9:00 AM - 9:15 AM
[BCG06-01] Independent component analysis of sedimentary rocks through geologic time: Coevolution of the Earth’s surface environment with life
Keywords:Independent, Banded iron formation, marine nutrients, Evolution of life
It is one of the most important issues in deciphering the history of the Earth to estimate the relationship between changes in the Earth's surface environment and the evolution of life. In particular, estimating ancient seawater composition is essential for understanding the history of the earth and life, because changes in seawater composition may be related to changes in the activities of early organisms. In general, prokaryotes use more Ni and Co-bearing enzymes than Cu and Zn-bearing enzymes, compared to eukaryotes. There is much debate as to whether this evolution of life is due to the intrinsic evolution of organisms or external factors; thus, clarifying this is thought to lead to deciphering the evolution of life. Therefore, estimating the changes in the marine nutrient content of the external factors is essential for understanding the evolution of life.
Two methods can be used to estimate the concentration of marine nutrients: model calculation and chemical analysis of rock samples. Anbar et al. (2008) showed the changes in trace element contents of seawater through geologic time based on a solubility model of trace elements in aqueous solutions. However, there is debate regarding the ocean redox state, sulfide ion concentration, and atmospheric oxygen concentration used in the model calculation; thus, the validity of the calculation is highly questionable. Therefore, empirical research based on geological records is required.
Ancient seawater compositions have been estimated using chemical sediments, particularly banded iron formations (BIFs). Konhauser et al. (2009) estimated that the early Archean seawater from 3.8 to 2.7 Ga was more enriched in Ni from the high Ni/Fe ratios of the early Archean BIFs, and suggested that the decrease in marine Ni content in the Neoarchean caused the decline of methanogens that require Ni-bearing coenzyme F430 and the relative prosperity of oxygenic photosynthetic organisms, resulting in Great Oxidation Event (GOE). However, there are many problems with the data on which they are based. In general, BIFs contain not only iron oxides, but also a large amount of Ni-rich terrigenous and pyroclastic detritus. In addition, the original minerals precipitated in the ocean were iron hydroxide minerals; thus, iron oxide in the BIFs was formed during diagenesis. However, these effects were not fully considered in previous studies. Aoki et al. (2018) proposed a method to estimate the Ni/Fe ratio of an iron oxide end-member using Zr content as an indicator of the amount of detritus. However, this method cannot quantitatively determine the composition of iron oxide end-members for BIFs with multiple origins of detritus.
We applied independent component analysis (ICA) to BIFs of eight different ages from the Archean to Neoproterozoic to estimate the chemical composition of the iron hydroxide end-member. We corrected the starting point of the vectors by adding a virtual component with a 100% SiO2 content and no other elements. The extracted components with high FeO, MgO+CaO, and TiO2+Al2O3 content were interpreted as iron hydroxide, carbonate, and detritus, respectively. Furthermore, the detrital components were classified into five to seven origins based on their chemical characteristics. The Ni, Co, and Cr contents of the iron hydroxide components were high from 3.9 to 3.0 Ga, gradually decreased between 3.0 and 2.2 Ga with temporal increase at 2.5 Ga, and were quite high after the Sturtian Snowball Earth, again. Assuming insignificant changes in their adsorption coefficients through geologic time, the fluctuations are interpreted as secular changes in seawater composition because the elements in the iron hydroxides are considered to be adsorbed to the iron hydroxides. The high Ni, Co, and Cr contents of seawater from 3.9 to 3.0 Ga were due to reduced conditions, active submarine hydrothermal activity, and strong silicification due to high CO2 levels, whereas their decrease after 3.0 Ga and before GOE was due to a decrease in silicification. The temporary increase at 2.5 Ga is thought to be due to enhanced hydrothermal activity because of the widespread distribution of the Neoarchean BIFs. It is considered that they decreased after the GOE. In addition, it is considered that the Neoproterozoic deep seawater was highly enriched in these elements due to reduction during the Sturtian Snowball Earth and the manganese shuttle from the oxic shallow seawater afterwards.
The estimated secular change of seawater composition is consistent with the use of trace elements by organisms, suggesting that biological evolution was promoted by the external factor.
Two methods can be used to estimate the concentration of marine nutrients: model calculation and chemical analysis of rock samples. Anbar et al. (2008) showed the changes in trace element contents of seawater through geologic time based on a solubility model of trace elements in aqueous solutions. However, there is debate regarding the ocean redox state, sulfide ion concentration, and atmospheric oxygen concentration used in the model calculation; thus, the validity of the calculation is highly questionable. Therefore, empirical research based on geological records is required.
Ancient seawater compositions have been estimated using chemical sediments, particularly banded iron formations (BIFs). Konhauser et al. (2009) estimated that the early Archean seawater from 3.8 to 2.7 Ga was more enriched in Ni from the high Ni/Fe ratios of the early Archean BIFs, and suggested that the decrease in marine Ni content in the Neoarchean caused the decline of methanogens that require Ni-bearing coenzyme F430 and the relative prosperity of oxygenic photosynthetic organisms, resulting in Great Oxidation Event (GOE). However, there are many problems with the data on which they are based. In general, BIFs contain not only iron oxides, but also a large amount of Ni-rich terrigenous and pyroclastic detritus. In addition, the original minerals precipitated in the ocean were iron hydroxide minerals; thus, iron oxide in the BIFs was formed during diagenesis. However, these effects were not fully considered in previous studies. Aoki et al. (2018) proposed a method to estimate the Ni/Fe ratio of an iron oxide end-member using Zr content as an indicator of the amount of detritus. However, this method cannot quantitatively determine the composition of iron oxide end-members for BIFs with multiple origins of detritus.
We applied independent component analysis (ICA) to BIFs of eight different ages from the Archean to Neoproterozoic to estimate the chemical composition of the iron hydroxide end-member. We corrected the starting point of the vectors by adding a virtual component with a 100% SiO2 content and no other elements. The extracted components with high FeO, MgO+CaO, and TiO2+Al2O3 content were interpreted as iron hydroxide, carbonate, and detritus, respectively. Furthermore, the detrital components were classified into five to seven origins based on their chemical characteristics. The Ni, Co, and Cr contents of the iron hydroxide components were high from 3.9 to 3.0 Ga, gradually decreased between 3.0 and 2.2 Ga with temporal increase at 2.5 Ga, and were quite high after the Sturtian Snowball Earth, again. Assuming insignificant changes in their adsorption coefficients through geologic time, the fluctuations are interpreted as secular changes in seawater composition because the elements in the iron hydroxides are considered to be adsorbed to the iron hydroxides. The high Ni, Co, and Cr contents of seawater from 3.9 to 3.0 Ga were due to reduced conditions, active submarine hydrothermal activity, and strong silicification due to high CO2 levels, whereas their decrease after 3.0 Ga and before GOE was due to a decrease in silicification. The temporary increase at 2.5 Ga is thought to be due to enhanced hydrothermal activity because of the widespread distribution of the Neoarchean BIFs. It is considered that they decreased after the GOE. In addition, it is considered that the Neoproterozoic deep seawater was highly enriched in these elements due to reduction during the Sturtian Snowball Earth and the manganese shuttle from the oxic shallow seawater afterwards.
The estimated secular change of seawater composition is consistent with the use of trace elements by organisms, suggesting that biological evolution was promoted by the external factor.