Editor’s note: The E-Geo News has been walking readers through the various steps required to turn physical data such as core samples into a functional geological map. This is the sixth installment in this series; core collection was covered in September 2022, core describing in November 2022, lab analysis in December 2022, outcrop scouting in February 2023, and historical data mining in May 2023.

November 2023

In the process of collecting, analyzing, and finally disseminating geologic data, earth scientists must also act as data scientists. At the IGWS, two non-geologists, GIS Developer Ben Romlein and GIS Analyst Dana Bissey, weave in and out of projects at varying times, delivering tools to mapping teams to help collect data more efficiently and then organizing and storing that data for later use.

If you’re not a data person, you might think of a database as a fancy Excel spreadsheet—which is sort of true. GIS databases, though, can also be viewed on a map. Tabular data is accompanied by spatial data that corresponds to points, lines, and/or polygons that reference features on a map.

For the team mapping the Jasper quad for STATEMAP, Bissey created an online map for data collection which geologists could take with them in the field on a tablet or a phone. The initial data came from an IGWS stratigraphic database holding decades worth of geologic details. Using this map, geologists in the field could see where previous geologists had marked outcrops or other points of interest, at which thicknesses and elevations they recorded group contacts and transitions, and other features. If their observations differ from the historic notes in the database, or if they go to a location that has no historical data associated with it, they can create new points and add notes, photos, and other attributes to the location which can then be synchronized back to the original database.

Romlein has been working on systems to maintain and continually refine data from this project and others.

“Ideally, we want to have all the data in the state from every project, to be up to date,” he said. Researchers are always working to collect and publish the highest quality data, but because of obstacles like staff turnover, differing project requirements, and timing constraints, it’s an ongoing challenge to make that data available both to other researchers in the Survey and to the public. In addition to all the data being collected up to today, the Survey also maintains records dating back almost 100 years. “In the past, we’ve found that in the early phases of a project, researchers would end up reproducing a lot of work done earlier because old project data is no longer available," he said. “Some of the ways we try to address this are with workflows that store data centrally and make it easier to update as research is conducted, rather than after a project is completed.” This reduces the likelihood of data being siloed or lost.

Back in the office, geologists import the collection of point data into the mapping program ArcGIS Pro to make an initial map draft. Geologist Robin Rupp uses many sources of historical data to create the bedrock surface (discussed in the May issue of the E-Geo News), and the fieldwork team creates another layer of subsurface data from the field. Romlein wrote a script to help merge those two data layers and to predict and interpolate subsurface rock unit elevations when raw data is sparse (for example, in areas that are impractical or impossible to study outcrops or drill a borehole). The output of this script is a set of contours that track the elevation where the rock unit being mapped intersects the bedrock surface.

“And that’s where the researcher’s work starts over,” he said, “because then they look at the contours, and they say, ‘Does this make sense with where I expect the rock to be?’ … That’s where they have to start using their own experience, expertise, and knowledge of earth science and find where they expect the elevations of the units they’re mapping go up and down … pulling those contours in various places.” Teams may need to run the tool and adjust the output many times.

Once the Jasper team is comfortable with the map geometry, the project will go back to Bissey. She’ll create a database specific to this project and format it according to the Geologic Map Schema (GeMS), a data standard created by the USGS. That involves another validation process to find possible data errors, which Bissey must fix.

The Jasper quad database will be published in 2024 alongside a printable geologic map. Consumers of that data don’t have to use the Survey’s map, though; they can create and customize their own by pulling the data into a program like ArcGIS. This will allow them to add other layers like roads and county lines or their own data and customize colors and symbols.

The overall goal is for this data to be used. Bissey, who used to work for a building department, has seen what happens when major construction project teams don’t check available bedrock data for their project area: “They hit bedrock super shallow and it knocked them back months and they had to drill through it. … Had they looked at the geology of the area and had someone tell them, ‘Yeah, you’re going to hit bedrock in 10 feet,’ that, for construction purposes planning, would have saved a lot of money.”