Investigation of Absorption Bands around 3.3 μm in CRISM Data

Absorptions in the range 3.1 μm to 3.6 μm are under the spotlight in the context of planetary research, because hydrocarbon molecules show absorption bands in this range. Consequently, even knowing that the Compact Reconnaissance Imaging Spectrometer for Mars (CRISM) was designed for the detection of mineralogical features on Mars’s surface, we exploited CRISM data in the range 3.2 μm to 3.4 μm to search for potential hydrocarbon compounds. To date, methane has been the only hydrocarbon detected on Mars. Therefore, we began our investigation into CRISM data in locations in which methane had been detected and where it could form due to the mineralogy of the specific site. The datasets chosen for this study included observation sites in the Oxia Planum, the Gale Crater, and Nili Fossae areas. We mapped the modified Gaussian model (MGM) to fit the CRISM data in order to extract the band parameters of the absorptions in the 3.3 μm spectral region. As a result, we found clusters of pixels with spectra that exhibited band centers between approximately 3.28 and 3.35 μm. The hydrocarbons showing absorptions in this range included polycyclic aromatic compounds as well as methane, ethane, and aliphatic compounds. We speculated that some absorptions of approximately 3.3 μm could be related to methane, so we calculated a theoretical lower limit of detection for each observation in the selected CRISM datasets. This was performed by simulating the CRISM spectra for the different sites, with diverse concentrations of CH4, using NASA’s Planetary Spectrum Generator online tool. These simulations established the relationship between the concentration and methane band depths, as detected by the CRISM. Methane band depths exceeding the thresholds varied from one observation to another, in the range of 0.0136 to 0.0237, which corresponded to a range of theoretically lower limits of concentration between 180 and 600 ppbv. Although we could not confirm or deny the occurrences of methane seepages or hydrocarbons in the investigated datasets, we demonstrated a possible method for searching for hydrocarbons in other CRISM data and for assessing a confidence limit in the detection of the methane band in CRISM data.