Inferring Titan Atmosphere Methane Content From Dragonfly Backshell Radiation Measurements

During the entry of the Dragonfly capsule into the atmosphere of Titan, a variety of instruments will record vehicle surface properties including radiation. Uncertainty in CN emission remains a principal limitation in modeling and interpreting radiative heating during Titan atmospheric entry. Prior work has shown that the ratio of integrated CN red and CN violet band radiances can, in principle, be used to infer the freestream methane mole fraction when only chemical kinetic uncertainty is considered. In this study, we extend that framework to include uncertainty in the heavy-particle excitation rates governing the CN(A–X) and CN(B–X) band systems. The rates of these excitation processes are difficult to measure experimentally and exhibit discrepancies of multiple orders of magnitude across available datasets. Using polynomial chaos expansion surrogate models coupled with Bayesian inference in PyMC, we examine how methane identifiability evolves as excitation-rate uncertainty bounds are systematically tightened. We find that current uncertainty levels produce broad, non-informative methane posteriors, even when radiative measurements are assumed noise-free, but that progressive reduction of the dominant excitation-rate uncertainties yields a clear transition toward identifiable and increasingly concentrated posteriors. Analyses at both peak convective and peak radiative heating conditions on the entry trajectory demonstrate the degree of uncertainty reduction required before meaningful methane retrieval becomes feasible. These results provide a quantitative mapping between excitation-rate precision and methane identifiability, underscoring the need for targeted laboratory and theoretical studies to enable robust methane inference during Dragonfly’s atmospheric entry.