Moho depth of the British Isles: a probabilistic perspective

Abstract

SUMMARY We present a new Moho depth model of the British Isles and surrounding areas from the most up-to-date compilation of Moho depth estimates obtained from refraction, reflection and receiver function data. We use a probabilistic, trans-dimensional and hierarchical approach for the surface reconstruction of Moho topography. This fully data-driven approach allows for adaptive parametrization, assessment of relative importance between different data-types and uncertainties quantification on the reconstructed surface. Our results confirm the first order features of the Moho topography obtained in previous work such as deeper Moho (29–36 km) in continental areas (e.g. Ireland and Great Britain) and shallower Moho (12–22 km) offshore (e.g. in the Atlantic Ocean, west of Ireland). Resolution is improved by including recent available data, especially around the Porcupine Basin, onshore Ireland and Great Britain. NE trending features in Moho topography are highlighted above the Rockall High (about 28 km) and the Rockall Trough (with a NE directed deepening from 12 to about 20 km). A perpendicular SE oriented feature (Moho depth 26–28 km) is located between the Orkney and the Shetland, extending further SW in the North Sea. Onshore, our results highlight the crustal thinning towards the N in Ireland and an E–W oriented transition between deep (34 km) and shallow (about 28 km) Moho in Scotland. Our probabilistic results are compared with previous models showing overall differences around ±2 km, within the posterior uncertainties calculated with our approach. Bigger differences are located where different data are used between models or in less constrained areas where posterior uncertainties are high.

Reflection seismic thermometry

“…Estimates of the Curie isotherm at its shallowest in the study area place it at similar average depths of ca.30 km (Fichler et al., 2011; Kubala et al., 2003). Under the graben, Moho depth does become shallow, up to 22 km in places (Licciardi et al., 2020). Referring to Figure 8a, modelled heatflow is higher towards the graben centre, which suggests some correlation.…”

A Bayesian Approach for Thermal History Reconstruction in Basin Modeling

“…These hyperparameters are effectively treated as model parameters to be inferred from the input data by the rjMcMC sampling. They provide an estimate of the posterior uncertainties on data errors and simultaneously act as weighting factors of different data types (Bodin, Salmon, et al., 2012; Licciardi et al., 2020). This is of key importance in our joint inversion scheme, as the relative importance of each data type (VR, AFT, BHT, and porosity data) is usually not known a priori.…”

“…Several authors have shown the advantages of transdimensional inference (see Sambridge et al., 2013, for a review). Examples of successful applications of this type of algorithm can be found in seismology (Bodin, Sambridge, et al., 2012; Piana Agostinetti et al., 2015; Ravenna & Lebedev, 2017), mixture modeling in geochronology (Jasra et al., 2006), geospatial analysis (Bodin, Salmon, et al., 2012; Licciardi et al., 2020), exploration geophysics (Ray et al., 2016), stratigraphy (Charvin et al., 2009), magnetotellurics (Mandolesi et al., 2018; Xiang et al., 2018), geoacoustics (Dettmer et al., 2010; Dosso et al., 2014), and borehole geophysics (Reading & Gallagher, 2013).…”

Subsurface temperature of the northern North Sea Basin

“…Estimates of the Curie isotherm at its shallowest in the study area place it at similar average depths of ca.30 km (Fichler et al, 2011;Kubala et al, 2003). Under the graben, Moho depth does become shallow, up to 22 km in places (Licciardi et al, 2020). Referring to Fig.…”