Looking for minor spectral features on 67P

Thursday, 18 December 2014
Stéphane Erard1, Fabrizio Capaccioni2, Gianrico Filacchione3, Cedric Leyrat4, Federico Tosi3, Maria Cristina De Sanctis3, Dominique Bockelée-Morvan1 and Pierre Drossart5, (1)Paris Observatory, Paris, France, (2)Organization Not Listed, Washington, DC, United States, (3)IAPS-INAF, Rome, Italy, (4)Paris Observatory Meudon, Meudon, France, (5)LESIA Observatoire de Paris, Meudon, France
Observations of 67P/Churyumov-Gerasimenko carried out by the VIRTIS spectrometer in the pre-landing phase of the Rosetta mission are analyzed to infer composition of the surface and coma [1]. Minerals, dust, ice mixtures and organic materials produce characteristic absorptions at the surface; similarly, typical emission features result from the de-excitation of the main gases in the coma. However, like all dark objects, the surface of 67P displays very reduced spectral contrast, and coma emissions are equally difficult to characterize at large solar distance.

The MULTiScale Unimixing Method for Extraction of Spectral Components (Mulțumesc) was devised to address this problem: extracting small spectral features in the presence of noise, and providing quantitative detection criteria [2]. The method does not assume any spectral characteristics or composition of the object, and only requires a realistic estimate of the noise level. It has previously been used to confirm marginal detections of ices on TNOs [2], and to derive an upper limit on the abundance of pyroxenes on Mercury [3].

The method is applied to the M–IR (1-5 µm, R~300) and H (2-5 µm, R~3000) channels of VIRTIS. In the case of the H channel, which is an échelle spectrometer, the method is applied to individual grating orders in order to preserve spectral resolution. Spectral features are required to have a minimum spectral and spatial extension in order to filter transient effects in the detector. The noise estimate is derived from the analysis of dark current measurements. Longward of ~3.5 µm, spectra of the surface are first inverted to corrected from thermal emission [4]; this is needed only to study broad features which could be significantly modified by the onset of the thermal emission. Results from the pre-landing phase will be presented at the meeting.

VIRTIS is supported by ASI, CNES and DLR. Support from the Rosetta and VIRTIS science, instrument, and operation teams is gratefully acknowledged.


[1] Coradini, A., et al. (2011) Science 334, 492-494 [2] Erard S. (2013) EPSC2013-520 [3] Erard S. et al. (2011) PSS 49, 1842-1852 [4] Erard S. (2014) EPSC2014-482