Geophysical Processing, Modelling and Inversion Tools

2D & 3D Geophysical Potential Field Tools

A suite of open-source tools for processing, modelling and inversion of gravity and magnetic data. Developed at the Centre for Exploration Targeting, University of Western Australia, with funded by the MinEx CRC, the West African eXploration Initiative (WAXI), EU Marie Skłodowska-Curie Actions and the Australian Research Council.

Tools & Resources

Δg High ρ Low ρ MPI Parallel Processing
Fortran / MPI

Tomofast-x v2.0

A powerful 3D parallel inversion platform for single-domain or joint inversion of gravity and magnetic data. Supports gravity gradiometry (FTG), multi-component magnetics, and magnetization vector inversion including remanence. Features wavelet compression of sensitivity kernels for memory efficiency, arbitrary surface topography, and petrophysical constraints. Scales from desktops to supercomputers.

Joint Inversion Wavelet Compression MPI Parallel Topography ADMM Constraints FTG Support
View on GitHub
RUN
QGIS Plugin

Tomofast-x-q

QGIS plugin to streamline Tomofast-x workflow. Load gravity or magnetic data (CSV/TIF), define mesh parameters, configure inversion settings, and generate parameter files. Supports IGRF field calculations, memory estimation, and can run inversions directly via WSL on Windows.

Mesh Generation Parameter Files IGRF Calculator WSL Support Memory Estimation
View on GitHub
import tomofast_utils # Convert model grid mesh = create_mesh ( nx, ny, nz ) # Visualize results plot_model ( model ) export_vtk ( "output.vtk" ) Python
Python Utilities

Tomofast-tools

Python utility scripts for Tomofast-x workflows. Create and convert model grids, add noise to synthetic data, export models to various formats, find magnetic vectors, convert Noddy models to Tomofast format, and visualize inversion results.

Mesh Creation Model Export Noddy Conversion Visualization Noise Addition
View on GitHub
1. Deposit 2. Fold 3. Fault 4. Intrude → Grav → Mag
Windows / Python

Noddy

Rapidly build complex 3D geological models through kinematic event histories and calculate resulting gravity and magnetic fields. Define sequences of folding, faulting, intrusion, and depositional events. Ideal for teaching structural geophysics and testing geological scenarios. Includes pynoddy Python wrapper for batch processing.

Kinematic Modelling Event Histories Forward Modelling pynoddy Teaching Tool
Visit Noddy Website
Nullspace Same Δg response GeoMos
Python

GeoMos Nullspace

Explore alternative geological scenarios through nullspace navigation. Based on the "nullspace shuttles" algorithm, this tool perturbs inversion models while maintaining the same gravity response. Applied to the Western Pyrenees case study. Integrates with Tomofast-x sensitivity matrices including wavelet compression.

Nullspace Shuttles Scenario Testing Tomofast-x Integration Model Uncertainty
View on GitHub
+ Trans-D Bayesian
Python

Trans-D Level Set

Trans-dimensional Bayesian inversion for 3D geometrical recovery of geological structures from gravity and magnetic data. Uses level set geometry with birth-death MCMC sampling to explore models with varying numbers of geological units, enabling data-driven determination of model complexity. Integrates with Tomofast-x sensitivity matrices.

Trans-Dimensional Level Set Birth/Death MCMC Bayesian Tomofast-x Uncertainty
View on GitHub
Raw TMI RTP After RTP Derivatives Tilt Angle Continuation Euler Wavelets PCA/ICA
QGIS & ArcGIS Pro

SGTool

Structural Geophysics Tool for potential field grid processing. Comprehensive suite including RTP/RTE, continuation, derivatives, tilt angle, analytic signal, bandpass filters, AGC, Euler deconvolution, WTMM wavelets, PCA/ICA analysis, and BSpline gridding. Supports WAXI/Agate Structural Geophysics courses.

RTP/RTE Euler Deconvolution Tilt Angle Wavelets PCA/ICA Gridding
View on GitHub
Magnetic Gravity GLCM t-SNE Atlas Haralick Textures Data Fusion t-SNE Interactive Geo-Linked Multi-scale
Python

t-SNE Atlas

Interactive geophysical feature discovery through Haralick texture-based data fusion and t-SNE dimensionality reduction. Integrates magnetic and gravity datasets into a unified representation space, enabling autonomous pattern recognition and navigation without predefined assumptions. Links clustered features back to geographic coordinates for intuitive exploration of geological relationships.

Feature Fusion t-SNE Haralick Textures Interactive Pattern Discovery Multi-scale
View Paper
Model Ensemble N=10 Entropy Voxet Low High Metrics • Histogram • Variogram • Connectivity • Wavelets • Topology Cardinality Entropy Dissimilarity Connectivity Wavelets Topology
Python

loopUI-0.1

Uncertainty indicators for 3D geological modelling. Provides local measures (cardinality, Shannon entropy, continuous entropy) and global dissimilarity metrics (histogram, semi-variogram, connectivity, multiple-point statistics, wavelets, topology) to quantify and visualize uncertainty across ensembles of voxet models. Supports both discrete lithological units and continuous property fields.

Uncertainty Entropy Dissimilarity Voxet Analysis Connectivity Topology
View Paper

Development Team

Researchers and developers contributing to the Tomofast-x ecosystem

Guillaume Pirot

Guillaume Pirot

Senior Research Fellow
Jérémie Giraud

Jérémie Giraud

Research Fellow
Leonardo Portes dos Santos

Leonardo Portes dos Santos

Adjunct Senior Research Fellow
Lizzie Bruce

Lizzie Bruce

PhD Candidate
Mark Jessell

Mark Jessell

Professor
Marko Vanić

Marko Vanić

PhD Candidate
Michel Nzikou Mamboukou

Michel Nzikou Mamboukou

Research Fellow
Vinicius Antunes Ferreira da Silva

Vinicius Antunes Ferreira da Silva

PhD Candidate
Vitaliy Ogarko

Vitaliy Ogarko

Research Fellow