Determining the cooling history of in situ lower oceanic crust—Atlantis Bank, SW Indian Ridge

The cooling history and therefore thermal structure of oceanic lithosphere in slow-spreading environments is, to date, poorly constrained. Application of thermochronometric techniques to rocks from the very slow spreading SW Indian Ridge provide for the first time a direct measure of the age and thermal history of in situ lower oceanic crust. Crystallization of felsic veins (∼850°C) drilled in Hole 735B is estimated at 11.93±0.14 Ma, based on U-Pb analyses of zircon by ion probe. This crystallization age is older than the ‘crustal age’ from remanence inferred from both sea surface and near-bottom magnetic anomaly data gathered over Hole 735B which indicate magnetization between major normal polarity chrons C5n.2n and C5An.1n (10.949-11.935 Ma). ⁴⁰Ar/³⁹Ar analyses of biotite give plateau ages between 11 and 12 Ma (mean 11.42±0.21 Ma), implying cooling rates of >800°C/m.y. over the first 500,00 years to temperatures below ∼330-400°C. Fission-track ages on zircon (mean 9.35±1.2 Ma) and apatite reveal less rapid cooling to <110°C by ∼7 Ma, some 4-5 m.y. off axis.Comprehensive thermochronometric data from the structurally intact block of gabbro between ∼700 and 1100 m below sea floor suggest that crust traversed by ODP Hole 735B mimics conductive cooling over the temperature range ∼900-330°C, characteristic of a 2-D plate-cooling model for oceanic lithosphere. In contrast, lower temperature chronometers (fission track on zircon, titanite, and apatite; T≤280°C) are not consistent with these predictions and record anomalously high temperatures for crust >700 m below sea floor at 8-10 Ma (i.e. 2-4 m.y. off axis). We offer two hypotheses for this thermal anomaly:

(i)

Off-axis (or asymmetric) magmatism that caused anomalous reheating of the crust preserved in Hole 735B. This postulated magmatic event might be a consequence of the transtension, which affected the Atlantis II transform from ∼19.5 to 7.5 Ma.

(ii)

Late detachment faulting, which led to significant crustal denudation (2.5-3 km removed), further from the ridge axis than conventionally thought.