61-3 Identifying robust tectonic features by mapping agreement among seismic tomography models: Quantile area maps as a tool for research on lithospheric and upper-mantle tectonics

Abstract:

Seismic tomography reveals velocity patterns in the lithosphere and upper mantle. However, different tomographic models commonly give incompatible images.  Seismologists use a vote-map procedure to identify where deep-mantle models agree.  Here we adapt it to the lithosphere and upper mantle, where tectonic questions live. Our procedure shows which features, either high-velocity or low-velocity, are seen by most or all models and can thus be used to develop a robust interpretation.

Existing vote-map procedures usually flag a fraction of cells that varies from model to model.  Our procedure instead uses a tunable fixed-fraction percentile, flagging the same fraction of cells in every model at every depth; we call the result a Quantile Area Map (QAM).  We sample each model, either globally or for some geographic area, down to some chosen depth.  That volume is divided into cells, whose size sets the resolution of the QAM.  If we are interested in high-velocity bodies, like subducted slabs, then for each depth level, we flag all the cells where the model velocity is among the highest x% (typically 30%) of all the cells at that depth, or conversely, for low-velocity bodies like hot-spot plumes, the lowest x%.

We present a map of model agreement: the value for each cell is not a shear-wave velocity but the vote, the number of models for which the velocity is either among the highest x%, or alternatively is among the lowest x%.  For example, if high-velocity bodies are being studied and 5 existing shear-wave models are being considered, a vote of 0 or 1 means none of the models, or just one, place this cell among the x% fastest, a vote of 2 or 3 means the voting is about even, while a vote of 4 or 5 indicates that this cell is reliably high-velocity.  Structures agreed upon in this way are robustly supported by seismic tomography, and are therefore suitable for interpretation.  The QAM can then be displayed as horizontal slices at various depths, or as cross sections.

Vote maps have rarely been employed by geologists.  Our QAM approach should be applicable to many tectonic studies, for example, testing the hypothesis that the West African Craton was partially emplaced beneath the Saharan Metacraton during the Neoproterozoic Panafrican Orogeny.

© Copyright 2026 The Geological Society of America (GSA), all rights reserved.