Geologically Current Directions of Motion of 53 Hotspots Estimated from Monte Carlo Inversion

We previously estimated geologically current rates of hotspot motion of 2 to 4 mm/yr from Monte Carlo inversion of the trend of 56 young tracks of hotspots.  Plate motions were constrained to consistency with the MORVEL set of plate relative angular velocities. To determine the average rate of motion, each realization randomly assigns a motion direction to each hotspot and a globally uniform rate of motion is imposed ranging from 0 to 15 mm/yr. We require the misfit for the solution set to lie in a range that is neither too small nor too large given objectively estimated uncertainties of observed hotspot trends. From one million realizations, only 21,749 (≈2%) gave an acceptable fit.

The set of successful solutions also contains information about what directions of hotspot motion produce misfits to the observed trends that are significantly better than those obtained assuming fixed hotspots.  For each hotspot we generate a Rose diagram showing the distribution of the direction of motion for the successful realizations.

We test the directions of motion for each hotspot using the Rayleigh test of uniformity.  Six of the 53 hotspots have a value of p > 0.05, which is not significantly different from a uniform distribution. The other 47 hotspots tend to move perpendicular to the plate-motion direction (p=5.1 × 10^–11 for the Rayleigh test applied to the set of hotspot-motion directions relative to the local plate motion direction).

Exceptions to this pattern occur for hotspots on ultra-slow-moving lithosphere.  Because they are sited on ultra-slow-moving lithosphere, the tracks of these hotspots may record the direction of motion of individual hotspots relative to the mean hotspot reference frame.  Examples of hotspot tracks on the Eurasian, Antarctic, and part of the Nubian plate, all sites of ultra-slow-moving lithosphere, will be examined and discussed.