In only a few short weeks, my fellowship project has undergone a series of substantial transformations. While the Fellowship course provided useful information about the inner-workings of digital humanities, having a chance to see my own project develop from start to (tentative) finish was an invaluable lesson about everything that goes into a work of digital scholarship.

The changes my project has undergone this summer are largely a result of my own naivety and my misguided belief that the mapping technology could accommodate imperfect datasets. Before submitting my proposal, I compiled a list of databases which tracked climate traced displacement throughout the 20th century. While these datasets came from a variety of sources, I hoped that I could combine them to form a single set of information which accurately depicted the climate diaspora. Instead, I discovered that datasets can rarely be melded together, and that ArcGIS and other mapping programs require an almost completely uniform set of data to create a comprehensible map. If there is one thing that I will take away from the fellowship experience, it is that a basic understanding of DH tools and programs is an absolute must before you even begin to formulate a project. With a better understanding of the capabilities of these programs, you can ensure that your chosen data can be presented in the way that you initially envisioned.

I am happy to report that through this fellowship, I now have a greater understanding of several digital skills, which I can use to inform any future projects. My greatest leap in understanding this summer took place through my continued use of QGIS. When my Fellowship partner Jay Bowen first showed me the program, it could not have looked more alien. But through continued practice, (and Jay’s bottomless pit of patience), I came to understand QGIS as a fairly user-friendly program that is adaptable to a variety of datasets. In using GIS, a scholar can generate migration paths, join csv data to vector layers, reproject, and convert data to a geojson format. From there, we were able to use Atom for scripting. Jay also showed me a brief tutorial on the program Leaflet, which we used as a Javascript library for building the interactive story map. Github will be used to host and publish the completed story map. We chose Github because it allows publishers to include short windows of background information to better explain the data and provide a human element to what would otherwise be dots on a map.

The digital storymap I created illustrates the immense migration that took place in the United States as a result of the 1930s Dust Bowl. My hope is to continue adding to this map or to create accompanying maps, which will eventually show the total number of climate migrations that occurred in the U.S. during the 20th century.