¹C. Wang,^1,2T. Usui,^3,4M. Melwani Daswani
Journal of Geophysical Research: Planets (in Press) Open Access Link to Article [https://doi.org/10.1029/2025JE009358]
¹Department of Earth and Planetary Science, Graduate School of Science, The University of Tokyo, Tokyo, Japan,
²Institute of Space and Astronautical Science, Japan Aerospace Exploration Agency, Sagamihara, Japan,
³The SETIInstitute, Mountain View, CA, USA,
⁴Earth‐Life Science Institute, Institute of Science Tokyo, Tokyo, Japan
Published by arrangement with John Wiley & Sons

Feldspar-rich rocks have increasingly been discovered on the martian surface. They may have been an important part of the ancient martian crust and may be related to Ca/Fe-rich carbonates (one of two types of carbonates on Mars and the other being Mg-rich carbonates), but compared to mafic rocks, their interaction with water on ancient Mars is poorly understood. We conducted 1-D thermochemical modeling to determine how mafic or feldspar-rich rock composition controls the products of aqueous alteration on ancient Mars, with a focus on carbonates, considering the effects of groundwater flow and alteration duration in low-temperature environments. Evaporation of the alteration fluid was also simulated. We found that protolith composition, fluid transport process, and duration of alteration together control the composition, abundance, and distribution of carbonates and other secondary minerals. A causal link may exist between feldspar-rich rocks and some Ca/Fe-carbonates on Mars: Dominantly Mg-rich carbonates form only from mafic protoliths, while Ca/Fe-carbonates can form from either a feldspar-rich protolith generally or from a mafic protolith with a short alteration process. Percolation of atmospheric CO2-equilibrated water also provides a mechanism to suppress surface carbonate formation, dissolve shallow subsurface carbonates, and bury them deeply underground. These idealized scenarios employ simplified assumptions (equilibrium precipitation, laboratory dissolution rates, and specified transport). Absolute timescales are uncertain, so we focus on robust qualitative controls. The simulations demonstrate that crustal heterogeneity can explain the observed dichotomy and that carbonate composition may indicate protolith composition where direct detection is difficult.