RIVERS IN TIME: AN INTEGRATED FRAMEWORK FOR RECONSTRUCTING TERRESTRIAL CLIMATE VARIABILITY IN DEEP TIME

The biogeochemical records from ancient alluvial basins facilitate quantitative reconstructions of terrestrial climate in deep time; however, climate shifts can alter the dynamics of rivers systems and biogeochemical proxy (e.g., leaf wax biomarkers, stable isotopes, soils) production therein. Freshly deposited river sediment integrates biogeochemical signals over temporal and spatial scales associated with sediment mobilized from the entire catchment; on stable floodplains, inherited signals in deposited sediment are gradually overprinted by the local signal during soil production. Thus, river and floodplain dynamics set the spatiotemporal scales to which paleoclimate reconstructions apply. If catchment-averaged, environmental signals may be integrated over more than ~10⁶ years; if local, they may be integrated over just ~10⁴ years.

We use experiments and field data to explore how river kinematics impact the scales of integration in the climate proxy record. Experimental results suggest that more stable distributary networks with larger ratios of water to sediment discharge favor the preservation of laterally discontinuous floodplain surfaces (with deeply weathered soils) that were in extended stasis. Distributive networks with greater mobility and higher relative sediment loads preserve abundant, laterally extensive floodplain surfaces (with weakly weathered soils) that represent short periods of stasis.

We characterize floodplain dynamics preserved in the White River Group (WRG), which spans the Eocene-Oligocene climate transition in central North America. WRG strata record a transition from perennial Eocene streams, with minimal discharge variability, stable floodplains and low avulsion frequencies to ephemeral, Oligocene streams, with significant discharge variability, dynamic floodplains and higher avulsion frequencies. Based on our analytical framework, we expect the proxy record (e.g., δ²H of organic biomarkers) in stable late Eocene floodplains to record local climate information; conversely, we expect the dynamic floodplains of Oligocene streams to record catchment-averaged information. The sedimentary record can, therefore, offer significant insight into the scales of relevance of terrestrial climate reconstructions during this interval.