Geosetta: Unearthing Geotechnical Data for the Benefit of All

As a public servant geotechnical engineer, I observed how costly and valuable geotechnical field explorations are to the success of every project. Often however, the field exploration is a long and complex process that is made less efficient by the difficulty of retrieving and utilizing nearby internal or external historic geotechnical datasets. As an employee of a public agency, we would often start our investigations for every project from scratch, given the difficulty of locating and utilizing historic geotechnical data. This is an ethical and technical failing of our profession. When we start subsurface explorations from scratch, not only do we often duplicate investigations, we are also likely defaulting to reapplying tradi- tional test methods that yield con- servative geotechnical parameters, resulting in overly conservative and potentially costly construction. To address this issue while working as a public servant, I started a research program with Yunfeng Zhang, Ph.D., of the University of Maryland, to build a tool that could compile internal historic geotechnical data and extract additional value from it, using machine learning. This tool was able to provide, upon request, nearby historic geotechnical data, estimated soil parameters and recommendations for geotechnical test methods best suited for a given project site. This prevented common field ex- ploration problems such as, requesting duplicate data, requesting test methods that were not applicable for the project geology, not realizing presence of karsts, missing permits for wetlands, private property and others. This tool allowed for more efficient field investigations with regards to time and money saved. There was one problem with this tool: it could only be used by internal employees of a single state's depart- ment of transportation (DOT) for which this data was compiled. I realized at that point that it would be of tremendous value to make this concept universally accessible to the geotechnical engi- neering profession. I left public service to devote time to building a platform that would host historic geotechnical data from publicly funded sources throughout the country. The tool I helped build as a public servant was the foundation of Geosetta, a nonprofit company dedicated to serving the geotechnical profession. The nonprofit structure was adopted to prevent acquisition by a for-profit company and to ensure that Geosetta remains an independent steward of our profession's valuable data. Geosetta provides a GIS (geographic information system) platform that enables users to view historic geo- technical data and extract value from it with machine learning-based predictions and visualization tools. This data is provided in an open and interchangeable format called DIGGS, which stands for Data Interchange for Geotechnical and Geoenvironmental Specialists, making it easy for users to integrate it into their workflows. Geosetta is collaborating with the American Society of Civil Engineers' Geo-Institute and other sponsors to extract public data at no cost to the agency and provide the profession with free access to the existing data and access to advanced tools through a subscription to support the nonprofit mission.