The Detection-versus-retrieval Challenge: Titan as an Exoplanet

Abstract Cassini observations of Titan’s atmosphere are exemplary benchmarks for exoplanet atmospheric studies owing to their precision, their spectral coverage, and our independent knowledge of Titan. We carry out atmospheric retrievals of Titan’s transmission spectrum to test the sensitivity of interpretations to the range of molecular species included in the retrieval (i.e., set a priori as “detectable”). We find that hydrocarbons (HCs) such as ethane or isopropane can be “retrieved” from Titan’s 3.3 μ m band. However, given that a multitude of HC species exhibit overlapping features in this region, such retrievals cannot be claimed as bona fide “detections,” particularly given the limited accuracy and availability of associated cross sections. The choices of HCs used in the retrieval process can vary the retrieved abundance of key absorbers like methane by ∼0.5 dex (a factor of ∼3). This underscores the broader issue that beyond the possible misidentification of molecular features (e.g., the inventory debate surrounding dimethyl sulfide in K2-18 b), the inventory of molecular species, often dictated by computational considerations, can bias the retrieved atmospheric parameters. We thus recommend a sensitivity analysis to assess the dependencies of atmospheric inferences on such a priori selections, in tandem with complementary information (e.g., chemistry models). Finally, we show an independent path to constrain the dominant atmospheric constituent, even when lacking observable absorption features (e.g., H 2 and N 2 ) through scale height.