Caldeira Lab Research:Paleoclimate and geochemical cycles

Antipodal Hotspot Pairs on the Earth

Michael R. Rampino & Ken Caldeira

Statistical analysis performed on observed data reveals the interesting fact that a large percentage of hotspots appear in nearly antipodal pairs -- much more than would be expected if their distribution was random. This paper is an investigation of what caused hotspots to form in this manner.

Rampino, M.R. and Caldeira, K., Antipodal hotspots on the Earth, Geophysical Research Letters 19, 2011–2014, 1992

Significance of the number of antipodal pairs found: To determine the statistical significance of the results, the amount of antipodal pairs found in observed data is compared to the amount of antipodal hotspots found in two simulations. In the H1 simulation, hotspots are distributed uniformly. In the H2 simulation, 1000 new "perturbed" distributions were made to determine the number of antipodal pairs in a random distribution. The number is generally higher in the observed data, implying that it is not a coincidence.


The results of statistical analyses performed on three published hotspot distributions suggest that significantly more hotspots occur as nearly antipodal pairs than is anticipated from a random distribution, or from their association with geoid highs and divergent plate margins. The observed number of antipodal hotspot pairs depends on the maximum allowable deviation from exact antipodality. At a maximum deviation of <700km 26% to 37% of hotspots form antipodal pairs in the published lists examined here, significantly more than would be expected from the general hotspot distribution. Two possible mechanisms that might create such a distribution include: (1) symmetry in the generation of mantle plumes, and (2) melting related to antipodal focusing of seismic energy from large-body impacts.