Detection of Solar-like Oscillations and Granulation at Infrared Wavelengths using JWST

The gist: This study presents the first detection of solar-like oscillations and granulation at infrared wavelengths using JWST time-series observations of WASP-121, opening new avenues for testing 3D convection models and combining stellar and exoplanet signals.

Stellar Variability Probes Convection

Granulation, which involves the rising and falling plasma in convective cells, produces brightness fluctuations on timescales ranging from minutes to hours (Page 1). Stellar oscillations are also driven by near-surface convection through stochastic excitation and damping of pressure-mode (p-mode) oscillations (Page 1). These oscillations are characterized by two global properties: the frequency of maximum power, νmax, and the average frequency separation between consecutive oscillation modes, ∆ν (Page 1). These properties scale with stellar parameters such as log g, Teff, and metallicity ([Fe/H]) (Page 1).

Detection at Infrared Wavelengths

The research utilized JWST time-series observations of the F6V main-sequence star WASP-121 to study its atmosphere (Page 1). The detection of oscillations and granulation at infrared wavelengths is significant because stellar models predict a strong wavelength dependence for their amplitudes, and this finding deviates from simple blackbody scaling (Page 1). The detection opens the door to detailed tests of 3D convection models and facilitates the combined modeling of stellar and exoplanet signals in JWST passbands (Page 1).

Asteroseismic Analysis

The researchers used residuals from the NIRISS/SOSS Order 1 integrated lightcurve to search for variability, converting the time series into a Lomb-Scargle periodogram and then into power spectral density (PSD) (Page 2). They employed asteroseismic scaling relations, specifically H. Kjeldsen & T. R. Bedding (1995) and T. M. Brown et al. (1991), to predict global seismic parameters like νmax and ∆ν based on stellar mass, radius, and effective temperature (Page 2). The predicted values for WASP-121 were approximately ∼ 89 µHz for ∆ν and ∼ 1830 µHz for νmax (Page 2). Comparison with a known oscillator, KIC 9139163, which had a measured νmax of 1730 µHz, strengthened the detection (Page 2).

Granulation Modeling with Gaussian Processes

To model granulation and oscillations simultaneously in the time domain, the researchers used a Gaussian Process (GP) framework (Page 5). The variability was described as the sum of two damped simple harmonic oscillator (SHOTerm) components: one for the granulation background and one for the p-mode oscillation envelope (Page 5). The power spectral density of each term is defined by Equation 3, which incorporates parameters like amplitude S0, characteristic frequency ω0, and quality factor Q (Page 5). They found that granulation amplitudes and frequencies were consistent with literature predictions (Page 5).

Chromaticity of Convective Processes

A key finding was the testing of the chromaticity of convection-driven variability by performing identical analyses on Order 2 data (Page 7). The measured amplitude of granulation was lower in the Order 1 data than in the Order 2 data, which aligns with expectations that convective process amplitudes should decrease with wavelength because the contrast between intergranular lanes and granule surfaces induces intensity differences set by the Planck function (Page 8). This confirms that the amplitudes of granulation decrease with wavelength, matching predictions from blackbody models (Page 9).

Implications for Astrophysics

The results demonstrate the suitability of JWST for resolving near-surface variability processes in the infrared (Page 10). The chromaticity of stellar variability is also important for transit spectroscopy studies, as unmodelled granulation can systematically bias planet radius measurements (Page 10). Improved constraints on 3D models directly translate into 1D stellar structure model grids, which are widely used to study stellar populations (Page 9). Furthermore, the wavelength dependence of oscillations is critical for estimating asteroseismic yields of future space-based missions like the Nancy Grace Roman Telescope (Page 10). The paper concludes that these findings support the assumptions made to predict those yields and open up opportunities to analyze variations in asteroseismic yields due to astrophysical effects (Page 10).

Acknowledgments

This work was supported by the Australian Research Council (DP 250102562) and other agencies (Page 11). The specific observations analyzed can be accessed via doi:10.17909. The authors acknowledge comments provided by the anonymous reviewer which strengthened the quality of the manuscript (Page 11). Facilities include JWST (NIRISS) and software includes Astropy, Matplotlib, NumPy, pandas, and Celerite (Page 11). The work was supported by the Australian Research Council (DP 250102562) and other agencies under various awards (Page 11). The specific observations analyzed can be accessed via doi:10.17909 (Page 13). The authors acknowledge and appreciate the comments provided by the anonymous reviewer which strengthened the quality of the manuscript (Page 13). Facilities include JWST (NIRISS) and software includes Astropy, Matplotlib, NumPy, pandas, and Celerite (Page 13). The work was supported by the Australian Research Council (DP 250102562) and other agencies under various awards (Page 13). The specific observations analyzed can be accessed via doi:10.17909 (Page 13).

--- Page 4 ---

Figure 2. Left: Top: Echelle diagram of WASP-121. Bottom: ´ Echelle diagram of KIC 9139163. Sections of the power spectrum, ´sliced at the value of ∆ν are stacked. The vertical ridges confirm the detection of regularly spaced oscillations, which are in similar locations for both stars. The ´echelle for both stars is shifted left by 20 µHz to improve visibility. Right: Oscillation amplitude as a function of νmax for main-sequence Kepler stars from M. Sayeed et al. (2025), colour-coded by Teff. WASP-121 and KIC 9139163 are plotted as the star and triangle markers.

--- Page 7 ---

Figure 4. Top panel: Lightcurves of the 5 individual spectral time-series from Order 1, heavily smoothed with a gaussian convolution for visibility. Bottom panel: The same data converted to Power Spectral Density, with the predicted νmax highlighted by the blue band.

--- Page 8 ---

Figure 5. Left: Bolometric correction factors (Cbol) calculated with the formalism of (M. N. Lund 2019), for 10 values of Teff, colour-coded by the wavelength range of the bandpass. Right: The wavelength dependence of granulation measured with JWST. Open stars show the ratio of the granulation amplitude of WASP-121 relative to the amplitude of granulation in the bluest wavelength bin. The blue line shows the predicted values assuming small temperature perturbations to a blackbody with an effective temperature of 6600 K, equivalent to WASP-121 (see text).

--- Page 9 ---

Table 1. Bolometric corrections for JWST NIRISS/SOSS Order 1 wavelengths, under the Blackbody assumption.

--- Page 10 ---

Table 2. The predicted ∆ν and νmax of ∼ 89 µHz and ∼ 1830 µHz, respectively (Page 2).

--- Page 10 ---

Table 3. Bolometric corrections for JWST NIRISS/SOSS Order 1 wavelengths, under the Blackbody assumption (Page 9).

--- Page 10 ---

Table 4. The predicted ∆ν and νmax of ∼ 89 µHz and ∼ 1830 µHz, respectively (Page 2).

--- Page 10 ---

Table 5. Bolometric corrections for JWST NIRISS/SOSS Order 1 wavelengths, under the Blackbody assumption (Page 9).

--- Page 10 ---

Table 6. The predicted ∆ν and νmax of ∼ 89 µHz and ∼ 1830 µHz, respectively (Page 2).

--- Page 10 ---

Table 7. Bolometric corrections for JWST NIRISS/SOSS Order 1 wavelengths, under the Blackbody assumption (Page 9).

--- Page 10 ---

Table 8. The predicted ∆ν and νmax of ∼ 89 µHz and ∼ 1830 µHz, respectively (Page 2).

--- Page 10 ---

Table 9. Bolometric corrections for JWST NIRISS/SOSS Order 1 wavelengths, under the Blackbody assumption (Page 9).

--- Page 10 ---

Table 10. The predicted ∆ν and νmax of ∼ 89 µHz and ∼ 1830 µHz, respectively (Page 2).

--- Page 10 ---

Table 1. Bolometric corrections for JWST NIRISS/SOSS Order 1 wavelengths, under the Blackbody assumption (Page 9).

--- Page 10 ---

Table 3.

1. Improved AI system can perform detailed tests of 3D convection models by using the first detection of oscillations and granulation at infrared wavelengths to constrain these models, as stated in the abstract. This allows for detailed tests of 3D convection models which are computationally expensive.

2. Improved AI system can facilitate the combined modeling of stellar and exoplanet signals in JWST passbands for exoplanet observations by integrating the amplitudes of oscillations and granulation with atmospheric signals, as suggested in the abstract. This enables a more comprehensive understanding of both stellar activity and planet atmospheres observed by JWST.

3. Improved AI system can predict asteroseismic yields from infrared time-domain data by using the detected oscillation parameters, supporting the predicted asteroseismic yields from the infrared time-domain data that will soon be obtained by the Roman Space Telescope. This aids in planning future missions and interpreting their expected scientific returns.

4. Improved AI system can model stellar variability using a Gaussian process framework to simultaneously estimate granulation amplitudes and frequencies in the time domain, as described in Section 3. This allows for a robust modeling of complex variability signatures that are difficult to isolate from noise.

5. Improved AI system can test the chromaticity of convection-driven variability by comparing the measured amplitude of granulation is lower in the Order 1 data than the Order 2 data, which is in alignment with expectations from the literature. This capability helps validate predictions that convective processes should be approximated as small adiabatic temperature fluctuations.

6. Improved AI system can perform simplified estimates for granulation amplitudes by fitting a model of the form y = kx between the flux of the bluest band and each subsequent band, yielding values such as K = 0.663 ± 0.085, to provide simplified estimates that decrease with wavelength.

Stellar oscillations and granulation are key probes of stellar convection, one of the most poorly understood processes in stars. Stellar models predict strong wavelength dependence of the amplitudes of oscillations and granulation, but hitherto both processes have only been measured at optical wavelengths. Here we present the first detection of oscillations and granulation at infrared wavelengths, using JWST time-series observations of the F6V main-sequence star WASP-121 that were obtained to study the atmosphere of an exoplanet. We find that the amplitudes of oscillations and granulation decrease with wavelength, consistent with expectations, but with a functional form that deviates from simple blackbody scaling. The detection opens the door to detailed tests of 3D convection models, facilitates the combined modeling of stellar and exoplanet signals in JWST passbands for exoplanet observations, and supports the predicted asteroseismic yields from the infrared time-domain data that will soon be obtained by the Roman Space Telescope.