Abstract
Measuring the abundances of carbon and oxygen in exoplanet atmospheres is considered a crucial avenue for unlocking the formation and evolution of exoplanetary systems^{1,2}. Access to the chemical inventory of an exoplanet requires highprecision observations, often inferred from individual molecular detections with lowresolution spacebased^{3,4,5} and highresolution groundbased^{6,7,8} facilities. Here we report the mediumresolution (R ≈ 600) transmission spectrum of an exoplanet atmosphere between 3 and 5 μm covering several absorption features for the Saturnmass exoplanet WASP39b (ref. ^{9}), obtained with the Near Infrared Spectrograph (NIRSpec) G395H grating of JWST. Our observations achieve 1.46 times photon precision, providing an average transit depth uncertainty of 221 ppm per spectroscopic bin, and present minimal impacts from systematic effects. We detect significant absorption from CO_{2} (28.5σ) and H_{2}O (21.5σ), and identify SO_{2} as the source of absorption at 4.1 μm (4.8σ). Bestfit atmospheric models range between 3 and 10 times solar metallicity, with subsolar to solar C/O ratios. These results, including the detection of SO_{2}, underscore the importance of characterizing the chemistry in exoplanet atmospheres and showcase NIRSpec G395H as an excellent mode for timeseries observations over this critical wavelength range^{10}.
Similar content being viewed by others
Main
We obtained a singletransit observation of WASP39b using the NIRSpec^{11,12} G395H grating on 30–31 July 2022 between 21:45 and 06:21 UTC using the Bright Object Time Series mode. WASP39b is a hot (T_{eq} = 1,120 K), lowdensity giant planet with an extended atmosphere. Previous spectroscopic observations have shown prominent atmospheric absorption by Na, K and H_{2}O (refs. ^{3,4,13,14,15}), with tentative evidence of CO_{2} from infrared photometry^{4}. Atmospheric models fitted to the spectrum have inferred metallicities (amount of heavy elements relative to the host star) from 0.003 to 300 times solar^{3,15,16,17,18,19,20}, which makes it difficult to ascertain the formation pathway of the planet^{21,22}. The host, WASP39, is a G8type star that shows little photometric variability^{23} and has nearly solar elemental abundance patterns^{24}. The quiet host and extended planetary atmosphere make WASP39b an ideal exoplanet for transmission spectroscopy^{25}. The transmission spectrum of WASP39b was observed as part of the JWST Transiting Exoplanet Community Director’s Discretionary Early Release Science (JTEC ERS) Program^{26,27} (ERS1366; principal investigators Natalie M. Batalha, Jacob L. Bean and Kevin B. Stevenson), which uses four instrument configurations to test their capabilities and provide lessons learned for the community.
The NIRSpec G395H data were recorded with the 1.6″ × 1.6″ fixed slit aperture using the SUB2048 subarray and NRSRAPID readout pattern, with spectra dispersed across both the NRS1 and NRS2 detectors. Over the roughly 8h duration of the observation, a total of 465 integrations were taken, centred around the 2.8h transit. We obtained 70 groups per integration, resulting in an effective integration time of 63.14 s. During the observation, the telescope experienced a ‘tilt event’, a spontaneous and abrupt change in the position of one or more mirror segments, causing changes in the point spread function (PSF) and hence jumps in flux^{28}. The tilt event occurred midtransit, affecting approximately three integrations and resulted in a noticeable step in the flux time series, the size of which is dependent on wavelength (Fig. 1 and Methods). The tilt event also affects the PSF, with the full width at half maximum (FWHM) of the spectral trace showing a stepfunctionlike shape (see Extended Data Figs. 2 and 3).
We produced several reductions of the observations using independent analysis pipelines (see Methods). For each reduction, we created broadband and spectroscopic light curves in the ranges 2.725–3.716 μm for NRS1 and 3.829–5.172 μm for NRS2 using 10pixelwide bins (≈0.007 μm, median resolution R ≈ 600), excluding the detector gap between 3.717–3.823 μm. The light curves show a settling ramp during the first ten integrations (≈631.4 s), with a linear slope across the entire observation for NRS1. We otherwise see no substantial systematic trends and achieve improvements in precision from raw uncorrected to fitted broadband light curves of 1.63 to 1.03 times photon noise for NRS1 and 1.95 to 1.31 times for NRS2. The flux jump caused by the mirrortilt event could be corrected by detrending against the spectral trace x and y positional shifts, normalizing the light curves or fitting the light curves with a step function (see Methods). We produced several fits from each set of light curves, resulting in a total of 11 independently measured transmission spectra. Figure 1 demonstrates that our spectroscopic light curves achieve precisions close to photon noise, with a median precision of 1.46 times photon noise across the full wavelength range (see Extended Data Fig. 4).
We show transmission spectra from several combinations of independent reductions and lightcurvefitting routines in Fig. 2, along with the weighted average of all 11 transmission spectra with the unweighted mean uncertainty produced by our analyses (see Methods). We find that using different combinations of reduction and fitting methods results in consistent transmission spectra (see Methods and Extended Data Fig. 5). Although we see some artefacts at the edges of the detectors (see Fig. 3, bottom panel) that may be caused by uncharacterized systematics, these only affect a small number of wavelength bins. Our resulting averaged NIRSpec G395H spectrum shows increased absorption towards bluer wavelengths short of 3.7 μm and a prominent absorption feature between 4.2 and 4.5 μm, along with a smalleramplitude absorption feature at 4.1 μm and a narrow feature around 4.56 μm.
We compared the weightedaverage G395H transmission spectrum to three grids of 1D radiative–convective–thermochemical equilibrium (RCTE) atmosphere models of WASP39b (see Methods and Extended Data Table 2), containing a total of 10,308 model spectra. The bestfit models from each grid provide a reduced chisquare per data point (χ^{2}/N) of 1.08–1.20 for our 344datapoint transmission spectrum (Fig. 3). The increased absorption at blue wavelengths across NRS1 is consistent with absorption from H_{2}O (at 21.5σ; see Methods), whereas the large bump in absorption between 4.2 and 4.5 μm (ref. ^{29}) can be attributed to CO_{2} (28.5σ). H_{2}O and CO_{2} are expected atmospheric constituents for nearsolar atmospheric metallicities, with the CO_{2} abundance increasing nonlinearly with higher metallicity^{30}. The spectral feature at 4.56 μm (3.3σ) is unidentified at present but does not correlate with any obvious detector artefacts and is reproduced by several independent analyses. The absorption feature at 4.1 μm is also not seen in the RCTE model grids. After an exhaustive search for possible opacity sources (S.M. Tsai et al., manuscript in preparation), described in the corresponding NIRSpec PRISM analysis^{31}, we interpret this feature as SO_{2} (4.8σ), as it is the best candidate at this wavelength.
Although SO_{2} would have volume mixing ratios (VMRs) of less than 10^{−10} throughout most of the observable atmosphere in thermochemical equilibrium, coupled photochemistry of H_{2}S and H_{2}O can produce SO_{2} on giant exoplanets, with the resulting SO_{2} mixing ratio expected to increase with increasing atmospheric metallicity^{32,33,34}. We find that a VMR of approximately 10^{−6} of SO_{2} is required to fit the spectral feature at 4.1 μm in the transmission spectrum of WASP39b, consistent with lowerresolution NIRSpec PRISM observations of this planet^{31} and previous photochemical modelling of supersolar metallicity giant exoplanets^{34,35}. Figure 4 shows a breakdown of the contributing opacity sources for the lowest χ^{2}/N bestfit model (PICASO 3.0) with VMR = 10^{−5.6} injected SO_{2}. The inclusion of SO_{2} in the models results in an improved χ^{2}/N and is detected at 4.8σ (see Methods), confirming its presence in the atmosphere of WASP39b.
We also look for evidence of CH_{4}, CO, H_{2}S and OCS (carbonyl sulfide) because their nearsolar chemical equilibrium abundances could result in a contribution to the spectrum. We see no evidence of CH_{4} in our spectrum between 3.0 and 3.6 μm (ref. ^{23}), which is indicative of C/O < 1 (ref. ^{36}) and/or photochemical destruction^{34,37}. With regards to CO, H_{2}S and OCS, we were unable to conclusively confirm their presence with these data. In particular, CO, H_{2}O, OCS and our modelled cloud deck all have overlapping opacity, which creates a pseudocontinuum from 4.6 to 5.1 μm (see Figs. 3 and 4). Therefore, we were unable to unambiguously identify the individual contributions from CO and other molecules over this wavelength region at the resolution presented in this work.
Our models show an atmosphere enriched in heavy elements, with bestfit parameters ranging from 3 to 10 times solar metallicity, given the spacing of individual model grids (see Methods). The spectra also indicate C/O ratios ranging from subsolar to solar depending on the grid used, informed by the relative strength of absorption from carbonbearing molecules to oxygenbearing molecules. The interpretation of the relatively high resolution and precision of the G395H spectrum seems to be sensitive to the treatment of aerosols in the model, with one grid preferring 3 times solar metallicity when using a wavelengthdependent cloud opacity and physically motivated vertical cloud distribution^{38} but 10 times solar metallicity when assuming a grey cloud. In general, forward model grids fit the main features of the data but do not place statistically significant constraints on many of the atmospheric parameters (see Methods). Future interpretation of the JTEC ERS WASP39b data with Bayesian retrieval analyses will provide robust confidence intervals for these planetary properties and explore the degree to which these data are sensitive to modelling assumptions (for example, chemical equilibrium versus disequilibrium) and parameter degeneracies (for example, clouds versus atmospheric metallicity).
We are able to strongly rule out an absolute C/O ≥ 1 scenario (χ^{2}/N ≥ 3.97), which has previously been proposed for gasdominated accretion at wide orbital radii beyond the CO_{2} ice line at which the gas may be carbonrich^{39}. Our C/O upper limit, therefore, suggests that WASP39b may have either formed at smaller orbital radii with gasdominated accretion or that the accretion of solids enriched the atmosphere of WASP39b with oxygenbearing species^{2}. The level of metal enrichment (3–10 times solar) is consistent with similar measurements of Jupiter and Saturn^{40,41}, potentially suggesting coreaccretion formation scenarios^{42}, and is consistent with upper limits from interiorstructure modelling^{43}. These NIRSpec G395H transmission spectroscopy observations demonstrate the promise of robustly characterizing the atmospheric properties of exoplanets with JWST unburdened by substantial instrumental systematics, unravelling the nature and origins of exoplanetary systems.
Methods
Data reduction
We produced several analyses of stellar spectra from the Stage 1 2D spectral images produced using the default STScI JWST Calibration Pipeline^{44} (‘rateints’ files) and by means of customized runs of the STScI JWST Calibration Pipeline with userdefined inputs and processes for steps such as the ‘jump detection’ and ‘bias subtraction’ steps.
Each pipeline starts with the raw ‘uncal’ 2D images that contain grouplevel products. As we noticed that the default superbias images were of poor quality, we produced two customized runs of the JWST Calibration Pipeline, using either the default bias step or a customized version. The customized step built a pseudobias image by computing the median pixel value in the first group across all integrations and then subtracted the new bias image from all groups. We note that the poor quality of the default superbias images affects NRS1 more notably than NRS2, and this method could be revised once a better superbias is available.
Before ramp fitting, both our standard and custom bias step runs of the edited JWST Calibration Pipeline ‘destriped’ the grouplevel images to remove socalled ‘1/f noise’ (correlated noise arising from the electronics of the readout pattern, which appears as column striping in the subarray images^{11,12}). Our grouplevel destriping step used a mask of the trace 15σ from the dispersion axis for all groups within an integration, ensuring that a consistent set of pixels is masked within a ramp. The median values of nonmasked pixels in each column were then computed and subtracted for each group.
The results of our customized runs of the JWST Calibration Pipeline are a set of custom grouplevel destriped products and custom biassubtracted grouplevel destriped products. In both cases, the rampjump detection threshold of the JWST Calibration Pipeline was set to 15σ (as opposed to the default of 4σ), as it produced the most consistent results at the integration level. In both custom runs of the JWST Calibration Pipeline, all other steps and inputs were left at the default values.
For all analyses, wavelength maps from the JWST Calibration Pipeline were used to produce wavelength solutions, verified against stellar absorption lines, for both detectors. The midintegration times in BJD_{TDB} were extracted from the image headers for use in producing light curves. None of our datareduction pipelines performed a flatfield correction, as the available flat fields were of poor quality and unexpectedly removed portions of the spectral trace. In general, we found that 1/f noise can be corrected at either the group or integration levels to similar effect; however, correction at the group level with a repeated columnbycolumn cleaning step at the integration level probably results in cleaner 1D stellar spectra. This was particularly true for NRS2, owing to the limited number of columns in which the unilluminated region on the detector extends both above and below the spectral trace, as shown in Extended Data Fig. 1.
Below we detail each of the independent datareduction pipelines used to produce the time series of stellar spectra from our G395H observations.
ExoTiCJEDI pipeline
We used the Exoplanet Timeseries Characterisation  JWST Extraction and Diagnostics Investigator (ExoTiCJEDI^{45}) pipeline on our custom grouplevel destriped products, treating each detector separately. Using the dataquality flags produced by the JWST Calibration Pipeline, we replaced any pixels identified as bad, saturated, dead, hot, low quantum efficiency or no gain value with the median value of surrounding pixels. We also searched each integration for pixels that were spatial outliers from the median of the surrounding 20 pixels in the same row by 6σ (to remove permanently affected ‘bad’ pixels) or outliers from the median of that pixel in the surrounding ten integrations in time by 20σ (to identify highenergy shortterm effects such as cosmic rays) and replaced the outliers with the median values. To obtain the trace position and FWHM, we fitted a Gaussian to each column of an integration, finding a median standard deviation of 0.7 pixels. A fourthorder polynomial was fitted through the trace centres and the widths, which were smoothed with a median filter, to obtain a simple aperture region. This region extended 5 times the FWHM of the spectral trace, above and below the centre, corresponding to a median aperture width of 7 pixels. To remove any remaining 1/f and background noise from each integration, we subtracted the median of the unilluminated region in each column by masking all pixels that were 5 pixels away from the aperture. For each integration, the counts in each row and column of the aperture region were summed using an intrapixel extraction, taking the relevant fractional flux of the pixels at the edge of the aperture and crosscorrelated to produce xpixel and ypixel shifts for detrending (see Extended Data Fig. 2). On average, the xpixel shift represents movement of 1 × 10^{−4} and 8 × 10^{−6} of a pixel for NRS1 and NRS2, respectively. The aperture column sums resulted in 1D stellar spectra with errors calculated from photon noise after converting from data numbers using the gain factor. This reduction is denoted hereafter as ExoTiCJEDI [V1].
We produced further 1D stellar spectra from both the custom grouplevel destriped product and custom biassubtracted grouplevel destriped products using the ExoTiCJEDI pipeline as described above, but with further cleaning by repeating the spatial outliers step. The reduction with further cleaning using the custom grouplevel destriped products is hence denoted as ExoTiCJEDI [V2] and the reduction with further cleaning using the custom biassubtracted grouplevel destriped products is hence denoted as ExoTiCJEDI [V3].
Tiberius pipeline
We used the Tiberius pipeline, which builds on the LRGBEASTS spectral reduction and analysis pipelines^{15,46,47}, on our custom grouplevel destriped products. For each detector, we created badpixel masks by manually identifying hot pixels in the data. We then combined them with pixels flagged as greater than 3σ above the defined background. Before identifying the spectral trace, we interpolated each column of the detectors onto a grid 10 times finer than the initial spatial resolution. This step reduces the noise in the extracted data by improving the extraction of flux at the subpixel level, particularly where the edges of the photometric aperture bisect a pixel. We also interpolated over the bad pixels using their nearestneighbouring pixels in x and y.
We traced the spectra by fitting Gaussians at each column and used a running median, calculated with a moving box with a width of five data points, to smooth the measured centres of the trace. We fitted these smoothed centres with a fourthorder polynomial, removed points that deviated from the median by 3σ and refitted with a fourthorder polynomial. To remove any residual background flux not captured by the grouplevel destriping, we fitted a linear polynomial along each column, masking the stellar spectrum. This was defined by an aperture with a width of 4 pixels centred on the trace. We also masked an extra 7 pixels on either side of the aperture so that the background was not fitting the wings of the stellar PSF and we clipped any pixels in the background that deviated by more than 3σ from the mean for that particular column and frame. After removing the background in each column, the stellar spectra were then extracted by summing within a 4pixelwide aperture and correcting for pixel oversampling caused by the interpolation onto a finer grid, as described above. The uncertainties in the stellar spectra were calculated from the photon noise before background subtraction.
transitspectroscopy pipeline
We used the transitspectroscopy pipeline^{48} on the ‘rateints’ products of the JWST Calibration Pipeline, treating each detector separately. The trace position was found from the median integration by crosscorrelating each column with a Gaussian function, removing outliers using a median filter with a 10pixelwide window and smoothing the trace with a spline. We removed 1/f noise from the ‘rateints’ products by masking all pixels within 10 pixels from the centre of the trace and calculating and removing the median value from all columns. We then used optimal extraction^{49} to obtain the 1D stellar spectra, with a 5pixelwide aperture above and below the trace. This allowed us to treat bad pixels and cosmic rays that had not been accounted for or masked in the ‘rateints’ products in an automated fashion. To monitor systematic trends in the observations, we also calculated the trace centre as described above and the FWHM for all integrations. The FWHM was calculated at each column and at each integration by first subtracting each column to half the maximum value in it, with a spline used to interpolate the profile. The roots of this profile were then found to estimate the FWHM.
Eureka! pipeline
We used two customized versions of the Eureka! pipeline^{50}, which combines standard steps from the JWST Calibration Pipeline with an optimal extraction scheme to obtain the time series of stellar spectra.
The first Eureka! reduction used our custom grouplevel destriped products and applied Stages 2 and 3 of Eureka! Stage 2, a wrapper of the JWST Calibration Pipeline, followed the default settings up to the flat fielding and photometric calibration steps, which were both skipped. Stage 3 of Eureka! was then used to perform the background subtraction and extraction of the 1D stellar spectra. We started by correcting for the curvature of NIRSpec G395H spectra by shifting the detector columns by whole pixels, to bring the peak of the distribution of the counts in each column to the centre of our subarray. To ensure that this curvature correction was smooth, we computed the shifts in each column for each integration from the median integration frame in each segment and applied a running median to the shifts obtained for each column. The pixel shifts were applied with periodic boundary conditions, such that pixels shifted upwards from the top of the subarray appeared at the bottom after the correction, ensuring no pixels were lost. We applied a columnbycolumn background subtraction by fitting and subtracting a flat line to each column of the curvaturecorrected data frames, obtained by fitting all pixels further than six pixels from the central row. We also performed a double iteration of outlier rejection in time with a threshold of 10σ, along with a 3σ spatial outlierrejection routine, to ensure that bad pixels were not biasing our background correction. These outlierrejection thresholds were selected to remove clear outliers in the data and provide a balance with the background subtraction step. We performed optimal extraction using an extraction profile defined from the median frame, the central nine rows of our subarray (four rows on either side of the central row). We also measured the vertical shift in pixels of the spectrum from one integration to the other using crosscorrelation and the average PSF width at each integration, obtained by adding all columns together and fitting a Gaussian to the profile to estimate its width. This reduction is henceforth denoted as Eureka! [V1].
The second Eureka! reduction (Eureka! [V2]) used the ‘rateints’ outputs of the JWST Calibration Pipeline and applied Stage 2 of Eureka! as described above, with a modified version of Stage 3. In this reduction, we corrected the curvature of the trace using a spline and found the centre of the trace using the median of each column. We removed 1/f noise by subtracting the mean from each column, excluding the region 6 pixels away from the trace, sigmaclipping outliers at 3σ. We extracted the 1D stellar spectra using a 4pixelwide aperture on either side of the trace centre.
Limbdarkening
Limbdarkening is a function of the physical structure of the star that results in variations in the specific intensity of the light from the centre of the star to the limb, such that the limb looks darker than the centre. This is because of the change in depth of the stellar atmosphere being observed. At the limb of the star, the region of the atmosphere being observed at slant geometry is at higher altitudes and lower density, and thus lower temperatures, compared with the deeper atmosphere observed at the centre of the star, at which hotter, denser layers are observed. The effect of limbdarkening is most prominent at shorter wavelengths, resulting in a more Ushaped light curve compared with the flatbottomed light curves observed at longer wavelengths. To account for the effects of limbdarkening on the timeseries light curves, we used analytical approximations for computing the ratio of the mean intensity to the central intensity of the star. The most commonly used limbdarkening laws for exoplanet transit light curves are the quadratic, squareroot and nonlinear fourparameter laws^{51}:
Quadratic:
Squareroot:
Nonlinear fourparameter:
in which I(1) is the specific intensity in the centre of the disk, u_{1}, u_{2}, s_{1}, s_{2}, c_{1}, c_{2}, c_{3} and c_{4} are the limbdarkening coefficients and μ = cos(γ), in which γ is the angle between the line of sight and the emergent intensity.
The limbdarkening coefficients depend on the particular stellar atmosphere and therefore vary from star to star. For consistency across all of the lightcurve fitting, we used 3D stellar models^{52} for T_{eff} = 5,512 K, log(g) = 4.47 cgs and Fe/H = 0.0, along with the instrument throughput to determine I and μ. As instrument commissioning showed that the throughput was higher than the premission expectations^{53}, a custom throughput was produced from the median of the measured ExoTiCJEDI [V2] stellar spectra, divided by the stellar model and Gaussian smoothed.
For the limbdarkening coefficients that were held fixed, we used the values computed using the ExoTiCLD^{54,55} package, which can compute the linear, quadratic and threeparameter and fourparameter nonlinear limbdarkening coefficients^{51,56}. To compute and fit for the coefficients from the squareroot law, we used previously outlined formalisms^{57,58}. We highlight that we do not see any dependence in our transmission spectra on the limbdarkening procedure used across our independent reductions and analyses.
Lightcurve fitting
From the time series of extracted 1D stellar spectra, we created our broadband transit light curves by summing the flux over 2.725–3.716 μm for NRS1 and 3.829–5.172 μm for NRS2. For the spectroscopic light curves, we used a common 10pixel binning scheme within these wavelength ranges to generate a total of 349 spectroscopic bins (146 for NRS1 and 203 for NRS2). We also tested wider and narrower binning schemes but found that 10pixelwide bins achieved the best compromise between the noise in the spectrum and showcasing the abilities of G395H across analyses. In our analyses, we treated the NRS1 and NRS2 light curves independently to account for differing systematics across the two detectors. To construct the full NIRSpec G395H transmission spectrum of WASP39b, we fitted the NRS1 and NRS2 broadband and spectroscopic light curves using 11 independent lightcurvefitting codes, which are detailed below. When starting values were required, all analyses used the same system parameters^{37}. In many of our analyses, we detrended the raw broadband and spectroscopic light curves using the timedependent decorrelation parameters for the change in the FWHM of the spectral trace or the shift in xpixel and ypixel positions (Extended Data Fig. 2). We also used various approaches to account for the mirrortilt event, which we found to have a smaller effect at longer wavelengths (Extended Data Fig. 3).
Using fitting pipeline 1, we measured a centre of transit time (T_{0}) of T_{0} = 2,459,791.612039 ± 0.000017 BJD_{TDB} and T_{0} = 2,459,791.6120689 ± 0.000021 BJD_{TDB} computed from the NRS1 and NRS2 broadband light curves, respectively; most of the fitting pipelines obtained T_{0} within 1σ of the quoted uncertainty.
For each of our analyses, we computed the expected photon noise from the raw counts taking into account the instrument read noise (16.18 e^{−} on NRS1 and 17.75 e^{−} on NRS2), gain (1.42 for NRS1 and 1.62 for NRS2) and the background counts (which are found to be negligible after cleaning) and compared it to the final signaltonoise ratio in our light curves (see Fig. 1). We also determine the level of white and red noise in our spectroscopic light curves by computing the Allan deviation^{59}, which is used to measure the deviation from the expected photon noise by binning the data into successively smaller bins (that is, fewer data points per bin) and calculating the signaltonoise ratio achieved^{60}. Extended Data Fig. 4 shows the Allan deviation for three of the 11 reductions performed on the data (see the ExoTiCISM noise_calculator function^{54}).
Although there is a general consensus across each of the data analyses, by comparing the results of each fitting pipeline, we were better able to evaluate the impact of different approaches to the data reduction, such as the removal of bad pixels. For future studies, we recommend the application of several pipelines that use differing analysis methods, such as the treatment of limbdarkening, systematic effects and noise removal. No single pipeline presented on its own can fully evaluate the measured impact of each effect, given the differing strategies, targets and potential for chance events such as a mirror tilt with each observation. In particular, attention should be paid to 1/f noise removal at the group versus integration levels for observations with fewer groups per integration than this study.
Below, we detail each of our 11 fitting pipelines and summarise them in Extended Data Table 1.
Fitting pipeline 1: ExoTiCJEDI
We fitted the broadband and spectroscopic light curves produced from the ExoTICJEDI [V3] stellar spectra using the leastsquares optimizer, scipy.optimize lm (ref. ^{61}). We simultaneously fitted a batman transit model^{62} with a constant baseline and systematics models for data pretilt and posttilt event, fixing the centre of transit time T_{0}, the ratio of the semimajor axis to stellar radius a/R_{⋆} and the inclination i to the broadband lightcurve bestfit values. The systematics models included a linear regression on x and y, for which x and y are the measured trace positions in the dispersion and crossdispersion directions, respectively. We accounted for the tilt event by normalizing the light curve pretilt by the median pretransit flux and normalizing the light curve posttilt by the median posttransit flux. We discarded the first 15 integrations and the three integrations during the tilt event. Fourteenpixel columns were discarded owing to outlier pixels directly on the trace. We fixed the limbdarkening coefficients to the fourparameter nonlinear law.
Fitting pipeline 2: Tiberius
We used the broadband light curves generated from the Tiberius stellar spectra and fitted for the ratio of the planet to stellar radii R_{p}/R_{⋆}, as well as i, T_{0}, a/R_{⋆}, the quadratic law limbdarkening coefficient u_{1} and the systematics model parameters, the xpixel and ypixel shifts, FWHM and sky background, with the period P, the eccentricity e and u_{2} fixed. We used uniform priors for all the fitted parameters. Our analytic transit lightcurve model was generated with batman. We fitted our broadband light curve with a transit + systematics model using a Gaussian process (GP)^{63,64}, implemented through george^{65}, and a Markov chain Monte Carlo method, implemented through emcee^{66}. For our Tiberius spectroscopic light curves, we held a/R_{⋆}, i and T_{0} fixed to the values determined from the broadband lightcurve fits and applied a systematics correction from the broadband lightcurve fit to aid in fitting the mirrortilt event. We fitted the spectroscopic light curves using a GP with an exponential squared kernel for the same systematics detrending parameters detailed above. We used a Gaussian prior for a/R_{⋆} and uniform priors for all other fitted parameters.
Fitting pipeline 3: Aesop
We used transit light curves from the ExoTiCJEDI [V1] stellar spectra and fit the broadband and spectroscopic light curves using the leastsquares minimizer LMFIT^{67}. We fitted each light curve with a twocomponent function consisting of a transit model (generated using batman) multiplied by a systematics model. Our systematics model included the xpixel and ypixel positions on the detector (x, y, xy, x^{2} and y^{2}). To capture the amplitude of the tilt event in our systematics model, we carried out piecewise linear regression on the outoftransit baseline pretilt and posttilt. We first fit the broadband light curve by fixing P and e and fitting for T_{0}, a/R_{⋆}, i, R_{p}/R_{⋆}, stellar baseline flux and systematic trends using wide uniform priors. For the spectroscopic light curves, we fixed T_{0}, a/R_{⋆} and i to the bestfit values from the broadband light curve and fit for R_{p}/R_{⋆}. We held the nonlinear limbdarkening coefficients fixed.
Fitting pipeline 4: transitspectroscopy
We fit the broadband and spectroscopic light curves produced from the transitspectroscopy stellar spectra, running juliet^{68} in parallel with the lightcurvefitting module of the transitspectroscopy pipeline^{48} with dynamic nested sampling through dynesty^{69} and analytical transit models computed using batman. We fit the broadband light curves for NRS1 and NRS2 individually, fixing P, e and ω and fitting for the impact parameter b, as well as T_{0}, a/R_{⋆}, R_{p}/R_{⋆}, extra jitter and the mean outoftransit flux. We also fitted two linear regressors, a simple slope and a ‘jump’ (a regressor with zeros before the tilt event and ones after the tilt event), scaled to fit the data. We fitted the squarerootlaw limbdarkening coefficients using the Kipping sampling scheme. We fitted the spectroscopic light curves at the native resolution of the instrument, fixing T_{0}, a/R_{⋆} and b. We used the broadband lightcurve systematics model for the spectroscopic light curve, with wide uniform priors for each wavelength bin, and set truncated normal priors for the squarerootlaw limbdarkening coefficients. We also fitted a jitter term added in quadrature to the error bars at each wavelength with a loguniform prior between 10 and 1,000 ppm. We computed the mean of the limbdarkening coefficients by first computing the nonlinear coefficients from ATLAS models^{70} and passing them through the SPAM algorithm^{71}. We binned the data into 10pixelwavelength bins after fitting the nativeresolution light curves.
Fitting pipeline 5: ExoTEP
We fitted the transit light curves from the Eureka! [V1] stellar spectra using the ExoTEP analysis framework^{72,73,74,75}. ExoTEP uses batman to generate analytical lightcurve models, adds an analytical instrument systematics model along with a photometric scatter parameter and fits for the bestfit parameters and their uncertainties using emcee. Before fitting, we cleaned the light curves by running ten iterations of 5σ clipping using a running median of window length 20 on the flux, xpixel and ypixel shifts and the ‘ydriftwidth’ data product from Eureka! Stage 3 (the average spatial PSF width at each integration). Our systematics model consisted of a linear trend in time with a ‘jump’ (constant offset) after the tilt event. The ‘ydriftwidth’ was used before the fit to locate the tilt event. We used a running median of ‘ydriftwidth’ to search for the largest offset and flagged every data point after the tilt event so that they would receive a constant ‘jump’ offset in our systematics model. We also removed the first point of the tilt event in our fits, as it was not captured by the ‘jump’ model. We fitted the broadband light curves, fitting for R_{p}/R_{⋆}, photometric scatter, T_{0}, b, a/R_{⋆}, the quadratic limbdarkening coefficients and the systematics model parameters (normalization constant, slope in time and constant ‘jump’ offset). We used uninformative flat priors on all the parameters. The orbital parameters were fixed to the bestfit broadband light curve values for the subsequent spectroscopic lightcurve fits.
Fitting pipeline 6
We fitted transit light curves from the ExoTiCJEDI [V1] stellar spectra using a custom lmfit lightcurvefitting code. The final systematic detrending model included a batman analytical transit model multiplied by a systematics model consisting of a linear stellar baseline term, a linear term for the xpixel and ypixel shifts and an exponential ramp function. The tilt event was accounted for by decorrelating the light curves with the ypixel shifts, using a (1 + constant × yshift) term with the constant fitted for in each light curve. For the broadband lightcurve fits, we fixed P and fitted for T_{0}, i, R_{p}/R_{⋆}, a/R_{⋆}, xpixel and ypixel shifts and the exponential ramp amplitude and timescale. We fixed the nonlinear limbdarkening coefficients. For the spectroscopic lightcurve fits, we fixed the values of T_{0}, i and a/R_{⋆} and the exponential ramp timescale to the broadband lightcurvefit values, and fitted for R_{p}/R_{⋆}, the xpixel and ypixel shifts and the ramp amplitude. Wide, uniform priors were used on all the fitting parameters in both the broadband and spectroscopic lightcurve fits.
Fitting pipeline 7
We fitted transit light curves from the Eureka! [V2] stellar spectra, using PyLightcurve (ref. ^{75}) to generate the transit model with emcee as the sampler. We calculated the nonlinear fourparameter limbdarkening coefficients using ExoTHETyS (ref. ^{76}), which relies on PHOENIX 2012–2013 stellar models^{77,78}, and fixed these in our fits to the precomputed theoretical values. Our full transit + systematics model included a transit model multiplied by a secondorder polynomial in the time domain. We accounted for the tilt event by subtracting the mean of the last 30 integrations of the pretransit data from the mean of the first 30 integrations of the posttransit data, to account for the jump in flux, shifting the posttransit light curve upwards by the jump value. We fitted for the systematics (the parameters of the secondorder polynomial), R_{p}/R_{⋆} and T_{0}. We used uniform priors for all the fitted parameters. We adopted the root mean square of the outoftransit data as the error bars for the lightcurve data points to account for the scatter in the data.
Fitting pipeline 8
We used the transit light curves generated from the ExoTiCJEDI [V1] stellar spectra. We fit the broadband light curves with a batman transit model multiplied by a secondorder systematics model as a function of xpixel and ypixel shifts. We fixed both of the quadratic limbdarkening coefficients for each wavelength bin. We fitted for R_{p}/R_{⋆}, i, T_{0} and a/R_{⋆}, using wide uninformed priors, and ran our fits using emcee. For the spectroscopic lightcurve fits, we fixed i and a/R_{⋆} to the broadband lightcurve bestfit values and fitted for an extra error term added in quadrature.
Fitting pipeline 9
We used the transit light curves from the ExoTiCJEDI [V1] stellar spectra. We fixed both of the quadratic limbdarkening coefficients and fitted the light curves with a batman transit model multiplied by a systematics model of a secondorder function of xpixel and ypixel shifts. We fixed the bestfit broadband lightcurve values for T_{0}, a/R_{⋆} and i for the spectroscopic lightcurve fits and fitted for R_{p}/R_{⋆} using emcee for each 10pixel bin, with the walkers initialized in a tight cluster around the bestfit solution from a Levenberg–Marquardt minimization. For both the broadband and spectroscopic light curves, we also fit for an extra perpoint error term.
Fitting pipeline 10
We fitted the transit light curves from the ExoTiCJEDI [V2] stellar spectra and performed our model fitting using automatic differentiation implemented with JAX (ref. ^{79}). We used a GP systematics model with a timedependent Matérn (ν = 3/2) kernel and a variable whitenoise jitter term. The mean function consists of a linear trend in time plus a sigmoid function to account for the drop in measured flux that occurred midtransit owing to the mirrortilt event. For the transit model, we used the exoplanet package^{80}, making use of previously developed lightcurve models^{81,82}. For the GP systematics component, a generalization of the algorithm used by the celerite package^{83} was adapted for JAX. We fixed both of the quadratic limbdarkening coefficients. For the initial broadband lightcurve fit, both NRS1 and NRS2 were fitted simultaneously. All transit parameters were shared across both light curves, except for R_{p}/R_{⋆}, which was allowed to vary for NRS1 and NRS2 independently. We fitted for T_{0}, the transit duration b and both R_{p}/R_{⋆} values. For the spectroscopic lightcurve fits, all transit parameters were then fixed to the maximumlikelihood values determined from the broadband fit, except for R_{p}/R_{⋆}, which was allowed to vary for each wavelength bin. Uncertainties for the transit model parameters, including R_{p}/R_{⋆}, were assumed to be Gaussian and estimated using the Fisher information matrix at the location of the maximumlikelihood solution, which was computed exactly using the JAX automatic differentiation framework.
Fitting pipeline 11: Eureka!
We used transit light curves from the Eureka! [V2] timeseries stellar spectra with the opensource Eureka! code to estimate the bestfit transit parameters and their uncertainties using a Markov chain Monte Carlo method fit to the data (implemented by emcee). A linear trend in time was used as a systematics model and a step function was used to account for the tilt event. We fixed a/R_{⋆}, i, T_{0} and the time of the tilt event to the bestfit values from the NRS1 broadband light curve, with the three integrations during the tilt event clipped from the light curve. We fitted for R_{p}/R_{⋆}, both quadratic limbdarkening coefficients, the linear time trend and the magnitude of the step from the tilt event, with uniform priors for both the magnitude of the step and the limbdarkening coefficients.
Transmission spectral analysis
On the basis of the independent lightcurve fits described above, we produced 11 transmission spectra from our NIRSpec G395H observations using several analyses and fitting methods. Extended Data Table 1 shows a breakdown of the different steps used in each fitting pipeline. In this work, three different 2D spectral image products were used, producing seven different 1D stellar spectra. Eleven different fitting pipelines using five different limbdarkening methods were then applied. Each of these fitting pipelines resulted in an independent analysis of the observations and 11 comparative transmission spectra. Extended Data Fig. 5 details comparative information for all 11 analyses to quantify their similarities and differences.
We computed the standard deviation of the 11 spectra in each wavelength bin and compared this to the mean uncertainty obtained in that bin. The average standard deviation in each bin across all fitting pipelines was 199 ppm, compared with an average uncertainty of 221 ppm (which ranged from 131 to 625 ppm across the bins). The computed standard deviation in each bin across all pipelines ranged from 85 to 1,040 ppm, with greater than 98% of the bins having a standard deviation lower than 500 ppm. We see an increase in scatter at longer wavelengths, with the structure of the scatter following closely with the measured stellar flux, for which throughput beyond 3.8 μm combines with decreasing stellar flux. The unweighted mean uncertainty of all 11 transmission spectra follows a similar structure to the standard deviation, with the uncertainty increasing at longer wavelengths. The uncertainties from each fitting pipeline are consistent to within 3σ of each other, with the uncertainty per bin typically overestimated compared with the mean uncertainty across all reductions.
From all 11 transmission spectra, we computed a weightedaverage transmission spectrum using the transit depth values from all reductions in each bin weighted by 1/variance (1/σ^{2}, in which σ is the uncertainty on the data point from each reduction). For this weightedaverage transmission spectrum, the unweighted mean of the uncertainties in each bin was used to represent the error bar on each point. By using the weighted average of all 11 independently obtained transmission spectra, we therefore do not apply infinite weight to any one reduction in our interpretation of the atmosphere. Although the weighted average could be sensitive to any one spectrum with underestimated uncertainties, we find that our uncertainties are typically overestimated compared with the average. Similarly, we chose to use the mean rather than the median of the transmission spectral uncertainties, as this results in a more conservative estimate of the uncertainties in each bin. We find that all of the 11 transmission spectra are within 2.95σ of the weightedaverage transmission spectrum without applying offsets.
We calculated normalized transmission spectrum residuals for each fitting pipeline by subtracting the weightedaverage spectrum and dividing by the uncertainty in each bin. We generated histograms of the normalized transmission spectrum residuals and used the mean and standard deviation of the residuals to compute a normalized probability density function (PDF). We performed a Kolmogorov–Smirnov test on each of the normalized residuals and found that they are all approximately symmetric around their means, with normal distributions. This confirms that they are Gaussian in shape, with the null hypothesis that they are not Gaussian strongly rejected in the majority of cases (see Extended Data Fig. 5).
The PDFs of the residuals indicate three distinct clusters of computed spectra based on their deviations from the mean and their spreads. The first cluster is negatively offset by less than 200 ppm and corresponds to fitting pipelines that used extracted stellar spectra and that underwent further cleaning steps (for example, ExoTiCJEDI [V3]). The second cluster is positively offset from the mean by about 120 ppm and contains most of the transmission spectra produced. We see no obvious trends in this group to any specific reduction or fitting process. The final cluster is centred around the mean but has a broad distribution, suggesting a larger scatter both above and below the average transmission spectrum. This is probably the result of uncleaned outliers or marginal offsets between NRS1 and NRS2. These transmission spectra demonstrate that the 11 independent fitting pipelines are able to accurately reproduce the same transmission spectral feature structures, further highlighting the minimized impact of systematics on the timeseries light curves. We suspect that the minor differences resulting from different reduction products and fitting pipelines are linked to the superbias and treatment of 1/f noise. We anticipate that the impacts of these will be improved with new superbias images, expected to be released soon by STScI, and with more detailed investigation into the impact of 1/f noise at the group level beyond the scope of this work.
Model comparison
To identify spectral absorption features, we compared the resulting weightedaverage transmission spectrum of WASP39b to several 1D RCTE atmosphere models from three independent model grids. Each forward model is computed on a set of common physical parameters (for example, metallicity, C/O ratio, internal temperature and heat redistribution), shown in Extended Data Table 2. Additionally, each model grid contains different prescriptions for treating certain physical effects (for example, scattering aerosols). Although each grid contains different opacity sources from varying line lists (see Extended Data Table 2), they each consider all of the main molecular and atomic species^{84}. Each model transmission spectrum from the grids was binned to the same resolution as that of the observations to compute the χ^{2} per data point, with a wavelengthindependent transit depth offset as the free parameter. In general, the forward model grids fit the main features of the data but are unable to place statistically significant constraints on many of the atmospheric parameters, owing to both the finite nature of the forward model grid spacing^{13} and the insensitivity of some of these parameters to the 3–5μm transmission spectrum of WASP39b (for example, >100 K differences in interior temperature provided nearly identical χ^{2}/N).
ATMO
We used the ATMO RCTE grid^{85,86,87,88}, which consists of model transmission spectra for different day–night energy redistribution factors, atmospheric metallicities, C/O ratios, haze factors and grey cloud factors with a range of line lists and pressurebroadening sources^{88}. In total, there were 5,160 models. Within this grid, we find the bestfit model to have 3 times solar metallicity, with a C/O ratio of 0.35 and a grey cloud opacity 5 times the strength of H_{2} Rayleigh scattering at 350 nm and a χ^{2}/N = 1.098 for N = 344 data points and only fitting for an absolute altitude change in y.
PHOENIX
We calculated a grid of transmission spectra using the PHOENIX atmosphere model^{89,90,91}, varying the heat redistribution of the planet, atmospheric metallicity, C/O ratio, internal temperature, the presence of aerosols and the atmospheric chemistry (equilibrium or rainout). Opacities used include the BT2 H_{2}O line list^{92}, as well as HITRAN for 129 other main molecular absorbers^{93} and Kurucz and Bell data for atomic species^{94}. The HITRAN line lists available in this version of PHOENIX are often complete only at room temperature, which may be the cause of the apparent shift in the CO_{2} spectral feature compared with the other grids that primarily use HITEMP and ExoMol lists. This shift is the cause of the difference in χ^{2} between PHOENIX and the other model grids. In total, there were 1,116 models. Within this grid, the bestfit model has 10 times solar metallicity, a C/O ratio of 0.3, an internal temperature of 400 K, rainout chemistry and a cloud deck top at 0.3 mbar. The bestfit model has a χ^{2}/N = 1.203 for N = 344 data points.
PICASO 3.0 and Virga
We used the opensource radiative–convective equilibrium model PICASO 3.0 (refs. ^{95,96}), which has its heritage in the Fortranbased EGP mode^{97,98}, to compute a grid of 1D pressure–temperature models for WASP39b. The opacity sources included in PICASO 3.0 are listed in Extended Data Table 2. Of the 29 molecular opacity sources included, the line lists of notable molecules used were: H_{2}O (ref. ^{99}), CO_{2} (ref. ^{100}), CH_{4} (ref. ^{101}) and CO (ref. ^{102}). The parameters varied in this grid of models include the interior temperature of the planet (T_{int}), atmospheric metallicity, C/O ratio and the daysidetonightside heat redistribution factor (see Extended Data Table 2), with correlatedk opacities^{98,103}. In total, there were 192 cloudfree models. We include the effect of clouds in two ways. First, we postprocessed the pressure–temperature profile using the cloud model Virga^{95,104}, which follows from previously developed methodologies^{38}, in which we included three condensable species (MnS, Na_{2}S and MgSiO_{3}). Virga requires a vertical mixing parameter, K_{zz} (cm^{2} s^{−1}), and a vertically constant sedimentation efficiency parameter, f_{sed}. In general, f_{sed} controls the vertical extent of the cloud opacity, with low values (f_{sed} < 1) creating large, vertically extended cloud decks with small particle sizes. In total, there were 3,840 cloudy models. The best fit from our grid with Virgacomputed clouds has 3 times solar metallicity, solar C/O (0.458) and f_{sed} = 0.6, which results in a χ^{2}/N = 1.084.
As well as the grid fit, we also use the PICASO framework to quantify the featuredetection significance. In this method, we are able to incorporate clouds on the fly using the fitting routine PyMultiNest^{105}. We fit for each of the grid parameters using a nearestneighbour technique and a radius scaling to account for the unknown reference pressure, giving five parameters in total. When fitting for clouds, we either fit for K_{zz} and f_{sed} in the Virga framework (seven parameters in total) or we fit for the cloudtop pressure corresponding to a grey cloud deck with infinite opacity (six parameters in total). These results are described in the following section.
Featuredetection significance
From the chemical equilibrium results of the single bestfit models, the molecules that could potentially contribute to the spectrum based on their abundances and 3–5μm opacity sources are H_{2} and He (via continuum) and CO, H_{2}O, H_{2}S, CO_{2} and CH_{4}. More minor sources of opacity with VMR abundances <1 ppm are molecules such as OCS and NH_{3}. For example, removing H_{2}S, NH_{3} and OCS from the single bestfit PICASO 3.0 model increases the chisquare value by less than 0.002. Therefore, we focus on computing the statistical significance of only H_{2}O, SO_{2}, CO_{2}, CH_{4} and CO.
To quantify the statistical significance, we performed two different tests. First, we used a Gaussian residual fitting analysis, as used in other JTEC ERS analyses^{23,29,31}. In this method, we subtracted the bestfit model without a specific opacity source from the weightedaverage spectrum of WASP39b, isolating the supposed spectral feature. We then fit a threeparameter or fourparameter Gaussian curve to the residual data using a nested sampling algorithm to calculate the Bayesian evidence^{106}. For H_{2}O and CO, the extra transit depth offset parameter for the Gaussian fit was necessary to account for local mismatch of the fit to the continuum, whereas only a mean, standard deviation and scale parameter were required for a residual fit to the other molecules. We then compared this to the Bayesian evidence of a flat line to find the Bayes factor between a model that fits the spectral feature versus a model that excludes the spectral feature. These fits are shown in Extended Data Fig. 6.
Although the Gaussian residual fitting method is useful for quantifying the presence of potentially unknown spectral features, it cannot robustly determine the source of any given opacity. We therefore used the Bayesian fitting routine from PyMultiNest in the PICASO 3.0 framework to refit the grid parameters, while excluding the opacity contribution from the species in question. Then, we compared the significance of the molecule through a Bayes factor analysis^{107}. Those values are shown in Extended Data Table 3.
We find significant evidence (>3σ) for H_{2}O, CO_{2} and SO_{2}. In general, the two methods only agree well for molecules whose contribution has a Gaussian shape (that is, SO_{2} and CO_{2}). For example, for CO_{2}, we find decisive 28.5σ and 26.9σ detections for the Bayes factor and Gaussian analysis, respectively. Similarly, for H_{2}O, we find 21.5σ and 16.5σ detections, respectively. The evidence for SO_{2} is less substantial, but both methods give significant detections of 4.8σ and 3.5σ, respectively. Although the Gaussian fitting method found a broad 1μmwide residual in the region of CO (that is, >4.5 μm), its shape was unlike that seen with the PRISM data^{31}. CO remained undetected with the Bayesian fitting analysis and therefore we are unable to robustly confirm evidence of CO. Similarly, no evidence for CH_{4} was found^{23}. Gaussian residual fitting in the region of CH_{4} absorption only found a very broad inverse Gaussian and so is not included in Extended Data Table 3.
SO_{2} absorption
We performed an injection test with the PICASO bestfit model in the PyMultiNest fitting framework to determine the abundance of SO_{2} required to match the observations. We add SO_{2} opacity using the ExoMol line list^{108}, without rerunning the RCTE model to selfconsistently compute a new climate profile. Fitting for the cloud deck dynamically, without SO_{2}, produces a single best estimate of 10 times solar metallicity, subsolar C/O (0.229), resulting in a marginally worse χ^{2}/N = 1.11. With SO_{2}, the single best fit tends back to 3 times solar metallicity, solar C/O. This suggests that cloud treatment and the exclusion of spectrally active molecules have an effect on the resultant physical interpretation of bulk atmospheric parameters. Ultimately, if we fit for SO_{2} in our PyMultiNest framework with the Virga cloud treatment, we obtain 3 times solar metallicity, solar C/O, log SO_{2} = −5.6 ± 0.1 (SO_{2} = 2.5 ± 0.65 ppm) and χ^{2}/N = 1.02, which is our single bestfit model (shown in Fig. 4). For context, an atmospheric metallicity of 3–10 times solar would provide a thermochemical equilibrium abundance of 72–240 ppm H_{2}S, the presumed source for photochemically produced SO_{2} (ref. ^{36}).
To confirm the plausibility of SO_{2} absorption to explain the 4.1μm spectral feature, we also computed models with prescribed, vertically uniform SO_{2} VMRs of 0, 1, 5 and 10 ppm using the structure from the bestfit PHOENIX model (10 times solar metallicity, C/O = 0.3). We calculated ad hoc spectra using the gCMCRT radiative transfer code^{109} with the ExoMol SO_{2} line list^{108} (see Extended Data Fig. 7). Linearly interpolating the models with respect to the SO_{2} abundance and performing a Levenberg–Marquardt regression gave a bestfit value of 4.6 ± 0.67 ppm. Inserting this abundance of SO_{2} into the bestfit PHOENIX model improves the χ^{2}/N from 1.2 to 1.08.
Future atmospheric retrievals can provide a more statistically robust measurement for the SO_{2} abundance and add extra information from the similar absorption seen in the PRISM transmission spectrum^{29,31}.
4.56μm feature
A 0.08μmwide bump in transit depth centred at 4.56 μm is not fit by any of the model grids. This feature, picked up by the resolution of G395H, is not clearly seen in other ERS observations of WASP39b. Following the same Gaussian residual fitting procedure as described above, we found a feature significance of 3.3σ (see Extended Data Fig. 6). To identify possible opacity sources in the atmosphere of WASP39b that might be the cause of this absorption, we compared the feature with CH_{4} (ref. ^{110}), C_{2}H_{2} (ref. ^{111}), C_{2}H_{4} (ref. ^{112}), C_{2}H_{6} (ref. ^{113}), CO (ref. ^{114}), CO_{2} (ref. ^{100}), CS_{2} (ref. ^{113}), CN (ref. ^{115}), HCN (ref. ^{116}), HCl (ref. ^{113}), H_{2}S (ref. ^{117}), HF (ref. ^{118}), H_{3}^{+} (ref. ^{119}), LiCl (ref. ^{115}), NH_{3} (ref. ^{120}), NO (ref. ^{114}), NO_{2} (ref. ^{113}), N_{2}O (ref. ^{114}), N_{2} (ref. ^{121}), NaCl (ref. ^{122}), OCS (ref. ^{113}), PH_{3} (ref. ^{123}), PN (ref. ^{124}), PO (ref. ^{125}), SH (ref. ^{126}), SiS (ref. ^{127}), SiH_{4} (ref. ^{128}), SiO (ref. ^{129}), the X–X state of SO (ref. ^{130}), SO_{2} (ref. ^{108}), SO_{3} (ref. ^{108}) and isotopologues of H_{2}O, CH_{4}, CO_{2} and CO, but did not find a convincing candidate that showed opacity at the correct wavelength or the correct width. The narrowness of the feature suggests that it could be a very distinct Qbranch, in which the rotational quantum number in the ground state is the same as the rotational quantum number in the excited state. However, of the molecules we explored, there were no candidates with a distinct Qbranch at this wavelength whose Pbranch and Rbranch did not obstruct the neighbouring CO_{2} and continuumlike CO + H_{2}O opacity.
We also note that many of these species lack hightemperature linelist data, making it difficult to definitively rule out such species. For example, OCS, SO and CS_{2} are available in HITRAN2020 (ref. ^{113}) but not in ExoMol^{131}. Furthermore, if photochemistry is important for WASP39b, as indicated by the presence of SO_{2}, then there may be many species out of equilibrium that may contribute to the transit spectrum, some of which do not have hightemperature opacity data at present (such as OCS, NH_{2} or HSO). Future observations over this wavelength region of this and other planets may confirm or refute the presence of this unknown absorber.
Data availability
The data used in this paper are associated with JWST programme ERS 1366 (observation #4) and are available from the Mikulski Archive for Space Telescopes (MAST; https://mast.stsci.edu). Science data processing version (SDP_VER) 2022_2a generated the uncalibrated data that we downloaded from MAST. We used JWST Calibration Pipeline software version (CAL_VER) 1.5.3 with modifications described in the text. We used calibration reference data from context (CRDS_CTX) 0916, except as noted in the text. All the data and models presented in this publication can be found at https://doi.org/10.5281/zenodo.7185300.
Code availability
The codes used in this publication to extract, reduce and analyse the data are as follows; STScI JWST Calibration Pipeline^{44} (https://github.com/spacetelescope/jwst), Eureka!^{50} (https://eurekadocs.readthedocs.io/en/latest/), ExoTiCJEDI^{45} (https://github.com/ExoTiC/ExoTiCJEDI), juliet^{68} (https://juliet.readthedocs.io/en/latest/), Tiberius^{15,46,47}, transitspectroscopy^{48} (https://github.com/nespinoza/transitspectroscopy). Furthermore, these made use of batman^{62} (http://lkreidberg.github.io/batman/docs/html/index.html), celerite^{83} (https://celerite.readthedocs.io/en/stable/), chromatic (https://zkbt.github.io/chromatic/), dynesty^{69} (https://dynesty.readthedocs.io/en/stable/index.html), emcee^{66} (https://emcee.readthedocs.io/en/stable/), exoplanet^{80} (https://docs.exoplanet.codes/en/latest/), ExoTEP^{72,73,74}, ExoTHETyS^{76} (https://github.com/uclexoplanets/ExoTETHyS), ExoTiCISM^{54} (https://github.com/ExoTiC/ExoTiCISM), ExoTiCLD^{55} (https://exoticld.readthedocs.io/en/latest/), george^{65} (https://george.readthedocs.io/en/latest/), JAX^{79} (https://jax.readthedocs.io/en/latest/), LMFIT^{67} (https://lmfit.github.io/lmfitpy/), PyLightcurve^{75} (https://github.com/uclexoplanets/pylightcurve), PyMC3 (ref. ^{132}) (https://docs.pymc.io/en/v3/index.html) and Starry^{81} (https://starry.readthedocs.io/en/latest/), each of which use the standard Python libraries astropy^{133,134}, matplotlib^{135}, numpy^{136}, pandas^{137}, scipy^{61} and xarray^{138}. The atmospheric models used to fit the data can be found at ATMO^{85,86,87,88}, PHOENIX^{89,90,91}, PICASO^{95,96} (https://natashabatalha.github.io/picaso/), Virga^{95,104} (https://natashabatalha.github.io/virga/) and gCMCRT^{109} (https://github.com/ELeeAstro/gCMCRT).
References
Öberg, K. I., MurrayClay, R. & Bergin, E. A. The effects of snowlines on C/O in planetary atmospheres. Astrophys. J. Lett. 743, L16 (2011).
Mordasini, C., van Boekel, R., Mollière, P., Henning, T. & Benneke, B. The imprint of exoplanet formation history on observable presentday spectra of hot Jupiters. Astrophys. J. 832, 41 (2016).
Sing, D. K. et al. A continuum from clear to cloudy hotJupiter exoplanets without primordial water depletion. Nature 529, 59–62 (2016).
Wakeford, H. R. et al. The complete transmission spectrum of WASP39b with a precise water constraint. Astron. J. 155, 29 (2018).
Alam, M. K. et al. The Hubble Space Telescope PanCET program: an optical to infrared transmission spectrum of HATP32Ab. Astron. J 160, 51 (2020).
Birkby, J. L. Exoplanet atmospheres at high spectral resolution. Preprint at https://arxiv.org/abs/1806.04617 (2018).
Line, M. R. et al. A solar C/O and subsolar metallicity in a hot Jupiter atmosphere. Nature 598, 580–584 (2021).
Pelletier, S. et al. Where is the water? Jupiterlike C/H ratio but strong H_{2}O depletion found on τ Boötis b using SPIRou. Astron. J 162, 73 (2021).
Faedi, F. et al. WASP39b: a highly inflated Saturnmass planet orbiting a late Gtype star. Astron. Astrophys. 531, A40 (2011).
Batalha, N. E. & Line, M. R. Information content analysis for selection of optimal JWST observing modes for transiting exoplanet atmospheres. Astron. J 153, 151 (2017).
Jakobsen, P. et al. The NearInfrared Spectrograph (NIRSpec) on the James Webb Space Telescope. I. Overview of the instrument and its capabilities. Astron. Astrophys. 661, A80 (2022).
Birkmann, S. M. et al. The NearInfrared Spectrograph (NIRSpec) on the James Webb Space Telescope. IV. Capabilities and predicted performance for exoplanet characterization. Astron. Astrophys. 661, A83 (2022).
Fischer, P. D. et al. HST hotJupiter transmission spectral survey: clear skies for cool Saturn WASP39b. Astrophys. J. 827, 19 (2016).
Nikolov, N. et al. VLT FORS2 comparative transmission spectroscopy: detection of Na in the atmosphere of WASP39b from the ground. Astrophys. J. 832, 191 (2016).
Kirk, J. et al. LRGBEASTS: transmission spectroscopy and retrieval analysis of the highly inflated Saturnmass planet WASP39b. Astron. J. 158, 144 (2019).
Barstow, J. K., Aigrain, S., Irwin, P. G. & Sing, D. K. A consistent retrieval analysis of 10 hot Jupiters observed in transmission. Astrophys. J. 834, 50 (2017).
Pinhas, A., Madhusudhan, N., Gandhi, S. & MacDonald, R. H_{2}O abundances and cloud properties in ten hot giant exoplanets. Mon. Not. R. Astron. Soc. 482, 1485–1498 (2019).
Tsiaras, A. et al. A population study of gaseous exoplanets. Astron. J 155, 4 (2018).
Welbanks, L. et al. Mass–metallicity trends in transiting exoplanets from atmospheric abundances of H_{2}O, Na, and K. Astrophys. J. Lett. 887, L20 (2019).
Kawashima, Y. & Min, M. Implementation of disequilibrium chemistry to spectral retrieval code ARCiS and application to 16 exoplanet transmission spectra. Indication of disequilibrium chemistry for HD 209458b and WASP39b. Astron. Astrophys. 656, A90 (2021).
Shibata, S., Helled, R. & Ikoma, M. The origin of the high metallicity of closein giant exoplanets. Combined effects of resonant and aerodynamic shepherding. Astron. Astrophys. 633, A33 (2020).
Helled, R. & Morbidelli, A. in ExoFrontiers: Big Questions in Exoplanetary Science (ed. Madhusudhan, N.) (IOP Publishing, 2021).
Ahrer, E.M. et al. Early Release Science of the exoplanet WASP39b with JWST NIRCam. Nature https://doi.org/10.1038/s41586022055904 (2023).
Polanski, A. S., Crossfield, I. J., Howard, A. W., Isaacson, H. & Rice, M. Chemical abundances for 25 JWST exoplanet host stars with KeckSpec. Res. Notes AAS 6, 155 (2022).
Mullally, S. E., Rodriguez, D. R., Stevenson, K. B. & Wakeford, H. R. The Exo.MAST table for JWST exoplanet atmosphere observability. Res. Notes AAS 3, 193 (2019).
Stevenson, K. B. et al. Transiting exoplanet studies and community targets for JWST’s Early Release Science Program. Publ. Astron. Soc. Pac. 128, 094401 (2016).
Bean, J. L. et al. The Transiting Exoplanet Community Early Release Science Program for JWST. Publ. Astron. Soc. Pac. 130, 114402 (2018).
Rigby, J. et al. The science performance of JWST as characterized in commissioning. Preprint at https://arxiv.org/abs/2207.05632 (2023).
JWST Transiting Exoplanet Community Early Release Science Team. Identification of carbon dioxide in an exoplanet atmosphere. Nature https://doi.org/10.1038/s4158602205269w (2022).
Lodders, K. & Fegley, B. Atmospheric chemistry in giant planets, brown dwarfs, and lowmass dwarf stars: I. Carbon, nitrogen, and oxygen. Icarus 155, 393–424 (2002).
Rustamkulov, R. et al. Early Release Science of the exoplanet WASP39b with JWST NIRSpec PRISM. Nature https://doi.org/10.1038/s4158602205677y (2023).
Zahnle, K. et al. Atmospheric sulfur photochemistry on hot Jupiters. Astrophys. J. Lett. 701, L20 (2009).
Hobbs, R. et al. Sulfur chemistry in the atmospheres of warm and hot Jupiters. Mon. Not. R. Astron. Soc. 506, 3186–3204 (2021).
Tsai, S.M. et al. A comparative study of atmospheric chemistry with VULCAN. Astrophys. J. 923, 264 (2021).
Polman, J., Waters, L.B.F.M., Min, M., Miguel, Y. & Khorshid, N. H2S and SO2 detectability in hot Jupiters. Sulphur species as indicators of metallicity and C/O ratio. Astron. Astrophys. https://doi.org/10.1051/00046361/202244647 (in the press).
Moses, J. I., Madhusudhan, N., Visscher, C. & Freedman, R. S. Chemical consequences of the C/O ratio on hot Jupiters: examples from WASP12b, CoRoT2b, XO1b, and HD 189733b. Astrophys. J. 763, 25 (2012).
Moses, J. I. et al. Disequilibrium carbon, oxygen, and nitrogen chemistry in the atmospheres of HD 189733b and HD 209458b. Astrophys. J. 737, 15 (2011).
Ackerman, A. S. & Marley, M. S. Precipitating condensation clouds in substellar atmospheres. Astrophys. J. 556, 872 (2001).
Mousis, O., Aguichine, A., Helled, R., Irwin, P. G. J. & Lunine, J. I. The role of ice lines in the formation of Uranus and Neptune. Philos. Trans. R. Soc. A 378, 20200107 (2020).
Wong, M. H., Mahaffy, P. R., Atreya, S. K., Niemann, H. B. & Owen, T. C. Updated Galileo probe mass spectrometer measurements of carbon, oxygen, nitrogen, and sulfur on Jupiter. Icarus 171, 153–170 (2004).
Fletcher, L. N., Orton, G. S., Teanby, N. A., Irwin, P. G. J. & Bjoraker, G. L. Methane and its isotopologues on Saturn from Cassini/CIRS observations. Icarus 199, 351–367 (2009).
Pollack, J. B. et al. Formation of the giant planets by concurrent accretion of solids and gas. Icarus 124, 62–85 (1996).
Thorngren, D. & Fortney, J. J. Connecting giant planet atmosphere and interior modeling: constraints on atmospheric metal enrichment. Astrophys. J. Lett. 874, L31 (2019).
Bushouse, H. et al. JWST Calibration Pipeline (1.6.2). Zenodo https://doi.org/10.5281/zenodo.7041998 (2022).
Alderson, L., Grant, D., Wakeford, H. ExoTiC/ExoTiCJEDI: v0.1betarelease. Zenodo https://doi.org/10.5281/zenodo.7185855 (2022).
Kirk, J. et al. LRGBEASTS III: groundbased transmission spectrum of the gas giant orbiting the cool dwarf WASP80. Mon. Not. R. Astron. Soc. 474, 876 (2018).
Kirk, J. et al. ACCESS and LRGBEASTS: a precise new optical transmission spectrum of the ultrahot Jupiter WASP103b. Astronom. J. 162, 34 (2021).
Espinoza, N. TransitSpectroscopy (0.3.11). Zenodo https://doi.org/10.5281/zenodo.6960924 (2022).
Marsh, T. R. The extraction of highly distorted spectra. Publ. Astron. Soc. Pac. 101, 1032 (1989).
Bell, T. J. et al. Eureka!: An endtoend pipeline for JWST timeseries observations. J. Open Source Softw. 7, 4503 (2022).
Claret, A. A new nonlinear limbdarkening law for LTE stellar atmosphere models. Calculations for −5.0 ≤ log[M/H] ≤ +1, 2000 K ≤ T_{eff} ≤ 50000 K at several surface gravities. Astron. Astrophys. 363, 1081–1190 (2000).
Magic, Z., Chiavassa, A., Collet, R. & Asplund, M. The STAGGERgrid: a grid of 3D stellar atmosphere models. IV. Limb darkening coefficients. Astron. Astrophys. 573, A90 (2015).
Space Telescope Science Institute. JWST User Documentation (JDox) website. https://jwstdocs.stsci.edu/ (2016).
Laginja, I. & Wakeford, H. ExoTiCISM: a Python package for marginalised exoplanet transit parameters across a grid of systematic instrument models. J. Open Source Softw. 5, 2281 (2020).
Wakeford, H. & Grant, D. ExoTiC/ExoTiCLD: ExoTiCLD v2.1 Zenodo https://doi.org/10.5281/zenodo.6809899 (2022).
Sing, D. K. Stellar limbdarkening coefficients for CoRot and Kepler. Astron. Astrophys. 510, A21 (2010).
Kipping, D. M. Efficient, uninformative sampling of limbdarkening coefficients for twoparameter laws. Mon. Not. R. Astron. Soc. 435, 2152–2160 (2013).
Espinoza, N. & Jordán, A. Limb darkening and exoplanets – II. Choosing the best law for optimal retrieval of transit parameters. Mon. Not. R. Astron. Soc. 457, 3573–3581 (2016).
Allan, D. W. Statistics of atomic frequency standards. Proc. IEEE 54, 221–230 (1966).
Pont, F., Zucker, S. & Queloz, D. The effect of red noise on planetary transit detection. Mon. Not. R. Astron. Soc. 373, 231–242 (2006).
Virtanen, P. et al. SciPy 1.0: fundamental algorithms for scientific computing in Python. Nat. Methods 17, 261–272 (2020).
Kreidberg, L. batman: BAsic Transit Model cAlculatioN in Python. Publ. Astron. Soc. Pac. 127, 1161 (2015).
Gibson, N. P. et al. A Gaussian process framework for modelling instrumental systematics: application to transmission spectroscopy. Mon. Not. R. Astron. Soc. 419, 2683–2694 (2012).
Gibson, N. P. Reliable inference of exoplanet lightcurve parameters using deterministic and stochastic systematics models. Mon. Not. R. Astron. Soc. 445, 3401–3414 (2014).
Ambikasaran, S., ForemanMackey, D., Greengard, L., Hogg, D. W. & O’Neil, M. Fast direct methods for Gaussian processes. IEEE Trans. Pattern Anal. Mach. Intell. 38, 252–265 (2015).
ForemanMackey, D. et al. emcee v3: a Python ensemble sampling toolkit for affineinvariant MCMC. J. Open Source Softw. 4, 1864 (2019).
Newville, M. et al. LMFIT: nonlinear leastsquare minimization and curvefitting for Python. Zenodo https://doi.org/10.5281/zenodo.11813 (2014).
Espinoza, N., Kossakowski, D. & Brahm, R. juliet: a versatile modelling tool for transiting and nontransiting exoplanetary systems. Mon. Not. R. Astron. Soc. 490, 2262–2283 (2019).
Speagle, J. S. DYNESTY: a dynamic nested sampling package for estimating Bayesian posteriors and evidences. Mon. Not. R. Astron. Soc. 493, 3132–3158 (2020).
Kurucz, R. L. Model atmospheres for g, f, a, b, and o stars. Astrophys. J. Suppl. Ser. 40, 1–340 (1979).
Howarth, I. D. On stellar limb darkening and exoplanetary transits. Mon. Not. R. Astron. Soc. 418, 1165–1175 (2011).
Benneke, B. et al. Spitzer observations confirm and rescue the habitablezone superEarth K218b for future characterization. Astrophys. J. 834, 187 (2017).
Benneke, B. et al. A subNeptune exoplanet with a lowmetallicity methanedepleted atmosphere and Miescattering clouds. Nat. Astron. 3, 813–821 (2019).
Benneke, B. et al. Water vapor and clouds on the habitablezone subNeptune exoplanet K218b. Astrophys. J. Lett. 887, L14 (2019).
Tsiaras, A. et al. Detection of an atmosphere around the superEarth 55 Cancri e. Astrophys. J. 820, 99 (2016).
Morello, G. et al. ExoTETHyS: tools for exoplanetary transits around host stars. J. Open Source Softw. 5, 1834 (2020).
Claret, A., Hauschildt, P. H. & Witte, S. New limbdarkening coefficients for PHOENIX/1D model atmospheres. I. Calculations for 1500 K ≤ T_{eff} ≤ 4800 K Kepler, CoRot, Spitzer, uvby, UBVRIJHK, Sloan, and 2MASS photometric systems. Astron. Astrophys. 546, A14 (2012).
Claret, A., Hauschildt, P. H. & Witte, S. New limbdarkening coefficients for PHOENIX/1D model atmospheres. II. Calculations for 5000 K ≤ T_{eff} ≤ 10 000 K Kepler, CoRot, Spitzer, uvby, UBVRIJHK, Sloan, and 2MASS photometric systems. Astron. Astrophys 552, A16 (2013).
Bradbury, J. et al. JAX: Autograd and XLA. Astrophysics Source Code Library. https://ascl.net/2111.002 (2021).
ForemanMackey, D. et al. exoplanetdev/exoplanet: exoplanet v0.5.0. Zenodo https://doi.org/10.5281/zenodo.4737444 (2020).
Luger, R. et al. STARRY: analytic occultation light curves. Astron. J 157, 64 (2019).
Agol, E., Luger, R. & ForemanMackey, D. Analytic planetary transit light curves and derivatives for stars with polynomial limb darkening. Astron. J 159, 123 (2020).
ForemanMackey, D., Agol, E., Ambikasaran, S. & Angus, R. Fast and scalable Gaussian process modeling with applications to astronomical time series. Astron. J 154, 220 (2017).
Sharp, C. & Burrows, A. Atomic and molecular opacities for brown dwarf and giant planet atmospheres. Astrophys. J. Suppl. Ser. 168, 140 (2007).
Tremblin, P. et al. Fingering convection and cloudless models for cool brown dwarf atmospheres. Astrophys. J. Lett. 804, L17 (2015).
Drummond, B. et al. The effects of consistent chemical kinetics calculations on the pressuretemperature profiles and emission spectra of hot Jupiters. Astron. Astrophys. 594, A69 (2016).
Goyal, J. M. et al. A library of ATMO forward model transmission spectra for hot Jupiter exoplanets. Mon. Not. R. Astron. Soc. 474, 5158 (2018).
Goyal, J. M. et al. A library of selfconsistent simulated exoplanet atmospheres. Mon. Not. R. Astron. Soc. 498, 4680 (2020).
Hauschildt, P. H., Allard, F. & Baron, E. The NextGen model atmosphere grid for 3000 ≤ T_{eff} ≤ 10,000 K. Astrophys. J. 512, 377 (1999).
Barman, T. S., Hauschildt, P. H. & Allard, F. Irradiated planets. Astrophys. J. 556, 885–895 (2001).
Lothringer, J. D. & Barman, T. The PHOENIX exoplanet retrieval algorithm and using H^{−} opacity as a probe in ultrahot Jupiters. Astron. J 159, 289 (2020).
Barber, R. J., Tennyson, J., Harris, G. J. & Tolchenov, R. N. A highaccuracy computed water line list. Mon. Not. R. Astron. Soc. 368, 1087–1094 (2006).
Rothman, L. S. et al. The HITRAN 2008 molecular spectroscopic database. J. Quant. Spectrosc. Radiat. Transfer 110, 533–572 (2009).
Kurucz, R. & Bell, B. Atomic line data. CDROM no. 23 (Smithsonian Astrophysical Observatory, 1995).
Batalha, N. E., Marley, M. S., Lewis, N. K. & Fortney, J. J. Exoplanet reflectedlight spectroscopy with PICASO. Astrophys. J. 878, 70 (2019).
Mukherjee, S., Batalha, N. E., Fortney, J. J. & Marley, M. S. PICASO 3.0: a onedimensional climate model for giant planets and brown dwarfs. Astrophys. J. 942, 71 (2023).
Fortney, J. J., Marley, M. S., Lodders, K., Saumon, D. & Freedman, R. Comparative planetary atmospheres: models of TrES1 and HD 209458b. Astrophys. J. 627, L69–L72 (2005).
Marley, M. S. et al. The Sonora brown dwarf atmosphere and evolution models. I. Model description and application to cloudless atmospheres in rainout chemical equilibrium. Astrophys. J. 920, 85 (2021).
Polyansky, O. L. et al. ExoMol molecular line lists XXX: a complete highaccuracy line list for water. Mon. Not. R. Astron. Soc. 480, 2597–2608 (2018).
Huang, X., Gamache, R. R., Freedman, R. S., Schwenke, D. W. & Lee, T. J. Reliable infrared line lists for 13 CO_{2} isotopologues up to E′=18,000 cm^{−1} and 1500 K, with line shape parameters. J. Quant. Spectrosc. Radiat. Transfer 147, 134–144 (2014).
Yurchenko, S. N., Amundsen, D. S., Tennyson, J. & Waldmann, I. P. A hybrid line list for CH_{4} and hot methane continuum. Astron. Astrophys. 605, A95 (2017).
Li, G. et al. Rovibrational line lists for nine isotopologues of the CO molecule in the X^{1}Σ^{+} ground electronic state. Astrophys. J. Suppl. Ser. 216, 15 (2015).
Lupu, R., Freedman, R., GharibNezhad, E., Visscher, C. & Molliere, P. Correlated k coefficients for H2He atmospheres; 196 spectral windows and 1460 pressuretemperature points. Zenodo https://doi.org/10.5281/zenodo.5590989 (2021).
Rooney, C. M., Batalha, N. E., Gao, P. & Marley, M. S. A new sedimentation model for greater cloud diversity in giant exoplanets and brown dwarfs. Astrophys. J. 925, 33 (2022).
Buchner, J. PyMultiNest: Python interface for MultiNest. Astrophysics Source Code Library. https://www.ascl.net/1606.005 (2016).
Skilling, J. Nested sampling. AIP Conf. Proc. 735, 395–405 (2004).
Trotta, R. Bayes in the sky: Bayesian inference and model selection in cosmology. Contemp. Phys. 49, 71–104 (2008).
Underwood, D. S. et al. ExoMol molecular line lists – XIV. The rotation–vibration spectrum of hot SO_{2}. Mon. Not. R. Astron. Soc. 459, 3890–3899 (2016).
Lee, E. K. et al. 3D radiative transfer for exoplanet atmospheres. gCMCRT: a GPUaccelerated MCRT code. Astrophys. J. 929, 180 (2022).
Hargreaves, R. J. et al. An accurate, extensive, and practical line list of methane for the HITEMP database. Astrophys. J. Suppl. Ser. 247, 55 (2020).
Chubb, K. L., Tennyson, J. & Yurchenko, S. N. ExoMol molecular line lists – XXXVII. Spectra of acetylene. Mon. Not. R. Astron. Soc. 493, 1531–1545 (2020).
Mant, B. P., Yachmenev, A., Tennyson, J. & Yurchenko, S. N. ExoMol molecular line lists – XXVII. Spectra of C_{2}H_{4}. Mon. Not. R. Astron. Soc. 478, 3220–3232 (2018).
Gordon, I. E. et al. The HITRAN2020 molecular spectroscopic database. J. Quant. Spectrosc. Radiat. Transfer 277, 107949 (2022).
Rothman, L. S. et al. HITEMP, the hightemperature molecular spectroscopic database. J. Quant. Spectrosc. Radiat. Transfer 111, 2139–2150 (2010).
Bernath, P. F. MoLLIST: molecular line lists, intensities and spectra. J. Quant. Spectrosc. Radiat. Transfer 240, 106687 (2020).
Harris, G. J., Tennyson, J., Kaminsky, B. M., Pavlenko, Y. V. & Jones, H. R. A. Improved HCN/HNC linelist, model atmospheres and synthetic spectra for WZ Cas. Mon. Not. R. Astron. Soc. 367, 400–406 (2006).
Azzam, A. A. A., Tennyson, J., Yurchenko, S. N. & Naumenko, O. V. ExoMol molecular line lists – XVI. The rotation–vibration spectrum of hot H_{2}S. Mon. Not. R. Astron. Soc. 460, 4063–4074 (2016).
Coxon, J. A. & Hajigeorgiou, P. G. Improved direct potential fit analyses for the ground electronic states of the hydrogen halides: HF/DF/TF, HCl/DCl/TCl, HBr/DBr/TBr and HI/DI/TI. J. Quant. Spectrosc. Radiat. Transfer 151, 133–154 (2015).
Mizus, I. I. et al. ExoMol molecular line lists – XX. A comprehensive line list for H_{3}^{+}. Mon. Not. R. Astron. Soc. 468, 1717–1725 (2017).
Coles, P. A., Yurchenko, S. N. & Tennyson, J. ExoMol molecular line lists – XXXV. A rotation–vibration line list for hot ammonia. Mon. Not. R. Astron. Soc. 490, 4638–4647 (2019).
Western, C. M. et al. The spectrum of N_{2} from 4,500 to 15,700 cm^{−1} revisited with PGOPHER. J. Quant. Spectrosc. Radiat. Transfer 219, 127–141 (2018).
Barton, E. J. et al. ExoMol molecular line lists V: the rovibrational spectra of NaCl and KCl. Mon. Not. R. Astron. Soc. 442, 1821–1829 (2014).
SousaSilva, C., AlRefaie, A. F., Tennyson, J. & Yurchenko, S. N. ExoMol line lists – VII. The rotation–vibration spectrum of phosphine up to 1500 K. Mon. Not. R. Astron. Soc. 446, 2337–2347 (2015).
Yorke, L., Yurchenko, S. N., Lodi, L. & Tennyson, J. Exomol molecular line lists – VI. A high temperature line list for phosphorus nitride. Mon. Not. R. Astron. Soc. 445, 1383–1391 (2014).
Prajapat, L. et al. ExoMol molecular line lists – XXIII. Spectra of PO and PS. Mon. Not. R. Astron. Soc. 472, 3648–3658 (2017).
Gorman, M. N., Yurchenko, S. N. & Tennyson, J. ExoMol molecular line lists XXXVI: X ^{2}Π – X ^{2}Π and A ^{2}Σ^{+} – X ^{2}Π transitions of SH. Mon. Not. R. Astron. Soc. 490, 1652–1665 (2019). (2019).
Upadhyay, A., Conway, E. K., Tennyson, J. & Yurchenko, S. N. ExoMol line lists XXV: a hot line list for silicon sulphide, SiS. Mon. Not. R. Astron. Soc. 477, 1520–1527 (2018).
Owens, A., Yachmenev, A., Thiel, W., Tennyson, J. & Yurchenko, S. N. ExoMol line lists – XXII. The rotation–vibration spectrum of silane up to 1200 K. Mon. Not. R. Astron. Soc. 471, 5025–5032 (2017).
Barton, E. J., Yurchenko, S. N. & Tennyson, J. ExoMol line lists – II. The rovibrational spectrum of SiO. Mon. Not. R. Astron. Soc. 434, 1469–1475 (2013).
Brady, R. P., Yurchenko, S. N., Kim, G. S., Somogyi, W. & Tennyson, J. An ab initio study of the rovibronic spectrum of sulphur monoxide (SO): diabatic vs. adiabatic representation. Phys. Chem. Chem. Phys. 24, 24076–24088 (2022).
Tennyson, J. et al. The 2020 release of the ExoMol database: molecular line lists for exoplanet and other hot atmospheres. J. Quant. Spectrosc. Radiat. Transfer 255, 107228 (2020).
Salvatier, J., Wiecki, T. V. & Fonnesbeck, C. Probabilistic programming in Python using PyMC3. PeerJ Comput. Sci. 2, e55 (2016).
The Astropy Collaboration et al.Astropy: a community Python package for astronomy. Astron. Astrophys. 558, A33 (2013).
The Astropy Collaboration et al.The Astropy Project: building an openscience project and status of the v2.0 core package. Astron. J 156, 123 (2018).
Hunter, J. D. Matplotlib: a 2D graphics environment. Comput. Sci. Eng. 9, 90–95 (2007).
Harris, C. R. et al. Array programming with NumPy. Nature 585, 357–362 (2020).
McKinney, W. pandas: a foundational Python library for data analysis and statistics. In PyHPC 2011: Python for High Performance and Scientific Computing Vol. 14, 1–9 (2011).
Hoyer, S. & Hamman, J. xarray: ND labeled arrays and datasets in Python. J. Open Res. Softw. 5, 10 (2017).
Espinoza, N. et al. Spectroscopic timeseries performance of JWST/NIRSpec from commissioning observations. Preprint at https://arxiv.org/abs/2211.01459 (2022).
Caffau, E., Ludwig, H. G., Steffen, M., Freytag, B. & Bonifacio, P. Solar chemical abundances determined with a CO5BOLD 3D model atmosphere. Sol. Phys. 258, 255–269 (2011).
Asplund, M., Grevesse, N., Sauval, A. J. & Scott, P. The chemical composition of the Sun. Preprint at https://arxiv.org/abs/0909.0948 (2022).
Lodders, K., Palme, H. & Gail, H.P. Abundances of the elements in the solar system. Preprint at https://arxiv.org/abs/0901.1149 (2022).
Acknowledgements
This work is based on observations made with the NASA/ESA/CSA JWST. The data were obtained from the Mikulski Archive for Space Telescopes at the Space Telescope Science Institute, which is operated by the Association of Universities for Research in Astronomy, Inc., under NASA contract NAS 503127 for the JWST. These observations are associated with programme JWSTERS01366. Support for programme JWSTERS01366 was provided by NASA through a grant from the Space Telescope Science Institute, which is operated by the Association of Universities for Research in Astronomy, Inc., under NASA contract NAS 503127. L.A. acknowledges funding from STFC grant ST/W507337/1 and from the University of Bristol School of Physics PhD Scholarship Fund.
Author information
Authors and Affiliations
Contributions
All authors played a substantial role in one or more of the following: development of the original proposal, management of the project, definition of the target list and observation plan, analysis of the data, theoretical modelling and preparation of this manuscript. Some specific contributions are listed as follows. N.M.B., J.L.B. and K.B.S. provided overall programme leadership and management. L.A. and H.R.W. led the efforts for this manuscript. D.K.S., E.M.R.K., H.R.W., I.J.M.C., J.L.B., K.B.S., L.K., M.L.M., M.R.L., N.M.B., V.P. and Z.K.B.T. made notable contributions to the design of the programme. K.B.S. generated the observing plan, with input from the team. E.S. and N.E. provided instrument expertise. B.B., E.M.R.K., H.R.W., I.J.M.C., J.L.B., L.K., M.L.M., M.R.L., N.M.B. and Z.K.B.T. led or coled working groups and/or contributed to important strategic planning efforts, such as the design and implementation of the prelaunch Data Challenges. A.L.C., D.K.S., E.S., N.E., N.P.G. and V.P. generated simulated data for prelaunch testing of methods. L.A., H.R.W., M.K.A., N.E.B. and J.D.L. contributed substantially to the writing of this manuscript, along with contributions in Methods from J.A.R., S.B., M.D., N.E., L.F., J.M.G., D.G., J.I., T.M.E., P.A.R. and N.L.W. L.A., H.R.W., M.K.A., J.A.R., S.B., M.D., N.E., L.F., D.G., J.I., T.M.E., P.A.R. and N.L.W. contributed to the development of dataanalysis pipelines and/or provided the dataanalysis products used in this analysis, that is, reduced the data, modelled the light curves and/or produced the planetary spectrum, with further contributions from J.Brande, T.D. and L.R.R. J.D.L., N.E.B., J.M.G., E.K.H.L. and R.H. generated theoretical model grids for comparison with data. H.R.W., J.D.L. and N.E.B. generated figures for this manuscript. M.L.M., K.D.C., N.P.G., L.K., M.L., J.I.M. and E.S. provided substantial feedback to the manuscript, coordinating comments from all other authors. T.D. is a LSSTC Catalyst Fellow, N.H.A. and A.D.F. are NSF Graduate Research Fellows, J.Kirk is an Imperial College Research Fellow, R.J.M., M.M., D.P., J.D.T. and L.W. are NHFP Sagan Fellows and B.V.R. is a 51 Pegasi b Fellow.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing interests.
Peer review
Peer review information
Nature thanks Norio Narita and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Peer reviewer reports are available.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Extended data figures and tables
Extended Data Fig. 1 The throughput and spectral trace for WASP39 across NRS1 and NRS2.
a, Normalized throughput of NRS1 and NRS2 detectors (as custom produced; see Methods, ‘Limbdarkening’), which shows the cutoff at short wavelengths. b, 2D spectral images of the trace produced from the ExoTiCJEDI [V1] reduction before cleaning steps. The aspect ratio has been stretched in the y direction to show the structure of the trace over the 32pixelwide subarray more clearly. The NRS2 spectral position is slightly offset from that of NRS1, as the NRS2 subarray was moved following commissioning to ensure that the centre of the spectral trace fell fully on the detector and did not fall off the topright corner^{139}.
Extended Data Fig. 2 Timedependent decorrelation parameters.
a, The change in the FWHM of the spectral trace at selected wavelengths. This change does not correspond to any highgain antenna movements and is attributed to a large mirrortilt event. These measurements demonstrate that the mirrortilt event has a wavelength dependence. Changes to the PSF have a larger impact at short wavelengths, as the PSF of the spectrum increases with wavelength^{139}. b,c, The change in the xpixel and ypixel position of the spectral trace as functions of time, respectively. Positional shifts are calculated by crosscorrelating the spectral trace with a template to measure subpixel movement on the detector. The yposition shift clearly shows a link to the mirrortilt event.
Extended Data Fig. 3 Normalized flux offset of the stellar baseline before and after the tilt event as a function of wavelength for NRS1 and NRS2.
Purple denotes NRS1 and orange denotes NRS2. The normalized flux offset is calculated per pixel by measuring the median flux in the stellar baseline before and after the transit and calculating the difference. These differences are then normalized by the beforetransit flux and plotted on a common scale. Overplotted are the data binned to a resolution of 10 pixels to match our presented transmission spectra (Fig. 2). We also show a linear fit to each detector to better quantify the decreasing tilt flux amplitude with increasing wavelength (NRS1 = −0.00073374x + 0.00707344, NRS2 = −0.00067165x + 0.00588128).
Extended Data Fig. 4 Normalized rootmeansquared binning statistic for three of the 11 reductions detailed in Methods.
In each subplot, the red line shows the expected relationship for perfect Gaussian white noise. The black lines show the observed noise from each spectroscopic light curve for pipelines 1, 3 and 5. To compare bins and noise levels, values for all bins in each pipeline are normalized by dividing by the value for a bin width of 1.
Extended Data Fig. 5 Comparison between all fitting pipelines performed on the spectroscopic light curves.
a, The underlying grey data points show the standard deviation between all transmission spectra per spectral bin. The black line shows the unweighted mean uncertainty on the transit depth per bin. Spikes in the uncertainties correspond to spectral bins with higher standard deviations, probably because of differences in pixelflagging or sigmaclipping at the lightcurve level. b, Gaussian PDFs of the normalized transmission spectrum residuals, showing the mean offset and the spread relative to the weightedaverage transmission spectrum. c, Histograms of the normalized transmission spectrum residuals aligned to zero by subtracting the mean of the distribution that was used to generate the PDF above. In panels b and c, the coloured lines and numbers correspond to the fitting pipeline used to obtain each transmission spectrum, as summarized in Extended Data Table 1. The dashed lines correspond to the fitting pipeline results presented in Fig. 2, demonstrating that they are drawn from across the distribution.
Extended Data Fig. 6 Gaussian versus flatline fits to the residual transmission spectrum for CO_{2}, H_{2}O, SO_{2} and the 4.56μm feature.
Shown after all other absorption from the bestfit model is subtracted from the data. Each of the Gaussian fits has a higher Bayesian evidence than the flatline fits, indicating a detection, although to varying degrees of significance.
Extended Data Fig. 7 Model transmission spectra of WASP39b with PHOENIX and gCMCRT with varying abundances of SO_{2}.
Model transmission spectra compared with the observed spectral feature at 4.1 μm in the G395H data. At wavelengths short of 3.95 μm, which is outside the SO_{2} band, all models overlap, further suggesting that the data can be explained by the presence of SO_{2} in the atmosphere. By interpolating these 10 times solar metallicity models, we find a bestfit SO_{2} abundance of 4.6 ± 0.67 ppm. With the bestfit PICASO 3.0 at 3 times solar metallicity, we find an SO_{2} abundance of 2.5 ± 0.65 ppm.
Supplementary information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Alderson, L., Wakeford, H.R., Alam, M.K. et al. Early Release Science of the exoplanet WASP39b with JWST NIRSpec G395H. Nature 614, 664–669 (2023). https://doi.org/10.1038/s41586022055913
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1038/s41586022055913
This article is cited by

A warm Neptune’s methane reveals core mass and vigorous atmospheric mixing
Nature (2024)

SO2, silicate clouds, but no CH4 detected in a warm Neptune
Nature (2024)

Sulfur dioxide in the midinfrared transmission spectrum of WASP39b
Nature (2024)

A benchmark JWST nearinfrared spectrum for the exoplanet WASP39 b
Nature Astronomy (2024)

Analogous response of temperate terrestrial exoplanets and Earth’s climate dynamics to greenhouse gas supplement
Scientific Reports (2023)
Comments
By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.