5:15 PM - 6:45 PM
[MGI28-P03] Analysis of mantle processes based on multivariate analysis of compositional data from mantle sections of Oman ophiolite
Keywords:ophiolite, compositional data analysis, independent component analysis, mantle, oman, trace element
Multivariate analysis methods such as principal component analysis and independent component analysis are useful for understanding igneous processes, and studies have been conducted to extract and quantify igneous processes from multidimensional chemical composition data (e.g., Ueki and Iwamori, 2017). However, previous studies using multivariate analysis to extract uppermost mantle processes from mantle peridotite are limited (e.g., Nishio et al., 2022), and have not been applied to the whole rock chemical composition of ophiolite. Ophiolite exhibits lithospheric section, and their mantle section can be viewed as a fossil of the uppermost mantle in early island arc environment. In this study, multivariate analysis using independent component analysis was applied to the trace chemical composition of the uppermost mantle peridotite of the Oman ophiolite to investigate potential mantle processes.
The multivariate analysis used whole rock chemical compositions (Ga, Rb, Sr, Y, Cs, Ba. La, Ce, Nd. Sm, Eu, Gd, Tb. Dy, Ho, Er, Yb, Lu, Hf) of 169 peridotite specimens collected from the Fizh massif in the northern part and the Wadi Tayin massif in the southern part of the Oman Ophiolite. Four components were extracted based on the contribution ratio of the principal component analysis. The chemical composition data exist in a simplex space with one dimension lower than the number of variables. The Linear transformation method, Additive Log Ratio, was used to solve this limitation. The calculations were normalized to the elemental concentrations in SiO2, and the natural logarithm was taken.
The components obtained from the analysis are uncorrelated and non-linear, i.e., they are independent components. Each independent component exhibits characteristics indicative of different mantle processes. The first independent component is positively correlated with olivine mode and Spinel-Cr# and negatively correlated with pyroxene mode and all trace elements. The second independent component has a weak positive correlation with olivine mode and a weak negative correlation with pyroxene mode. It also has a positive correlation with LILE, a positive correlation with incompatible HFSE, a positive correlation with LREE, and a negative correlation with HREE. The third independent component has a strong negative correlation with Sr and Ba. The fourth independent component strongly correlates positively with LILE for Cs, Rb, Ba, and Th.
Based on these results, the geological significance of four mantle processes is discussed. The first independent component is partial melting by decompression, the second is flux melting by supplying slab fluid, the third is enrichment of Sr and Ba by serpentinization at low temperatures, and the fourth is adding slab fluid. Plotting these components on geologic maps and cross-sections determined the spatial distribution of mantle processes. The degree of partial melting by decompression is low at the base of the mantle and high near Moho, and the degree of partial melting tends to be high in the high-temperature region. Flux melting occurred heterogeneously in the mantle and is strong near the High Refractory Zone (Kanke and Takazawa, 2014) and Moho. Enrichment of Sr and Ba by serpentinization tends to be high at the boundary between the low and intermediate temperature deformation zones. On the other hand, fluids enriched in LILE elements may have penetrated to a height near 1.7 km from the basement.
This study shows that four independent components can explain most of the mantle processes in the Oman ophiolite. In addition, the spatial distribution of mantle processes can be captured by linking these independent components with geographical information. These results confirm the effectiveness of multivariate analysis for the whole rock chemical composition of ophiolitic mantle peridotite based on compositional data analysis and independent component analysis.
1. Ueki, K., Iwamori, H., 2017. Geochemical differentiation processes for arc magma of the Sengan volcanic cluster, Northeastern Japan, constrained from principal component analysis. Lithos 290–291, 60–75.
2. Nishio, I., Itano, K., Waterton, P., Tamura, A., Szilas, K., Morishita, T., 2022. Compositional Data Analysis (CoDA) of Clinopyroxene From Abyssal Peridotites. Geochem Geophys Geosyst 23.
3. Kanke, N., Takazawa, E., 2014. A kilometre-scale highly refractory harzburgite zone in the mantle section of the northern Oman Ophiolite (Fizh Block): implications for flux melting of oceanic lithospheric mantle. Geological Society, London, Special Publications 392, 229–246.
The multivariate analysis used whole rock chemical compositions (Ga, Rb, Sr, Y, Cs, Ba. La, Ce, Nd. Sm, Eu, Gd, Tb. Dy, Ho, Er, Yb, Lu, Hf) of 169 peridotite specimens collected from the Fizh massif in the northern part and the Wadi Tayin massif in the southern part of the Oman Ophiolite. Four components were extracted based on the contribution ratio of the principal component analysis. The chemical composition data exist in a simplex space with one dimension lower than the number of variables. The Linear transformation method, Additive Log Ratio, was used to solve this limitation. The calculations were normalized to the elemental concentrations in SiO2, and the natural logarithm was taken.
The components obtained from the analysis are uncorrelated and non-linear, i.e., they are independent components. Each independent component exhibits characteristics indicative of different mantle processes. The first independent component is positively correlated with olivine mode and Spinel-Cr# and negatively correlated with pyroxene mode and all trace elements. The second independent component has a weak positive correlation with olivine mode and a weak negative correlation with pyroxene mode. It also has a positive correlation with LILE, a positive correlation with incompatible HFSE, a positive correlation with LREE, and a negative correlation with HREE. The third independent component has a strong negative correlation with Sr and Ba. The fourth independent component strongly correlates positively with LILE for Cs, Rb, Ba, and Th.
Based on these results, the geological significance of four mantle processes is discussed. The first independent component is partial melting by decompression, the second is flux melting by supplying slab fluid, the third is enrichment of Sr and Ba by serpentinization at low temperatures, and the fourth is adding slab fluid. Plotting these components on geologic maps and cross-sections determined the spatial distribution of mantle processes. The degree of partial melting by decompression is low at the base of the mantle and high near Moho, and the degree of partial melting tends to be high in the high-temperature region. Flux melting occurred heterogeneously in the mantle and is strong near the High Refractory Zone (Kanke and Takazawa, 2014) and Moho. Enrichment of Sr and Ba by serpentinization tends to be high at the boundary between the low and intermediate temperature deformation zones. On the other hand, fluids enriched in LILE elements may have penetrated to a height near 1.7 km from the basement.
This study shows that four independent components can explain most of the mantle processes in the Oman ophiolite. In addition, the spatial distribution of mantle processes can be captured by linking these independent components with geographical information. These results confirm the effectiveness of multivariate analysis for the whole rock chemical composition of ophiolitic mantle peridotite based on compositional data analysis and independent component analysis.
1. Ueki, K., Iwamori, H., 2017. Geochemical differentiation processes for arc magma of the Sengan volcanic cluster, Northeastern Japan, constrained from principal component analysis. Lithos 290–291, 60–75.
2. Nishio, I., Itano, K., Waterton, P., Tamura, A., Szilas, K., Morishita, T., 2022. Compositional Data Analysis (CoDA) of Clinopyroxene From Abyssal Peridotites. Geochem Geophys Geosyst 23.
3. Kanke, N., Takazawa, E., 2014. A kilometre-scale highly refractory harzburgite zone in the mantle section of the northern Oman Ophiolite (Fizh Block): implications for flux melting of oceanic lithospheric mantle. Geological Society, London, Special Publications 392, 229–246.