Correcting stellar contamination in transmission spectroscopy with contemporaneous monitoring: Application to GJ 1214 b

Transmission spectroscopy has emerged as an essential tool for characterising the atmospheres of transiting exoplanets. However, stellar surface inhomogeneities contaminate transmission spectra, which poses a major challenge, particularly to high-precision observations such as those from the (JWST). James Webb Space Telescope We developed and applied a set of practical methodologies to correct for stellar contamination in transmission spectroscopy data, thereby improving the accuracy of exoplanet atmospheric characterisation. We obtained coordinated multi-band photometry and high-resolution spectroscopy around JWST observations of GJ,1214,b and applied two correction frameworks: (i) an inversion with the StarSim code, in which we infer stellar activity parameters from the monitoring data and propagate them to corrections; and (ii) , a data-driven approach in which neural networks (NNs) are trained on a large suite of simulations to learn the mapping from the monitoring data to the corrections. StarSim Both frameworks agree within the uncertainties and indicate that the JWST observations of GJ,1214 occurred at a favourable rotation phase with low stellar contamination. For , the inferred corrections can be constrained at the ∼10--30 ppm level across the JWST wavelength range, implying post-correction residuals of the same order. If the same system were observed near activity maximum, the predicted contamination would be strongly chromatic and substantially affect atmospheric retrievals, highlighting the importance of coordinated ground-based monitoring and robust activity correction techniques to accurately interpret transmission spectra. The two approaches are complementary but serve different roles. The inversion yields physically interpretable parameters and surface maps, whereas marginalises more efficiently over surface-configuration degeneracies and provides tighter predictive corrections once trained and validated within the relevant simulation domain. The NNs carry an upfront training cost, but then enable fast predictions within their training distribution. Our pipeline can potentially be generalised to other active stars: with multi-band and multi-technique monitoring around JWST visits, it can provide activity-level forecasts for scheduling and activity corrections for atmospheric analyses.