Compositional Mapping of Ganymede with VLT/SPHERE using Markov Chain Monte Carlo Spectral Analysis

Datasets: Near-IR observations of Ganymede were acquired with the ground-based VLT/SPHERE instrument in 2015 and 2021, producing spectral cubes covering 0.95 to 1.65 m (at R30) and narrow-band images at 2.11 and 2.25 m. SPHERE has a high spatial resolution, with a pixel size of 7.46 mas/px, corresponding to 25 km/px at Jupiter. Accounting for diffraction, this allows features <150 km across to be resolved. The dataset has been mapped and photometrically corrected using the Oren-Nayar photometric correction, allowing accurate mapping up to emission angles of ~70. Galileo/NIMS observations with similar spatial coverage have also been analysed to provide a direct comparison to the SPHERE dataset. Spectral modelling: We analyse the mapped cubes by fitting to laboratory spectra from reference cryogenic libraries, including water ice, sulphuric acid, hydrated salts, and spectrally flat synthetic spectra. Our fitting routine uses Markov Chain Monte Carlo (MCMC) techniques to model observed spectra, producing a posterior distribution of fitted abundance values for each endmember at each observed location. These posterior distributions can be sampled to calculate the best estimate abundance and its associated uncertainty for each endmember and class of endmembers, allowing more detailed investigation of the confidence of different detections. Results: Initial results show strong similarities where the SPHERE and NIMS datasets overlap and are consistent with previously observed compositional features. These include Ganymedes younger brighter terrain having higher water ice abundances and the older terrain having a high abundance of spectrally flat (i.e., grey) material. This spectrally flat material has a uniformly low albedo and is consistent with higher silicate abundance in Ganymedes old dark terrain. This near-IR spectral modelling complements longer wavelength IR and microwave observations from Juno. The full 2021 SPHERE observing campaign will enable compositional mapping of almost all (~95%) of Ganymedes surface at ~100km spatial resolution. Acknowledgments: We would like to thank the Royal Society for supporting this work.