Modern Approaches to Digital Technologies for Geologic Fieldwork

Over the past two centuries, from William Smith's field investigation in England and Wales during the early 19^th century to the end of the 20^th century, the basic methodologies of field mapping and data collection remained essentially unchanged. Smith traveled around the British Isles on horseback and railway measuring the orientation of rock layers, recording observations in a field book, and using that data to hand draw a geologic map (Winchester, 2001). A visit to a capstone geology field camp in the late 20^th century would have revealed that field methods were still centered on paper-based mapping and data recording. In contrast, the broad availability of mobile digital devices in the early 21^st century has radically changing how we approach data collection in the field.

Today, geoscience professionals map and collect data in the field using approaches that our predecessors likely never envisioned. Modern field data collection tools include apps on smartphones and iPads, drones that collect high-resolution imagery, and digital outcrop models assembled with LiDAR and/or photogrammetry (e.g. Bemis et al., 2014; Cawood et al., 2017; Pavlis and Mason, 2017). These new tools have encouraged the development of new approaches to field mapping and data collection, including the use of digital compasses to measure geologic features, crowdsourcing field data collection, and the assembly and analyses of very large field datasets.

The advent of mobile devices, such as smart phones with iOS and Android operating systems, has put digital field data collection within the capabilities of almost everyone. The presence of internal sensors in these devices, such as accelerometers, gyroscopes, and magnetometers, led tech-savvy geoscientists to realize that smart phones could be used as digital compasses (McCarthy et al., 2009; Lee et al., 2013, among others). The incorporation of GPS chips in most mobile devices meant that these devices could be used for geologic mapping in the field. Mobile apps, such as Fieldmove Clino (Midland Valley, 2017) and Stereonet Mobile (Allmendinger et al., 2017), incorporate built-in digital compasses that enable fast and efficient acquisition of orientation measurements. Recent statistical analyses suggest that the precision of digital geologic compasses in iPhones and iPads is similar to that of traditional analogue compasses, such Brunton Pocket Transits (Whitmeyer et al., in review). Android devices are more variable and, at present, not as reliable for professional fieldwork.

The near-universal availability of handheld mobile devices has likewise enabled a new "crowdsourcing" approach to collecting big datasets of publicly-accessible information. A key component of crowdsourcing is the use of a group of relative novices to collect data that was traditionally the purview of professionals (Estellés-Arolas and González-Ladrón-de-Guevara, 2012). Over the past several years we have utilized undergraduate geoscience students as a "crowd" of novice mappers during a digital field mapping exercise. We collected students' digital field data into a large dataset that grew each year as we incrementally progressed across a several kilometer-square field area (Whitmeyer and De Paor, 2014). Several years of data collection produced a dense dataset of outcrop bedding orientation measurements, color-coded by lithology (Fig. 1). The density of this crowd-sourced dataset is impressive, but close examination of the map reveals several areas with conflicting lithologic information (e.g. colored dots and symbols that appear outside regions of similarly-colored polygons). Resolution of the inconsistencies in crowd-sourced field data is a challenge that we are currently addressing via comparison of the dense novice dataset with a less-dense but more accurate expert dataset (Whitmeyer et al., in review). Ultimately, we envision that the next fundamental development in field data collection for geologic maps will be the development of statistical filters and algorithms for large field datasets, in order to identify spurious and/or conflicting data.

The examples above highlight how modern field data collection methods that use digital and mobile devices are faster and more efficient that traditional analogue methods. In addition, they can produce much larger datasets via crowdsourcing approaches. However, these new approaches also require new techniques to verify the accuracy and precision of field data, such as statistical methods more commonly applied to laboratory analyses. Ultimately, these new techniques will need to be incorporated into field methods courses in undergraduate geoscience curricula.