Towards a global TiO2 map of the lunar surface: machine learning predictions from integrated spectral datasets
Journal of Logic and Computation · 02.11.2026 · 36 (8)
DOI: 10.1093/logcom/exag053 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Logic and Computation, online erschienen am 13.11.2026. Verfasst von Julia Fernández Díaz und 5 weiteren Autor:innen. Die Arbeit nennt 23 Literaturangaben.
Abstract (Originalsprache)
This study presents a global mapping of titanium dioxide (TiO$_{2}$) concentration on the lunar surface using machine learning and spectral data integration. A predictive model based on multivariate adaptive regression splines was trained by combining remote sensing data from the Clementine mission, compositional analyses of Apollo lunar samples and laboratory spectra from the RELAB database. The consistency between Clementine and RELAB datasets enabled the establishment of a robust relationship between spectral reflectance and TiO$_{2}$ content. Model performance was independently validated using a dataset of 34 lunar samples analysed with data from the Moon Mineralogy Mapper instrument aboard Chandrayaan-1. The validated model was applied to generate a global TiO$_{2}$ distribution map, allowing the analysis of spatial heterogeneity across different lunar geological units. The results highlight the potential of integrating spectrophotometric data and machine learning techniques for extensive-scale lunar mineralogical mapping, with implications for understanding regolith evolution and supporting future exploration and in-situ resource utilization.
Zitieren
APA
Díaz, J. F., Cuervo-Fernández, E., Lasheras, F. S., Rodríguez, J. G., Iglesias, S., & Juez, F. J. D. C. (2026). Towards a global TiO2 map of the lunar surface: machine learning predictions from integrated spectral datasets. Journal of Logic and Computation, 36(8). https://doi.org/10.1093/logcom/exag053
Vancouver
Díaz JF, Cuervo-Fernández E, Lasheras FS, Rodríguez JG, Iglesias S, Juez FJDC. Towards a global TiO2 map of the lunar surface: machine learning predictions from integrated spectral datasets. Journal of Logic and Computation. 2026;36(8). doi:10.1093/logcom/exag053
BibTeX
@article{Daz2026,
author = {Díaz, Julia Fernández and Cuervo-Fernández, Eva and Lasheras, Fernando Sánchez and Rodríguez, Javier Gracia and Iglesias, Santiago and Juez, Francisco Javier de Cos},
title = {Towards a global TiO2 map of the lunar surface: machine learning predictions from integrated spectral datasets},
journal = {Journal of Logic and Computation},
year = {2026},
volume = {36},
number = {8},
doi = {10.1093/logcom/exag053}
}
EUVIMED