This is a Preprint and has not been peer reviewed. This is version 3 of this Preprint.
Decoupling Kinetic and Hydrological Controls on Global Silicate Weathering via a Stratified Observational Factorial Framework
Downloads
Supplementary Files
Authors
Arpan Halder 
Abstract
Continental silicate weathering acts as the primary geological sink for atmospheric
carbon dioxide over multimillion-year timescales, yet whether weathering fluxes are
fundamentally governed by thermodynamic temperature limits or hydrological runoff
remains fiercely debated. Resolving this controversy globally has been hindered by the
strong natural collinearity of terrestrial temperature and precipitation. Here, we
decouple these drivers by applying a stratified 3×3 factorial response-surface
framework to 5,292 harmonised modern river sediment samples compiled from global
geochemical databases and WorldClim grids. Two-way analysis of variance reveals
that temperature exerts a substantially larger main effect (F = 163.34, partial η2 =
0.058) than precipitation (F = 87.33, partial η2 = 0.032). Furthermore, significant
interaction curvature (p < 0.00001) uncovers an asymptotic saturation plateau in warm,
hyper-humid extremes: shifting from moderate to high rainfall in the warmest tier yields
no alteration gain (ΔCIA = −0.77, p = 0.42; ΔCIX = −0.44, p = 0.58). Molar cation
preservation ratios confirm that this plateau is driven by primary feldspar bypass
resulting from monsoonal regolith stripping, establishing critical non-linear constraints
for Earth System Models.
DOI
https://doi.org/10.31223/X5PN6F
Subjects
Design of Experiments and Sample Surveys, Geochemistry
Keywords
Silicate Weathering, Chemical Index of Alteration (CIA), Factorial ANOVA, Carbon-Silicate Cycle, Earth System Models, Paleoclimatology, Paleoclimatology, Chemical Index of Alteration (CIA), Factorial ANOVA, Carbon-Silicate Cycle, Earth System Models
Dates
Published: 2026-09-17 16:37
Last Updated: 2026-09-25 17:22
Older Versions
License
CC BY Attribution 4.0 International
Additional Metadata
Conflict of interest statement:
the author declares no competing interests related to this work
Data Availability:
The harmonized, DoE-ready global river sediment dataset generated and analyzed in this study, along with Python scripts for ANOVA are publicly archived on the Zenodo repository under DOI: 10.5281/zenodo.22760321. Raw benchmark compilations are openly accessible via Zenodo Record 6066701 (Deng et al., 2022) and Record 4485795 (Müller et al., 2021).
Metrics
Views: 374
Downloads: 68
There are no comments or no comments have been made public for this article.