A Gradient in Cooling Rate Beneath the Moho at the Oman Ophiolite: Fresh Insights into Cooling Processes at Mid-Ocean Ridges from REE-Based Thermometry

Abstract:

The Wadi Tayin massif in the southern Oman ophiolite has a more than 10 km thick mantle section and is believed to have formed in a mid-ocean ridge like environment with an intermediate to fast spreading rate. Previously, [1] used major element geothermometers to investigate spatial variations in temperatures recorded in mantle peridotites and observed that samples near the paleo-Moho have higher closure temperatures than samples at the base of the mantle section. Motivated by these observations, we measured major and trace elements in orthopyroxene and clinopyroxene in peridotites from depths of ~1-8km beneath the Moho to determine closure temperatures of REE in the samples using the REE-in-two-pyroxene thermometer [2].

Clinopyroxene are depleted in LREE and have REE concentrations that vary depending on distance from the Moho. Samples nearer the Moho have lower REE concentrations than those deeper in the section (e.g., chondrite normalized Yb ranges from ~1.5 at the Moho to 4 at 8km depth), consistent with near fractional melting along a mantle adiabat. Orthopyroxene are highly depleted in LREE but measurements of middle to heavy REE have good reproducibility.

We find that REE-in-two-pyroxene temperatures decrease with increasing distance from the Moho, ranging from 1325±10°C near the Moho to 1063±24°C near the base of the mantle section. Using methods from [3], we calculate cooling rates of >1000°C/Myr near the Moho, dropping to rates of <10°C/Myr at the bottom of the section. The faster cooling rate is inconsistent with conductive cooling models. Fast cooling of the mantle lithosphere could be facilitated by infiltration of seawater to or beneath the petrologic Moho. This can explain why abyssal peridotites from ultra-slow spreading centers (which lack a crustal section) have cooling rates comparable to those of Oman peridotites [3].

[1] Hanghøj et al. (2010), JPet 51(1-2), 201-227.

[2] Liang et al. (2013), GCA 102, 246-260.

[3] Dygert & Liang (2015), EPSL 420, 151-161.