School of Earth Sciences - Theses
Permanent URI for this collection
Search Results
1 - 1 of 1
-
ItemThe composition of altered oceanic crust: Implications for mantle evolution( 2017)The geochemical characterisation of oceanic crust, in its dual role as a mantle-derived melt and a subducted material, is an essential component in understanding the evolution of the Earth’s mantle. Interaction of newly formed crust with seawater, however, producing ‘altered oceanic crust’ (AOC), provides a degree of complexity in this undertaking. This study aims to conduct a global survey of trace element and Sr-Nd-Hf-Pb isotopic compositions of AOC by compiling high quality literature data and adding new analyses where appropriate to address identified gaps in the record. Ten representative DSDP/ODP/IODP sites were selected for the new analyses. In total 93 samples for homogeneous portions and 19 samples for heterogeneous portions of AOC were analysed. These results are combined with all available data for these ten sites as well as geochemical data from other locations that also cored AOC samples. Prior to any assessment of alteration effects, variations due to magmatic processes are considered. Comparison of alteration-insensitive elements and isotope ratios of AOC, grouped by rock type, crustal age and ocean basin (i.e. spreading rate) reveals the relative importance of these parameters on AOC composition. The results show that rock type provides the first order control on trace element variations. Alteration-insensitive trace element compositions are not correlated with either crustal age or ocean basin/spreading rate, whereas Nd and Hf isotope ratios do show trends with crustal ages, mirroring the evolution of the mantle source over the past 170 Ma. In addition, the compositions of lower crustal rocks from slow-spreading ridges are significantly more variable than those from fast-spreading ridges. This is because the lower crust at slow-spreading ridges is generated from multiple magma pulses compared to that at fast-spreading ridges that generally forms more homogeneous magmas derived from larger magma chambers In terms of alteration effects compositional variations in the upper and lower crust are considered separately because of their very different nature. The alteration in homogeneous parts of the upper AOC varies with crustal age. This is consistent with the correlation between the oxidation state, which plays an important role in upper crust alteration, and crustal age. The alteration in homogeneous parts of the lower AOC, however, is better linked with spreading rate. In heterogeneous parts of AOC, such as breccias and veins, compositional variation is principally dependent on the phases themselves instead of crustal age or spreading rate. The compiled results, with the controls on compositional variation revealed, are used to calculate the global compositions of eight AOC types, namely fast-upper AOC, slow-upper AOC, fast-lower AOC, slow-lower AOC, fast-total AOC, slow-total AOC, global AOC and global AOC-plume free. Site medians are calculated first and the crustal age- and spreading rate-weighted averages are determined for upper and lower crust, respectively. The global AOC compositions are then calculated using the proportion of fast- and slow-spreading crust worldwide. Also, the compositions of plume-free global AOC are also estimated by removing some of plume-influenced sites. The results are then compared with previous estimates of AOC composition revealing some differences. The results of this investigation encompass a greater number of drill sites with a more detailed understanding of the controls on variation, and are therefore likely a more accurate representation than previous estimations. The newly derived AOC compositions can be used to test models for the involvement of subducted AOC in generating Ocean Island Basalt (OIB) compositional variation, in particular the origin of the ‘high-Mu’ (HIMU) signature. As an example a preliminary test was performed using source and recycling ages of 3 and 2 Ga, respectively. The model results show that 1) subduction modification is essential for AOC to generate HIMU signals and 2) the upper crustal components are most likely to generate HIMU signals in all isotope systems after subduction modification (dehydration) with adjusted mobility of Th. However, a more detailed determination of element mobility is required in order to obtain more reliable modelling results. Intriguingly, four analysed samples have extremely unradiogenic Pb isotope compositions, unlike anything previously recovered from the ocean floor. These are compared with similar signals observed in peridotites but show different trends, indicating a different source or magmatic origin. They might also be caused by the occurrence of secondary sulfides, but this aspect requires further investigation.