Enhanced Rock Weathering–Induced Carbon Dioxide Removal in Flooded Rice Paddies: Mineral-Based Monitoring from Field Experiments in Japan

Enhanced Rock Weathering (ERW) has attracted increasing attention as a carbon dioxide removal (CDR) approach, yet its implementation depends on monitoring, reporting, and verification (MRV) strategies that can reliably detect and attribute CO₂ removal under different field conditions. Flooded rice paddies are considered a potentially favorable environment for ERW due to continuous water fluxes, but strong hydrological and biogeochemical dynamics also complicate signal detection. Thus, two-year field experiments were conducted at four rice paddy sites in Japan with contrasting silicon (Si) supply capacities. Finely ground basalt was applied prior to cultivation at rates of 0, 100 (5 wt%), or 200 (10 wt%) t ha^-1. ERW of reactive minerals was quantified by directly tracking temporal changes in Ca-plagioclase using quantitative X-ray powder diffraction (XRPD). These mineral-based estimates were compared with cation-based estimates derived from X-ray fluorescence (XRF) measurements of total Ca loss. XRPD analyses revealed that plots amended with 10 wt% basalt exhibited significant reductions in Ca-plagioclase within one year after application across all sites, whereas such reductions were not consistently observed in the 5 wt% plots. These results provided direct field evidence of in situ mineral weathering under flooded conditions. Estimated CDR potentials for all basalt-amended plots derived from XRPD ranged from 1.1 to 9.3 t CO₂ ha^-1 yr^-1 and were broadly consistent with, but systematically higher than, XRF-based estimates. This discrepancy likely reflects the influence of external Ca inputs that can mask Ca depletion in XRF-based approach, as well as the uncertainties associated with estimated Ca-bearing minerals from idealized mineral stoichiometries in XRPD-based calculations. Notably, both methods consistently indicated higher weathering rates and CDR potentials at sites with lower initial Si availability. In parallel, basalt application increased plant-available Si and rice straw Si uptake over two growing seasons and was associated with reduced proportions of immature grains. However, the persistence of minerals after one year underscores the need for multi-year assessments. Overall, this study demonstrates that mineralogical monitoring provides a robust MRV pathway for ERW in dynamic paddy systems and that site selection is critical for reliable detection and attribution of CDR signals under field conditions.