Global crustal thickness from neural network inversion of surface wave data

Abstract

S U M M A R YWe present a neural network approach to invert surface wave data for a global model of crustal thickness with corresponding uncertainties. We model the a posteriori probability distribution of Moho depth as a mixture of Gaussians and let the various parameters of the mixture model be given by the outputs of a conventional neural network. We show how such a network can be trained on a set of random samples to give a continuous approximation to the inverse relation in a compact and computationally efficient form. The trained networks are applied to real data consisting of fundamental mode Love and Rayleigh phase and group velocity maps. For each inversion, performed on a 2 • × 2 • grid globally, we obtain the a posteriori probability distribution of Moho depth. From this distribution any desired statistic such as mean and variance can be computed. The obtained results are compared with current knowledge of crustal structure. Generally our results are in good agreement with other crustal models. However in certain regions such as central Africa and the backarc of the Rocky Mountains we observe a thinner crust than the other models propose. We also see evidence for thickening of oceanic crust with increasing age. In applications, characterized by repeated inversion of similar data, the neural network approach proves to be very efficient. In particular, the speed of the individual inversions and the possibility of modelling the whole a posteriori probability distribution of the model parameters make neural networks a promising tool in seismic tomography.

Moho Density Contrast in Central Eurasia from GOCE Gravity Gradients

“…For the least-squares estimation, the weight matrix was chosen to be the identity matrix. We then repeated the computation for the Moho parameter T0 taken from the seismic models GSMM [41] and M13 [42]. Since these two models provide information only on the Moho depth, we adopted the value ∆ρ0 from CRUST1.0.…”

Modelling Moho depth in ocean areas based on satellite altimetry using Vening Meinesz–Moritz’ method

“…Much research using seismic surveys to estimate the global crust-mantle boundary has been made in the last decades. For example, Shapiro and Ritzwoller (2002) and Meier et al (2007) compiled global Moho models based purely on seismic data analysis, and Lebedev et al (2013) estimated the Moho depth using seismic surface waves. For global studies the most frequently used crustal models are the CRUST2.0 (Bassin et al 2000) and CRUST1.0 model, the latter compiled with a 1 9 1 arc-degree spatial resolution.…”

Visualizing the Structure of the Earth’s Lithosphere on the Google Earth Virtual-Globe Platform

“…The implementation of the proposed framework can be divided into four key steps: (1) exporting lithospheric information from LITHO1.0; (2) encoding global lithospheric models in KML format; (3) representing the global lithosphere with two scales; and (4) visualizing and disseminating the global lithospheric structure on Google Earth. The step-by-step execution is explained below.…”

“…It is of the order of 100 km thick and comprises the crust and the uppermost solid mantle [1]. Understanding the structure of the Earth's lithosphere is particularly useful because it provides a key to understanding a broad range of applications including improving whole-mantle tomography, defining and understanding crust-mantle interaction, monitoring seismicity at regional or global scales, and understanding the linkage and interaction between the atmosphere and the Earth's deep interior [2][3][4][5]. Over the years, a number of global models with various levels of detail, such as 3SMAC [6], CRUST 5.1 [2], CRUST 2.0 [7], CRUST 1.0 [8], and LITHO1.0 [5,9], have been presented to depict structural features and property parameters of all or part of the Earth's lithosphere.…”