Active methanogenesis during the melting of Marinoan snowball Earth

Geological evidence indicates that the deglaciation of Marinoan snowball Earth ice age (~635 Myr ago) was associated with intense continental weathering, recovery of primary productivity, transient marine euxinia, and potentially extensive CH4 emission. It is proposed that the deglacial CH4 emissions may have provided positive feedbacks for ice melting and global warming. However, the origin of CH4 remains unclear. Here we report Ni isotopes (δ60Ni) and Yttrium-rare earth element (YREE) compositions of syndepositional pyrites from the upper most Nantuo Formation (equivalent deposits of the Marinoan glaciation), South China. The Nantuo pyrite displays anti-correlations between Ni concentration and δ60Ni, and between Ni concentration and Sm/Yb ratio, suggesting mixing between Ni in seawater and Ni from methanogens. Our study indicates active methanogenesis during the termination of Marinoan snowball Earth. This suggests that methanogenesis was fueled by methyl sulfides produced in sulfidic seawater during the deglacial recovery of marine primary productivity.


Supplementary Notes 1. Geological Background and Sample Description
The South China Block (SCB) consists of the Yangtze Block in the modern northwest and Cathaysia Block in the modern southeast 1,2 . The amalgamation of the two blocks occurred at ~830-820 Ma, followed by a rifting and thermos-subsidence cycle in late Neoproterozoic. The upper glacial deposits are represented by the Nantuo Formation 5 . In the Yangtze Block, the Nantuo Formation shortens from 2000 meters of basin facies in the southeast (present orientation) to a few meters in thickness of shelf facies in the northwest. An U-Pb age of 635 ± 0.6Ma was reported from the top of Nantuo Formation, thus the Nantuo Formation corresponds with the Marinoan glacial deposits 6,7 . The Nantuo Formation unconformably overlies Liantuo/Chengjiang formations in shallow facies, while it conformably overlies the interglacial deposits (the Datangpo Formation or the top of the Fulu Formation) in deep water facies. The Nantuo Formation is conformably overlain by a 3-6 meter thick cap carbonate in the basal Doushantuo Formation.
Two glacial episodes were recognized in the Nantuo Formation: The first glacial interval is recorded by massive coarse-grained diamictite in the lower part of Nantuo Formation, while the second glacial episode is represented by the re-appearance of diamictite in the middle parts of the Nantuo Formation 4 . Approximately 10-meter-thick siltstone/shale were deposited between the two units of glacial deposits, suggesting an non-glacial period within the Marinoan glaciation. In the upper most of the Nantuo Formation, ~10s meter-thick pebbly sandstone/siltstone units represents the deglacial period of the Marinoan glaciation which was supported by isotopically heavy Mg values (δ 26 Mg≈1.0‰), which suggests an intense continental weathering event in the end of Nantuo glaciation 2 . Besides, the topmost of the Nantuo Formation also contain massive pyrite concretions which were pervasively distributed in the SCB.

Supplementary Notes 2.Calculation for Mixing Model
The calculation for mixing model is based on mass balance models. 8,9 Let c1, c2 and δ1, δ2 to be the concentration and δ value for two end members respectively. The elemental mass balance and isotopic mass balance can be written as: Here c and δ is the concentration and δ value for mixed system, f is the fraction for the first end member. Rearranging equation S1 we get: As a result, the δ and c in the mixing model are inversely proportional.

Supplementary Notes 3. Calculation for Rayleigh Distillation
The Rayleigh isotope distillation model is used to describe a gradual fractionation process for a closed system. 8,10 It assumes a first order reaction while the rate factor for different isotope is unequal. Let A and B be different isotopes of an element, the rate equations for the reaction is written by: Here the rate factor ≠ , [A] and [B] represents the concentration for A and B respectively.
Here we define the fractionation factor as the ratio of the rate constant: Rearranging the equation, we get: Then integrating: Here we can define the fraction of original mass remaining f as: Replace B by 60 Ni, A by 58 Ni, then the formula could be written by: Here δ 60 Ni 0 means the δ 60 Ni value for initial material, δ 60 Ni r means the δ 60 Ni value for remaining Ni.
In a kinetic fractionation model for closed system, assuming the pyrite precipitate from seawater, the relationship between δ 60 Nipy, δ 60 Nir, and δ 60 NiSW can be written as: (1 − ) ⋅ δ 60 Ni py + ⋅ δ 60 Ni r = δ 60 Ni SW (S12) Here δ 60 Nipy, δ 60 Nir, and δ 60 NiSW stand for the average δ 60 Ni value for pyrite, residual seawater and original seawater respectively. Replacing the δ 60 Nir, then we get: Rearranging the equation, we get: In this work, the MQ H2O used here were produced by Milli-Q Element system (Millipore, USA (1) Removing Ca by concentrated HCl 1ml Bio-Rad AG50W-X8 cation exchange resin is used for this step. The resin was first washed by 2 x 5ml MQ H2O, 2 x 5ml 6mol/L HCl then 2 x 5ml MQ H2O, then conditioning with 1ml concentrated HCl twice. First, the sample was loaded onto the resin with 1ml solution in concentrated HCl. Ni starts eluting as soon as loaded to the resin, and it could be quantitatively collected using 6 x 1ml concentrated HCl to wash the resin. Residue elements in the resin could be completely washed off using 5 x 1ml 6mol/L HCl. The recovery rate for Ni is higher than 99.9%.
Around 95% of Ca, 90% of Fe would be removed in this step.
(2) Removing Fe, Al, Ti by HF-HNO3 mixed solution The same column with step (1) was used for this step. The washing step is also same with step (1), but here 2 x 1ml 0.5mol/L HF + 1 mol/L HNO3 mixed solution was used for conditioning. The sample was loaded to the resin using 0.5ml solution in 0.5mol/L HF + 1 mol/L HNO3. Fe, Al, Ti would be washed off from the resin immediately as a complex with fluorinion. After completely washing Fe, Al, Ti using 5 x 0.5ml 0.5mol/L HF + 1 mol/L HNO3 mixed solution, the residue elements including Ni was completely collected using 5 x 1ml 6mol/L HCl. The recovery rate for Ni is higher than 99.9%, while the removal of Fe, Al, Ti, Na is around 99%.
(3) Removing Fe, Mn by HCl-acetone solution The column and the washing procedure for this step is same with that for step (1  Fe would be adsorbed in the resin as complex with chloridion, while Ni would be eluted immediately.
After loading 3 x 0.5ml 6mol/L HCl, Ni would be completely washed out from the resin. The recovery rate for Ni is higher than 99.9% for this step.
In this work, samples were first digested using mixed solution of concentrated HCl and HNO3.       The MREE* is defined as (2*Sm N )/(La N +Yb N ). The red circle, red square and red diamond correspond to Tongle, Yazhai and Datan section respectively. All REE data are normalized by shale. No relationship occurs in all relationship above, suggesting the contamination for Ni and REE for basin samples is weak. Supplementary Figure S6. δ 60 Ni value for geological reference material. The red circle represents data in this work. The green circle represents published data from the same laboratory 11 . The black square, black triangle and black diamond are literature value 12,14,15 . The error bar is defined as 2SD (two standard deviation). The δ 60 Ni for geological reference material in our work are close to literature value, which certify the accuracy for nickel isotopic measurement.