Skip to main content

A novel Augmented Reality Sandbox for Outreach and Education in Geoscience

Seeing Geology is hard! To build an intuitive understanding of complex 3 dimensional structures in the subsurface requires a lot of training. To make this experience more accessible, we present here a concept for haptic interaction with geological models in 3-D, through a link of advanced geological modeling with an augmented reality (AR)-sandbox. Augmented Reality Sandboxes are interactive devices in which a sculptable sand surface is constantly scanned by a depth camera and a computed image is projected back onto the sand. Such AR Sandboxes are a popular tool for geoscience education and outreach. However, existing functionality is commonly limited to the visualization of topography with contour lines and colors and lacks subsurface informations. We present Open-AR-Sandbox, an open-source, python based Augmented Reality sandbox designed specifically for the use in geoscience education and outreach. In addition to the visualization of topography it implements a variety of modules to interactively explore geological concepts and processes such as the relation of topography and the geological subsurface, mass transport simulation or interactive geological mapping exercises. Our solution utilizes the powerful open-source geological modelling Software GemPy which makes it easy to create custom geological models for the use in the sandbox and even makes it possible to change model parameters on the fly. With the help of computer vision, optical markers placed on the sand are automatically recognized to display for example arbitrary cross sections through a model or setting the location of a landslide. Further control elements and information such as parameter sliders, buttons or interactive 3d views of models and topography can be accessed conveniently on an additional screen or touchscreen. Both Open-AR-Sandbox ( as well as GemPy ( are written in Python and released under a LGPL-3.0 open-source License. Contributions to these projects as well as comments from the community are welcome.


Daniel Escallón Botero (1), Simon Virgo (1,2) & Florian Wellmann (1,2)
RWTH Aachen University, Germany (1); Teranigma Solutions GmbH (2)
GeoUtrecht 2020