Thermogenic carbon release from the Central Atlantic magmatic province caused major end-Triassic carbon cycle perturbations

Abstract

The Central Atlantic magmatic province (CAMP), the end-Triassic mass extinction (ETE), and associated major carbon cycle perturbations occurred synchronously around the Triassic-Jurassic (T-J) boundary (201 Ma). Negative carbon isotope excursions (CIEs) recorded in marine and terrestrial sediments attest to the input of isotopically light carbon, although the carbon sources remain debated. Here, we explore the effects of mantle-derived and thermogenic carbon released from the emplacement of CAMP using the long-term ocean-atmosphere-sediment carbon cycle reservoir (LOSCAR) model. We have tested a detailed emission scenario grounded by numerous complementary boundary conditions, aiming to model the full extent of the carbon cycle perturbations around the T-J boundary. These include three negative CIEs (i.e., Marshi/Precursor, Spelae/Initial, Tilmanni/Main) with sharp positive CIEs in between. We show that a total of ∼24,000 Gt C (including ∼12,000 Gt thermogenic C) replicates the proxy data. These results indicate that thermogenic carbon generated from the contact aureoles around CAMP sills represents a credible source for the negative CIEs. An extremely isotopically depleted carbon source, such as marine methane clathrates, is therefore not required. Furthermore, we also find that significant organic carbon burial, in addition to silicate weathering, is necessary to account for the positive δ¹³C intervals following the negative CIEs.

Keywords: C cycle modeling; C cycle perturbations; Central Atlantic magmatic province; end-Triassic extinction.

Fig. 1.

(A) Compilation of available high-precision U–Pb ages of CAMP rocks (–21) demonstrates a total duration of ∼800 ky and potentially four main phases of CAMP activity (early, main, late, and final). The three oldest high- and low-Ti Brazilian sills overlap with the main phase, while the youngest high-Ti sill marks the onset of the late phase. Note that the age for the North Mountain Basalt represents an average value after refs. –. (B) Timeline of selected CAMP rocks from the Amazonas, Solimões, and Newark basins, a composite carbon isotope curve (9) and pCO₂ data (2) spanning the T–J boundary after ref. . See SI Appendix, Fig. S1 for detailed T–J boundary δ¹³C_carb and δ¹³C_org curves. The main CAMP phase, the latest high-Ti Brazilian sill and the Preakness basalt coincide with the Marshi (M) (CIE 1), Spelae (S) (CIE 2), and Tilmanni (T) (CIE 3) negative excursions, respectively. BUT, Butner diabase; PAL, Palisades sill; PRB, Preakness basalt. The blue dashed line represents the initial (pre-CAMP) pCO₂ value, and blue arrows represent the increases in pCO₂.

Fig. 2.

Model response of atmospheric pCO₂ (A) and δ¹³C of shallow ocean sediments (B; mean value of Atlantic, Indian, Pacific, and Tethys shallow ocean boxes) to a CAMP emission scenario including five pulses of carbon release (Table 1). The increases in δ¹³C following each negative excursion and decreases in pCO₂ after each pCO₂ peak reflect organic carbon burial and/or silicate weathering (SI Appendix, Fig. S8). The gray outlines/star symbols represent the range of observed δ¹³C_carb values (4, 11, 12, 54) (SI Appendix, SI Text) and pCO₂ data (2, 3) (light gray, ref. ; dark gray, ref. 3).

Fig. 3.

Interpretation of CAMP sill emplacement and thermogenic carbon release in three main phases in the Amazonas and Solimões basins, Brazil. The first phase of sill emplacement includes both low- and high-Ti sills intruding the upper and lower Paleozoic series, releasing 4,800 Gt of mixed (50/50) inorganic and organic thermogenic carbon. The second and third phases include high-Ti sills intruding the lower Paleozoic series, releasing 4,800 and 2,500 Gt of organic thermogenic carbon, respectively. The cross-section is not drawn to scale.