Asking the pikas politely…
Dissolved lithium isotope ratios (δ⁷Li_diss) trace the balance between primary silicate dissolution and secondary clay formation, but in mixed-lithology watersheds this signal is complicated by source mixing superimposed on process-driven isotopic fractionation. Disentangling these contributions remains a key challenge for interpreting riverine δ⁷Li_diss as a weathering proxy. Here we present δ⁷Li_diss and solute geochemistry data from 54 stream sites across two mid-summer baseflow campaigns (July 2024 and July 2025) in the East River watershed (Colorado, USA), a headwater system spanning shale, sandstone, and granodiorite. We apply lithologic mapping, thermodynamic modeling, and DEM-derived landscape analysis to identify the primary controls on δ⁷Li_diss variability. Across the full dataset, dissolved δ⁷Li_diss spans 7.7 to 25.9‰ and shows more scatter against Li/Na (another tracer of weathering congruency) than would be expected for a small, geologically coherent watershed (R² = 0.176, p = 0.0021, n = 51). A much tighter relationship emerges when restricting to shale-dominated catchments (R² = 0.579, p < 0.0001, n = 20), demonstrating that lithologic heterogeneity is a major source of scatter and that lithologic homogeneity is a prerequisite for using solute ratios as weathering proxies. Thermodynamic analysis reveals concurrent feldspar undersaturation and clay supersaturation across all lithologic classes, consistent with ongoing incongruent weathering. Batch fractionation calculations indicate that a median of 66% of released Li from silicates is sequestered into secondary phases (interquartile range: 52–78%), with an effective fractionation factor (α = 0.9772; Δ = -22.8‰) that varies systematically by lithologic class. Mancos Shale areal fraction correlates strongly with flow path geometry metrics in both campaigns (R² = 0.52–0.57, p < 0.0001), raising the possibility that lithology shapes δ⁷Li_diss partly by organizing hydrologic routing rather than through mineralogy alone. Disentangling these two distinct influences of lithology will require further work. Nevertheless, our results show that lithologic heterogeneity generates substantial δ⁷Li_diss variability in this upland, mixed-lithology system.
Items connected by shared entities, co-authorship, citations, or semantic similarity.