Search

The Ariel Space Mission will observe a large and diverse sample of exoplanetary atmospheres in the 0.5 to 7.8-micron range of the electromagnetic spectrum. As part of the Ariel observing programme, a shallow Reconnaissance survey (Tier 1) will provide transiting and eclipse spectroscopy on about 1000 targets, with low spectral resolution but sufficient SNR to identify the signature of molecular species. The wealth of information provided by this survey will be the basis for promoting targets for re-observation to reach sufficient SNR at higher spectral resolution. At the same time, these low spectral resolution observations are not suitable for estimating molecular abundances with an appropriate confidence level. Therefore, it is paramount to develop special data analysis techniques to extract their information content. This work investigates using the abundance posteriors from spectral retrieval as an unbiased metric to assess the presence of a molecule up to a certain threshold. The experimental dataset comprises simulated Tier 1 transmission spectra for about 300 targets from the Ariel Mission Reference Sample produced using the Alfnoor software. We use the TauRex 3 retrieval framework to run spectral retrievals on each “observed” spectrum, and we compute the probability that the spectra bear a molecule by integrating the posteriors above a specified threshold of molecular concentration. We find that the retrieved probabilities correlate with the abundances in the forward models and that this method is statistically reliable and has considerable predictive power and diagnostic ability. The predictive power is not significantly affected by adding molecules in the fitted composition that are not present in the forward models, while omitting molecules should be discouraged as it can lead to biased results.

ExoSim 2 is a time-domain simulator for exoplanet observations. The software can simulate exoplanetary transit, eclipse and phase curve observations from ground and space-based instruments. Such simulation can capture temporal effects, such as correlated noise and systematics on the light curve. The simulator will produce spectral images like those produced by an actual observation. ExoSim 2 has been developed for the Ariel Space Mission, to assess the impact of astronomical and instrumental systematic on astrophysical measurement, and to prepare the data reduction pipeline against realistic data sets. ExoSim 2 output can be utilised by different data reduction methods, not only to determine the best pipeline strategy to remove the systematics in the measurements but also to assess the confidence level of retrieved quantities. ExoSim 2 is a refactored version of ExoSim: an end-to-end simulator that models noise and systematics in a dynamical simulation. The first version of ExoSim (Sarkar et al. 2020) was developed for the Ariel Space Mission, then adapted to the James Webb Telescope and presented to the community as JexoSim (Sarkar et al. 2019 and Sarkar et al. 2021). ExoSim 2 is meant to be easier to use than its predecessor and largely customizable. It is completely written in Python, tested against Python 3.7+, and follows the object-oriented philosophy. It comes with an installer, documented examples, a comprehensive guide, and almost every part of the code can be replaced by a user-defined function, which allows the user to include new functionalities to the simulator. We believe that ExoSim 2 is a versatile tool, which can be used for the development of instruments other than Ariel, or to assess the impact of different astronomical or instrumental systematics.