THE FUTURE OF PUBLISHED PAPERS: IMPROVING REPRODUCIBILITY PRACTICES FOR USGS PUBLICATIONS

Some of the major cornerstones of modern science are the ideas of reproducibility and reusability. Ideally, once a paper is published, a reader should be able to verify the author’s work and conclusions. For many disciplines, including the geosciences, published works lose valuable context and data each year to preventable factors including outdated technology, deleted files, and poor digital practices, making replication of work difficult or impossible. In order to increase the reproducibility of scientific publications, future USGS geoscientists must take steps to increase the credibility of their published works, including but not limited to publishing and citing all final and intermediate data, publishing and citing any and all software, curating proper and useful metadata to improve the findability of these resources, outlining procedure and methods to an extent which would improve the reproducibility and credibility of the author’s conclusions, and using digital object identifiers (DOIs) and Open Researcher and Contributor identifiers (ORCIDs) to persistently identify various inputs and authors. In order to ensure that these solutions would impact the usefulness of the research process, multiple published manuscripts were assessed based on a reproducibility rubric developed by the authors. The rubric helps to determine how easily research inputs, such as methods, data, software, and samples can be found and accessed. Ultimately, these reproducibility practices will contribute to the geoscientist’s ability to build off of the research of those before them, decreasing the cost and time required to do research, and creating a more ideal and efficient scientific process. This work is supported by the USGS and UNAVCO Geo-Launchpad Intern Program.