SILICATE WEATHERING INTENSITY IN FLOODPLAINS ACROSS THE PALEOCENE-EOCENE THERMAL MAXIMUM INFLUENCED BY PALEO-LANDSCAPE POSITION

We present new geochemical and isotopic data on fluvial deposits that span the Paleocene-Eocene thermal maximum in the Bighorn Basin (Wyoming, USA) to quantify silicate weathering intensities throughout the hyperthermal event: a 190 kyr-long event where global average temperatures increased by 5-8 °C and atmospheric pCO₂ doubled. We determine silicate weathering intensities by comparing the Li isotope compositions of the clay-sized (< 2 µm) fraction in pedogenically-modified mudstones to those of shales and igneous conglomerate clasts that crop out in the basin. Mixing models using immobile element ratios (Ti, Cs, Zr, Al) are employed to determine the proportion of shale and igneous source rocks for each mudstone.

Our results point to the influence of source rock mineralogy on predicted silicate weathering intensities. Mixing models show that the proportion of shale among the clay-sized fraction in mudstones covers a wide range (0 to 87%), corresponding largely with their paleo-landscape positions. Mudstones that are interpreted to be deposited far from the active channel have distinctly high proportions of shale whereas mudstones closer to the active channel have distinct lower ones, suggesting that hydrodynamic sorting of sediments during overbank deposition separates igneous and shale sediments. Additionally, the Li isotope data show clay δ⁷Li values comprising a narrow range (-2.3 to -0.4‰ L-SVEC) and source rock δ⁷Li values comprising a much large one (shales spanning -2.7 to +4.1‰ and conglomerate clasts spanning 4.8 to 9.5‰). When source rock proportions are considered alongside Li isotope data, we determine that samples with high proportions of shale have higher Δ⁷Li_clay-source values whereas those with high proportions of igneous sediments have lower Δ⁷Li_{clay-source­} values. This finding suggests that samples closer to active channels undergo more intense silicate weathering than those that are distal, countering common thought where distal portions of floodplains that are less susceptible to erosion are thought to be locations where more intense silicate weathering occurs.

Lastly, despite the differences in predicted silicate weathering intensity, we observe a consistent range of weathering intensities before, during, and after the body of the PETM. This finding may support the idea that silicate weathering in floodplains is weakly responsive to climatic perturbations and thus silicate weathering in hillslopes enabled climate to return to its pre-PETM state.