Surface ocean warming and acidification driven by rapid carbon release precedes Paleocene-Eocene Thermal Maximum

Abstract

The Paleocene-Eocene Thermal Maximum (PETM) is recognized by a major negative carbon isotope (δ¹³C) excursion (CIE) signifying an injection of isotopically light carbon into exogenic reservoirs, the mass, source, and tempo of which continue to be debated. Evidence of a transient precursor carbon release(s) has been identified in a few localities, although it remains equivocal whether there is a global signal. Here, we present foraminiferal δ¹³C records from a marine continental margin section, which reveal a 1.0 to 1.5‰ negative pre-onset excursion (POE), and concomitant rise in sea surface temperature of at least 2°C and a decline in ocean pH. The recovery of both δ¹³C and pH before the CIE onset and apparent absence of a POE in deep-sea records suggests a rapid (< ocean mixing time scales) carbon release, followed by recovery driven by deep-sea mixing. Carbon released during the POE is therefore likely more similar to ongoing anthropogenic emissions in mass and rate than the main CIE.

Fig. 1.. Regional map of core drilled within the Mid-Atlantic Coastal Plain [modified from (19)].

Trace element, stable isotope, and boron isotope data presented here from Bass River (BR), Cambridge-Dorchester (CamDor), and SDB are marked by star symbols. Core holes Ancora (AN), Millville (MV), and Wilson Lake (WL) discussed in this study are marked by circle symbols.

Fig. 2.. SDB carbon isotope (δ¹³C) bulk carbonate, planktonic, and benthic foraminifera records.

Planktonic foraminifera Mg/Ca estimated ocean temperatures, and error envelope includes uncertainty in Mg/Ca seawater composition. Average δ¹¹B of single specimen benthic foraminifera (C. alleni), with 2 SE associated with the mean for each depth and the uncertainty in our internal reference material PS69/318-1. Vertical error bars of δ¹¹B represent the full sample depth range of the multiple individual measurements represented by the average δ¹¹B value. Highlighted in yellow is the POE and in blue is the body of the CIE.

Fig. 3.. POE interval at SDB carbon isotope (δ¹³C) bulk carbonate, planktonic, and benthic foraminifera records.

Single specimen δ¹¹B of benthic foraminifera (C. alleni) is represented by solid symbols, and the multiple specimen average δ¹¹B is represented by open symbols. δ¹¹B data presented include 2 SE associated with the mean for each depth and the uncertainty in our internal reference material PS69/318-1. Vertical error bars of δ¹¹B represent the full sample depth range of the multiple individual measurements represented by the average δ¹¹B value. pH reconstruction based on analysis of average δ¹¹B in benthic foraminifera (C. alleni). An initial surface Atlantic Ocean pH = 7.89 was assumed, and error uncertainty was propagated using a Monte Carlo approach.

Fig. 4.. LOSCAR ΔpH output from a suite of simulations in which mass (left: y axis, in hundreds of Gt of carbon; right: y axis, in Gt of CO₂) and duration (x axis, in hundreds of years) were systematically varied.

The complete ΔpH output complying with a duration of carbon release from century to millennia is included. All simulations were started from equilibrium atmospheric CO₂ of 500 ppm. The ΔpH color scale is adjusted to the limits consistent to the ΔpH (−0.08 to 0.35) based on boron isotopes with an initial surface Atlantic Ocean pH = 7.89 (table S2). Projected total anthropogenic carbon emissions relative to preindustrial (1850) in 2020 and 2050 from the SSP (shared socioeconomic pathway) scenarios (80) are superimposed on the LOSCAR ΔpH output array.

Fig. 5.. Simplified schematic of the carbon exchange between the atmosphere and ocean DIC reservoirs during the POE onset and recovery phases.

Approximate preindustrial reservoir carbon stocks in PgC (1 PgC = 10¹⁵ gC) and annual carbon exchange fluxes are noted in parentheses (PgC year⁻¹) from Ciais et al. (81). LOSCAR experiments in this study have slightly different reservoir sizes and fluxes to better represent early Paleogene ocean biogeochemistry conditions. (Left) The injection of depleted carbon into the atmosphere resulted in elevated CO₂ levels, negative δ¹³C_DIC POE, ocean acidification, and warming. A POE duration of century to millennia limits the addition of CO₂ and its δ¹³C signal to the surface ocean. Inset: Carbon evasion and warming are observed in coastal waters along the Mid-Atlantic Coastal Plain (SDB; star); vertical scale represents meters water depth. (Right) Ocean overturning circulation eventually dilutes the depleted δ¹³C and low pH surface ocean signal with the larger deep-ocean carbon reservoir.