Rotation-Activity Relation of Ca II and Mg I Infrared Emission Lines of Young Stars

arXiv:2104.02910v1 [astro-ph.SR] 7 Apr 2021 ’noa&Mziel 19)cntutdfu evolutionary four constructed (1994) Mazzitelli & D’Antona 1 ino vltoaytak.Jn i 20)assumed (2007) Kim & Jung tracks. α evolutionary of tion α rae ihrda itnet h htshr.I nac- an such In lines permitted H photosphere. emit as atoms the region, to chromospheric distance tive in- radial gradually with chromosphere the creases of temperature The corona. Introduction 1 Yamashita Mai Emissi Infrared I Stars Mg Young and II Ca of Relation Rotation-Activity Günther M. H. Dupree, K. A. Wolk, J. Systems S. Stellar by Stars, Edited Cool on Workshop Cambridge 20.5th The ment dy- the by generated is which process. field, namo magnetic the by driven hpbtencrmshrcln tegh n h Rossby the and strengths number, line chromospheric between ship for lines stars. to emission PMS e chromospheric expected Yamashita strong Actually, are confirmed (2020) they al. activities. 2015), chromospheric Bouvier high are & have stars (Gallet (PMS) rotating sequence rotators fast pre-main a fast Because that considered chromospheric lines. strong widely emission showing is thus dynamo, It strong has 1996). star al. et nas current (= a enetnieyeaie o ansqec stars. adequate the main-sequence constructed for (1984) al. examined et extensively Noyes been has nvriyo yg (Japan), Hyogo of University ( α hoopeei h einbtenpoopeeand photosphere between region the is Chromosphere ntednm rcs,teCroi oc =rttoa mo- rotational (= force Coriolis the process, dynamo the In orva h tutr ftecneto oe h relation- the zone, convection the of structure the reveal To ≡ 0 = α soeo h otciia aaeeso h calcula- the on parameters critical most the of one is iiglength mixing × orva h ealo h nenlsrcue h relation the structure, internal the of detail the reveal To Abstract msinlnso 0PSsas hi qiaetwdh r c are widths equivalent Their ( stars. ratio PMS 60 of lines emission eune(AS tr.Hwvr eas all because However, stars. (ZAMS) sequence oainlperiod rotational st pl h aemtost r-ansqec PS star (PMS) pre-main-sequence to methods same the apply to is 10 ta.(00 netgtdterltosi between relationship the investigated (2020) al. et bopincmoet h gIln sdtce sa emissio an as detected is line I Mg the component, investiga absorption we (UCLES), Spectrograph Echelle London College eemnto fteadequate the of determination oe o h M tr.W oie htM msinlnsat lines emission I Mg that noticed We stars. PMS the for model hrfr,teadequate the Therefore, n aI.I scamdta hoopei ciiyis activity chromospheric that claimed is It II. Ca and . 5 ovcinvlct)blne ihteLrnzforce Lorentz the with balances velocity) convection − n osrce h vltoaytak,whereas tracks, evolutionary the constructed and 4 × . 1 R N h gIln sntstrtdyti testrtdrgm fo regime saturated "the in yet saturated not is line I Mg The . antcsrnt est fpam)(Baliu- plasma) of density / strength magnetic ′ R .Te5 M tr have stars PMS 54 The ). ≡ 1 ociItoh Yoichi , ovcietroe time turnover convective l/ rsuesaeheight scale pressure P oainlperiod rotational ovcietroe time turnover convective / 2 τ ainlAtooia bevtr fJapan of Observatory Astronomical National c o M tr a edtrie ymauigo their of measuring by determined be can stars PMS for 1 τ ue Takagi Yuhei , c o M tr.Uigteaciedt fteAnglo-Australia the of data archive the Using stars. PMS for N R P < H 10 τ p τ c 1 = ) − c , τ R 1 c oe with model . ,hsbe xesvl xmndfrmi-eunesas T stars. main-sequence for examined extensively been has ), ′ 0 n h Sun the and , a auae against saturated was n show and 2 . 9 N . R n tegh fteC Iifae rpe (IRT; triplet infrared II Ca the of strengths and (1) hpbtencrmshrcatvt n h osynumber, Rossby the and activity chromospheric between ship t e h gIln f5 ASsas fe utatn photosph subtracting After stars. ZAMS 52 of line I Mg the ted R netdit h msinln oselrblmti lumin bolometric stellar to line emission the into onverted 1 ′ n dniyteaporaemdlof model appropriate the identify and s iei 5ZM tr,whose stars, ZAMS 45 in line n 8808 ∼ uiamlclrcod h roi B1 soito,the association, 1c OB Orionis the cloud, molecular Auriga rcswith tracks tr n dniyteaporaemdlof model appropriate the identify and stars structure. stellar on α with (IRT; Reduction Data and Observations 2 have stars PMS that expected is it Thus small 2007). Kim & (Jung rvdsdffrn vltoaytak.Freape PMS a ( as example, identified For is star tracks. evolutionary different provides ( e ove 03 n ogrcnetv unvrtime, turnover convective longer and 2013) Bouvier period, & rotation let shorter both have WTTS) and (CTTS h BDrdsmvn ru.Temdu-adhigh- and medium- with The conducted were group. spectroscopy moving resolution Doradus AB the tr yusing by stars used (2020). methods al. et reduction Yamashita the data in presented the with is of here obtained description detailed data used. A also were Archival VLT/X-Shooter and VLT/UVES, Subaru/HDS. Keck/HIRES, and Observa- Astronomical tory Nishi-Harima at Telescope/MALLS h CaNa”rgo,the region, the cloud, ”Cha-Near” molecular association, the Hydrae Perseus TW the the Taurus- association, the Scorpius groups; Upper moving five or clouds molecular four 0 10 . soeo h otfnaetlbtuepoe parameters unexplored but fundamental most the of one is h olo u oki oapyteNys ehdt PMS to method Noyes’ the apply to is work our of goal The nti td,teC Iifae rpe msinlines emission triplet infrared II Ca the study, this In eas netgtdteM ieo ,K ye5 ZAMS 52 type G K, F, of line I Mg the investigated also We 035 sa pial hncrmshrcln,aporaefor appropriate line, chromospheric thin optically an is Å − h aI msinlns,i.e. lines", emission II Ca the r 4 α N . λ N 2 8498 R 0 = − slrea h maximum the as large as R twsntpsil oetmt h appropriate the estimate to possible not was it , n ciechromosphere. active and 2 . 5 , . α M 5 8542 ept ( despite , R 0 = 3 . MgI ′ ⊙ m 9 , , 8662 . 10 1 5 values. . , nl-utainTlsoe(AAT)/the Telescope Anglo-Australian M 0 5 1 1 . eecp AT/h University (AAT)/the Telescope n . 2 − )aeivsiae o 0PSstars PMS 60 for investigated are Å) M 2 , ⊙ 1 , R 10 . 5 Myr 3 ⊙ β MgI ′ , 8 , 2 10 R Myr 1 yr itrsmvn ru,and group, moving Pictoris . 0 ′ τ sbetween is − λ ,bigascae with associated being ), epciey Different respectively. , c ftezr-g main- zero-age the of ya vltoaytrack evolutionary an by ) η 1 8498 o hm Yamashita them. for . 6 ega forwork our of goal he hmeeni cluster, Chamaeleontis nLnsof Lines on yta with that by ) N < , 8542 R 10 < , τ − 8662 c 10 5 M stars PMS . m 2 . 9 N osity − and eric R 0 α Å) . τ (= 8 P Nayuta c . 2 = (Gal- . τ 0 α c . Mai Yamashita, Yoichi Itoh & Yuhei Takagi

University College London Echelle Spectrograph (UCLES) archive spectra. The PI is S. C. Marsden, date of the observations are 2000-03-17, 18, 19, 2001-01-06, 07, 08, and 2001-02- 11, 12. The wavelength coverage was between 3522 Å and 9386 Å. The integration time for each object was between 300 s and 1200 s. A detailed description of the observations is presented in Marsden et al. (2009). We selected the objects from single stars in IC 2391 (30 ± 5Myr; Barrado y Navas- cuesetal.2004)andIC2602(50±5Myr; Stauffer et al. 1997) clusters, whose membership was confirmed by Marsden et al. (2009). We used the Image Reduction and Analysis Facility (IRAF) software package1 for data reduction. Overscan subtraction, bias subtraction, flat fielding, removal of scattered light, removal of cosmic rays, extraction of a spectrum, wavelength calibration using a Tr-Ar lamp, continuum normalization, combining spectra, and subtracting the photospheric absorption component were conducted for all the spectra obtained by UCLES. The MgI emission componentis always buried by the photospheric absorption as shown in Fig. 1. In this study, inactive stars with a spectral type similar to that of the target Figure 1: The procedures of the spectral subtaction of the were used as template stars. We reduced UCLES spectra of photospheric component for the PMS star, RECX 11. The ob- HD 16673 (F6V), α Cen A (G2V), α Cen B (K1V). We also observed spectrum of RECX 11 is shown in the top of the panel tained the VLT archive spectra of ten inactive stars choosen from the inactive stars library (Yee, Petigura & von Braun with a solid line. The dotted line is the spectrum of fitted inactive template star. The difference between the observed 2017) and UVES POP, namely ζ Ser (F2IV), HD 3861 (F8V), and template spectra is shown in the bottom of the panel, HD 1388 (G0V), HD 156846 (G1V), HD 109749 (G3V), HIP where the Ca II IRT narrow line and the Mg I narrow line 94256 (G5V), HD 190360 (G7V), HD 217107 (G8V), HD 16160 appear in emission. The observed spectrum of RECX 11 and (K3V), and HD 165341B (K4V). As a result of selecting inac- that of the template star are shown shifted by for dis- tive stars having similar metalicity with IC 2391 (0.00±0.01; +1.5 play purposes. Randich et al. 2001) and IC 2602 (−0.01 ± 0.02; D’Orazi & Randich 2009), their metalicity is −0.11 ≤ [Fe/H] ≤ 0.31. For the correction of the rotational broadening, the spectra of the template stars were convolved with a Gaussian kernel RX J1147.7-7842are broad but not strong (EQW < 10 Å). All to match the width of the absorption lines of each object. By EQWs of the narrow emission lines are weaker than 5 Å. using high S/N spectra of inactive field stars, we carefully Fig. 2 shows the Mg I emission line at λ8808 Å of ZAMS subtracted the photospheric absorption component. We con- stars after subtracting the photospheric absorption. Before sidered that the subtraction was reliable if nearby faint pho- subtracting the photospheric absorption component, the Mg tospheric absorption lines, such as Ni I (6644, 6728, 7525 Å), I emission component is always buried by the photospheric Mn I (6014, 6017, 6022 Å), V I (6039 Å), and Ti I (7523 Å), absorption. As a result of the data reduction, the Mg I line disappeared in the spectrum after the subtraction. Finally, we is detected as an emission line in 45 ZAMS stars. The Mg measured the equivalent widths (EQW) of the Mg I emission I line of these ZAMS stars shows narrow emission, indica- line. Before measuring the EQWs, the continuum component tive of chromospheric origin. Their EQWs range from 0.02 Å of the spectra was added to unity. To obtain the EQWs of the to 0.63 Å. The Mg I emission lines of 33 ZAMS stars have Mg I emission lines, the area of the emission profile was di- EQWs less than 0.1 Å. 5 ZAMS stars (Cl* IC2391 L32, VXR rectly integrated. The typical EQW errors were estimated by PSPC 03A, VXR PSPC 16A, VXR PSPC 45A, [RSP95] 85) do multiplying the standard deviation of the continuum by the not show the Mg I line in emission or absorption, but like wavelength range of the Mg I emission line of the five ZAMS simply continuum component. stars. Avoiding any emission and absorption lines, the standard deviation of the continuum was determined by averag- 4 Discussion − − ing over the value in λ8798 8802 Å and λ8813 8819 Å. 4.1 Chromospheric activity and mass accretion rate 3 Results We compared the strengths of the Ca II IRT emission lines and mass accretion rates to discuss whether the chromo- Seven PMS stars have broad emission lines of Ca II λ8498 Å sphere is activated by mass accretion from the protoplan- − (FWHM > 100 km · s 1), while most PMS stars exhibit nar- etary disk. Mohanty et al. (2005) investigated the chro- − row emission lines (FWHM ≤ 100 km · s 1). The emission mospheric activity of CTTSs, very low-mass young stars lines of DG Tau, DL Tau, and DR Tau are broad and strong (0.075 ≤ M∗ < 0.15 M⊙) and young brown dwarfs (M∗ ≤ (EQW ∼ 50 Å), while those of RY Tau, SU Aur, RECX 15, and 0.075M⊙). They selected ”accretors” by applying a number of accretion diagnostics, such as an Hα10% width ≥ 1IRAF is distributed by the National Optical Astronomy Observatories, 200 km · s−1. For the accretors, the surface flux of the Ca which are operated by the Association of Universities for Research in As- II emission line, ′ , showed a positive correlation with tronomy, Inc., under cooperative agreement with the National Science Foun- Fλ8662 dation. their mass accretion rate, M˙ , for approximately 4 orders of

2 Zenodo, 2021 The 20.5th Cambridge Workshop on Cool Stars, Stellar Systems, and the Sun

′ Figure 3: Surface flux of the Ca II emission line, Fλ8662,asa function of mass accretion rate, M˙ . The open circles show the objects with a broad Ca II IRT emission line (FWHM > Figure 2: Examples of the emission profiles of the Mg I line 100 km · s−1) and the filled circles represent the objects with at 8807.8 Å. Photospheric absorption lines have already been a narrow Ca II IRT emission line (FWHM ≤ 100 km · s−1). subtracted. The cross symbols show the CTTSs, very low-mass young stars (0.075 ≤ M∗ < 0.15 M⊙) and young brown dwarfs (M∗ ≤ 0.075M⊙) studied in Mohanty et al. (2005). magnitude. Hence, they claimed that the mass accretion rate can be estimated using the strength of the broad components of the Ca II IRT emission lines. In our work, M˙ were taken from several studies, which as large as the maximum R′ of ZAMS stars (Marsden et al. −0.8 ′ were estimated from the amount of veiling. For objects 2009). The PMS stars have NR < 10 and constant R whose veiling had not been determined, we referred M˙ esti- against NR; i.e., the Ca II IRT emission lines of the PMS stars mated with an EQW of the Hα emission line. The EQWs of are saturated. The chromosphere of these stars is considered the Ca II emission lines are converted into F ′. The method to be completely filled by the active region. of calculating F ′ is described in Yamashita et al. (2020). Because all R′ of Ca II IRT emission lines were saturated We see two groups amongthe PMSstars (Fig. 3). ThePMS against NR, it was not possible to estimate the appropriate stars showing a broad Ca II IRT emission line have high mass τc model for the PMS stars. PMS stars with unsaturated Ca −7 −1 accretion rate (M˙ & 10 M⊙ · yr ). Those are classified II IRT emission lines should be investigated. The Ca II IRT as accretors based on the diagnosis of Mohanty et al. (2005). emission lines are not saturated for the ZAMS stars with −0.8 On the other hand, the PMS stars showing a narrow emis- NR > 10 . However no PMS star is expected to have −0.8 ′ ˙ NR > 10 due to their fast rotation and long τc. sion have log Fλ8662 ∼ 3 with a flat distribution to M. How- ever, we note that the Ca II IRT emission lines of PMS stars As a consequence, other optically thin chromospheric vary with time. In this work, BP Tau shows narrow emissions emission line should be observed. As claimed in Linsky (EQW = 1.15 Å), whereas showed broad emission features (2017), the column densities of Ca II are very large in late- (EQW = 7.8 Å) in Mohanty et al. (2005). type stars. Moreover, Ca II H, K lines heavily obscured by in- terstellar medium due to their short wavelength. In addition 4.2 Rotation-Activity Relation of Ca II emission lines to this, their large optical depth are pointed out in Hamann & Persson (1992). In case of He I emission line, it is diffi- We examined the relationship between the ratio of the sur- cult to discuss where is the emitting region. Mg II h, k lines face flux of the Ca II IRT emission line to the stellar bolomet- ′ are expected to blend each other in PMSs because they often ric luminosity, R , and NR of the 60 PMS stars (Fig. 4). For rotate fast (Neuhauser et al. 1995). calculating NR, we estimated τc of the PMS stars using pre- main sequence evolutionary tracks presented in Jung & Kim 4.3 Rotation-Activity Relation of Mg I emission lines (2007). In this track, τc of low-mass stars near the main sequence was not calculated. For such objects, we applied the We noticed that Mg I line at 8808 Å is a relatively optically approximation of Noyes et al. (1984). A detailed description thin and isolated chromospheric line. This Mg I emission line ′ of the calculation of NR and R is presented in Yamashita et has ever been observed only for one object, the Sun. The al. (2020). solar observations suggest that it forms in the chromosphere Only three PMS stars show broad and strong emissions, in- about 500 km above the photosphere (Fleck et al. 1994). dicative of large mass accretion. Most of the PMS stars have We examined the relationship between the ratio of the sur- narrow and weak emissions. This indicates that their chro- face flux of the Mg I emission line to the stellar bolometric ′ mospheric activity is induced by the dynamo process. All of luminosity, RMgI, and NR of the ZAMS stars (Fig. 5). In this ′ −4.2 their Ca II IRT emission lines have R ∼ 10 , which is work, we refereed NR calculated by Marsden et al. (2009).

Zenodo, 2021 3 Mai Yamashita, Yoichi Itoh & Yuhei Takagi

lar radius estimated using Stefan-Boltzmann’s law with the photospheric luminosity, and Teff of the objects in Gaia DR2. For object whose Teff is not listed in Gaia DR2, we used the luminosity and Teff listed in Marsden et al. (2009); VXR PSPC 50A, VXR PSPC 80A, [RSP95] 10, [RSP95] 15A, [RSP95] 45A, [RSP95] 88A, [RSP95] 8A, and Cl* IC2602 W79. F was mul- tiplied by the EQW of the Mg I emission lines. F ′ = F × EQW, (5) ′ ′ 4 For calculating R , F are divided by σTeff . ′ ′ F R = 4 , (6) σTeff where σ is Stefan-Boltzmann’s constant. The dependence of the surface flux upon the Teff of the objects is eliminated by this calculation. The typical uncertainty of R′ shown in Fig. 5 is estimated with the error of EQW of five ZAMS stars, Figure 4: Relationship between the ratio of the surface flux of ignoring the uncertainties of other parameters such as T . the Ca II IRT line to the stellar bolometric luminosity, R′ and eff We found that the Mg I line is still not saturated in the the Rossby number, NR (Yamashita et al. 2020). The filled cir- −1.8 −0.8 cles represent the PMS stars with narrow Ca II IRT emission region of 10 < NR < 10 , which is the saturated lines and the open circles show the PMS stars with broad Ca II region for Ca II IRT emission lines. According to the discus- IRT emission lines. The cross symbols represent ZAMS stars sion by Marsden et al. (2009), the chromospheric saturation in the young open clusters IC 2391 and IC 2602 (Marsden et is suggested to be caused by the similar to coronal saturation. al. 2009). Suppose if the emitting region of Ca II are completely filled, the unsaturation of the Mg I emission line suggests that the area of the Mg I emitting region may be smaller than that of Ca II. Otherwise Mg I is optically thinner than Ca II. We ex- They used P for 21 ZAMS stars or v sin i for 31 ZAMS stars pected that the Mg I line is useful for estimating appropriate into Eq. (2), and calculated NR. τc of PMS stars. P 2πR∗ NR ≡ = < sin i> (2) τc τcv sin i We calculated the ratio of the surface flux of the emission line to the stellar bolometric luminosity, R′, for the Mg I emission line. For calculating the surface flux F ′, a bolometric continuum flux per unit area at a stellar surface, F , was calculated at first. We used the i-band mag (the AB system) of the UCAC4 Catalogue (Zacharias et al. 2013). For nine objects, namely VXR PSPC 45A, VXR PSPC 80A, [RSP95] 10, [RSP95] 15A, [RSP95] 35, [RSP95] 43, [RSP95] 45A, [RSP95] 79, [RSP95] 8A, [RSP95] 88A, their i-band mag are not listed in the UCAC4 Catalogue (Zacharias et al. 2013), so that we used the i-band mag listed in Marsden et al. (2009). F is given as f 2 log = − × mi∗, (3) f0 5 d 2 Figure 5: The relationship between the ratio of the surface F = f × , (4) flux of theMg I line (λ8807.8 Å) to the stellar bolometric lu- R∗ ′ minosity, RMgI and the Rossby number, NR , of ZAMS stars where f is the bolometric continuum flux of the object per (filled circles). The cross symbols represent R′ of Ca II IRT unit area as observed on the Earth. mi∗ is the apparent mag- emission lines of the ZAMS stars (Marsden et al. 2009). nitude of the object in the i-band. The bolometric continuum flux per unit area under mi = 0mag (the AB system) con- −12 −2 −1 dition, f0, is 1.852 × 10 W · m · Å (Fukugita et al. 1996). d denotes the distance of an object from the Earth 5 Conclusion (Gaia DR2: Bailer-Jones et al. 2018). Unfortunately, the dis- 1. Seven PMS stars (DG Tau, DL Tau, DR Tau, RY Tau, tance of VXR PSPC 50A,VXR PSPC 80A,[RSP95]10, [RSP95] SU Aur, RECX 15, and RX J1147.7-7842) have broad Ca 15A,[RSP95]45A,[RSP95]88A,[RSP95]8A,and [RSP95]42C II IRT emission lines. It is suggested that their broad are not listed in Gaia DR2, so we substituted that of IC 2391 emissions result from heavy mass accretion from their and IC 2602 (145pc; van Leeuwen 2007). R∗ is the stel- protoplanetary disks.

4 Zenodo, 2021 The 20.5th Cambridge Workshop on Cool Stars, Stellar Systems, and the Sun

2. Most PMS stars have narrow Ca II IRT emission lines Stauffer, J. R., Hartmann, L. W., Prosser, C. F., et al. 1997, As- similar to zero-age main-sequence stars in young open trophys. J., 479, 776 clusters. The emissions of these objects indicate no cor- van Leeuwen, F. 2007, Astron. Astrophys., 474, 653 relation with the mass accretion rate. The ratio of the Yee, S. W., Petigura, E. A., & von Braun, K. 2017, Astrophys. surface flux of the Ca II IRT emission lines to the stellar J., 836, 77 bolometric luminosity, R′, of these objects are as large Yamashita, M., Itoh, Y., & Takagi, Y. 2020, PASJ, 72, 80 as the largest R′ of the cluster members. Most PMS Zacharias, N., Finch, C. T., Girard, T. M., et al. 2013, Astro- stars have chromospheric activity similar to zero-age phys. J., 145, 1 main-sequence stars. The chromosphere of these stars are completely filled by the Ca II emitting region. 3. The Mg I line is not saturated yet in "the saturated −1.6 regime for the Ca II emission lines", i.e. 10

Zenodo, 2021 5