A failure to reject: Testing the correlation between large igneous provinces and deep mantle structures with EDF statistics

Absolute reconstructions of large igneous provinces (LIPs) for the past 300 Ma reveal a remarkable spatial pattern suggesting that almost all LIPs have erupted over the margins of the two large-scale structures in the Earth's lower mantle commonly referred to as the Large Low Shear-wave Velocity Provinces (LLSVPs). This correlation suggests that mantle plumes that have triggered LIP eruptions rose from the margins of LLSVPs, implying long-term stability of these structures and suggesting that they may be chemically distinct from the bulk of the mantle. Yet, some researchers consider the LLSVPs to be purely thermal upwellings, arguing that the observed distribution of LIPs can be explained by plumes randomly forming over the entire areas of LLSVPs. Here we examine the correlation between the LIPs and LLSVPs using nonparametric statistical tests, updated plate reconstructions, and a large number of alternative definitions of LLSVPs based on seismic tomography. We show that probability models assuming plume sources originating at the margins of LLSVPs adequately explain the observed distribution of reconstructed LIPs. In contrast, we find strong evidence against the models seeking to link LIPs with plumes randomly forming over the entire LLSVP areas. However, the hypothesis proposing that the correlation can be explained by plumes randomly forming over a larger area of slower-than-average shear wave velocities in the lowermost mantle cannot be ruled out formally. Our analysis suggests that there is no statistically sound reason for questioning the hypothesis that the LIPs correlate with the margins of LLSVP globally.