Dürkefälden, Antje (2019) Insights into the origin and evolution of intraplate magmatism in the Caribbean (initial stage of the Galápagos hotspot) and the South Atlantic (Discovery and Shona hotspot tracks). (PhD/ Doctoral thesis), Christian-Albrechts-Universität, Kiel, Germany, XIII, 316 pp.

	Text Dissertation_Antje_Dürkefälden_2019.pdf - Published Version Restricted to Repository staff only Download (17MB) | Contact
Preview	Text (Short version) Dissertation_Antje_Dürkefälden_2019_shortversion.pdf Available under License Creative Commons: Attribution 4.0. Download (935kB) | Preview

Abstract

According to the classic mantle plume model, the evolution of a mantle plume comprises an initial plume head stage followed by a later plume tail stage. My thesis aims at deciphering the origin and evolution of submarine structures resulting from intraplate magmatism in two different areas: 1) the Caribbean Large Igneous Province (CLIP) formed during the plume head stage and 2) the Discovery Rise and adjacent structures in the South Atlantic representing the plume tail stage. The CLIP is assumed to have formed during the plume head stage of the Galápagos plume. Whereas this initial stage typically leads to the generation of huge continental flood basalt provinces and oceanic plateaus, the CLIP does not form such a coherent plateau. Instead, it consists of numerous oceanic flood basalt fragments accreted by tectonic processes on land around the margins of the LIP. In addition, large parts of the Caribbean seafloor are likely to be covered by CLIP sequences. In contrast to the better accessible subaerial fragments, the submarine sequences are poorly explored. Thus, my thesis addresses the detailed geochemical and geochronological investigation of two large submarine structures in the Caribbean Sea, the Beata Ridge and the Lower Nicaraguan Rise (LNR), to get new insights into the origin and evolution of the CLIP. My studies show for the first time that parts of the Beata Ridge formed during the main magmatic CLIP stage at ~89 Ma by widespread extrusive activity, contrasting previous assumptions that the ridge dominantly represents an intrusive complex formed during a later magmatic phase at ~76 Ma. My 40 Ar/ 39 Ar age dating yielded ages of ~92-77 Ma and thus cover both phases confirming long-term volcanism. Stratigraphically controlled sampling shows a high geochemical variability on a small scale (<100 km) with trace element and radiogenic isotope compositions ranging from depleted to enriched. These observations point to the preservation of enriched and depleted mantle source components and small-scale heterogeneities within the plume head. A rough trend to more depleted compositions from older to younger samples indicates that the depleted component may have become more pronounced with time. All analyzed samples from the LNR are volcanic contradicting the suggestion of some authors that it belongs to the continental Chortís block in Central America. Instead, it is likely to be part of the CLIP. The majority of samples show strongly depleted IV incompatible element and radiogenic isotope compositions, and only a few samples have more enriched compositions similar to those found throughout the CLIP region. Combined with previously published seismic data and ages of ~81 Ma for similar depleted rocks drilled on the Hess Escarpment bordering the LNR, these results suggest that the enriched rocks represent the main CLIP event, whereas the depleted rocks were generated by second-stage melting ~10 Ma after the main stage. Lithospheric thinning probably led to upwelling of still hot, residual plume head material. The plume tail stage explored in my second study area in the South Atlantic typically leads to the formation of linear hotspot tracks. In contrast to the Tristan-Gough hotspot track, the other structures in this area do not form such linear seamount chains but are more irregular. As they are poorly investigated so far, I produced comprehensive geochemical and geochronological datasets for volcanic rocks from the Discovery Rise and from the Richardson Seamount, Agulhas Ridge and Meteor Rise in order to test if the structures are related to the Discovery and Shona hotspots, respectively. Thus far, these two plumes have largely been identified based on geochemical anomalies along the southern Mid-Atlantic Ridge (SMAR) indicating plume-ridge interaction. My new 40 Ar/ 39 Ar age data for the Discovery Rise show an age progression in the direction of plate motion from 23-40 Ma supporting a mantle plume origin. The rise consists of two parallel seamount chains. These display differences in incompatible element and Sr-Nd-Hf-Pb radiogenic isotope data indicating a spatial geochemical zonation of the Discovery plume. The northern chain has compositions similar to those of the nearby Gough subtrack and characteristic for the enriched DUPAL anomaly, a large geochemical domain in the Southern Hemisphere. In contrast, the southern chain is compositionally more enriched representing an extreme DUPAL-like component. These differences in composition are reflected in lavas from the close-by SMAR pointing to a geochemical zonation of the plume for ~40 Ma. The Richardson Seamount, Agulhas Ridge and Meteor Rise show geochemical characteristics ranging from Gough-like compositions to compositions similar to those for the Shona geochemical anomaly along the SMAR. My new 40 Ar/ 39 Ar age data range from 83-72 Ma, and combined with published age data from the Meteor Rise and Shona Ridge, they indicate that the structures roughly become younger toward the SMAR. These results suggest that the structures represent the hotspot track of the Shona plume. V My new geochemical data additionally provide new insights into the origin of the DUPAL anomaly. They indicate that detached or delaminated subcontinental lithospheric mantle (SCLM) and/or lower continental crust recycled through the lower mantle could represent the source of the anomaly. Furthermore, the South Atlantic plumes are located above the margins of the African Large Low Shear Velocity Province (LLSVP), and the geochemical zonation of the Tristan-Gough plume has been attributed to sampling of material from the LLSVP (DUPAL-like Gough component) and the ambient lower mantle (Tristan component). Similar DUPAL-like compositions for the other South Atlantic plumes suggest that these plumes also sample material from the LLSVP. The occurrence of an additional extreme DUPAL-like component in the southern Discovery Rise samples, however, points to a third lower mantle reservoir besides the Gough-type reservoir and the ambient lower mantle. Thus, the distinct geochemical domains may alternatively be independent of the LLSVP.

Actions (login required)

View Item