IN048-08
LoopStructural 1.0: Time aware geological modelling

Thursday, 17 December 2020: 05:58
Virtual
Lachlan Grose, Monash University, School of Earth, Atmosphere and Environment, Melbourne, VIC, Australia, Laurent Ailleres, Monash University, School of Earth Atmosphere and Environment, Melbourne, VIC, Australia, Gautier Laurent, Univ. Orléans, CNRS, BRGM, ISTO, UMR 7327, F-45071, Orléans, France, Michael Hillier, Geological Survey of Canada, Ottawa, ON, Canada and Mark Jessell, The University of Western Australia, Centre for Exploration Targeting (School of Earth Sciences), Crawley, WA, Australia
Recorded Presentation
Abstract:
Building 3D geological modelling has previously been limited to expensive commercial software with limited or no programmatical interfaces. These packages generally have limited access to the parameters controlling the fundamental algorithms used for 3D geological modelling. Usually only one interpolation algorithm is provided by modelling packages. This means that modellers are forced to use only one type of interpolation algorithm and are unable to use the algorithm most suitable to their problem. To make comparisons between interpolation algorithms a modeller will have to transfer data between proprietary data structures and initialise models using package specific approaches.

In this contribution we introduce LoopStructural, a new open source 3D geological modelling python package (www.github.com/Loop3d/LoopStructural). LoopStructural provides a generic API for 3D geological modelling applications harnessing the core python scientific libraries Pandas, Numpy and Scipy. Six different interpolation algorithms including 3 discrete interpolators and 3 polynomial trend interpolators, can be used from the same model design. This means that different interpolation algorithms can be mixed and matched within a geological model allowing for different geological objects e.g. different conformable foliations, fault surfaces, unconformities to be modelled using different algorithms. We use LoopStructural to produce regional scale models comparing the surfaces produced by different interpolation algorithms. Our results show that for complex terranes using polynomial trend interpolation algorithms produces surfaces that more closely fit local observations but produce artefacts in poorly constrained areas e.g. subsurface or where outcrops are missing. Discrete interpolators usually produce smoother models with more geologically reasonable surfaces away from datapoints at the expense of higher misfit for local observations.

<< Previous Abstract | Next Abstract >>