Year One: Flint Water Crisis
In this year, our fellows focused on projects relating to various aspects of the Flint Water Crisis. The first semester was spent delving into the complex systems that contributed to the Flint Water Crisis and the city’s recovery following the water contamination. In the second semester, two groups formed to a) identify and map networks from emailed communications between individuals involved in the water crisis and b) utilize the aftermath of the crisis to explore how data science can have unexpected and lasting repercussions on the communities they are designed to help. Much of the work of this cohort continued on in future years, with the following publications emerging from the work that began in this PFI.
Nadel P, Smith KM. Reconstructing decision-making dynamics during public health crises by applying data science to public records. J Public Health Policy. 2024 Dec 18. doi: 10.1057/s41271-024-00540-y. Epub ahead of print. PMID: 39695218.
Rabb, N., Knox, C., Nadgir, N., Islam, S. (2024). A Tale of Two Cities: Information Diffusion During Environmental Crises in Flint, Michigan and East Palestine, Ohio. In: Cherifi, H., Rocha, L.M., Cherifi, C., Donduran, M. (eds) Complex Networks & Their Applications XII. COMPLEX NETWORKS 2023. Studies in Computational Intelligence, vol 1144. Springer, Cham. https://doi.org/10.1007/978-3-031-53503-1_15
Year Two: Defense Against the Dark (Data) Arts
This PFI was inspired by a challenge common to today’s media: despite the same data availability, drastically different conclusions are reached and utilized to support opposing policies. This year began with an analysis of gun violence data, which had been interpreted by both right-leaning and left-leaning media to reach different conclusions. Following an in-depth study of this dataset and phenomenon, the group researched similar cases of potentially deceptive data management, raising questions about whether or not data can ever be ‘objective’ and focusing on developing a critical eye towards data. The projects during this PFI introduced students to the world of data journalism.
Year Three: Integrated Assessment Models in Setting Climate Direction
Instead of a particular topic, PFI 3 focused on a class of models called Integrated Assessment Models, which are frequently used to set climate policy in the USA and globally. These models are often challenging to understand, run, and explore, often generating terabytes of data at one time. In the fall semester, students utilized the High Performance Computing Cluster to build and run the Global Change Assessment Model (GCAM), which was developed by the Joint Global Change Research Institute (JGCRI). The students split into two groups, which explored one project representative of more typical GCAM uses, and a project that pushed GCAM to explore how conflict and catastrophe can (or perhaps, cannot) be integrated. This pushed the students to ask critical questions about models and their use in setting global agendas.
Year Four: Cascading Hazards
It’s not an unfamiliar story: climate change leads to drought in southern california, leading to uncontrollable wildfires that now leave communities at risk of debris and mudslides due to unpredictable winter downpours. Despite living with patterns of cascading hazards, many of these connections are still uncharacterized and not very well understood. The goal of the current PFI is to explore how cascading patterns are manifesting and developing in multiple sectors. This PFI is still evolving as the year progresses.