New Maps of Lunar Surface Oxide Abundances and Mg# Using an Optimized Ensemble Learning Algorithm

The oxide abundance and Mg# (Mg/(Mg + Fe)) of the lunar surface are critical for understanding the Moon's petrology and evolution. Previous studies primarily relied on a single remote sensing data source and traditional regression algorithms for lunar oxide abundance inversion, potentiall...

Full description

Saved in:
Bibliographic Details
Main Authors: Chaofa Bian, Kefei Zhang, Yunzhao Wu, Suqin Wu, Yu Lu, Yabo Duan, Huajing Wu, Zhenxing Zhao, Wei Wu
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10919024/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849697462138175488
author Chaofa Bian
Kefei Zhang
Yunzhao Wu
Suqin Wu
Yu Lu
Yabo Duan
Huajing Wu
Zhenxing Zhao
Wei Wu
author_facet Chaofa Bian
Kefei Zhang
Yunzhao Wu
Suqin Wu
Yu Lu
Yabo Duan
Huajing Wu
Zhenxing Zhao
Wei Wu
author_sort Chaofa Bian
collection DOAJ
description The oxide abundance and Mg# (Mg/(Mg + Fe)) of the lunar surface are critical for understanding the Moon&#x0027;s petrology and evolution. Previous studies primarily relied on a single remote sensing data source and traditional regression algorithms for lunar oxide abundance inversion, potentially compromising the accuracy of chemical composition results. In this study, multisource data, specifically KAGUYA multiband imager (MI) data and Diviner Christiansen feature (CF) products, referred to as MI-CF, combined with an ensemble learning algorithm optimized by improved particle swarm optimization (IPSO), were utilized to produce new maps of six oxides (FeO, TiO<sub>2</sub>, Al<sub>2</sub>O<sub>3</sub>, CaO, MgO, and SiO<sub>2</sub>) and Mg# at a resolution of 59 m/pixel within the Moon&#x0027;s 65&#x00B0; N/S range. The models were constructed using oxide abundances measured from lunar samples collected by the Apollo, Luna, and Chang&#x0027;E missions. Among the models tested, the SXL algorithm (stacking of support vector machine regression, extreme gradient boosting, and linear regression), which was selected from a stack of 2 or 3 out of six typical algorithms, achieved the highest inversion accuracy. The SXL algorithm&#x0027;s accuracy was further enhanced using IPSO with nonlinear weights and learning factors, resulting in a significant improvement over previous studies. The results confirm that the IPSO-SXL approach effectively improves the accuracy of lunar oxide inversion. Furthermore, the study demonstrates that using MI-CF data together yields better results than using MI or CF data alone. The oxide and Mg# distributions in three craters from different geological units&#x2014;Highlands, SPA basin, and Maria&#x2014;indicate different magmatic evolution processes, providing new insights into lunar geological evolution and volcanism.
format Article
id doaj-art-843d0ab5d3dc4222b3dcd7f19bec4132
institution DOAJ
issn 1939-1404
2151-1535
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
spelling doaj-art-843d0ab5d3dc4222b3dcd7f19bec41322025-08-20T03:19:12ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352025-01-01189119913410.1109/JSTARS.2025.354969110919024New Maps of Lunar Surface Oxide Abundances and Mg# Using an Optimized Ensemble Learning AlgorithmChaofa Bian0https://orcid.org/0000-0001-6979-349XKefei Zhang1https://orcid.org/0000-0001-9376-1148Yunzhao Wu2https://orcid.org/0000-0001-8408-1204Suqin Wu3https://orcid.org/0000-0002-0994-402XYu Lu4https://orcid.org/0000-0003-1884-1724Yabo Duan5https://orcid.org/0009-0002-8370-1277Huajing Wu6https://orcid.org/0000-0002-7248-680XZhenxing Zhao7Wei Wu8School of Environment Science and Spatial Informatics, China University of Mining and Technology, Xuzhou, ChinaSchool of Environment Science and Spatial Informatics, China University of Mining and Technology, Xuzhou, ChinaPurple Mountain Observatory, Chinese Academy of Sciences, Nanjing, ChinaSchool of Environment Science and Spatial Informatics, China University of Mining and Technology, Xuzhou, ChinaPurple Mountain Observatory, Chinese Academy of Sciences, Nanjing, ChinaSchool of Environment Science and Spatial Informatics, China University of Mining and Technology, Xuzhou, ChinaSchool of Environment Science and Spatial Informatics, China University of Mining and Technology, Xuzhou, ChinaState Key Laboratory of Space Weather, National Space Science Center, Chinese Academy of Sciences, Beijing, ChinaSchool of Marine Technology and Geomatics, Jiangsu Ocean University, Lianyungang, ChinaThe oxide abundance and Mg# (Mg/(Mg + Fe)) of the lunar surface are critical for understanding the Moon&#x0027;s petrology and evolution. Previous studies primarily relied on a single remote sensing data source and traditional regression algorithms for lunar oxide abundance inversion, potentially compromising the accuracy of chemical composition results. In this study, multisource data, specifically KAGUYA multiband imager (MI) data and Diviner Christiansen feature (CF) products, referred to as MI-CF, combined with an ensemble learning algorithm optimized by improved particle swarm optimization (IPSO), were utilized to produce new maps of six oxides (FeO, TiO<sub>2</sub>, Al<sub>2</sub>O<sub>3</sub>, CaO, MgO, and SiO<sub>2</sub>) and Mg# at a resolution of 59 m/pixel within the Moon&#x0027;s 65&#x00B0; N/S range. The models were constructed using oxide abundances measured from lunar samples collected by the Apollo, Luna, and Chang&#x0027;E missions. Among the models tested, the SXL algorithm (stacking of support vector machine regression, extreme gradient boosting, and linear regression), which was selected from a stack of 2 or 3 out of six typical algorithms, achieved the highest inversion accuracy. The SXL algorithm&#x0027;s accuracy was further enhanced using IPSO with nonlinear weights and learning factors, resulting in a significant improvement over previous studies. The results confirm that the IPSO-SXL approach effectively improves the accuracy of lunar oxide inversion. Furthermore, the study demonstrates that using MI-CF data together yields better results than using MI or CF data alone. The oxide and Mg# distributions in three craters from different geological units&#x2014;Highlands, SPA basin, and Maria&#x2014;indicate different magmatic evolution processes, providing new insights into lunar geological evolution and volcanism.https://ieeexplore.ieee.org/document/10919024/Chang'e-5 (ce-5)christiansen feature (CF)improved ensemble learning algorithmKAGUYA multiband imager (MI)lunar oxide
spellingShingle Chaofa Bian
Kefei Zhang
Yunzhao Wu
Suqin Wu
Yu Lu
Yabo Duan
Huajing Wu
Zhenxing Zhao
Wei Wu
New Maps of Lunar Surface Oxide Abundances and Mg# Using an Optimized Ensemble Learning Algorithm
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Chang'e-5 (ce-5)
christiansen feature (CF)
improved ensemble learning algorithm
KAGUYA multiband imager (MI)
lunar oxide
title New Maps of Lunar Surface Oxide Abundances and Mg# Using an Optimized Ensemble Learning Algorithm
title_full New Maps of Lunar Surface Oxide Abundances and Mg# Using an Optimized Ensemble Learning Algorithm
title_fullStr New Maps of Lunar Surface Oxide Abundances and Mg# Using an Optimized Ensemble Learning Algorithm
title_full_unstemmed New Maps of Lunar Surface Oxide Abundances and Mg# Using an Optimized Ensemble Learning Algorithm
title_short New Maps of Lunar Surface Oxide Abundances and Mg# Using an Optimized Ensemble Learning Algorithm
title_sort new maps of lunar surface oxide abundances and mg using an optimized ensemble learning algorithm
topic Chang'e-5 (ce-5)
christiansen feature (CF)
improved ensemble learning algorithm
KAGUYA multiband imager (MI)
lunar oxide
url https://ieeexplore.ieee.org/document/10919024/
work_keys_str_mv AT chaofabian newmapsoflunarsurfaceoxideabundancesandmgusinganoptimizedensemblelearningalgorithm
AT kefeizhang newmapsoflunarsurfaceoxideabundancesandmgusinganoptimizedensemblelearningalgorithm
AT yunzhaowu newmapsoflunarsurfaceoxideabundancesandmgusinganoptimizedensemblelearningalgorithm
AT suqinwu newmapsoflunarsurfaceoxideabundancesandmgusinganoptimizedensemblelearningalgorithm
AT yulu newmapsoflunarsurfaceoxideabundancesandmgusinganoptimizedensemblelearningalgorithm
AT yaboduan newmapsoflunarsurfaceoxideabundancesandmgusinganoptimizedensemblelearningalgorithm
AT huajingwu newmapsoflunarsurfaceoxideabundancesandmgusinganoptimizedensemblelearningalgorithm
AT zhenxingzhao newmapsoflunarsurfaceoxideabundancesandmgusinganoptimizedensemblelearningalgorithm
AT weiwu newmapsoflunarsurfaceoxideabundancesandmgusinganoptimizedensemblelearningalgorithm