betaFeedback

WorksPhysics and Astronomy

Comparative analysis of two delta Scuti stars approaching the TAMS using MESA-GYRE models

Abstract

This study presents a comparative asteroseismic analysis of two late main-sequence \dss\ with slow to moderate rotation, TIC 274211446 and TIC 16053262. Both are also characterised by showing an intense emission in the Hα_α line. Based on the first 35 extracted frequencies from short-cadence photometric data from the TESS mission, with no significant pulsations out of the [9, 40] \cd range, I estimated the large frequency separation. I calibrated evolutionary and adiabatic oscillation models using the \mesa\ and \gyre\ codes under the Traditional Approximation of Rotation (TAR). I used the observed large separation and spectroscopic temperature to constrain the models. Finally, I have used kernel curve fitting (KDE) to find the best profiles for both stars. With a hydrogen fraction of 21%\% in the convective core, TIC~274211446 is more evolved than TIC~16053262, which still retains 30%\% of its core fuel. The former has a large separation ΔνΔν = (2.7±\pm0.1) \cd, and points to a model with M = (1.84±\pm0.07) M⊙_{\odot}, a radius of R = (2.67±\pm0.07) R⊙_{\odot}, and an effective temperature of \teff = (7308±\pm71) K. The latter, with ΔνΔν = (3.25±\pm0.1) \cd, points to a more compact model with M = (1.72±\pm0.07)M⊙_{\odot}, R = (2.28±\pm0.05)R⊙_{\odot}, and \teff = (7298±\pm59) K. The equatorial rotation velocities are (28±\pm1) \kms\ and (47±\pm13) \kms, respectively. TIC~274211446 has an age of (1070±\pm127) \myr, and TIC~16053262 is (1145±\pm165) \myr\ old.

Introduction

Between the main sequence (MS) and the instability strip δ\delta Sct are found. They are Population I stars with spectral types between A2 and F2 and masses ranging from 1.5 to 2.5 M⊙_{\odot} [8, 39, 1]. They can show dozens of radial and non-radial pulsation modes, with periods spanning from 15 minutes to 8 hours [8, 48, 22, 21]. Not all of them are acoustic (pp-modes); buoyancy modes (gg-modes) have also been detected in δ\delta Sct [48, 3]. With space-based data, primarily from Kepler (Bourque et al. 2010) and TESS [38], the number of known δ\delta and γ\gamma Dor has increased from a few hundred to thousands. Since the initial data provided by the CoRoT mission [2], periodicities have been observed in the pulsation spectra of δ\delta Sct stars [16, 14, 17, 33, 27]. The large frequency separation, common to the asymptotic regime in solar-like stars [45], has also become visible in δ\delta Sct stars, particularly in younger individuals [5]. Its scaling relation with stellar density [44, 14, 17] makes it a valuable parameter for constraining models. These authors also demonstrated that this relationship is independent of stellar rotation. However, estimating this parameter in stars that have evolved off the main sequence is difficult, primarily for two reasons: first, the avoided-crossing between pp-modes and gg-modes; and second, the splitting of non-radial modes due to rotation, which is typically moderate to high in these stars.

Regarding avoided-crossings, I am going to describe it based on Aerts et al. [1]. The Brunt-Väisälä frequency, NN, is a measure of the stability of the internal gas layers of a star against convective movements. It is given by the Eq. (1):

N2=gHP(∇ad−∇+χμ∇μ)(1)N^2 = \frac{g}{H_P}\left(\nabla_{\mathrm{ad}}-\nabla+\chi_\mu\nabla_\mu\right) \tag*{(1)}

where gg is the local gravity, HP=−P/(dP/dr)H_P=-P/(\mathrm{d}P/\mathrm{d}r) is the pressure-height scale, ∇=(dln⁡T/dln⁡P)\nabla=(\mathrm{d}\ln T/\mathrm{d}\ln P) is the real thermal gradient, ∇ad\nabla_{\mathrm{ad}} is the adiabatic thermal gradient, ∇μ=(dln⁡μ/dln⁡P)\nabla_\mu=(\mathrm{d}\ln\mu/\mathrm{d}\ln P) is the mean molecular weight, χμ=(∂ln⁡ρ/∂ln⁡μ)\chi_\mu=(\partial\ln\rho/\partial\ln\mu) is the sensitivity of the equation of state to changes in chemical composition, and ρ\rho, PP are, respectively, the local density and pressure in the stellar interior.

Within the convective core, the plasma is homogeneously mixed (∇μ=0\nabla_\mu=0), and the thermal gradient is nearly adiabatic (∇≈∇ad\nabla\approx\nabla_{\mathrm{ad}}). Then N2≤0N^2\leq0 and the gg-modes cannot propagate here. At the boundary, as one transitions from the convective zone to the radiative zone, a very steep gradient or a discontinuous jump occurs, depending on whether or not overshooting is active. This creates a high, narrow peak in NN, a composition bump. Outside the core, in the radiative zone, the composition stabilizes (∇μ=0\nabla_\mu=0), but transport is radiative (∇<∇ad\nabla<\nabla_{\mathrm{ad}}), so N2N^2 becomes positive. This is where gg-modes are born and propagate.

On the other hand, the parameter that best characterizes non-radial acoustic modes is the Lamb frequency SℓS_\ell, the minimum frequency at which an acoustic wave can propagate horizontally at a given depth within the star. It is given by the Eq. (2):

Sℓ=ℓ(ℓ+1)cs2r2(2)S_\ell=\frac{\ell(\ell+1)c_s^2}{r^2} \tag*{(2)}

where ℓ\ell is the spherical grade, csc_s is the local sound speed and rr is the local radii. It depends critically on ℓ\ell and rr. As r→0r\to0, Sℓ→∞S_\ell\to\infty. For this reason, acoustic waves never reach or cross the center of the star. The high Lamb frequency acts as a mirror, reflecting them back outward.

In order to separate pp-modes from gg-modes, one must compare the calculated frequencies, ω\omega, with these two limits, NN and SℓS_\ell, throughout the radii of the star. When ω<N\omega<N and ω<Sℓ\omega<S_\ell, the mode behaves like a buoyancy wave. It reaches its maximum amplitude near the convective core. When ω>N\omega>N and ω>Sℓ\omega>S_\ell, the mode behaves like an acoustic wave, reaching its maximum amplitude in the outer layers. If N<ω<SℓN<\omega<S_\ell or Sℓ<ω<NS_\ell<\omega<N, the wave cannot propagate, and its amplitude decreases exponentially.

As stars evolve along the MS, after having consumed a large fraction of the available hydrogen and approaching the TAMS, NN increases. In this phase of stellar evolution, both types couple to produce mixed modes pulsations. They behave as gg-modes deep within the star and as pp-modes when they reflect off the core and propagate back towards the surface [11, 29, 10, 1].

Secondly, these objects are typically fast rotators [40]. As their rotation velocity increases, non-radial modes split into the azimuthal order mm according to ωn,ℓ,m=ωn,ℓ+m(1−Cn,ℓ)Ω\omega_{n,\ell,m}=\omega_{n,\ell}+m(1-C_{n,\ell})\Omega, where Ω\Omega is the rotation frequency and Cn,ℓC_{n,\ell} is the Ledoux constant, accounting for the influence of the Coriolis force. For slow rotators, the Ledoux constant is approximately zero, and the multiplets are symmetric around the central m=0m=0 frequency [1]. However, the scenario becomes more complex in moderately and rapidly rotating stars, where multiplet symmetry breaks down [20] (Mirosh 2022).

In summary, the presence of mixed modes and densely interleaved multiplets within the acoustic regime constitute the two primary reasons why mode identification is challenging in late-MS δ\delta Sct stars, difficulting a precise characterisation of them using 1D models [30].

This is a comparative study of two similar F0-type stars with effective temperatures around 7300 K and a surface gravity around 3.9, using 1D MESA/GYRE models. This is possible because both of them are not rapid rotators, and their spectra appear to be free of gg- and mixed modes. Both stars are also notable for being emitters in the Hα\alpha line.

The paper is structured as follows: Sect. 2 introduces the two analysed stars; Sect. 3 explains the frequency analysis conducted for each target, which determines the large frequency separation and identifies several low-order radial modes; Sect. 4 details the properties of the computed MESA/GYRE grid and the statistical method for constraining the models; Sect. 5 presents the discussion of the results; and finally, Sect. 6 lists the main conclusions of this work.

The targets

The stars studied here are TYC 67-812-1 and BD+14 1714. Hereafter, they are referred to by their identifier in the TESS catalogue [42], TIC 274211446 and TIC 16053262, respectively. The key parameters are shown in Table 1.

TIC IDRA (deg)DE (deg)pp (mas)DD (pc)TmagT_{\rm mag}E(B−V)E(B-V)[Fe/H]TeffT_{\rm eff} (K)log⁡(L/L⊙)\log(L/L_{\odot})MM (M⊙M_{\odot})RR (R⊙R_{\odot})ρˉ\bar{\rho} (ρˉ⊙\bar{\rho}_{\odot})log⁡g\log g
27421144654.29±0.0454.29\pm0.044.85±0.034.85\pm0.032.18±0.042.18\pm0.04454±9454\pm910.147±0.00610.147\pm0.0060.174±0.0270.174\pm0.0270.080±0.0080.080\pm0.0087309±1367309\pm1361.19±0.041.19\pm0.041.65±0.271.65\pm0.272.45±0.142.45\pm0.140.11±0.040.11\pm0.043.88±0.093.88\pm0.09
16053262114.54±0.04114.54\pm0.0414.29±0.0414.29\pm0.043.67±0.043.67\pm0.04270±3270\pm38.840±0.0078.840\pm0.0070.016±0.0040.016\pm0.004−0.009±0.009-0.009\pm0.0097297±1097297\pm1091.13±0.021.13\pm0.021.65±0.281.65\pm0.282.29±0.082.29\pm0.080.14±0.040.14\pm0.043.93±0.083.93\pm0.08

Table 1. Parameters of the studied δ\delta Sct stars.

Below, in this paragraph, I present the astrometric and kinematic data obtained from Gaia DR3 [13, 25]. TIC 274211446 is located in the Taurus constellation, at ≃454\simeq454 pc (≃1,479\simeq1{,}479 light-years). It resides within the Local Arm (Orion Arm), pointing towards the outer region of the Milky Way. Three-dimensionally, it is positioned roughly three times deeper than the main core of the Taurus Molecular Cloud Complex (≃140\simeq140 pc). Its radial velocity (+9.72+9.72 km s−1^{-1}) indicates a net recession caused by galactic differential rotation. Its proper motion (pmra =−5.665=-5.665 mas/yr, pmdec =−14.976=-14.976 mas/yr) translates to a tangential velocity of 75.9 km s−1^{-1}. With a total 3D space velocity of 76.5 km s−1^{-1}, it is classified as a mature field star in a slightly eccentric orbit. TIC 16053262 is located in the Gemini constellation, at ≃270\simeq270 pc (≃880\simeq880 light-years), oriented towards the leading edge of the local disc rotation. It is in a low-density interstellar dust zone, near the outer boundaries of the Local Bubble. Its radial velocity is minimal (+2.41+2.41 km s−1^{-1}), demonstrating that it shares almost precisely the same orbit and velocity as the Sun. Its proper motion (pmra =−11.360=-11.360 mas/yr, pmdec =−0.494=-0.494 mas/yr) yields a tangential velocity of 53.9 km/s. Its 3D velocity of 54.0 km s−1^{-1} is typical for the thin disc, moving orderly and parallel to the galactic plane. The Renormalised Unit Weight Error (RUWE) is less than 1.4 in both cases (1.14 for TIC 274211446 and 0.97 for TIC 16053262), with the data fitting well to the solution of a single, isolated star.

There are hardly any references to these stars in the literature. In the most recent study Zhang et al. [49], they used spectroscopic data from LAMOST to include them as candidates for magnetically active stars, due to their intense emission in the Hα_{\alpha} line. The study demonstrated a correlation between the frequency of flare events and magnetic spots and the emission line intensity measured by LAMOST. The Hα_{\alpha} line width values are very similar in both cases: EW(Hα_{\alpha}) =(−0.305±0.017)=(-0.305\pm0.017) nm for TIC 274211446, and EW(Hα_{\alpha}) =(−0.292±0.019)=(-0.292\pm0.019) nm for TIC 16053262, which overlap with the -0.3 nm critical threshold for formal classification as an emission-line star. The ratio between the minimum width and the maximum width on that line is 0.992 for TIC 274211446; it is not available for TIC 16053262. Since the emission is practically constant, they were ruled out as magnetically active stars. Alternatively, the strong emission in the Hα_{\alpha} line can only be explained by the presence of a remanent circumstellar disk around both stars. FGK stars surrounded by a debris disk with ages exceeding one billion years have been reported (Trilling et al. [47]).

The frequency analysis

Both stars were monitored by TESS at short cadence (120 s). TIC 274211446 was observed in sector 31 and TIC 16053262 in sector 7. I downloaded the PDCSAP_FLUX light curves, which are corrected for instrumental effects, available in the Mikulski Archive for Space Telescopes.1 Figure 1 show them. No signals can be seen that would allow us to infer any type of magnetic activity or atmospheric flares in both stars.

Light curves. Top: TIC 274211446. Bottom: TIC 16053262.

Figure 1. Light curves. Top: TIC 274211446. Bottom: TIC 16053262.

I used MultiModes2 for frequency extraction (Pamos Ortega et al. [31]). Figure 2 shows the resulting periodograms. Zooming in on the light curves on day 21 of observation reveals a dominant pulsation with a frequency of approximately 13 d−1^{-1} in TIC 274211446, whereas the dominant pulsation in TIC 16053262 has a frequency of around 16 d−1^{-1}. Both periodograms are similar, with the significant frequency range spanning approximately between 9 and 30 d−1^{-1}. This indicates a predominance of pp-modes, albeit within a range where they may overlap with low order gg- and mixed-modes. Despite being late-MS stars located near the red edge of the instability strip, I have analysed them as pure δ\delta Sct stars, given that there is no presence of significant modes below 5 d−1^{-1}, the boundary that separates them from γ\gamma Dor stars (Grigahcène et al. [19]).

Light curves and periodograms for TIC 274211446 and TIC 16053262

Figure 2. Light curves and periodograms. Left: TIC 274211446. Right: TIC 16053262.

The upper panel of Figure 3 shows that TIC 274211446 has a periodogram nearly four times denser than that of TIC 16053262. Even although TIC 274211446 is more distant and is located in a more obscured environment than TIC 16053262, as confirmed by the higher reddening value, E(B-V), of the former. The lower panel of Figure 3 displays a plateau in the residual periodograms obtained upon completing the extraction of significant frequencies with signal to noise ratio of SNR >4.0>4.0. This plateau is more pronounced in the star with the higher frequency density, as predicted by Barceló Forteza et al. [4]. Furthermore, as explained by these authors, this feature serves as an indicator of the evolutionary stage of the star within the MS.

Compared frequency density in both periodograms and comparison of residuals

Figure 3. Top panel: Compared frequency density in both periodograms. Bottom panel: Comparison of the residuals following the extraction of significant frequencies (SNR > 4).

The Table A.1 shows the frequencies, amplitudes, phases, and SNRs of the first 35 extracted frequencies with MultiModes. Prior to the periodicity analysis, it was necessary to identify those modes that may be linear combinations of the highest-amplitude modes, adopting the Rayleigh resolution, approximately 0.041 d−10.041\ \mathrm{d}^{-1}, as the tolerance threshold. Furthermore, for TIC 274211446, two peaks emerge that are suspected to be aliases of one of the highest-amplitude frequencies at 11.839 d−111.839\ \mathrm{d}^{-1}. One of them, peak f26f_{26} at 10.839 d−110.839\ \mathrm{d}^{-1}, shows an exact shift of 1 d−11\ \mathrm{d}^{-1}. The other, peak f7f_{7} at 17.848 d−117.848\ \mathrm{d}^{-1}, displays a shift of ≈6 c/d\approx6\ \mathrm{c/d} and shares the same phase value. Despite its considerable amplitude, ranking seventh in the list, I excluded it from the spectra periodicity analysis. Additionally, I discarded the f27f_{27} peak that could represent a modulation of a neighbouring higher-amplitude peak, a common feature in these stars according to Bowman et al. [7]. Regarding TIC 16053262, two modes of low amplitudes emerge with frequencies below 5 d−15\ \mathrm{d}^{-1}. These two low amplitude modes, f26f_{26} and f29f_{29}, were also excluded from the periodicity analysis. Finally, out of the first 35 extracted frequencies, 24 passed the selection criteria for TIC 274211446 and 29 for TIC 16053262.

The Figure 4 shows the analysis of the periodicities of the frequencies. I used the Fourier transform (FT), the autocorrelation (AC) and the histogram of frequency differences (HFD), following the techniques of García Hernández et al. [15]; Ramón-Ballesta et al. [37]. A prominent peak emerges around 2.7 d−12.7\ \mathrm{d}^{-1} in the FT of TIC 274211446. It may be the large frequency separation, a value consistent with the star’s evolutionary stage. It is not so obvious in the case of TIC 116053262. The FT shows two peaks, one around 2.2 d−12.2\ \mathrm{d}^{-1} and the other around twice this value. None of these values are consistent with the star’s evolutionary stage. Considering that it should be placed somewhere between the two peaks, I have estimated a value of 3.25 d−13.25\ \mathrm{d}^{-1} for the second star. The Échelle diagrams (ED) do not help, at first glance, in determining the large frequency separation. Vertical alignments between modes of the same degree and consecutive orders are not readily apparent. However, as explained in Sect. 5, they become relevant when compared against models.

Large separation estimates for TIC 274211446 and TIC 16053262 through the Fourier transform, autocorrelation, and histogram of frequency differences

Figure 4. Large separation estimate through the Fourier transform(FT), the autocorrelation (AC) and the histogram of frequency differences (HFD). Left: TIC 274211446. Right: TIC 16053262.

The grid of models and the methodology

I used the MESA code, r-26.04.1 (Paxton et al. [35, 36]; Paxton [34]), to compute evolutionary models up to the TAMS. The Table 2 shows the main grid parameters: masses ranging from 1.501.50 to 2.00 M⊙2.00\ \mathrm{M}_{\odot}, with a step size of 0.020.02.

ParameterRangeStep
MM (M⊙M_{\odot})[1.50; 2.00]0.02 M⊙M_{\odot}
ZZ[0.008; 0.024]0.004
Ω/Ωc\Omega/\Omega_{\mathrm{c}}[0.10; 0.20]0.05
α\alpha1.8Fixed

Table 2. Parameters of the stellar models grid computed with the MESA code.

M⊙M_{\odot}. The metallicity, ZZ, ranges between 0.008 and 0.024, with a step size of 0.004. I have fixed the mixing-length parameter at α=1.8\alpha= 1.8 [10]. Overshooting was also included, as it is crucial at this evolutionary stage and key to determining the true age of the star, whose MS lifetime is extended by the injection of fresh hydrogen into the core. I have set values of f=0.015f = 0.015 and f0=0.005f_{0} = 0.005 for the core, and f=0.002f = 0.002 and f0=0.001f_{0} = 0.001 for the outer envelope [9]. As in my previous work [30], where I computed models of δ\delta Sct stars near the red edge of the instability strip, I have opted to retain the default boundary conditions for the Eddington atmosphere model. Rotation was activated at the ZAMS, with initial angular velocity values relative to the critical velocity ranging from 0.10 to 0.20, with a step size of 0.05. This allowed the stars to achieve equatorial rotation velocities between approximately 0 and 70 km s−1^{-1}. MESA uses the shellular treatment of rotation [26], which does not take into account the centrifugal deformation of the star. Although this 1D treatment breaks down at very high rotation rates, it is enough for the rotation rates discussed here. The transport of angular momentum is implemented via a diffusion approximation [36]. Regarding mesh resolution, I adopted the default spatial resolution value (mesh_delta_coeff = 1.0). However, I forced the code to implement finer time steps, particularly as it approaches the TAMS, by setting time_delta_coeff to 0.5. Consequently, changes in effective temperature and surface gravity are reduced, providing GYRE with a larger number of models at this stage for more precise frequency calculations.

Regarding the oscillation models, I used GYRE 9.1.1 [46]. Considering the frequency range in both stars, I computed only pp-modes with radial orders ranging from n=1n = 1 to n=8n = 8, spherical degrees ℓ=0\ell= 0, 1, 2, and azimuthal orders between m=−2m = -2 and m=+2m = +2. Within the &OSC namelist, I have adopted the standard Christensen-Dalsgaard outer boundary condition, outer_bound = JCD and variables_bound = JCD. I used the traditional approximation of rotation (TAR). For pp-modes, the traditional perturbation methods have been shown to be reliable up to about Ω/Ωc=0.4\Omega/\Omega_{\mathrm{c}} = 0.4 [24].

The numerical scheme for solving the equations was MAGNUS_GL2, and a linear frequency scan (grid_type = LINEAR) was employed. The frequency search window was restricted to values between 9 and 40 d−1\mathrm{d}^{-1}, with a scan resolution of 200 points. Regarding the mesh resolution parameters in the &grid namelist, I used the following values: wosc=10w_{\mathrm{osc}} = 10, wexp=2w_{\mathrm{exp}} = 2, wctr=10w_{\mathrm{ctr}} = 10 and nitermax=5n_{\mathrm{itermax}} = 5.

Once the model frequencies were computed, I calculated the large frequency separation of low orders (Eq. (3)) by averaging the differences between modes of consecutive radial orders across the different degrees and azimuthal orders, νnℓm\nu_{n\ell m}:

Δνnℓm=νnℓm−νn−1ℓm(3)\Delta\nu_{n\ell m} = \nu_{n\ell m} - \nu_{n-1\ell m} \tag*{(3)}

The Figure 5 shows the HR displaying the 28,126 computed MESA models older than 20 Myr and with Xc>0.045X_{\mathrm{c}} > 0.045, avoiding models near the ZAMS and very close to the TAMS. Highlighted in orange are 1,602 models that I selected to calculate the oscillation frequencies with GYRE, with effective temperatures and surface gravities within the uncertainty boxes of both stars (Table 1).

HR diagram of the MESA models showing MESA models and GYRE models

Figure 5. HR of the MESA models. Highlighted in orange is the selection used to compute the oscillation models with GYRE.

After estimating the large frequency separation, with the subsequent support of the ED and the physical consistency of the models (Sect. 5), I used it, along with the spectroscopic temperature, to constrain them. With the selected models, I plotted the histograms and the Kernel Density Estimation curves (KDE) to identify the best possible profile through a chi-squared minimisation process. As previously implemented in [31, 32, 30], I used a weighting factor to give greater weight to models with larger time steps, thereby compensating for the lower number of models produced in each evolutionary track.

Results

As discussed in Sect. 3, estimating the large frequency separation was challenging, given that the EDs did not initially clarify the results obtained from the other three techniques. It remains necessary to provide clearer evidences to robustly support the estimation of the large frequency separation, upon which this study is based. This is only possible once I have the candidate models against which the observations can be compared.

A visual inspection of the main extracted modes (Figure 6 and Figure 7) has also contributed to estimating the large separation and identifying the primary radial modes in both stars. Starting with the 35 extracted highest-amplitude frequencies, I removed linear combinations as well as those suspected of being aliases or amplitude modulations of the primary mode (Sect. 3). The wider gaps between one group of frequencies and the next point to an estimate of the large separation, marked by a dotted grey line running vertically across all the panels in both figures. Furthermore, in order to compare the observations with the models, in the bottom panels I have indicated the low-order radial modes from the best-fitting models using red squares.

Periodogram and extracted frequencies of TIC 274211446

Figure 6. From top to bottom: Periodogram of TIC 274211446. The first 35 extracted frequencies with the highest amplitudes. Frequencies remaining after removing the linear combinations. Frequencies remaining after removing the detected aliases. The red squares represent the positions of the primary radial modes.

Frequency spectra and extracted modes for TIC 16053262

Figure 7. The same as Figure 6 with TIC 16053262.

The Figure 8 illustrates the comparison between the observed ED (upper panels) and the model ED (lower panels), with TIC 274211446 on the left and TIC 16053262 on the right. I have highlighted some of the lowest-order radial modes, connected by a black dotted line. In order to assist in the visualisation and interpretation of the diagrams, for both the observed and model plots, the boundaries between consecutive orders have been represented by a red dotted line. In the model diagrams, the radial modes n=4n = 4 and n=5n = 5 have been connected by a black dashed line. This shift between orders to the opposite side of the diagram is not a physical discontinuity within the star; it is a geometric consequence of the module operator because of the pronounced deviation from the asymptotic regime. Regarding TIC 274211446, there is agreement between the observed and model-predicted pulsation spectra concerning the presence of three modes to the right of the first radial overtone. As the model suggests, these three modes may constitute the ℓ=1\ell=1 triplet of order 1, or at least two of its components, marked as f110f_{11 0} and f11+1f_{11+1}. They are the first (f1f_1) and the third (f3f_3) peaks with highest amplitudes in the periodogram (Table A.1), with a frequency difference of 0.118 d−10.118\ \mathrm{d}^{-1}. Using this value together with the observed radii, R=2.45±0.14 R⊙R = 2.45\pm0.14\ R_{\odot} (Table 1), the result is an equatorial rotation velocity between 21 and 24 km s−1\mathrm{km\ s}^{-1}. The best-fitting model has a radii of R=2.67±0.07 R⊙R = 2.67\pm0.07\ R_{\odot} and an equatorial rotation velocity of around 28±1 km s−128\pm1\ \mathrm{km\ s}^{-1}. The observed ED shows the presence of three modes near the first two radial modes, in particular, the 9.3 c/d mode to the left of the fundamental mode. These do not appear in the model’s power spectra, possibly because they are g- or mixed modes, which were not calculated in the models. The observed ED does not align well with the model’s ED when considering frequencies above the second order. The number of extracted frequencies and subsequently selected is insufficient to achieve a more accurate fit. However, the peak obtained in the FT (Figure 4) is sufficiently pronounced to lend robustness to the estimate of the large separation. Regarding TIC 16053262, there is a better agreement between the positions of the observed first radial modes and those corresponding to the models. The positions of the first overtone and the non-radial modes flanking it are in strong agreement. The peaks marked in the observed ED as f11−1f_{11-1} and f11+1f_{11+1} are 0.941 d−1^{-1} separated. It corresponds with a rotation frequency of 0.471 d−1^{-1}. Its observed radii is R=2.29±0.08 R⊙R = 2.29 \pm0.08\,R_{\odot}, then its equatorial rotation velocity is between 53 and 56 km s−1^{-1}. The best fitting model has a radii of R=2.28±0.05 R⊙R = 2.28 \pm0.05\,R_{\odot} and an equatorial rotation velocity of 47±1347 \pm13 km s−1^{-1}.

Observed and modelled échelle diagrams for TIC 274211446 and TIC 16053262

Figure 8. Top panel: Observed ED. Bottom panel: ED of the best model. Left: TIC 274211446. Right: TIC 16053262.

The left panels of the Figure 9 shows the evolution of the differential rotation as a function of the mass fraction for the best models corresponding to TIC 274211446 (top left) and TIC 16053262 (bottom left). In both cases, the gradient in the rotation frequency between the convective core and the radiative envelope increases as the models evolve on the MS. The right panels shows the differential rotation profile of the respective best fitting models for both stars. The gradient in the rotation frequency is larger in the model correspondig to TIC 16053262, but it never exceeds 17% of the critical velocity. This finding supports the use of the TAR approximation in gyre to compute the oscillation models, allowing me to account solely for the Coriolis force and neglect the centrifugal effects that deform the star, which would otherwise require a non-perturbative treatment of rotation. Although the models are physically consistent, this differential rotation profiles should be treated with caution, as I lack observations to support them. There are no g-modes at low frequencies, nor enough acoustic multiplets in low radial orders, that would allow for probing the deepest interior of either star, as in the studies of Kurtz et al. [23]; Saio et al. [41]. These researchs point to a rigid-body rotation model, with a null rotational gradient between the core and the envelope. The authors suggest a higher efficiency in the transport of angular momentum in the cases of KIC 11145123 and KIC 9244992. In these two cases, the researchers were able to use g- and p-multiplets to estimate the rotation of the core and the envelope, respectively. In this work, I did not have access to the core through the visibility of g-modes; consequently, the models have limitations about the physics of angular momentum transport.

Evolution of differential rotation and rotation-frequency profiles for the optimal models

Figure 9. Left panel: Evolution of the differential rotation for the best models in the late-MS corresponding to the studied stars. Top: TIC 274211446. Bottom: TIC 16053262. Top right: Differential rotation profile of the optimal models as a function of depth. Bottom right: Profile of the ratio between the rotation frequency and the critical rotation frequency of the optimal models as a function of depth.

The Figure 10 displays the period–luminosity relation from Ziaali et al. [50]. The corresponding magnitudes, MVM_{\mathrm{V}}, have been calculated with the Eq (4):

P-L diagram in $\delta$ Sct stars

Figure 10. P-L diagram in δ\delta Sct stars [50]. Black continuous line represents the fundamental one, and the dashed line the first overtone.

MV=V+5log⁡(p1000)+5−AV,(4)M_{\mathrm{V}} = V + 5 \log\left(\frac{p}{1000}\right) + 5 - A_{\mathrm{V}}, \tag*{(4)}

where VV is the apparent magnitude, AV=3.1 E(B−V)A_{\mathrm{V}} = 3.1\,E(B-V) is the reddening and pp is the parallax in mas (Table 1).

In both cases, the extracted highest-amplitude frequency, f1f_1, lies in close proximity to the first overtone. The pulsation constant, QQ, is defined by Eq. (5):

Q=Poscρρ⊙=1f1ρρ⊙(5)Q = P_{\mathrm{osc}} \sqrt{\frac{\rho}{\rho_{\odot}}} = \frac{1}{f_1} \sqrt{\frac{\rho}{\rho_{\odot}}} \tag*{(5)}

where Posc=1/f1P_{\mathrm{osc}}=1/f_1 is the pulsation period, given in days, and ρ\rho is the mean density. Regarding TIC 274211446, f1=13.05f_1 = 13.05 d−1^{-1} (Table A.1) and ρ/ρ⊙=0.11±0.04\rho/\rho_{\odot}= 0.11\pm0.04 (Table 1), then Q=0.025Q=0.025 d, typical of the first radial overtone [43, 12]. Regarding TIC 16053262, f1=16.477f_1 = 16.477 d−1^{-1} and ρ/ρ⊙=0.14±0.04\rho/\rho_{\odot}= 0.14\pm0.04, resulting in Q=0.023Q=0.023 d, a value that deviates from the first radial overtone. Given that it is a Low Amplitude Delta Scuti star (LADS), which may exhibit amplitude modulation over time [7], taking f8f_8 as the first radial overtone instead of f1f_1, the value is Q=0.025Q=0.025 d.

The Figure 11 shows what happens to the frequencies of the pp-modes of the models as they propagate toward the interior of the star. The left panel shows the Brunt-Väisälä and the Lamb frequencies for the spherical degrees ℓ=1\ell=1 and ℓ=2\ell=2, with solid-lines representing TIC 274211446 and dashed-lines representing TIC 16053262. The profile of the gg-mode cavity is very similar in both stars. The overshooting bump at the boundary of the convective core is slightly wider for the model corresponding to TIC 274211446. The most important feature in this diagram is how the ℓ=1\ell=1 Lamb frequency (blue lines) penetrates the gg-mode cavity within the frequency range scanned by gyre. This can be seen more clearly in the expanded view of the top-right panel of the figure. The frequencies ranging 9 to 25 d−1^{-1} are sensitive to the avoided-crossing phenomenon.

Propagation diagrams for the non-radial modes and zoom in the scanning window of GYRE

Figure 11. Left: Propagation diagrams for the non-radial modes. Right: Zoom in the scanning window of GYRE, between 9 y 40 d−1^{-1}.

The normalised inertia of a mode, EnormE_{\mathrm{norm}}, is a measure of its kinetic energy integrated over the entire star, which depends on the density of the medium, ρ(r)\rho(r), the displacement of the mass, ξ(r)\xi(r) and the spherical degree, ℓ\ell. The vast majority of the calculated pp-modes follow a continuous downward trend in inertia as frequency increases (Figure 12). This means

Inertia diagrams for the best models for TIC 274211446 and TIC 16053262

Figure 12. Inertia diagrams as a function of the frequency. The dashed blue line connects the ℓ=1\ell=1, m=0m=0 modes.

that high-order modes, which are excited closer to the surface, and thus in regions of lower density, require less energy to excite. The blue dashed line connects the ℓ=1\ell=1, m=0m=0 modes. The discontinuity observed in certain low-order ℓ=1\ell=1 modes reveals the hybrid nature of these frequencies. They approach the core so closely that they acquire a higher inertia value, as if it were a gg-mode rather than a pp-mode. No further discontinuities of this type appear in the diagram until reaching orders n=5n=5, 6 at spherical degree ℓ=2\ell=2, at frequencies around 30 d−1^{-1}. Returning to the propagation diagram in the Figure 11, it corresponds to values found within the overshooting bump. Unfortunately, there are not enough visible modes in the observed spectra, around the value of 30 c/d, to estimate the core frequency rotation by comparing the results with what these models predict. Furthermore, these modes have the added difficulty of being highly sensitive to the value of the overshooting. Characterizing both stars using these modes would require a grid of models with the overshooting as a free parameter.

Except for these two discontinuities, the models are quite consistent with those of a pure δ\delta Sct star. However, it would be necessary to calculate at least the first low order gg-modes, as well as some mixed modes [18], if the aim is to accurately identify the non-radial acoustic modes. This lies beyond the aims of this work, which is to reliably characterize both stars from a comparative perspective.

Performing the statistical treatment of the models as described in Sect. 4, I obtain the Table 3 with the weighted mean results for each parameter, accompanied by their 1-sigma uncertainties. The Figure 13 displays the histograms and KDE curves for the age, mass, radius, metallicity, surface gravity and luminosity. The position of the optimal model is also represented by a vertical dashed line for each parameter.

TIC IDZZlog⁡(L/L⊙)\log(L/L_{\odot})M(M⊙)M (M_{\odot})R(R⊙)R (R_{\odot})ρˉ (ρˉ⊙)\bar{\rho}\,(\bar{\rho}_{\odot})log⁡g\log gTeffT_{\mathrm{eff}} (K)vv (km s−1^{-1})XcX_{\mathrm{c}}Age (Myr)
2742114460.017 ± 0.0051.26 ± 0.031.84 ± 0.072.67 ± 0.070.14 ± 0.013.85 ± 0.027308 ± 7128 ± 10.21 ± 0.041070 ± 127
160532620.017 ± 0.0051.12 ± 0.031.72 ± 0.072.28 ± 0.050.21 ± 0.013.96 ± 0.027298 ± 5947 ± 130.30 ± 0.051145 ± 165

Table 3. Parameters of the seismically constrained models with their associated 1σ1\sigma uncertainties.

Histograms and KDE curves for age, mass, radius, metallicity, luminosity, and surface gravity

Figure 13. Histograms and KDE curves for age, mass, radius, metallicity, luminosity, and surface gravity. Left: TIC 274211446. Right: TIC 16053262.

Conclusions

This pair of stars represent two rare cases where, despite being so evolved within the MS, they display a frequency spectra clean of gg-modes, with a minimum frequency that lies at the limit where the star can be analysed as a pure δ\delta Sct star. A rotation ranging from slow to moderate makes it possible to characterise them using 1D-models. However, it is necessary to expand the list of calculated modes using gyre to include lower order gg- and mixed-modes, to more clearly identify additional non-radial modes. By this way, it will be possible to narrow the range of uncertainty for the model parameters. The overall frequency analysis reveals that TIC 274211446 is more evolved than TIC 16053262, as its spectrum is slightly shifted towards lower frequencies. The former has a lower value of its core hydrogen fraction, Xc=0.21X_{\mathrm{c}} = 0.21, indicating that its core has already contracted significantly, placing it much closer to the TAMS than TIC 116053262, with Xc=0.30X_{\mathrm{c}} = 0.30. This one is a more compact and older star. Both stars slightly exceed 1 Gyr, reaching the limit where 1D-models can reliably characterise this type of star, according to Pamos Ortega [30].

Determining the projected rotation speed, vsini, and therefore the inclination ii, would help to unravel the visibility geometry of the detected modes. It would explain why there is greater visibility of modes in TIC 274211446 than in TIC 116053262. I think the first star is being observed from an angle between polar and equatorial, which favors the visibility of radial and non-radial modes, both zonal and sectoral, while the second may be seen from an equatorial view, where zonal modes are cancelled due to the north-south symmetry, when integrating the light over the entire disk.

More detailed spectroscopic observations would help determine whether the atmospheres of both stars are homologous or not. In Sect. 2 I pointed out that both stars may be surrounded by a remanent circumstellar disk, which explains the intense emission in the Hα\alpha line, having ruled out a magnetic activity. The disc could produce a photometric veiling effect by equally diluting the amplitudes of all pulse modes, even obscuring those reaching the surface with less amplitude. Geometric cancellation added to this photometric veiling effect could be the reason why the frequency spectrum of TIC 116053262 is four times less dense than that of TIC 274211446. Of course, without forgetting reasons related to the efficiency of the κ\kappa-mechanism in exciting acoustic oscillations.

Data availability

The MESA-GYRE models and the complete tables containing the parameters of the modes extracted for both stars are available upon request.

Acknowledgements

This research has made use of the SIMBAD database, operated at CDS, Strasbourg, France. This paper includes data collected by the TESS mission, which were obtained from the Mikulski Archive for Space Telescopes (MAST) at the Space Telescope Science Institute. STScI is operated by the Association of Universities for Research in Astronomy, Inc., under NASA contract NAS5-03127. Support for MAST for TESS data is provided by the NASA Office of Space Science via grant NNX16AR83G and by other grants from the New York University.

First 35 extracted frequencies of both stars

TIC 274211446TIC 16053262
IDf(d−1)f(\mathrm{d}^{-1})A(mmag)ϕ\phiSNRModeIDf(d−1)f(\mathrm{d}^{-1})A(mmag)ϕ\phiSNRMode
f1f_113.0502.590.42157.80f1f_116.4771.490.65177.46
f2f_29.6011.780.28112.98f100f_{100}f2f_221.0011.340.98171.16
f3f_312.8681.640.14101.51f3f_317.7061.100.27137.21
f4f_411.8391.620.83101.20f4f_418.2320.830.60102.99
f5f_522.2351.090.6667.14f5f_511.7150.720.0991.08f100f_{100}
f6f_624.2411.050.6465.96f600f_{600}f6f_613.9770.841.0090.33
f7f_717.8480.960.8259.726 d−1\mathrm{d}^{-1} alias of f4f_4f7f_722.2430.580.7271.74
f8f_820.9680.880.3355.19f8f_815.1520.530.7366.24f200f_{200}
f9f_99.3120.850.6151.68f9f_916.3490.460.2255.55
f10f_{10}11.0940.750.1447.01f10f_{10}19.7290.370.5346.32
f11f_{11}11.2120.700.3642.73−f1+f6-f_1+f_6f11f_{11}21.0460.380.7345.01
f12f_{12}18.6730.660.7642.57f12f_{12}16.8600.360.8444.78
f13f_{13}15.9270.620.9938.30f13f_{13}10.5260.340.2542.86−f5+f7-f_5+f_7
f14f_{14}15.8230.500.5731.26f14f_{14}29.6090.260.9732.15
f15f_{15}17.3310.500.8531.13f15f_{15}25.3190.210.0926.22
f16f_{16}12.7820.510.2230.48f16f_{16}22.0570.200.7225.14
f17f_{17}18.4420.490.3730.45f400f_{400}f17f_{17}16.7420.180.2121.96
f18f_{18}22.6710.490.4130.33f1+f9f_1+f_9f18f_{18}15.6730.170.4320.27−f6+f14-f_6+f_{14}
f19f_{19}26.8190.470.5028.86f19f_{19}12.1870.160.8119.88
f20f_{20}21.4220.440.2227.85f2+f4f_2+f_4f20f_{20}16.0790.110.7413.88
f21f_{21}13.6380.410.5425.64f21f_{21}26.3120.110.5113.83
f22f_{22}18.1290.410.1125.59f22f_{22}15.8170.110.3713.31
f23f_{23}23.8480.370.8423.13f10+f16f_{10}+f_{16}f23f_{23}25.0210.100.7312.74
f24f_{24}15.4600.340.0322.96f24f_{24}20.0490.090.4011.82
f25f_{25}24.7260.360.5922.50f3+f4f_3+f_4f25f_{25}16.5940.090.5311.68
f26f_{26}10.8390.350.1622.231 d−1\mathrm{d}^{-1} alias of f4f_4f26f_{26}1.4470.090.7910.76
f27f_{27}12.0910.350.8221.95Modulation of f4f_4f27f_{27}18.0240.080.8910.62
f28f_{28}27.9740.340.4721.52f9+f12f_9+f_{12}f28f_{28}18.7660.080.509.96f300f_{300}
f29f_{29}12.5740.350.2121.50f200f_{200}f29f_{29}3.3840.080.529.48
f30f_{30}14.2460.340.3721.20f19−f29f_{19}-f_{29}f30f_{30}20.5510.080.629.34
f31f_{31}15.0680.340.7520.26f31f_{31}13.8650.070.639.08
f32f_{32}15.2550.320.2819.69f300f_{300}f32f_{32}21.2220.070.978.86
f33f_{33}26.3760.300.6418.59f10+f32f_{10}+f_{32}f33f_{33}14.8760.060.677.94
f34f_{34}19.2850.310.4918.38f34f_{34}12.5050.060.147.61
f35f_{35}13.3040.290.1018.00f35f_{35}24.2570.060.307.53

Table A.1. Frequencies, amplitudes, phases, SNRs and some identified modes of the first 35 extracted frequencies of both stars.

References

  1. [1]Aerts, C., Christensen-Dalsgaard, J., & Kurtz, D. W. 2010, Astroseismology (Springer Science+Business Media B.V.)
  2. [2]Baglin, A., Auvergne, M., Barge, P., et al. 2006, in ESA Special Publication, Vol. 1306, The CoRoT Mission Pre-Launch Status - Stellar Seismology and Planet Finding, ed. M. Fridlund, A. Baglin, J. Lochard, & L. Conroy, 33
  3. [3]Balona, L. A., Daszyńska-Daszkiewicz, J., & Pamyatnykh, A. A. 2015, Monthly Notices of Royal Astronomy Society, 452, 3073
  4. [4]Barceló Forteza, S., Pascual-Granado, J., Suárez, J. C., et al. 2024, Monthly Notices of Royal Astronomy Society, 535, 2189
  5. [5]Bedding, T. R., Murphy, S. J., Hey, D. R., et al. 2020, Nature, 581, 147
  6. [7]Bowman, D. M., Kurtz, D. W., Breger, M., Murphy, S. J., & Holdsworth, D. L. 2016, Monthly Notices of Royal Astronomy Society, 460, 1970
  7. [8]Breger, M. 2000, in Astronomical Society of the Pacific Conference Series, Vol. 210, Delta Scuti and Related Stars, ed. M. Breger & M. Montgomery, 3
  8. [9]Claret, A. & Torres, G. 2019, Astrophysical Journal, 876, 134
  9. [10]Dupret, M. A., Grigahcène, A., Garrido, R., Gabriel, M., & Scuflaire, R. 2005, Astronomy & Astrophysics, 435, 927
  10. [11]Dziembowski, W. 1997, in IAU Symposium, Vol. 181, Sounding Solar and Stellar Interiors, ed. J. Provost & F.-X. Schmider, ISBN0792348389
  11. [12]Fitch, W. S. 1981, Astrophysical Journal, 249, 218
  12. [13]Gaia Collaboration, Vallenari, A., Brown, A. G. A., et al. 2023, Astronomy & Astrophysics, 674, A1
  13. [14]García Hernández, A., Martín-Ruiz, S., Monteiro, M. J. P. F. G., et al. 2015, Astrophysical Journal, 811, L29
  14. [15]García Hernández, A., Moya, A., Michel, E., et al. 2009, Astronomy & Astrophysics, 506, 79
  15. [16]García Hernández, A., Moya, A., Michel, E., et al. 2013, Astronomy & Astrophysics, 559, A63
  16. [17]García Hernández, A., Suárez, J. C., Moya, A., et al. 2017, Monthly Notices of Royal Astronomy Society, 471, L140
  17. [18]Gautam, A., Murphy, S. J., & Bedding, T. R. 2025, arXiv e-prints, arXiv:2507.03561DOI
  18. [19]Grigahcène, A., Antoci, V., Balona, L., et al. 2010, Astrophysical Journal, 713, L192
  19. [20]Guo, Z., Bedding, T. R., Pamyatnykh, A. A., et al. 2024, Monthly Notices of Royal Astronomy Society, 535, 2927
  20. [21]Guzik, J. A. 2021, Frontiers in Astronomy and Space Sciences, 8, 55
  21. [22]Holdsworth, D. L., Smalley, B., Gillon, M., et al. 2014, Monthly Notices of Royal Astronomy Society, 439, 2078
  22. [23]Kurtz, D. W., Saio, H., Takata, M., et al. 2014, Monthly Notices of Royal Astronomy Society, 444, 102
  23. [24]Lignières, F., Rieutord, M., & Reese, D. 2006, Astronomy & Astrophysics, 455, 607
  24. [25]Lindegren, L., Klioner, S. A., Hernández, J., et al. 2021, Astronomy & Astrophysics, 649, A2
  25. [26]Meynet, G., Ekstrom, S., Maeder, A., et al. 2013, in Lecture Notes in Physics, Berlin Springer Verlag, ed. M. Goupil, K. Belkacem, C. Neiner, F. Lignières, & J. J. Green, Vol. 865 (Lecture Notes in Physics, Volume 865. ISBN 978-3-642-33379-8. Springer-Verlag Berlin Heidelberg), 3
  26. [27]Michel, E., Dupret, M.-A., Reese, D., et al. 2017, in European Physical Journal Web of Conferences, Vol. 160, European Physical Journal Web of Conferences (EDP), 03001
  27. [29]Nowakowski, R. M. & Dziembowski, W. A. 2003, Astrophysics and Space Science, 284, 273
  28. [30]Pamos Ortega, D. 2025, Astronomy & Astrophysics, 703, A31
  29. [31]Pamos Ortega, D., García Hernández, A., Suárez, J. C., et al. 2022, Monthly Notices of Royal Astronomy Society, 513, 374
  30. [32]Pamos Ortega, D., Mirouh, G. M., García Hernández, A., Suárez Yanes, J. C., & Barceló Forteza, S. 2023, Astronomy & Astrophysics, 675, A167
  31. [33]Paparó, M., Benkő, J. M., Hareter, M., & Guzik, J. A. 2016, Astrophysical Journal, 822, 100
  32. [34]Paxton, B. 2019, Modules for Experiments in Stellar Astrophysics (MESA)
  33. [35]Paxton, B., Bildsten, L., Dotter, A., et al. 2011, Astrophysical Journals, 192, 3
  34. [36]Paxton, B., Cantiello, M., Arras, P., et al. 2013, Astrophysical Journals, 208, 4
  35. [37]Ramón-Ballesta, A., García Hernández, A., Suárez, J. C., et al. 2021, Monthly Notices of Royal Astronomy Society, 505, 6217
  36. [38]Ricker, G. R., Winn, J. N., Vanderspek, R., et al. 2014, in Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, Vol. 9143, Proceedings of the SPIE, 914320
  37. [39]Rodríguez, E. & Breger, M. 2001, Astronomy & Astrophysics, 366, 178
  38. [40]Royer, F., Zorec, J., & Gómez, A. E. 2007, Astronomy & Astrophysics, 463, 671
  39. [41]Saio, H., Kurtz, D. W., Takata, M., et al. 2015, Monthly Notices of Royal Astronomy Society, 447, 3264
  40. [42]Stassun, K. G. 2019, VizieR Online Data Catalog, IV/38
  41. [43]Stellingwerf, R. F. 1979, Astrophysical Journal, 227, 935
  42. [44]Suárez, J. C., García Hernández, A., Moya, A., et al. 2014, Astronomy & Astrophysics, 563, A7
  43. [45]Tassoul, M. 1980, Astrophysical Journals, 43, 469
  44. [46]Townsend, R. H. D. & Teitler, S. A. 2013, Monthly Notices of Royal Astronomy Society, 435, 3406
  45. [47]Trilling, D. E., Bryden, G., Beichman, C. A., et al. 2008, Astrophysical Journal, 674, 1086
  46. [48]Uytterhoeven, K., Moya, A., Grigahcène, A., et al. 2011, Astronomy & Astrophysics, 534, A125
  47. [49]Zhang, L.-y., Su, T., Misra, P., et al. 2023, Astrophysical Journals, 264, 17
  48. [50]Ziaali, E., Bedding, T. R., Murphy, S. J., Van Reeth, T., & Hey, D. R. 2019, Monthly Notices of Royal Astronomy Society, 486, 4348

Paper details

vlnc:W_ezg7kc3a
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 5

downloads · 0

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

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents