Rotation and activity in late-type members of the young cluster ASCC 123 Free

ABSTRACT

ASCC 123 is a little-studied young and dispersed open cluster. Recently, we conducted the first research devoted to it. In this paper, we complement our previous work with Transiting Exoplanet Survey Satellite (TESS) photometry for the 55 likely members of the cluster. We pay special attention to seven of these high-probability members, all with FGK spectral types, for which we have high-resolution spectra from our preceding work. By studying the TESS light curves of the cluster members, we determine the rotational period and the amplitude of the rotational modulation for 29 objects. The analysis of the distribution of the periods allows us to estimate a gyrochronological age for ASCC 123 similar to that of the Pleiades, confirming the value obtained in our previous investigation. A young cluster age is also suggested by the distribution of variation amplitudes. In addition, for those stars with spectroscopic data, we calculate the inclination of their rotation axis. These values appear to follow a random distribution, as already observed in young clusters, with no indication of spin alignment. However, our sample is too small to confirm this on more solid statistical grounds. Finally, for these seven stars, we study the level of magnetic activity from the Hα and Ca ii H&K lines. Despite the small number of data points, we find a correlation of the Hα and Ca ii flux with Rossby number. The position of these stars in flux–flux diagrams follows the general trends observed in other active late-type stars.

1 INTRODUCTION

Stellar rotation and magnetic activity are properties strongly related to the evolution of stars, from the pre-main to the post-main sequence (MS). During the MS phase, the position of a star on the Hertzsprung–Russell (HR) diagram changes very little, so it is not an efficient indicator of stellar age. Conversely, the rotation rate and the level of magnetic activity, which is closely related to the rotation speed, change significantly over the life of a star in the MS due to the magnetic braking. The latter is caused by the angular momentum carried off by magnetized stellar wind.

Since the pioneering work of Skumanich (1972), who discovered the empirical law that bears his name, of the decline of the equatorial rotation speed with the inverse square root of the star’s age, the rotation rate and magnetic activity have been used to estimate the age of stars in the solar neighbourhood (e.g. Barry, Cromwell & Hege 1987; Soderblom, Duncan & Johnson 1991; Barnes 2003; Barnes, Spada & Weingrill 2016; Stanford-Moore et al. 2020; and references therein). Another widely used age indicator is the atmospheric abundance of lithium (e.g. Jeffries 2014; Gutiérrez Albarrán et al. 2020; and references therein).

On the other hand, rotation and activity can also be used as additional criteria, along with the kinematics, lithium abundance, and the position in the HR diagram, for evaluating the probability for a star to be member of an open cluster. This can be of great help in the case of stars in the halo of young clusters or very dispersed clusters. This is the case of ASCC 123, a young and poorly studied cluster. ASCC 123 was discovered by Kharchenko et al. (2005) based on Hipparcos proper motions and BV archival photometry. They found 24 likely members spread over a large region of more than 2° on the sky. They found a low reddening, E(B − V) = 0.10, placed the cluster at a distance of 250 pc, and estimated an age of about 260 Myr. The parameters of this cluster were subsequently reanalyzed by Yen et al. (2018) on the basis of the Gaia DR1/TGAS and HSOY data (Altmann et al. 2017). They confirmed the low reddening, E(B − V) = 0.097, a distance d = 243.5 pc, and revised the cluster age as τ = 130 Myr. Cantat-Gaudin et al. (2018), from the Gaia DR2 astrometric and photometric data, reported 55 cluster members and estimated a distance of 233.1 ± 5.5 pc.

In our previous work (Frasca et al. 2019, hereafter Paper I), we studied some of the brightest candidate members of ASCC 123 using high-resolution spectra obtained with HARPS-N at the Telescopio Nazionale Galileo (TNG) with the aim of determining their physical parameters, radial and rotational velocities, and the abundances of some elements. The chemical composition of this cluster turned out to be compatible with the Galactic trends in the solar neighbourhood. The HR and colour–magnitude diagrams (CMDs), as well as the lithium abundance and the Hα emission allowed us to infer an age in the range 100–200 Myr, i.e. similar to that of the Pleiades.

In the present paper, we use ground-based and Transiting Exoplanet Survey Satellite (TESS) photometry for the members of ASCC 123 to investigate their photospheric activity and study the distribution of their rotation periods. We use also the HARPS-N spectra already gathered to derive the level of magnetic activity from the Hα and Ca ii H&K lines and the projected rotational velocity (vsin i) to derive the inclination of the rotation axes. The paper is organized as follows. In Section 2, we present the observations and the criteria followed to select our targets. In Section 3, we show the results of our work, describing the analysis carried out on both the photometric and spectroscopic data. We discuss our results and compare our findings with other clusters in Section 4. Finally, Section 5 summarizes the main results and presents our conclusions.

2 OBSERVATIONS AND DATA REDUCTION

2.1 Spectroscopy

In this paper we investigate the chromospheric activity for the single, late-type stars studied in Paper I that are bona-fide members (i. e. with a membership probability Prob = 1) according to Cantat-Gaudin et al. (2018). For these seven FGK stars, we follow the same numbering used in our previous work (ID). We use high-resolution spectra taken in 2018 and 2019 with GIARPS (GIANO-B & HARPS-N; Claudi et al. 2017) at the 3.6 m TNG telescope. The reader is referred to Paper I for a description of the data and their reduction. The analysis of these spectra with the code ROTFIT, aimed at the determination of the atmospheric parameters and vsin i, can be also found in Paper I.

Table 1 reports the ID of the observed targets along with the TESS Input Catalog (TIC, Stassun et al. 2019; Paegert et al. 2022) identifier, equatorial coordinates (RA, DEC), effective temperature (T_eff), spectral type (SpT), and vsin i, which were already derived in Paper I. The masses (M_*) and radii (R_*), reported in columns 10 and 11 of Table 1, have been derived from the comparison of the position of these stars in the HR diagram (fig. 8 in Paper I) with the parsec isochrones (age = 155 Myr) and evolutionary tracks (Bressan et al. 2012). The rotation period (P_rot) measured from the analysis of TESS light curves (Section 3.1) and the inclination (i) of the rotation axis, calculated as described in Section 4.3, are also quoted in the last columns of Table 1.

2.2 Photometry

Space-born accurate photometry was obtained with NASA’s TESS (Ricker et al. 2015) for all the cluster members identified by Cantat-Gaudin et al. (2018), 55 stars in total. Our targets were observed by TESS in sector 16, between 2019 September 12 and 2019 October 6, and sector 17, between 2019 October 8 and 2019 November 02. The observations in two consecutive sectors allowed us to obtain nearly uninterrupted sequences (with a gap of about 1.4 d) of high-precision photometry lasting about 50 d, with a cadence of 30 min. Being ASCC 123 a sparse and nearby cluster, there is no severe star crowding, also taken the large pixel size of TESS (21″) into account. Indeed, searching in the Gaia DR3 catalogue (Gaia Collaboration 2022), no star with a comparable magnitude (ΔG < 2 mag) can be found within a radius of 21″ around the position of each target. Therefore, we expect no relevant flux contamination from nearby sources. We downloaded the TESS light curves (Huang et al. 2020) from the Mikulski Archive for Space Telescopes (MAST)¹ archive and used the simple aperture photometry flux (SAP) or the pre-search data conditioning SAP flux, where long-term trends have been removed, whenever available. For the targets fainter than about V = 14 mag, we used the TESS light curves extracted with a point spread function-based approach (PATHOS, Nardiello et al. 2019).

We have also performed multiband BVR_CI_C photometric observations with the facility imaging camera² at the 0.91 m telescope of the M. G. Fracastoro station (Serra La Nave, Mt. Etna, 1735 m a.s.l.) of the Osservatorio Astrofisico di Catania (OACT, Italy). We observed the three G-type members in our sample, namely, S 214, S 435, and S 517 during nine nights from October 22 2019 to January 16 2020, collecting from 23 to 29 data points per each filter and per each star. Time series data covering about 3–4 h could be acquired only on October 23 and November 4 and 20 2019. We decided to start observing S 214 and S 517 because of their large vsin i, which would imply a rotation period short enough and a modulation amplitude large enough to be able to detect brightness variations with observations from the ground. S 435 was included as a test target because of its spectral type similar to the other two fast-rotating stars. We did not consider the fast-rotating F4V stars S 39 and S 554 for which we expected small variation amplitudes that are hardly detectable from the ground. We used exposure times of 60, 30, 15, and 15 s for the B, V, R_C, and I_C band, respectively. The data were reduced by subtracting master darks taken with the same exposure times as the science images and by dividing them by master flats. Aperture photometry with a radius of 6 pixels (≃4″) was performed for the unsaturated stars in the field of view of 11.2′ × 11.2′ of each of the observed targets. Standard stars in the open cluster NGC 7790 (Stetson 2000) were also observed in the nights with the best photometric conditions to calculate the zero-points and transformation coefficients to the Johnson–Cousins system. With the latter, we calculated the BVR_CI_C magnitudes for a number of non-variable stars in each target field that have been used as comparison for doing ensemble photometry of the monitored stars. Although the precision and cadence from this ground-based photometry is not comparable to that of TESS, these multiband data are useful to determine average magnitudes and colours for these three targets. Since in the meantime TESS had started to observe the sky region containing ASCC 123 providing a very precise photometry (although with no colour information), we decided not to continue with the observations from the ground.

The main results presented in this paper are based on the TESS photometry only. For the sake of completeness, we report the results of ground-based photometry in Appendix  A.

3 DATA ANALYSIS AND RESULTS

3.1 TESS light curves

The main information we can get from the TESS light curves is the rotation period, P_rot, of the star, which can be measured thanks to the rotational modulation of the star’s brightness produced by an uneven distribution of cool photospheric spots. The amplitude of the photometric variations is also a useful parameter related to the magnetic activity.

To measure P_rot, we have applied a periodogram analysis (Scargle 1982) and the clean deconvolution algorithm (Roberts, Lehar & Dreher 1987) to the TESS light curves of the members of ASCC 123. For the stars with periods longer than a few days, we merged the data of sectors 16 and 17 to expand the time base so as to improve the precision of the period determination. Ultrafast rotators, such as S 214 and S 517, display a substantial starspot evolution during the 50-d time baseline, so that for these stars we have preferred to analyse independently sectors 16 and 17, obtaining always very similar values of P_rot in the two epochs.

We retrieved the TESS light curves for the 55 candidate members of the cluster ASCC 123 according to Cantat-Gaudin et al. (2018) but we were able to determine the rotation period only for 29 of them. Indeed, we rejected the stars with Gaia magnitude G > 16.5, whose TESS photometry is too noisy for a meaningful measure of the rotation period. Moreover, we discarded the objects that we recognized as double-lined binaries (SB2) in Paper I. All the periods are reported in Table 2. For illustrative purposes, the TESS light curve and the result of the period search for S 214 is shown in Fig. 1. The light curves of the remaining FGK members studied in Paper I are displayed in Figs B1–B7. The modulation of the star brightness produced by the spots and star rotation is clearly visible for all the targets; exceptions are the two hottest sources. Interestingly, the light curve of S 39 clearly displays two dips reminiscent of a planet transit with a time separation of about 20.31 d (see also Fig. B2). When we started this investigation, this star was not known to host exoplanets and was not included in the list of TESS Objects of Interest. We have then proposed this source for the TESS Follow-up Observing Program and we are planning ground-based observations, whose results will be presented in a forthcoming work. To search for the rotational period of S 39, we excluded the two transits from the TESS light curve obtained merging those of sector 16 and 17 before applying the period analysis.

We note that the cleaned periodograms display for all the stars but S 39, a clear power peak (marked with a red vertical line in each plot) without secondary peaks of comparable amplitude. This period thus represents the rotational one. Even the hottest star, S 554, displays a significant rotational modulation, although with a very low amplitude of only ∼0.007 mag, which is clearly shown by the phased light curve displayed in the right inset panel of Fig. B4. As mentioned above, the only doubtful case is S 39, for which the highest power in the periodogram is found at a period 4.86 d, which is clearly inconsistent with the rapid rotation of the star indicated by the high value of the projected rotation velocity, vsin i ≃ 49 km s⁻¹. However, the periodogram displays other peaks at about 7 d (which is still more inconsistent with the vsin i) and at 1.67 d (marked with a blue vertical line in the left inset panel of Fig. B1); we consider the last as a possible P_rot value and list it in Table 1 with a colon.

3.2 Chromospheric emission

The level of chromospheric activity can be evaluated from the emission in the core of the Hα and Ca ii H and K lines (see e.g. Frasca & Catalano 1994; Frasca et al. 2010, 2019; and references therein). To this end, we subtracted photospheric templates from the observed spectra of the targets, to remove the underlying photospheric absorption lines so as to leave as residual the chromospheric emission in the cores of Hα and Ca ii lines. The photospheric templates are made with synthetic BT-Settl spectra (Allard, Homeier & Freytag 2012) for Ca ii lines, and spectra of non-active stars for the Hα; indeed the latter are better reproducing the core of the Hα line in absence of a significant chromospheric contribution. Since the signal-to-noise ratio in the Ca ii H & K region is quite low, ranging from 3 (for the coolest members) to 38 (for the hottest one), we have degraded the original resolution (R = 115 000) of the HARPS-N spectra and the synthetic spectra to R = 42 000, which is the same of the ELODIE templates used for the determination of the atmospheric parameters and the subtraction of the photospheric Hα template (Paper I). Furthermore, the photospheric templates have been aligned in wavelength with the target spectra by means of the cross-correlation function and have been rotationally broadened by the convolution with a rotational profile with the vsin i of the target star (Table 1).

The observed spectra (black lines) and the photospheric templates (red lines) are displayed in Fig. 2 for the Ca ii H & K region. Fig. B8 shows the Hα profiles of the targets (black dots) overlaid to the photspheric templates (red lines) and the difference of the two (blue lines). The excess Hα equivalent width (EW), W_Hα, has been obtained by integrating the residual Hα emission profile, which is highlighted by the green hatched areas in the difference spectra in each box of Fig. B8. Analogously, for each target, the difference between observed and template spectrum in the Ca ii region leaves emission excesses in the cores of the H and K lines, which have been integrated obtaining the EWs, W_CaII-H and W_CaII-K. It is worth noticing that the hottest star in our sample, S 554 (T_eff = 6871 K), does not show any filling in the Hα core and only a tiny filling in the cores of the Ca ii lines (upper right panels in Figs B8 and 2). For S 39 (T_eff = 6667 K), a small residual emission is detected in the core of the Hα and also in the Ca ii lines. However, unlike the other stars of lower temperature, no reversal in emission emerges in the cores of the H and K lines for these two F4-type stars. This is in line with what is expected on the basis of the reduced chromospheric activity in these stars with shallow convective envelopes and the difficulty of detecting low chromospheric emission against a large continuum flux.

As a more effective indicator of chromospheric activity, we computed the surface flux in each of the chromospheric lines, F_line. For the Ca ii  K line, this reads as

where F₃₉₃₃ is the flux at the continuum at the centre of the Ca ii K line per unit stellar surface area, which is evaluated from the BT-Settl spectra (Allard et al. 2012) at the stellar temperature and surface gravity of the target. We evaluated the flux error by taking into account the error of W_CaII-K and the uncertainty in the continuum flux at the line centre, F₃₉₃₃, which is estimated considering the errors of T_eff and log g.

We computed surface fluxes for the other chromospheric diagnostics in the same way as for the Ca ii K line. The EWs and fluxes are reported in Table 3.

4 DISCUSSION

4.1 Rotational periods

The distribution of rotation periods as a function of the stellar mass is a very useful tool for characterizing the evolutionary stage of a cluster (e.g. Barnes 2003; and references therein). To this aim, we plot in Fig. 3(a) the periods listed in Table 2 versus the colour index (G_BP − G_RP)₀, which has been dereddened using the value A_V = 0.13 mag derived in Paper I and the extinction relations in Wang & Chen (2019). In the same plot we also show the periods derived by Rebull et al. (2016) for a sample of Pleiades members observed with Kepler–K2. To compare our period distribution with previous works devoted to other clusters, we also use the colour index (V − K_s)₀ in Fig. 3(b). Despite the small data sample, the rotation periods for the candidate members of ASCC 123 roughly follow the distribution of the Pleiades, with the presence of both fast and slow rotators for G-type stars. We note that, among the three solar-like stars, S 214 and S 517 lie in the fast-rotating C sequence (according to the notation of Barnes 2003), while S 435 falls on the I sequence of the Pleiades, for which the period increases steadily with the colour index up to (G_BP − G_RP)₀ ≈ 1.5 mag, and then decreases reaching the C sequence at (G_BP − G_RP)₀ ≈ 3 mag. We do not find fast rotators (in the lower C sequence) with colour 1 < (G_BP − G_RP)₀ < 2.5, i.e. approximately in the spectral-type range G3–M3. However, the presence of the two fast-rotating solar-type stars strongly supports a young age for the cluster, because the C sequence of older clusters like NGC 3532 (age ∼300 Myr) displaces towards cooler objects (e.g. Barnes 2003; Fritzewski et al. 2021) and practically disappears for the still older Hyades or Praesepe clusters (Barnes 2003; Rebull et al. 2022), i.e. at age ≈ 600 Myr. Indeed, the hottest stars in NGC 3532 with P_rot < 1 d have a (V − K_s)₀ ≃ 3 (Fritzewski et al. 2021). Another cluster in the age range between the Pleiades and the Hyades is M 48. For this cluster, with an age of about 450 Myr, Barnes et al. (2015) found no G- or early K-type fast-rotating stars. A younger cluster showing a clear C sequence reaching the G-type stars is M 34 (age ∼ 250 Myr; Ianna & Schlemmer 1993). Therefore, the gyrochronological age of ASCC 123 should not be greater than 250–300 Myr.³

As a further check, we have also plotted in Fig. 3, rotational isochrones calculated for slow-rotator sequences by Spada & Lanzafame (2020) and expressed as a function of stellar mass or B − V in their Table A1. For this purpose, we have converted the B − V colour index into V − K_S and G_BP − G_RP according to the calibrations of Mamajek.⁴ As apparent, the I sequences of both ASCC 123 and the Pleiades agree with the theoretical isochrone of Spada & Lanzafame (2020) with the smallest age in their model (τ = 100 Myr).

On the other hand, a much younger age seems to be ruled out by the clear gap between fast and slow rotators observed for ASCC 123. Indeed, this gap is not visible in very young (≈16 Myr) stellar populations, such as the Upper Centaurus-Lupus and Lower Centaurus-Crux (Rebull et al. 2022). The most discrepant object, with respect to the Pleiades diagram, is TIC 420123691, for which we measure a period of 13.4 d that places it much higher than the Pleiades I sequence. This object could be a non-member or a close binary.

4.2 Variation amplitudes

Another proxy for the magnetic activity level is the amplitude of photometric variations induced by starspots (see e.g. Messina 2021; See et al. 2021). The variation amplitudes of our sources, which are reported in Table 2, were measured on their TESS light curves, converting fluxes into magnitudes and rejecting 5σ outliers. We excluded long-term linear trends by taking suitable data chunks and took the largest variation amplitude. We show the amplitude of variation as a function of the (G_BP − G_RP)₀ colour index in Fig. 4(a), where the same data for the Pleiades (Rebull et al. 2016) are also plotted for comparison. TIC 420123691, with an amplitude of 0.44 mag, is the most discrepant object also in this diagram. A spectroscopic follow-up would be very useful to better clarify its nature. A similar plot with the (V − K_s)₀ colour in the x-axis is shown in Fig. 4(b), where we overlay the amplitudes at the 80th percentile of three clusters (Upper Sco, Pleiades, and Praesepe) in bins of the colour index as derived by Messina (2021, see his table 1). For the Pleiades, we plot the 80th percentile amplitudes of both the fast-rotating (0.1 < P_rot < 1.0 d) and slow-rotating stars (3 < P_rot < 9 d), while for the Praesepe, we do not have the sequence of fast rotators in this colour range, so we plot only the amplitude of slow rotators (P_rot > 9 d). Indeed, for the Praesepe stars with P_rot > 9 d, Messina (2021) reports the 80th percentile data for all the bins of (V − K_s)₀ in the range 1–6 mag. However, we note that the amplitudes of the Praesepe stars with 3 < P_rot < 9 d for the three (V − K_s)₀ bins for which they could be determined are basically the same as those in the longest period range. The comparison of the variation amplitudes with the loci of the three clusters further indicates that ASCC 123 is clearly older than Upper Sco (age ≈ 10 Myr; Feiden 2016) and younger than Praesepe, suggesting an age similar to that of the Pleiades.

One of the main ingredients of dynamo mechanisms generating magnetic fields and operating in stellar interiors is the rotation rate. Therefore, a correlation of magnetic field intensity (or proxies of magnetic field) and the rotation period is expected. Indeed, the correlation with P_rot has been widely documented in the literature for several diagnostics of magnetic activity at chromospheric and coronal levels (see e.g. Pizzolato et al. 2003; Cardini & Cassatella 2007; Reiners, Schüssler & Passegger 2014; Frasca et al. 2016). As regards the photospheric activity, traced, e.g. by the variation amplitude, correlation with P_rot has been found by Reinhold & Hekker (2020), who subdivided their very large sample of stars with periods and amplitudes detected from Kepler–K2 light curves in several subsamples with 200 K temperature bins. They found a general decrease of variability with increasing rotation period for all temperature bins, although the data show large scatter. We plot the variation amplitude as a function of P_rot in Fig. 5(a). No clear decrease of the amplitude with increasing P_rot is visible either for ASCC 123 or the Pleiades. This result is supported by the low Spearman’s correlation coefficient, both for the Pleiades (ρ ≃ −0.02) and ASCC 123 (ρ ≃ +0.12, even excluding the discrepant star with P_rot = 13.4 d). However, we note that our data include stars with very different T_eff values (and consequently different internal structure). As pointed out in several works, including some of those mentioned above, a better variable related to the dynamo action is the Rossby number, R_O = P_rot/τ_c, defined as the ratio of the rotation period and the convective turnover time τ_c. The latter is not a directly measurable variable, but can be derived from theoretical models for MS stars or from calibrations as a function of temperature or colour indices. In our case, we have used the empirical relation proposed by Wright et al. (2011, equation 10) as a function of V − K_S. Some correlation, albeit with large scatter, is found between amplitude and R_O (Fig. 5b), with Spearman’s coefficients ρ ≃ −0.16 and −0.13 for ASCC 123 and the Pleiades, respectively.

4.3 Stellar spin orientation

For the stars studied in Paper I, for which we measured accurate values of vsin i, and thanks to the rotational periods, P_rot, and stellar radii, R_*, which are reported in Table 1, we can derive the inclination of the rotation axis as

We found a value of sin i < 1 for all the stars, with the exception of S 39. For it, the peak at 1.67 d is marginally consistent (considering the errors) with the value of vsin i|$\simeq \,$| 49 km s⁻¹ measured in Paper I, implying an inclination close to 90°, but the TESS light curve phased with this period does not show any clear rotational modulation. We therefore consider this period an uncertain value. However, if the spin axis is nearly parallel to the orbital axis of the presumed transiting object, the inclination of the rotation axis should be very close to 90°, making 1.67 d a possible rotation period for this star. The values of inclination of the rotation axis for the seven stars with HARPS-N spectra are also listed in Table 1.

The statistical sample is too small to study the distribution of the inclinations of the stellar spins and investigate their degree of alignment, as pointed out, e.g. by Corsaro et al. (2017), whose simulations show that a sample 2–3 times larger is required for obtaining a high significance of any eventual spin alignment. However, we wanted to compare the values of i that we find for the seven stars in Table 1 with those expected from a random 3D distribution of the rotation axes, in a similar way to what Corsaro et al. (2017) did for giant stars in the old open clusters NGC 6791 and NGC 6819. In their case, for both clusters the i distribution was concentrated towards low values, i.e. it was very different from that expected from a random distribution of the rotation axes (see fig. 1 in their paper), suggesting a significant alignment of the stellar spins. In our case (Fig. 6), the values of inclination are all greater than 50 degrees and are compatible with a random distribution or, in any case, they do not allow us to draw firm conclusions. High- or medium-resolution spectra for a significant number of cluster members, as expected from future multi-object surveys like WEAVE (Jin et al. 2023), are needed to better investigate this aspect. We note that Healy, McCullough & Schlaufman (2021) find that isotropic spins or moderate alignment are both consistent with the sin i values they derived for members of Pleiades and Praesepe clusters. In a recent work, Healy et al. (2023) found eight out of the 10 investigated clusters having spin–axis orientations consistent with isotropy and two clusters whose distributions can be better described by an aligned fraction of stars combined with an isotropic distribution. However, they suggest this result can be influenced by systematic errors on i and by the poor statistics.

4.4 Chromospheric activity

Correlations between diagnostics of chromospheric activity and P_rot, or the Rossby number, have often been reported in the literature (e.g. Douglas et al. 2014; Frasca et al. 2016; Han et al. 2023; and references therein). For rapidly rotating stars, a saturated regime, in which the activity level is constant with P_rot, is clearly shown by the coronal X-ray emission (e.g. Pizzolato et al. 2003; Wright et al. 2011) but it is also seen at chromospheric level (e.g. Douglas et al. 2014; Newton et al. 2017). In particular, Douglas et al. (2014) found that Hα activity is saturated for R_O < 0.11 for members of Hyades and Praesepe, while Newton et al. (2017) found saturation in their sample of nearby field M-type stars for R_O < 0.2.

Despite the small number of stars with measured Hα and Ca ii fluxes, we wanted to see if there is any hint of a correlation with rotation in our data. As an example, we show F_CaII-K as a function of R_O in Fig. 7, which clearly shows the decline of flux with increasing R_O. The only discrepant point is that of the hottest star, S 554. Similar results are also obtained with Ca ii H and Hα as evidenced by the Pearson’s correlation coefficients ρ = −0.65, −0.71, and −0.74, for the Hα, Ca ii K, and H lines, respectively. Apart from the two F4V sources (S 39 and S 554), only the two ultrafast G-type stars (S 214 and S 517) should have a saturated chromospheric emission, on the basis of their Rossby number. These are the only stars in Table 1 for which a value of X-ray flux is reported in the literature. The values of F_X = 3.36 × 10⁻¹³ and 3.65 × 10⁻¹³ erg  cm⁻² s⁻¹ reported by Freund et al. (2022) translate, with the Gaia parallaxes and our values of T_eff and R_* (Table 1), into log (L_X/L_bol) values of −3.36 and −3.31 for S 214 and S 517, respectively. These values are close to the saturation level of −3.13 reported by Wright et al. (2011).

To further investigate the behaviour of the magnetic activity at chromospheric level, we have compared the different diagnostics by means of flux–flux diagrams. These diagrams are shown, in a logarithm scale, in Fig. 8 along with the power–law relationships found by Martínez-Arnáiz et al. (2011) from their large sample of active FGKM stars. We note that all the members lie, within the error bars, over the flux–flux relation for the Ca ii K versus Ca ii H, while the Hα fluxes, which are displayed versus Ca ii K ones in the right-hand panel of Fig. 8, are more scattered. This is consistent with the results of Martínez-Arnáiz et al. (2011), who found a larger scatter in the Hα–Ca ii flux diagrams compared to diagrams based on different Ca ii lines. Furthermore, they distinguished two different regimes in the Hα–Ca ii flux diagrams, with the very active late-K and M-type stars significantly deviating from the main behaviour and lying in an upper branch. They argued that these are stars with a saturated X-ray flux. The stars in our sample are all earlier than those in the upper branch of the flux–flux diagrams of Martínez-Arnáiz et al. (2011) and only the ultrafast G-type stars S 214 and S 527 display an activity level close to saturation. The remaining GK-type stars should not be saturated, on the basis of their Rossby number and the above mentioned correlations. However, we note that the two K-type stars, F1 and F2, lie right on the average power–law relation (red dotted line in Fig. 8), while the remaining stars are placed systematically lower, but are closer to the lower branch fit (blue dashed line in Fig. 8). The latter was calculated by Martínez-Arnáiz et al. (2011) by excluding the stars with saturated X-ray and Hα emission from the fit.

5 SUMMARY AND CONCLUSIONS

In this work, we have revisited ASCC 123. It is a little-studied young cluster, with no detailed spectroscopic investigation apart our recent study (Frasca et al. 2019). In that paper, we determined the main properties of the cluster, such as distance, age, and chemical composition, by combining Gaia data with high-resolution spectroscopy taken with HARPS-N at the TNG telescope. In this work, we have focused on rotation and magnetic activity. With this aim, we retrieved TESS photometry for all the 55 cluster members reported in Cantat-Gaudin et al. (2018). Among them, seven FGK-type stars were observed spectroscopically in our previous work. We performed the analysis of the light curves of the cluster members obtaining the rotational periods and amplitudes for 29 objects. By comparing the distributions of period and amplitude versus colour index with those of some well-studied clusters, we infer a gyrochronological age for ASCC 123 similar to that of the Pleiades, which is consistent with our precedent estimates based on the isochrone fitting and abundance of lithium. No clear correlation between the variation amplitude and P_rot is observed, while a Spearman’s coefficient ρ = −0.16 between amplitude and Rossby number suggests only a marginal correlation of these variables. Additionally, for the seven stars with spectra, we have derived the inclination of the rotational axis. Although our sample is not statistically significant, the distribution of inclinations seems to be random, which would be in agreement with that observed in other young open clusters. Finally, we studied the level of magnetic activity of these stars from the Hα and Ca ii H&K lines. We find that the chromospheric activity indicators display a better correlation with the Rossby number [ρ in the range (−0.65, −0.74)]. Moreover, they follow the general trends observed in other active FGKM stars in the flux–flux diagrams.

ACKNOWLEDGEMENTS

We thank the anonymous referee for a careful reading of the manuscript and valuable comments and suggestions. This research used the facilities of the Italian Center for Astronomical Archive (IA2) operated by Istituto Nazionale di Astrofisica (INAF) at the Astronomical Observatory of Trieste. This research has made use of the SIMBAD database and VizieR catalogue access tool, operated at CDS, Strasbourg, France.

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.

This paper includes data collected by the TESS mission, which are publicly available from the MAST. Funding for the TESS mission is provided by the NASA’s Science Mission Directorate. This work has been supported by the PRIN-INAF 2019 STRADE (Spectroscopically TRAcing the Disk dispersal Evolution) and by the Large Grant INAF YODA (YSOs Outflow, Discs, and Accretion).

Based on observations made with the Italian TNG operated on the island of La Palma by the Fundación Galileo Galilei of the INAF at the Observatorio del Roque de los Muchachos, as part of the Large Program ‘Stellar Population Astrophysics’. Based on photometry collected at the INAF–Osservatorio Astrofisico di Catania.

DATA AVAILABILITY

The spectroscopic and ground-based photometric data underlying this paper will be shared on reasonable request to the corresponding author. TESS photometric data are available at https://archive.stsci.edu/.

Footnotes

References

APPENDIX A: GROUND-BASED PHOTOMETRY

As mentioned in Section 2.2, we collected multiband BVR_CI_C photometry of the three G-type stars S 214, S 435, and S 517 at OACT. Despite the lower photometric precision, compared with TESS data, and the irregular sampling of our photometry, we were able to detect the rotational modulation and derive the period for the two ultrafast rotators (S 214 and S 517). The light curves in the BVR_CI_C bands of S 214 are shown in Fig. A1. The peak at a rotation period of 0.564 d is the highest one in the cleaned power spectrum of each band and is also recovered from the TESS photometry of this source (see Section 3.1). The amplitudes of the BVR_CI_C light curves are about 0.11, 0.15, 0.08, and 0.07 mag, respectively. The average magnitudes of these stars in the Johnson–Cousins bands from our OACT photometry are listed in Table A1.

APPENDIX B: ADDITIONAL FIGURES