Lacustrine 87Sr/86Sr as a tracer to reconstruct Milankovitch forcing of the Eocene hydrologic cycle

• Milankovitch-forced changes in Eocene precipitation recorded in carbonate 87Sr/86Sr.
• Low 87Sr/86Sr during high lake levels indicates influx of western source waters.
• Western source waters likely derived from Milankovitch-forced alpine snowpack.
• Coupled low frequency (Milankovitch) and high frequency (ENSO) climate variability.

The Green River Formation (GRF) provides one of the premier paleoclimate archives of the Early Eocene Climatic Optimum (∼50 Ma), representing the apex of the early Cenozoic greenhouse climate. Rhythmic lake-level variability expressed in the GRF has inspired numerous hypotheses for the behavior of the Eocene hydrologic cycle, including its linkage to astronomical forcing, solar variability, and the El Niño Southern Oscillation (ENSO). However, the lack of sufficient proxy data to document atmospheric water-mass transport and the geographic pattern of evaporation/precipitation/runoff has made it difficult to discriminate between different models for astronomical forcing. Variable 87Sr/86Sr ratios of bedrock that encompass the GRF provide an opportunity to reconstruct the spatial expression of the Eocene hydrologic cycle and its linkage to lake level. Here Sr isotope data from the Wilkins Peak Member, a rhythmic succession that has been demonstrated to record Milankovitch forcing of lake levels, indicate that high lake levels reflect an increased proportion of runoff from less radiogenic rocks west of the basin, eliminating a number of the existing astronomical-forcing hypotheses. The 87Sr/86Sr variability is consistent with a change in mean ENSO state, which is predicted by climate models to be linked to orbital-insolation. Thus, the 87Sr/86Sr data reveal a coupling of high frequency (ENSO) and low frequency (astronomical) climate variability, and also predict the existence of sizable astronomically-forced alpine snowpack during the last greenhouse climate. More broadly, this study demonstrates the utility of 87Sr/86Sr as a powerful tool for reconstructing the deep-time hydrologic cycle.