Visualization and Sharing of 3D Digital Outcrop Models to Promote Open Science

High-resolution 3D data sets, such as digital outcrop models (DOMs), are increasingly being used by geoscientists to supplement field observations and enable multiscale and repeatable analysis that was previously difficult, if not impossible, to achieve using conventional methods. Despite an increasing archive of DOMs driven by technological advances, the ability to share and visualize these data sets remains a challenge due to large file sizes and the need for specialized software. Together, these issues limit the open exchange of data sets and interpretations. To promote greater data accessibility for a broad audience, we implement three modern platforms for disseminating models and interpretations within an open science framework: Sketchfab, potree, and Unity. Web-based platforms, such as Sketchfab and potree, render interactive 3D models within standard web browsers with limited functionality, whereas game engines, such as Unity, enable development of fully customizable 3D visualizations compatible with multiple operating systems. We review the capabilities of each platform using a DOM of an extensive outcrop exposure of Late Cretaceous fluvial stratigraphy generated from uninhabited aerial vehicle images. Each visualization platform provides end-users with digital access and intuitive controls to interact with large DOM data sets, without the need for specialized software and hardware. We demonstrate a range of features and interface customizability that can be created and suggest potential use cases to share interpretations, reinforce student learning, and enhance scientific communication through unique and accessible visualization experiences. INTRODUCTION High-resolution 3D digital models are becoming increasingly common data sets in academic and commercial applications. In the geosciences specifically, digital outcrop models (DOMs), or virtual outcrops, can provide geoscientists with photorealistic models that preserve spatial precision, dimensionality, and geometric relationships between geologic features that are inherently 3D and susceptible to distortion and/or loss of information when rendered in 2D (Bellian et al., 2005; McCaffrey et al., 2005; Jones et al., 2009). Digital 3D mapping approaches using DOMs have enabled geoscientists to perform supplemental measurements, correlations, and interpretations that are difficult or impossible to obtain with traditional methods (Figs. 1–2; Pavlis and Mason, 2017; Nesbit et al., 2018). Until recently, however, collection and use of digital data sets has been limited to specialists, due to hardware and software limitations. A number of methods are now available for collecting and processing 3D models (Hodgetts, 2013; Carrivick et al., 2016). In particular, structure-from-motion and multi-view stereo (SfM-MVS) photogrammetry software, commonly paired with uninhabited aerial vehicles (UAVs), enables geoscientists to produce photorealistic DOMs through a highly streamlined UAV-SfM workflow (Chesley et al., 2017; Nieminski and Graham, 2017; Pavlis and Mason, 2017; Nesbit and Hugenholtz, 2019). Related efforts have centered on the development of 3D software solutions with tools for geoscience applications. Custom software packages, such as Virtual Reality Geology Studio (VRGS; Hodgetts et al., 2007) and LIME (Buckley et al., 2019), offer users lightweight executable tools and opportunities to analyze and revisit data at multiple scales. Open source programs, such as Blender and CloudCompare, can be used for data exploration and measurement and have also integrated specific geoscience toolsets (e.g., Brodu and Lague, 2012; Dewez et al., 2016; Thiele et al., 2017). Although acquiring DOMs has become more straightforward, and various 3D analysis programs are available, dissemination of DOMs, interpretations, and results has remained a challenge due to software and file-size barriers. Specialty 3D programs are often hindered by product licensing and can involve a considerable learning curve to understand controls, file formats, and integrated tools. Furthermore, DOMs can easily exceed multiple gigabytes (GB) in size, which can be taxing on computational resources for rendering, file storage, and data transfer. With the growing collection of high-resolution DOMs and similar 3D data sets, there is a need for dedicated, intuitive, and accessible 3D visualization platforms. Given the challenges outlined above, we examined existing visualization solutions that could potentially enable sharing of DOMs and support open science through increased data accessibility. To provide a functional introduction to modern visualization platforms, we illustrate the capabilities and functionality of two web-based interfaces (Sketchfab and potree) and a crossplatform videogame engine (Unity) using a geologic case study. A DOM was produced through a UAV-SfM workflow for an extensive outcrop (1 km2) exposed within the badland landscape of Dinosaur Provincial Park (Alberta, Canada). Each visualization platform provides access to the large DOM through an intuitive lightweight interface without the need for high-end hardware, Visualization and Sharing of 3D Digital Outcrop Models to Promote Open Science GSA Today, v. 30, https://doi.org/10.1130/GSATG425A.1. Copyright 2020, The Geological Society of America. CC-BY-NC. specialized software, or transfer and storage of large files. This prompts an increased ability to share data sets, interpretations, and results with a wider community, expanding opportunities for scientific communication and open science education.