Department Logo for Earth, Planetary, and Space Sciences

Correlation of geochemical signals with MORB and OIB temperatures


Oct. 24, 2019, 3:30 p.m. - 4:30 p.m.
Slichter 3853

Presented By:
Xiyuan Bao
UCLA

See Event on Google. Subscribe to Calendar

Are hotspots hotter than ridges? Answering this question is fundamental to understanding the possible source region of hotspots vs mid-ocean ridges. Are hotspots sourced from compositionally distinct active (and therefore hotter) upwellings deep in the mantle, such as the large show shear wave velocity provinces (LLSVPs)? The geochemical signals of ocean island basalts (OIBs) found at hotspots seem to indicate that is the case. The 3He/4He ratios of OIBs is as high as 50 times the atmospheric ratio (Ra), while the typical signal for MORB is about 8 Ra. OIBs with Low ε -Neodymium, which characterize enriched sources seem to be correlated with proximity to LLSVPs. Previous studies of the excess potential temperature of hotspot lava based on the olivine geothermometer suggest that indeed hotspots are hotter than ridges by as much as ~300K. Lower excess temperatures for OIBs seem to indicate proximity to the ridge, and MORBs uninfluenced by hotspots have no excess temperature with respect to the ambient mantle. I re-examine the differences in temperature and geochemical signals between hotspots and ridges by converting seismic velocities to temperature using a state-of-the-art thermodynamic model. My preliminary results suggest that: 1) not all hotspots are hotter than ridges, indeed the two are virtually indistinguishable. Hotspots have a global mean only 5K higher than ridges (1660 vs 1655K); 2) hotspots with higher 3He/4He, buoyancy flux or lower 133Nd/134Nd tend to be hotter; 3) the range of hotspot temperatures is independent of their proximity to ridges, while the mean temperature of ridges close to hotspots are slightly higher than that of those far away; 4) the mean temperature values of both hotspots and ridges are hotter than reference adiabat of 1600K. At face value, the results imply larger compositional than temperature variations for both hotspots and ridges, an ambient mantle that melts at temperatures at the upper range of what is usually assumed, as much variability in ridges as in hotspots, and a tight degree of correlation with proximity to LLSVPs. There are however many caveats related to the inherent uncertainty in seismic tomography and the uncertain dynamical nature of plumes.