¹S.P. Goudy, ¹M. Telus, ²K. Nagashima, ²G.R. Huss
Geochimica et Cosmochimica Acta (in Press) Link to Article [https://doi.org/10.1016/j.gca.2026.04.014]
¹Earth and Planetary Sciences, University of California at Santa Cruz, 1156 High Street, Room A232, Santa Cruz, CA 95064, United States
²Hawaii Institute of Geophysics and Planetology, University of Hawai’i at Mānoa, 1680 East-West Road, POST Building, Office 602, Honolulu, HI 96822, United States
Copyright Elsevier

Here we present petrographic, O isotope, and C isotope data on calcite and petrographic and O isotope data on magnetite in Aguas Zarcas (CM2) and Miller Range (MIL) 13005 (CM1/2) in an effort to test CM aqueous alteration models. Our O and C isotope data for Aguas Zarcas and MIL 13005 are within the ranges reported in previous work for calcite in CMs. Using O isotope data from Δ17O-matched calcite and magnetite grains in our samples, we calculated equilibrium-model formation temperatures for the analyzed calcites in each meteorite. Combining our isotopic and temperature data with literature data, we sort the data into less altered (CM2) and more altered (CM1/2 and CM1) categories, and examine that data for differences between the categories by analyzing the δ18O-δ17O, δ18O-δ13C, and model formation temperature data of the categories. We find potential differing cluster patterns between δ18O and δ13C in calcites between our two CM alteration categories, and find that sparse extant temperature data imply that more altered CMs (types 1 and 1/2) may have undergone alteration at a lower average temperature than CM2s. We also find that the O isotopic compositions between CM1s, CM2s, and CM1/2s do not differ significantly. Through use of a novel mass-balance model, we infer a pre-alteration ice Δ17O within the range of 5.3–13.3‰. We find two generations of calcite formation in MIL 13005 with different proportions of their O being sourced from ices and anhydrous silicates, and one generation of calcites within Aguas Zarcas. We created concentration mixing models through an original machine-learning-based analytical approach, which show that the calcite data can be readily explained using three optimally determined C- and O-bearing sources.