MOONINDEX: A Python Library to Generate Spectral Indexes from the Moon Mineralogy Mapper (M3)

MOONINDEX: A PYTHON LIBRARY TO GENERATE SPECTRAL INDEXES FROM THE MOON MINERALGY MAPPER (M3). J. E. Suárez-Valencia1, A. P. Rossi2, F. Zambon3, C. Carli3, G. Nodjoumi3, 1Università degli Studi di Padova ([email protected]), 2Earthgraph GmbH, Istituto Nazionale di Astrofisica -INAF. Introduction: Spectral indexes are tools widely used to analyse the composition of planetary surfaces. many indexes have been formulated over the years to map the lunar surface, but there is no unified database for them. In this work we describe an Open‐Source Python package called MoonIndex, that recreates 38 indexes compiled from the literature, using data from the Moon Mineralogy Mapper (M3). M3 acquired data in the spectral interval between 0.45 and 3 μm, corresponding to the range where the major mafic minerals and water ice exhibit clear absorption features [1, 2]. A common method to approach spectral analysis is the use of spectral indexes or parameters, which are specific combinations of bands, or band operations, that highlight a specific portion of the spectrum and thus a mineralogical composition [3]. The spectral parameters are intrinsic to the mineral species, which means spectra are comparable regardless of the planetary body. In this work we compiled and recreated in Python 38 spectral indexes and RGB composites from the literature [4]. Data and methods: In this work we use the reflectance cubes of M3 retrieved from the Planetary Data System. The cubes have 83 bands, they have a spectral sampling of 0.02 μm between 0.5 and 1.5 μm, and 0.04 μm between 1.5 and 3 μm; and a spatial resolution of 110 m/pixel [5]. The cubes were map-projected in ISIS. MoonIndex follows several steps to generate the indexes. First, some anomalous bands are removed and the format of the data is adapted to Python, the user can also select a region of interest. Second, Gaussian and Fourier filters are applied to decrease the noise that affects the cubes. Third, the spectral continuum is removed using a convex hull or a second‐and‐first‐order fit. Fourth, the key parameters of the spectrum are retrieved (band centers, band depths, and band shoulders). Fifth, the indexes are calculated according to the formulation in the literature. Finally, an image file containing all the calculated indexes is created, which is ready to use in geographic information systems. We performed a test of the tool by processing cubes over Apollo Basin, Vallis Alpes, and Aristarchus Crater; which were all studied by previous authors using some of the indexes that we recreated in this work [6, 7, 8]. Results: The resemblance of our parameters with the ones in the literature varies. For the indexes that are formulated before the continuum removal, the results are consistent with the original formulations, with changes only in the spatial resolution and the noise patterns, both particular to differences in the instruments used. Nevertheless, for the indexes obtained after the continuum removal, we identified variations that are related to the methodologies used by every author. Figure 1: Comparison between the spectral indexes calculated in MoonIndex and the ones used by previous authors. In Figure 1 we compare our results for the band center and depth at 1 μm with the ones done by other authors [4], in our case we removed the continuum using the convex hull method, while the previous authors used the second‐and‐first‐order fit method. Figure 1a shows the ratio between the band centers, the major differences can be seen in red vertical lines and in the rims of big craters, these changes are re- lated to the filtering process and to shadows, meaning there are no major variations in the meaningful data. Major differences can be seen in the ratio of the band depth at 1 μm (Figure 1b). The red areas are widespread, and although most of them are due to the filtering of our data, changes are considerable in locations with clear signals of the surface. As the band area and asymmetry are both derived and linked to the band depth, our results also diverge in a similar way from the ones of other authors. This major discrepancy in the band depth compared to the band center is the result of using different continuum‐removal methods (Figure 2), which showcase the importance of this process during the analysis. Figure 2: Comparison of the spectral profiles using the two continuum removal methods. As for the RGB composites, the comparison presented certain difficulties. The source material from previous authors is not always available, therefore we could not properly configure certain parameters like the band stretch or the rendering method. Nevertheless, even when the specific colours and tonalities of the indexes may vary between works, the patterns of the geological features on the image and their differences remained identifiable. Conclusions: Spectral indexes are an easy and versatile way to approach the compositional analysis of the Moon. During our recreation of the spectral indexes in python we added certain improvements to the data of M3. The Gaussian and Fourier filtering proved useful to reduce the vertical striping typical of M3 cubes, allowing the retrieval of clearer spectra, especially from cubes that otherwise would be almost useless for geological interpretation. The fidelity of the reconstructed indexes varies for several reasons. The most important one is related to the use of the convex hull method to remove the continuum, opposite to the polynomial fits applied by previous authors. But other factors unreported in the literature likely affected the results, such as the preprocessing routines, filtering methods, or the visualization parameters of the composites. Nevertheless, despite some changes in tonalities and values, the reproduced indexes have a similar scientific meaning in all cases and highlight the same compositional properties as the original formulations. The indexes produced by MoonIndex are consistent with each other, but the methodologies and algorithms described in this work should be considered when comparing them with indexes from other works. MoonIndex is available for Python 3.10 and higher in the PyPI repository. The source code, exemplifying Jupyter notebooks, definition of functions, and workflows can be accessed via GitHub and Zenodo [4]. References: [1] Burns, R. G. (1993). Mineralogical applications of crystal field theory. https://doi.org/10.1017/cbo9780511524899. [2] Pieters, C. M., Hanna, K. L. D., Cheek, L. C., Dhingra, D., Prissel, T. C., Jackson, C. R., et al. (2014). The distribution of MG‐spinel across the moon and constraints on crustal origin. American Mineralogist, 99(10), 1893–1910. https://doi.org/10.2138/am‐2014‐4776. [3] Montero, D., Aybar, C., Mahecha, M. D., Martinuzzi, F., Sochting, M., & Wieneke, S. (2023). A standardized catalogue of spectral indices to advance the use of remote sensing in Earth system research. Scientific Data, 10(1), 197. https://doi.org/10.1038/s41597‐023‐02096‐0. [4] Suárez‐Valencia, J. E., Rossi, A. P., Zambon, F., Carli, C., & Nodjoumi, G. (2024). MoonIndex, an open‐source tool to generate spectral indexes for the Moon from M3 data. Earth and Space Science, 11, e2023EA003464.https://doi.org/10.1029/2023EA003464. [5] Green, R. O., Pieters, C., Mouroulis, P., Eastwood, M., Boardman, J., Glavich, T., et al. (2011). The Moon Mineralogy Mapper (M3) imagingspectrometer for lunar science: Instrument description, calibration, on‐orbit measurements, science data calibration and on‐orbit validation. Journal of Geophysical Research, 116, E00G19. https://doi.org/10.1029/2011JE003797. [6] Zambon, F., Carli, C., Altieri, F., Luzzi, E., Combe, J.‐P., Ferrari, S., et al. (2020). Spectral index and RGB maps—Beethoven, Rembrandt basins on Mercury, Apollo basin and Leibnitz and Von Karman craters regions on the Moon (p. 57). [7] Bretzfelder, J. M., Klima, R. L., Greenhagen, B. T., Buczkowski, D. L., Petro, N. E., & Day, M. (2020). Identification of potential mantle rocks around the lunar imbrium basin. Geophysical R search Letters, 47(22), e2020GL090334. https://doi.org/10.1029/2020GL090334. [8] Horgan, B. H. N., Cloutis, E. A., Mann, P., & Bell, J. F. (2014). Near‐infrared spectra of ferrous mineral mixtures and methods for their identification in planetary surface spectra. Icarus, 234, 132–154. https://doi.org/10.1016/j.icarus.2014.02.031