Projects.
The Mississippi River Flood of 1874.
Mapping aid requests and distribution for a major Reconstruction-era flood in Mississippi.
Description.
My project adds to the sparse literature surrounding the Flood of 1874, a significant flood in the Upper Mississippi River in the devastating aftermath of the Civil War. This flood led to Congress adopting a ‘levees-only’ policy, which indirectly contributed to the Great Mississippi River Flood of 1927, one of the worst natural disasters in the nation’s history. A spatial and anecdotal analysis increases understanding of the disaster and federal aid, foregrounds the dire conditions of flood victims, and may inform future disaster relief efforts.
I visualize the extent of the suffering caused by the Flood of 1874 and the subsequent federal aid response. Drawing from the Civil War and Reconstruction Governors of Mississippi (CWRGM) database, I read over 130 historical letters written to Mississippi governors during the Reconstruction area and extracted 5 data points for each letter. I categorized letters into three groups: those requesting aid, those containing receipts or confirmations of aid being delivered, and those containing correspondence between local and national government officials. I then created two webmaps showing the spatial distribution of aid requests and aid deliveries across Mississippi. Additionally, my paper explores the anecdotal evidence found in the letters to complement the quantitative findings.
I completed this research project during the 2025 Mapping Freedom NSF-REU program at the University of Southern Mississippi in Hattiesburg, MS. You can read more about the program at the Mapping Freedom website, and view my weekly blog and webmaps on the personal site I created as a deliverable.
Deliverables.
Loading Viewer…
Loading Viewer…
Community Preservation in Iowa.
Quantitatively identifying communities of interest in Iowa with clustering algorithms.
Description.
This project adds to the literature on gerrymandering, communities of interest, and redistricting. Gerrymandering, or the practice of manipulating electoral district boundaries to gain a political advantage, has been and continues to be a significant obstacle to fair and equitable elections. Maps are regularly redrawn in 10-year redistricting cycles to promote fairness. The notion of “communities of interest” has recently emerged in redistricting efforts on all levels, broadly defined as groups of people, often geographically related and sharing common characteristics or interests, that should be maintained during redistricting. Quantitative methods for community identification compliment the existing qualitative research and may help inform future redistricting efforts.
I collected 1,062 individual demographic, social, economic, and housing statistics at the census tract level from the American Community Survey’s 2022 5-year estimate datasets, then narrowed down these data into summary statistics commonly used to define communities. Additionally, I collected geographical data such as electrical service areas, school district boundaries, and electoral maps, which identify regions with shared resources or similar infrastructure. The statistical data were aggregated and merged with 2022 census tract geographies to produce a shapefile, which was then converted into an undirected and unweighted dual graph. First, I used single-linkage hierarchical clustering with Euclidean distance to create geographical regions with similar summary statistics. Then, I clustered these regions using complete-linkage clustering with Modified Hausdorff Distance to find overlaps between these and the geographic boundary data. This created a heat map of overlapping regions, where areas with high overlap indicate strong candidates for communal preservation.
I completed this work for my Mentored Advanced Project (MAP) at Grinnell College in Grinnell, IA during the summer of 2024. I then presented my work and poster at the 2025 Joint Mathematics Meeting (JMM) in Seattle, WA.
Deliverables.
Loading Viewer…