Patterns and Mechanisms of Northeast Pacific Temperature Response to Pliocene Boundary Conditions

Ocean‐atmosphere dynamics in the north Pacific play an important role in the global climate system and influence hydroclimate in western North America. However, changes to this region's mean climate under increased atmospheric greenhouse gas concentrations are not well understood. Here we present new alkenone‐based records of sea surface temperature (SST) from the northeast Pacific from the mid‐Piacenzian warm period (approximately 3.3–3.0 Ma), an interval considered to be an analog for near‐future climate under middle‐of‐the‐road anthropogenic emissions. We compare these and other alkenone‐based SST records from the north Pacific to fully‐coupled climate model simulations to examine the impact of mid‐Pliocene CO2 and other boundary conditions on regional climate dynamics and to explore factors governing model disagreement about regional temperature patterns. Model performance varies regionally, with Community Earth System Model 1.2 (CESM 1.2) and CESM2 performing best in regions with greater warming like the California Margin, though these models underestimate the warming evidenced in our new proxy record and others from the region. Single forcing simulations reveal a strong influence for prescribed land surface changes and higher CO2 levels on coastal warming patterns along the California Margin in CESM2. Furthermore, differences in shortwave and longwave radiation and circulation between the models, likely related to changes in the atmospheric component of the model, may play a key role in the ability of models to capture regionally‐varying patterns of Pliocene warmth. Regional patterns of temperature change inferred from geochemical records could therefore help to understand the impacts of different model parameterization schemes on regional climate patterns. Plain Language Summary As a result of anthropogenic carbon emissions, 21st century climate will not resemble anything experienced on Earth since thousands or, more likely, millions of years before instrumental records began. In order to prepare for and accurately predict future climate change, it is necessary to study past climates that might closely relate to our anticipated future scenario. In this study, we use organic geochemical “proxy” indicators of temperature to estimate temperature changes during a mid‐Pliocene warm period, an interval approximately 3 million years ago when atmospheric CO2 was close to or slightly below near‐future values. We compare these results to climate model simulat