Simulating Self-Assembly of Tornado Storm Chasers Using Agent-Based Modeling

Alex Moxon, “Simulating Self-Assembly of Tornado Storm Chasers Using Agent-Based Modeling”
Mentor: Paul Roebber, Mathematical Sciences

Given the risk to public health and safety and the limited ability to forecast their occurrence, tornadoes are of great research interest to atmospheric scientists. Most importantly, more data is needed to better understand their development and evolution, with the ultimate goal of better predictions. On days in which tornadoes are forecast, storm chasers travel to areas where these storms are expected to develop with the intention of obtaining these data. However, this goal is difficult to accomplish because of timing and location uncertainties, issues that are further complicated by the inherent safety risks that they present. This project uses agent-based modeling techniques to determine the most ideal locations and routes that storm spotters can take during a simulated tornadic thunderstorm event. This research will help us to understand whether the emergent behavior of a cluster of storm chasers would be sufficient to satisfy both safety and data collection goals, compared to more traditional, centrally controlled field experiment deployments. Some recent experiences with real tornado events suggest this possibility, and there is a need to test whether such a result was unusual or would be characteristic of this kind of activity. Such a result would influence the design of future scientific field experiments and would be of great interest to the community.


  1. Nice job Alex! I’m wearing my Atmo Club tornado-chasing shirt as I watch your presentation.

  2. Nicely done, Alex! I look forward to seeing how everything progresses in the next several months!

  3. A good presentation and an interesting topic. I wanna hear more about the agent’s rules when I run into you next. Well done and take care.

  4. Hi Alex – your presentation materials were easy to follow, even though this is not my area of expertise. Well done!

Leave a Reply

Your email address will not be published.