretion rate after the outburst(Fig. 1), which reduces continuum brightness of the disk midplane. Therefore, only the first few observational dataare used for disk analyses in this section.If V960 Mon has a Keplerian disk with a radially decreasing temperature, the longer wavelength traces the largerradius where the disk rotates more slowly. We fit these double-peaked line profiles by convolving standard stellar spectrawith a disk rotational profile and estimated the temperature and radius where the observed lines are formed. Thestandard stellar spectra are convolved with a disk rotational profile as below (Calvet et al. 1993; Hartmann & Kenyon1996; Hartmann 2009). φ (∆ v ) = h − (cid:0) ∆ vv max (cid:1) i − / , (1)where ∆ v is the velocity shift from the line center and v max is the maximum projected rotational velocity ( v max = vsini ).For the optical analysis, we observed several standard stars with the same observational setup as V960 Mon. Weperformed disk rotational convolution in steps of 1 km s − and obtained consistent results within 3 km s − intervals,therefore, the uncertainty of the fitting is ± − . The best-fit was determined by the chi-square minimizationfrom the fitting of the double peak/boxy lines, and the spectra of HD 219477 (G2 II-III) and HD 18474 (G5 III) fitthe best the spectra of V960 Mon. The best-fit results of optical double-peaked/boxy lines are found in the rangesof 35-44 km s − , and the average and standard deviation of v max is 40.3 ± − . However, there is additionaluncertainty in the fitting results because of the coarse grid of spectral types and luminosity classes for standard stars.Fig. 6 shows two example spectra of the best-fit results for each of the optical and NIR. We adopted the T eff ofHD 18474 (G5 III) as 5013 K from Liu et al. (2014). In the case of HD 219477 (G2 II-III), the T eff was unknown.Therefore, we calculated the T eff as about 5300 K by adopting T eff -(B-V) relation (Flower 1996; Torres 2010).In analyzing the NIR spectra of V960 Mon, we used the spectra of standard stars from the IGRINS Spectral Library(Park et al. 2018), and most of the double-peaked lines were fitted well by a K1-type (HD 94600 (K1 III), T eff ∼ − and obtainedconsistent results within 2 km s − intervals, therefore, the uncertainty of the fitting is ± − . The best-fitresults of NIR double-peaked/boxy lines are found in the ranges of 32.5-41.5 km s − , and the average and standarddeviation of v max is 36.3 ± − .If V960 Mon has a Keplerian disk with an inclination of 90, and the central protostellar mass of 0.75 M ⊙ (K´osp´al et al.2015), the observed double-peaked optical and NIR lines trace 88 ± R ⊙ and 109 ± R ⊙ of the disk, respectively.Hence, the temperature of the disk decreases from 5300-5000 K at 88 ± R ⊙ to 4600 K at 109 ± R ⊙ . These resultsshow that the disk features at the longer wavelengths trace the cooler outer part of the disk with lower rotationalvelocity. The estimated disk radii depend on the inclination of V960 Mon, and the detected disk features show pectra of V960 Mon ± ± R ⊙ and 15 ± ± R ⊙ , respectively.We measured the Half-Width at Half-Depth (HWHD, Petrov & Herbig 2008) of the Fe I I I I I I µ m, Fe I µ m, Fe I µ m, and Ca I µ m linesand list the measured values in Table 2. As discussed above, we measured the HWHD of each line only the firstfew observations for BOES (from February 2015 to October 2015) and all observations for IGRINS and plot theaverages and standard deviations in Fig. 7. The average HWHD of optical and NIR lines are about 50 ± − and39 ± − , respectively. The HWHD decreases with increasing wavelength, consistent with the origin in a Kepleriandisk where hotter inner material is rotating faster than cooler outer material.In addition, the IGRINS spectra clearly show strong CO absorption features at 2.293 µ m (Fig. 8), one of therepresentative characteristics of FUors (Hartmann & Kenyon 1996; Audard et al. 2014; Connelley & Reipurth 2018),which are produced against the heated midplane by the accretion burst, and the broadened CO features are causedby the Keplerian rotation. There is no significant variation in CO absorption features during our NIR observations.V960 Mon shows broader line widths (black line in Fig. 9) than the standard star (gray line; HD 44391, Park et al.2018) because of the disk rotation. The CO absorption features are reasonably well fitted, but not perfectly, bythe stellar spectrum of HD 207089 and HD 44391 convolved with a projected rotational velocity of 40 km s − and30 km s − for bandhead (red) and rovibrational lines (orange), respectively. This result suggests that the lower energytransitions of the CO overtone band are produced at larger radii where the disk rotates more slowly.3.3. Outflow/Jet Features
Emission lines are, in general, hardly detected in FUors, except H α P Cygni profile. The optical and NIR spectraof V960 Mon show emission lines of [S II ] 6731 ˚A and H µ m (Fig. 10). The mean S/N of the [S II ] 6731 ˚A andthe H µ m line are about 69 and 191, respectively. Takagi et al. (2018) also detected the [S II ] 6731 ˚A emissionline in V960 Mon. Before the detection of emission lines in V960 Mon, V2494 Cyg was the only FUor that showedemission line of the [S II ] 6731 ˚A (Magakian et al. 2013). The H µ m spectrum in Fig. 10 represents the firstdetection of this feature in a FUor spectrum.The [S II ] 6731 ˚A emission line is a well-known outflow/jet tracer in Class II objects (Hirth et al. 1997; Simon et al.2016). According to Hartmann (2009), the [S II ] 6731 ˚A emission line can be formed in the entrained gas accelerated bya highly collimated jet. The peak velocity of the [S II ] 6731 ˚A emission line is blue-shifted with respective to the systemicvelocity by 19 ± − . The [S II ] 6731 ˚A emission line has similar physical properties to the [Fe II ] 1.644 µ m, butthe [Fe II ] line has a higher critical density ( ∼ × cm − ) than that of the [S II ] ( ∼ × cm − ) (Reipurth et al.2000; Nisini et al. 2005; Hayashi & Pyo 2009). Since only the [S II ] 6731 ˚A emission was detected, we infer that theoutflow has a lower density.The H µ m emission line is also known as a tracer of outflows at the earlier stages of YSOs, in particular, Class I(Davis et al. 2003; Bally et al. 2007; Davis et al. 2010; Greene et al. 2010; Bally 2016). Generally, the [Fe II ] 1.644 µ memission line arises from fast shock ( >
30 km s − ) while the H µ m emission line arises from relatively slowershock ( <
25 km s − ; Hayashi & Pyo 2009); the mechanism of line formation mainly depends on the shock velocity. Inthe spectra of V960 Mon, only the H emission line was detected, and its peak velocity is 5 ± − . Therefore, theH line may be induced by C-shock. According to the NIR spectroscopic survey of FUors (Connelley & Reipurth 2018),the H µ m emission line was not found in bonafide FUors but mostly found in FUor-like objects and peculiarobjects. The majority of the FUor-like objects and peculiar objects in Connelley & Reipurth (2018) are classified asClass I sources in Connelley & Greene (2010), which imply that they still have surrounding envelope material. COMPARISON WITH HBC 722The analyses of disk features (Section 3.2) show that the optical spectrum traces warmer material at higher velocity(smaller radius) than the NIR spectrum traces. These results show evidence for Keplerian disk rotation which was alsofound in HBC 722 (Lee et al. 2015). Both of the FUors have pre-outburst data and have high-resolution spectroscopicmonitoring data in the optical and NIR after their outburst. Therefore, we compared the two FUors to characterizeV960 Mon. Table 3 lists data useful for comparisons between V960 Mon and HBC 722.The two FUors show disk features in the optical and NIR spectra, and the trend of velocity with wavelength is similar.The optical and NIR spectra of HBC 722 trace the disk radius of about 39 ± R ⊙ and 76 ± R ⊙ at temperatures of Park et al. about 5000 K and 3000 K, respectively, when the v max (Lee et al. 2015) is used. According to the SED modeling byGramajo et al. (2014), the inclination and mass of HBC 722 are 85 degrees (almost edge-on) and 1 M ⊙ , respectively.Therefore, the disk radii traced by the optical and NIR spectra and their corresponding temperatures in HBC 722, whichwere obtained using v max and 1 M ⊙ (Lee et al. 2015), are adopted to compare with those of V960 Mon. The uncertaintyof the estimated disk radius is calculated by error propagation adopting 10% error for mass and rotational velocity.The uncertainty of mass was adopted as the standard deviation of masses obtained by comparing the temperatureand radius of HBC 722 presented in Gramajo et al. (2014) with three different evolutionary models (Siess et al. 2000;Bressan et al. 2012; Baraffe et al. 2015). The standard deviation of the rotational velocities estimated in V960 Monis about 10% of the mean rotational velocity for both optical and NIR. Since we applied the same technique to findthe rotational velocity, we adopted the same uncertainty for the rotational velocity for HBC 722. If we use the v max of V960 Mon, the optical and NIR spectra trace a disk radius of about 88 ± R ⊙ and 109 ± R ⊙ with temperaturesof about 5300-5000 K and 4600 K, respectively. Therefore, V960 Mon is hotter than HBC 722 at the radii traced bythe optical and NIR spectra if the maximum projected rotational velocities are adopted.Fig. 11 shows the comparison of disk radius between V960 Mon (circle) and HBC 722 (square). The solid linesindicate the estimated disk radius of optical (black) and NIR (red) as a function of disk inclination by adopting thebest-fit rotational velocity of V960 Mon. When disk inclination is assumed as 45 (60) degrees, the optical and NIRspectra of V960 Mon trace a disk radius of about 44 ±
16 (66 ± R ⊙ and 54 ±
20 (81 ± R ⊙ , respectively. If diskinclination is about 22 (28) degrees (Caratti o Garatti et al. 2015), the observed spectra trace a disk radius of about12 ± ± R ⊙ and 15 ± ± R ⊙ , respectively.The bolometric luminosity ( L bol ) at the outburst stage of V960 Mon is about 48 L ⊙ (Connelley & Reipurth 2018)while that of HBC 722 is about 8.7 to 17 L ⊙ (K´osp´al et al. 2011, 2016; Connelley & Reipurth 2018) suggesting thatthe disk of V960 Mon can be hotter than that of HBC 722 at the outburst stage. Since the accretion luminosity ( L acc )dominates the L bol in FUors (Hartmann & Kenyon 1996; K¨onigl et al. 2011), L bol is proportional to the mass accretionrate ( ˙ M ): L bol ∼ L acc ∝ M × ˙ M (Hartmann & Kenyon 1996; Hartmann 2009). The higher L bol indicates higher ˙ M ,which implies that a massive accretion heating occurs and the disk midplane becomes hotter ( T disk ∝ ˙ M / , Zhu et al.2007; Hartmann 2009). In addition, the L bol of V960 Mon (4.8 L ⊙ ; K´osp´al et al. 2015) at the pre-outburst stage isalso higher than that of HBC 722 (0.85 L ⊙ ; K´osp´al et al. 2011).If the mass of the two FUors is similar to 1 M ⊙ (see Table 3), then the relatively higher L bol of V960 Mon meansa relatively higher ˙ M than that of HBC 722. Moreover, the upper limit of the disk mass of HBC 722 is about0.01-0.02 M ⊙ (Dunham et al. 2012; K´osp´al et al. 2016) while the circumstellar mass of V960 Mon is about 0.01-0.06 M ⊙ (K´osp´al et al. 2016). Since V960 Mon is known as Class II before its outburst (K´osp´al et al. 2015), we canassume that the circumstellar mass is dominated by the disk mass. Then, the disk masses of V960 Mon and HBC 722are similar. Even if the protostellar masses and the disk masses are similar, the ˙ M of V960 Mon is higher than thatof HBC 722.Another difference between the two FUors is the existence of emission line. The emission lines have been hardlydetected in FUors, but we detected three emission lines of the H α II ] 6731 ˚A and the H µ m inV960 Mon. The [S II ] 6731 ˚A and the H µ m emission lines are known as jet/outflow tracers. The [S II ] 6731 ˚Aemission line is often detected in Class II (Hirth et al. 1997; Simon et al. 2016), while the H µ m emission linein Class I (Davis et al. 2003; Bally et al. 2007; Bally 2016). However, from the SEDs of their pre-outburst stage, thetwo FUors are known as Class II (K´osp´al et al. 2011; Miller et al. 2011; K´osp´al et al. 2015, 2016), and their spectralindex ( α ) is also about -0.4 (HBC 722; Miller et al. 2011) and -0.5 (V960 Mon; K´osp´al et al. 2015) which are thetypical values of Class II. In previous studies, HBC 722 is classified as an evolved Class II because outflow featurewas not observed and its envelope mass is small (Green et al. 2011; Dunham et al. 2012). Of the three emission linesdetected in V960 Mon, only H α P Cygni profile was detected in HBC 722.The bolometric temperatures ( T bol ) is an evolutionary indicator (Myers & Ladd 1993; Chen et al. 1995), and the T bol of V960 Mon in the pre-outburst stage was about 1190 K (K´osp´al et al. 2015) while T bol of HBC 722 in thepre-outburst phase is unknown. Therefore, we estimated the T bol of HBC 722 at the pre-outburst stage by adoptingthe photometric data (Guieu et al. 2009; Rebull et al. 2011; Barentsen et al. 2014). The calculated T bol of HBC 722in the pre-outburst stage is about 1451 ±
11 K, higher than that of V960 Mon. The higher T bol indicates a moreevolved stage (Myers & Ladd 1993). Therefore, the lower T bol of V960 Mon than that of HBC 722 might indicate thatV960 Mon is in a relatively earlier evolutionary stage than HBC 722 in the Class II stage. pectra of V960 Mon CONCLUSIONSWe have conducted monitoring observations of V960 Mon with high-resolution (R ≥ II ] 6731 ˚A and the H µ m are detected in our observations, which are rarelyfound in FUors. The H µ m emission line in V960 Mon is detected in our observation for the first time.4. The comparison with HBC 722, which is a more evolved Class II object according to its optical and NIR spectra,suggests that the disk of V960 Mon is probably hotter than that of HBC 722, and V960 Mon is in a relatively earlierClass II stage than HBC 722. Park et al.
We acknowledge with thanks the variable star observations from the AAVSO International Database contributedby observers worldwide and used in this research. This work was supported by the National Research Founda-tion of Korea (NRF) grant funded by the Korea government (MSIT) (grant numbers: NRF-2017R1A2B4007147,2018R1A2B6003423, NRF-2019R1C1C1005224). This work is also supported by the Korea Astronomy and SpaceScience Institute under the R&D program supervised by the Ministry of Science, ICT and Future Planning. This workused the Immersion Grating Infrared Spectrometer (IGRINS) that was developed under a collaboration between theUniversity of Texas at Austin and the Korea Astronomy and Space Science Institute (KASI) with the financial supportof the US National Science Foundation under grants AST-1229522 and AST-1702267, of the University of Texas atAustin, and of the Korean GMT Project of KASI. This paper includes data taken at The McDonald Observatory ofThe University of Texas at Austin. These results made use of the Discovery Channel Telescope at Lowell Observatory.Lowell is a private, non-profit institution dedicated to astrophysical research and public appreciation of astronomyand operates the DCT in partnership with Boston University, the University of Maryland, the University of Toledo,Northern Arizona University and Yale University. REFERENCES
Armitage, P. J. 2011, ARA&A, 49, 195Audard, M., ´Abrah´am, P., Dunham, M. M., et al. 2014,Protostars and Planets VI, 387Bakos, G. A. 1971, JRASC, 65, 222Bally, J. 2016, ARA&A, 54, 491Bally, J., Reipurth, B., & Davis, C. J. 2007, in Protostarsand Planets V, ed. B. Reipurth, D. Jewitt, & K. Keil, 215Baraffe, I., Homeier, D., Allard, F., & Chabrier, G. 2015,A&A, 577, A42Barentsen, G., Farnhill, H. J., Drew, J. E., et al. 2014,VizieR Online Data Catalog, II/321Bastian, U., & Mundt, R. 1985, A&A, 144, 57Bressan, A., Marigo, P., Girardi, L., et al. 2012, MNRAS,427, 127Calvet, N., Hartmann, L., & Kenyon, S. J. 1993, ApJ, 402,623Caratti o Garatti, A., Garcia Lopez, R., Ray, T. P., et al.2015, ApJL, 806, L4Chen, H., Myers, P. C., Ladd, E. F., & Wood, D. O. S.1995, ApJ, 445, 377Cohen, M., & Kuhi, L. V. 1979, ApJS, 41, 743Connelley, M. S., & Greene, T. P. 2010, AJ, 140, 1214Connelley, M. S., & Reipurth, B. 2018, ApJ, 861, 145Davis, C. J., Gell, R., Khanzadyan, T., Smith, M. D., &Jenness, T. 2010, A&A, 511, A24Davis, C. J., Whelan, E., Ray, T. P., & Chrysostomou, A.2003, A&A, 397, 693Dunham, M. M., Arce, H. G., Bourke, T. L., et al. 2012,ApJ, 755, 157Dunham, M. M., Evans, II, N. J., Terebey, S., Dullemond,C. P., & Young, C. H. 2010, ApJ, 710, 470Ellerbroek, L. E., Podio, L., Kaper, L., et al. 2013, A&A,551, A5 Flower, P. J. 1996, ApJ, 469, 355Gaia Collaboration, Brown, A. G. A., Vallenari, A., et al.2018, A&A, 616, A1Gramajo, L. V., Rod´on, J. A., & G´omez, M. 2014, AJ, 147,140Green, J. D., Evans, II, N. J., K´osp´al, ´A., et al. 2011,ApJL, 731, L25Greene, T. P., Barsony, M., & Weintraub, D. A. 2010, ApJ,725, 1100Guieu, S., Rebull, L. M., Stauffer, J. R., et al. 2009, ApJ,697, 787Hackstein, M., Haas, M., K´osp´al, ´A., et al. 2015, A&A, 582,L12Hanuschik, R. W. 2003, A&A, 407, 1157Hartmann, L. 2009, Accretion Processes in Star Formation:Second Edition (Cambridge University Press)Hartmann, L., Calvet, N., Gullbring, E., & D’Alessio, P.1998, ApJ, 495, 385Hartmann, L., Herczeg, G., & Calvet, N. 2016, ARA&A,54, 135Hartmann, L., & Kenyon, S. J. 1996, ARA&A, 34, 207Hayashi, M., & Pyo, T.-S. 2009, ApJ, 694, 582Herbig, G. H. 1966, Vistas in Astronomy, 8, 109—. 1977, ApJ, 217, 693—. 2009, AJ, 138, 448Herbig, G. H., Petrov, P. P., & Duemmler, R. 2003, ApJ,595, 384Hillenbrand, L. 2014, The Astronomer’s Telegram, 6797Hirth, G. A., Mundt, R., & Solf, J. 1997, A&AS, 126, 437Jurdana-ˇSepi´c, R., & Munari, U. 2016, NewA, 43, 87Kenyon, S. J., Hartmann, L. W., Strom, K. M., & Strom,S. E. 1990, AJ, 99, 869 pectra of V960 Mon Kim, K.-M., Jang, B.-H., Han, I., et al. 2002, Journal ofKorean Astronomical Society, 35, 221K¨onigl, A., Romanova, M. M., & Lovelace, R. V. E. 2011,MNRAS, 416, 757K´osp´al, ´A., ´Abrah´am, P., Mo´or, A., et al. 2015, ApJL, 801,L5K´osp´al, ´A., ´Abrah´am, P., Acosta-Pulido, J. A., et al. 2011,A&A, 527, A133—. 2016, A&A, 596, A52Kovtyukh, V. V. 2007, MNRAS, 378, 617Lee, J.-E., Kang, W., Lee, S.-G., et al. 2011, Journal ofKorean Astronomical Society, 44, 67Lee, J.-E., Park, S., Green, J. D., et al. 2015, ApJ, 807, 84Lee, J.-J., & Gullikson, K. 2017, igrins/plp v2.2.0-alpha.4, ,, doi:10.5281/zenodo.438353Liu, Y. J., Tan, K. F., Wang, L., et al. 2014, ApJ, 785, 94Maehara, H., Kojima, T., & Fujii, M. 2014, TheAstronomer’s Telegram, 6770Magakian, T. Y., Nikogossian, E. H., Movsessian, T., et al.2013, MNRAS, 432, 2685Miller, A. A., Hillenbrand, L. A., Covey, K. R., et al. 2011,ApJ, 730, 80Myers, P. C., & Ladd, E. F. 1993, ApJL, 413, L47Nisini, B., Bacciotti, F., Giannini, T., et al. 2005, A&A,441, 159Park, C., Jaffe, D. T., Yuk, I.-S., et al. 2014, in Proc. SPIE,Vol. 9147, Ground-based and Airborne Instrumentationfor Astronomy V, 91471DPark, S., Lee, J.-E., Kang, W., et al. 2018, ApJS, 238, 29Petrov, P. P., & Herbig, G. H. 1992, ApJ, 392, 209—. 2008, AJ, 136, 676 Pyo, T.-S., Oh, H., Yuk, I.-S., Kim, H., & Davis, C. J.2015, The Astronomer’s Telegram, 6901Rebull, L. M., Guieu, S., Stauffer, J. R., et al. 2011, ApJS,193, 25Reipurth, B., & Aspin, C. 2010, in Evolution of CosmicObjects through their Physical Activity, ed. H. A.Harutyunian, A. M. Mickaelian, & Y. Terzian, 19–38Reipurth, B., & Connelley, M. S. 2015, The Astronomer’sTelegram, 6862Reipurth, B., Yu, K. C., Heathcote, S., Bally, J., &Rodr´ıguez, L. F. 2000, AJ, 120, 1449Shu, F. H. 1977, ApJ, 214, 488Siess, L., Dufour, E., & Forestini, M. 2000, A&A, 358, 593Simon, M. N., Pascucci, I., Edwards, S., et al. 2016, ApJ,831, 169Takagi, Y., Honda, S., Arai, A., et al. 2018, AJ, 155, 101Tody, D. 1986, in Proc. SPIE, Vol. 627, Instrumentation inastronomy VI, ed. D. L. Crawford, 733Tody, D. 1993, in Astronomical Society of the PacificConference Series, Vol. 52, Astronomical Data AnalysisSoftware and Systems II, ed. R. J. Hanisch, R. J. V.Brissenden, & J. Barnes, 173Torres, G. 2010, AJ, 140, 1158Wu, Y., Singh, H. P., Prugniel, P., Gupta, R., & Koleva, M.2011, A&A, 525, A71Yuk, I.-S., Jaffe, D. T., Barnes, S., et al. 2010, inProc. SPIE, Vol. 7735, Ground-based and AirborneInstrumentation for Astronomy III, 77351MZhu, Z., Espaillat, C., Hinkle, K., et al. 2009, ApJL, 694,L64Zhu, Z., Hartmann, L., Calvet, N., et al. 2007, ApJ, 669,483 Park et al.
Figure 1. pectra of V960 Mon Figure 2.
Time variation of wind features (top: H β α Figure 3.
Time variation of H β α α . Then, the blue-shifted absorption component of H α turned into emission. In the same time, blue-shifted absorption features of H β became narrower and shallower. Park et al.
Figure 4.
Boxy or double-peaked absorption line profiles of optical spectra. Different disk features are shown in rows, anddifferent observation dates are shown in columns. The SNR was the highest at the first observation (2015 February 11) rightafter the burst, and the SNR decreases due to continuous brightness decrease. Therefore, the first four observation data wereused for analysis. pectra of V960 Mon Figure 5.
Boxy or double-peaked absorption line profiles of NIR spectra. Different disk features are shown in rows, anddifferent observation dates are shown in columns.
Figure 6.
Double-peaked line profiles of Fe I I µ m (right). Black lines present the disk features ofV960 Mon. Red lines indicate the stellar spectra convolved with disk rotational profile. Optical spectrum (left) fit well withthe G5-type (HD 18474) stellar spectrum convolved with a projected rotational velocity of 44 ± − , while NIR spectrum(right) fit well with the K1-type (HD 94600) stellar spectrum convolved with a projected rotational velocity of 32 ± − .The rotational velocity of the double-peaked lines decreases with increasing wavelengths; optical spectrum trace warmer innerpart of the disk and NIR spectrum trace cooler outer part of the disk. Park et al.
Figure 7.
HWHD of double-peaked absorption features as a function of wavelength. Different colors present different lines.The HWHD becomes narrower as the wavelength becomes longer, which is consistent with the Keplerian disk rotation.
Figure 8.
The CO first overtone band transitions of V960 Mon observed with IGRINS. Different colors indicate differentobservation dates. There is no significant change in CO absorption features. pectra of V960 Mon Figure 9.
The CO overtone transitions in V960 Mon (black) and standard star HD 44391 (K0 Ib; gray). The best-fit stellarspectrum of HD 207089 (K0 Ib; red) and HD 44391 (orange) convolved with a disk rotational profile of 40 km s − and 30 km s − is presented, respectively. The spectral features of V960 Mon are much broader than those of standard star, while they arereasonably matched with the stellar spectra convolved with a disk rotational profile. Figure 10.
Emission lines of V960 Mon. Left and right panels show the [S II ] 6731 ˚A and the H µ m emission lines,respectively. Different colors present different observation dates. The [S II ] 6731 ˚A emission line was shown until 2017 March,because of the low S/N ( ≤
10) since 2017 December. The [S II ] 6731 ˚A line (FWHM ∼ ± − ) is broader than theH µ m line (FWHM ∼ ± − ). Park et al.
Figure 11.
Estimated disk radius as a function of disk inclination. Black and red lines indicate the calculated disk radius bythe best-fit rotational velocity of optical ( v max = 40.3 ± − ) and NIR ( v max = 36.3 ± − ), respectively. Circleand square symbols represent V960 Mon and HBC 722 (Lee et al. 2015), respectively. The open circles indicate the disk radiicorresponding to the inclination (22 and 28 degrees) suggested by Caratti o Garatti et al. (2015), and the filled circles denotethe disk radii corresponding to the inclinations of 45, 60, and 90 degrees. The uncertainty of the disk radius is calculated bythe error propagation. pectra of V960 Mon Table 1.
Observation Logs of BOES and IGRINS
Telescope Instrument Spectral Resolution Observation Date Exposure Time a Telluric Standard Star[UT] [sec]BOAO BOES b, ∗ · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · c × ABBA) HD 45380 · · · · · · · · · × ABBA) HD 53205 · · · · · · · · · × ABBA) HD 53205 · · · · · · · · · × ABBA) HD 45380 · · · · · · · · · × ABBA) HIP 30594 · · · · · · · · · × ABBA) HD 53205 · · · · · · · · · × ABBA) HD 53205DCT/Lowell · · · · · · × ABBA) HD 56525HJST/McDonald · · · · · · × ABBA) HR 2584DCT/Lowell · · · · · · × ABBA) HR 1578 a Total integration time of each target (exposure time × the number of exposures = total integration time). b Wavelength coverage of BOES: 3900-9900 ˚A c Wavelength coverage of IGRINS: H (1.49-1.80 µ m) and K (1.96-2.46 µ m) bands ∗ HD 219477 (G2 II-III) and HD 18474 (G5 III) were observed as template spectra.
Table 2.
HWHD of Double-Peaked Lines
Wavelength Element HWHD[˚A] [km s − ]5383 Fe I 55.5 ± ± ± ± ± ± ± ± ± Park et al.
Table 3.
Comparison
Instrument Mass Target v max Spectral Type Temperature Radius † M ⊙ [km s − ] [K] [ R ⊙ ]BOES 0.75 ± V960 Mon 40.3 ± ∗ / 5013 ± ∼ , HBC 722 ∗∗