A long history of equatorial upwelling in the Pacific Ocean

Wind-driven divergence of ocean currents in the eastern equatorial Pacific (EEP) gives rise to strong upwelling of cold, nutrient- and CO₂-rich deep waters, contributing to high primary productivity in this region. Every 2-7 years, ceased upwelling in the EEP during El Niño events impact global climate through the El Niño – Southern Oscillation teleconnections. High productivity in the EEP also strongly influences the global carbon cycle and the cycling of important nutrients. Yet the geological history of the equatorial upwelling system is unclear. One hypothesis argues that before 3 Ma, the thermocline in the EEP was so deep that upwelling was effectively shut down. This proposition is supported by SST reconstructions that show a much warmer EEP over the late Miocene – Pliocene. However, high mass accumulation rates (MAR) of biogenic opal throughout the entire EEP has been observed over 4.5-6.7 Ma, suggesting abundant nutrient supply probably associated with strong upwelling.

Here we report a novel approach to probe the strength of equatorial upwelling using seawater CO₂ reconstructions. Shipboard surveys show that during normal (upwelling) or La Niña (strong upwelling) conditions, surface seawater in the EEP has a much higher CO₂ content than atmospheric conditions. However, during El Niño events when rates of upwelling are nearly absent, CO_2(sw) is close to air-sea equilibrium. CO_2(sw) data collected near ODP Site 850 in the EEP demonstrate a strong correlation between excess CO₂ and the Nino3 index. For this study, we apply CO₂ reconstructions using the alkenone-pCO₂ method from Site 850 and Site 806 in the Pacific warm pool to compute excess CO₂ since ~6 Ma, assuming that Site 806 reflects air-sea equilibrium in CO₂. Our results show that the surface seawater at Site 850 always contains more CO₂ than Site 806, indicating a prevailing upwelling history in the EEP for the past 6 million years. Applying the modern excess CO₂-Nino 3 index relationships, we show that upwelling was probably strongest during the late Miocene – early Pliocene, coinciding with the highest MAR of biogenic opal, and then declined over the Plio-Pleistocene. These results suggest that the warm EEP SSTs in the Miocene – Pliocene reflect warmer source waters and those SSTs is no adequate indicator of upwelling intensity during this time interval.