Second-stage Caribbean Large Igneous Province volcanism: The depleted Icing on the enriched Cake

Abstract

A long-standing question has been whether the Lower Nicaraguan Rise in the western Caribbean Sea represents thinned continental crust or is part of the Caribbean Large Igneous Province (CLIP). During the R/V Meteor cruise M81/2, the northeastern part of the Lower Nicaraguan Rise including the Hess Escarpment was extensively sampled. Only volcanic and carbonate rocks were recovered and no metamorphic rocks or rocks with continental crustal affinities, indicating that at least the uppermost basement is volcanic and belongs to a volcanic province, probably the CLIP. We provide a comprehensive geochemical dataset, including major and trace elements and Sr-Nd-Hf-Pb double spike isotope ratios, for the volcanic rocks. The basalts are geochemically subdivided into two groups: 1) depleted and 2) enriched. The majority belong to the depleted group with 6–9 wt% MgO and highly depleted incompatible element (e.g. (La/Yb)_N = 0.15–0.76, (La/Sm)_N = 0.21–0.65) compositions and depleted radiogenic isotope (e.g. εNd_i = 8.9–11.0, εHf_i = 13.6–16.5, ²⁰⁶Pb/²⁰⁴Pb_i = 18.28–18.98) ratios. The subordinate enriched group basalts have 5–8 wt% MgO and were solely recovered from the lower part of the central Hess Escarpment. These rocks possess enriched incompatible element ((La/Sm)_N = 1.15–1.60, (La/Yb)_N = 1.40–2.57) and isotopic (εNd_i = 7.52–8.34, εHf_i = 12.6–12.9, ²⁰⁶Pb/²⁰⁴Pb_i = 18.87–19.07) ratios resembling E-MORB. The depleted group is distinct from Pacific and Atlantic MORB based on combined ²⁰⁶Pb/²⁰⁴Pb, ¹⁴³Nd/¹⁴⁴Nd and ¹⁷⁶Hf/¹⁷⁷Hf data. It is similar, but not identical to the highly depleted Gorgona komatiites, thought to be generated during the CLIP event, and depleted Galápagos hotspot track lavas accreted to the Pacific margin of Central America. Ages of ~81 Ma from rocks drilled on the Hess Escarpment postdate the main CLIP phase (95-83 Ma), and indicate that at least the northeastern part of the Nicaraguan Rise formed during a later volcanic event. Lithospheric thinning after the main CLIP phase is suggested to have triggered upwelling of still anomalously hot, partially-melted or unmelted CLIP mantle material to form the younger depleted group of lavas. Our investigations imply that second-stage melting to generate depleted compositions during formation of oceanic large igneous provinces, as has been recently proposed for the Manihiki Plateau, is a more common process than originally thought.