Quantifying ecological impacts of mass extinctions with network analysis of fossil communities
By A. D. Muscente, Anirudh Prabhu, Hao Zhong, Ahmed Eleish, Michael B. Meyer, Peter Fox, Robert M. Hazen, and Andrew H. Knoll
Summarized by: Paul Ward. Paul Ward is a geology major at the University of South Florida. Currently, he is a senior. Once he earns his degree, he plans on taking the GIT and plans to work in the hydrology field. When he is not working on geology, he likes to go fossil hunting and cook.
What data were used: Data were collected using the Paleobiology Database on fossil occurrences, taxonomy, and diversity across mass extinction events through geologic time
Methods: Using network theory (essentially, it means we treat fossil occurrences as complex and interconnected-like how many fossils interacted together in paleoecosystems) and the Paleontological database of fossil occurrence, taxonomy, and diversity over time, they compiled all of this data to show co-occurrence of fossils with a custom application that was made in python, a coding language. The results were then analyzed in RStudio.
Results: The data that was acquired during the project was compiled to create a record of fossilized species from the paleontological database to determine how communities are affected by ecological change. Using this dataset, it was shown how communities rise and fall during a mass extinction event (figure 1). The data that was acquired also shows the different severities on ecology of each extinction: for example, the Permo-Triassic extinction had an extremely severe negative impact on ecology, whereas other extinctions were not nearly as severe. Through the data it was also observed that the Devonian extinction importance was underestimated in the severity of the event. The data showed that it is close in severity to the K-Pg extinction event where it was previously a whole rank lower than observed in this study.

Why is this study important: The significance of the data that was compiled shows us how the different taxa react to the severity of the extinction event and the selectivity that an event may have affected different communities compared to others. The data can also show us how these different extinctions affect ecological variation when compared (e.g., the Permo-Triassic had a negative impact on reef-building organisms, which when they go extinct, causes a significant ecological collapse).
The big picture: This data analysis is important for the larger paleobiology community, due to the ability to show trends that occurred in the different geologic ages. With this, what is known about the causes of previous extinction events can show how different species react to different adverse conditions. With the example of coral ecology, we can better estimate how Earth’s ecosystems will react to climate conditions today from anthropogenic influences.
Citation: Muscente, A. D., Prabhu, A., Zhong, H., Eleish, A., Meyer, M. B., Fox, P., Hazen, R., Knoll, A. (2018). Quantifying ecological impacts of mass extinctions with network analysis of fossil communities. Proceedings of the National Academy of Sciences of the United States, (20), 5217. https://doi-org.ezproxy.lib.usf.edu/10.1073/pnas.1719976115