Lunar Sample Geochemistry: Astromat Synthesis of 58,289 Analyses of Apollo, Luna and Chang'e 5 Moon Rocks and Soils

Provider
Astromat (EarthChem Library); Julien Simon (Hugging Face conversion)
License
CC-BY-4.0 (source)
Size
4 files
Format
JPEG, Parquet (in sampled files)
Stored at
Hugging Face (https://huggingface.co/datasets/juliensimon/lunar-sample-geochemistry)
Access
Free and public. No account with the data provider is needed.
DOI
10.60520/IEDA/113696
Category
Astronomy & Space
Source
https://doi.org/10.60520/IEDA/113696

About this dataset

The chemical composition of rocks and soils returned from the Moon, compiled from 980 peer-reviewed publications (1970 to 2025) by the Astromat Synthesis project (Astromat Synthesis Compilation: Lunar samples v.1, February 2025, EarthChem Library, DOI 10.60520/IEDA/113696) and converted here into one Parquet table. It holds 58,289 geochemical analyses of 14,379 unique samples from the Apollo missions, the Soviet Luna sample-return missions and China's Chang'e 5. Sample types: breccia (17,455 analyses), basalt (15,949), soil (12,703), crustal rocks (1,713), core samples (1,179) and unclassified rocks. The 55 columns cover sample metadata and references, major oxides in weight percent (SiO2, TiO2, Al2O3, FeO, MgO, CaO and others), trace elements and rare earth elements. Uses: classifying lunar basalts (high-Ti, low-Ti), studying the lunar crust, mantle and KREEP, comparing mare and highland compositions, and ground truth for orbital remote-sensing maps. Load with pandas or Hugging Face datasets (data/lunar_geochemistry.parquet). Astromat releases the compilation under CC BY 4.0; cite the Astromat compilation and the original publications.

How to load this dataset in Python

from datasets import load_dataset
ds = load_dataset("juliensimon/lunar-sample-geochemistry")
print(ds)

Data structure

File types in sampled files: JPEG, Parquet

Files

Example file: data/lunar_geochemistry.parquet

Analyze this dataset without downloading it

On Scigantic, Lunar Sample Geochemistry: Astromat Synthesis of 58,289 Analyses of Apollo, Luna and Chang'e 5 Moon Rocks and Soils opens in a hosted JupyterLab notebook with the data already mounted, so you can read files where they are stored instead of copying them first. Each notebook includes an AI assistant that can see the mounted files and write and run analysis code. Open the dataset page to launch a notebook, or browse the full catalog.

This page is also available as Markdown and as JSON.

Related Astronomy & Space datasets