Early Cenozoic evolution of topography, climate, and stable isotopes in precipitation in the North American Cordillera

Abstract

Paleoelevation reconstructions of the North American Cordillera inferred from the oxygen (δ^18^O) and hydrogen (δD) isotope ratios of terrestrial paleoclimate proxy materials (soils, ashes, lake sediments) suggest rapid north-to-south migration of topography in the early Cenozoic (pre-49 Ma to 28 Ma). The validation of this reconstruction relies on an accurate understanding of the δ^18^O~p~ and the associated regional climate change in response to the uplift of the western North America. Here we study this response using a global climate model (GCM) with explicit δ^18^O~p~ diagnostics (ECHAM5-wiso) focusing on the isotopic effects of different types of precipitation, vapor mixing, recycling and moisture source and compare the response against estimates made using a Rayleigh distillation models of moist adiabatic condensation (RDM). Four experiments are performed with Eocene topography inferred from terrestrial stable isotope paleoaltimetry records to investigate how southward propagation of topography affects regional climate (temperature, precipitation and circulation pattern) and δ^18^O~p~ over North America. Our experiments predict δ^18^O~p~ patterns that are broadly consistent with maps of temporally binned proxy δ^18^O and generally support an early Cenozoic north-to-south propagation of high topography in the North American Cordillera. They do not support the commonly made assumption that isotopic fractionation occurs primarily through rainout following Rayleigh distillation nor the application of modern empirical δ^18^O~p~ lapse rates to past environments. In our GCM simulations, precipitation processes and climate changes that are not captured by RDMs substantially affect δ^18^O~p~. These processes include shifts in local precipitation type between convective and large-scale rain and between rain and snow; intensification of low-level vapor recycling particularly on leeward slopes; development of air mass mixing and changes in wind direction and moisture source. Each of these processes can have significant (≥2‰) influences on δ^18^O~p~ that are comparable in magnitude to surface uplift of hundreds or even thousands of meters. In many regions, these processes fortuitously compensate each other, explaining the apparent agreement between ECHAM5-wiso and proxy δ^18^O and, more broadly, between RDM estimates and observed δ^18^O-elevation relationships. In some regions, compensation is incomplete, and as a result, ECHAM5-wiso δ^18^O~p~ does not agree with estimates from the RDM. In these regions, including the interior of the northern cordillera and the eastern flank of the southern Cordillera, moderate adjustments of paleoelevations may be in order.