The Future of Tornado *Field* Research

I would be remiss to not touch on the tragedy that befell the severe weather research community this past Friday, May 31st. We lost three good men and researchers–Tim Samaras, Paul Samaras, and Carl Young–to a powerful and destructive tornado. Of the deaths that day, most occurred in vehicles, and these three were no exception.

Until late Saturday, it appeared that the main lessons taken away from the day would be the impact of media on mass panic and evacuation in a major metro area, the beautiful structure of the multiple vortex tornado, and chasers getting too close. The last point was highlighted by video from Brandon Sullivan and his crew being pounded by hay bales and debris, Reed Timmer’s Dominator vehicle losing its top, and the Weather Channel’s Tornado Hunt vehicle being tossed and rolled with Mike Bettes and company inside.

There has been enough media coverage of the passing of Tim Samaras and crew that I do not want to add empty words lauding their work and demeanor when closer friends and more familiar colleagues have written such meaningful personal messages about them.

Another Purdue colleague touched on the topic that has been discussed quietly so far, out of respect for the deaths in the community before getting into bureaucratic issues. I’d like to follow up on his points on the future of tornado research. Derrick suggests (and I, along with others, agree) that the National Science Foundation may begin limiting funding of tornado field research in favor of modeling-based studies after this event. The incident alone may not push things this direction, but combined with furloughs, budget cuts, and insurance issues, the timing may be right. While I am no longer as well-connected as I was a year ago, I fear for the future of the DOW and other mobile radar programs, as well as the countless other severe weather field research projects.

My Master’s thesis utilized data collected in the field, primarily. My planned PhD work was to complement my observationally-based findings with numerical modeling work; this is a common research path to take. The TL;DR of my thesis is that there is a relationship between WSR-88D mid/upper-level velocities and DOW ground-level velocities that could potentially be used to provide real-time tornado intensity estimates.

Terrific, right?

In theory, yes. But despite the years of data collection from the DOW, I only was able to use a handful of events for my research. There are so many hiccups in field work that to maintain a consistent method I had to discard many of the cases the DOW collected data on. The major problem in publishing my work was the limited number of data points I had, and while I agreed with the reviewers that it was an issue, it was the best we could do with the available data.

I spent a lot of the time leading up to my qualifying exams (which, for the record, I passed) studying the symbiotic relationship between field work and modeling. Why can’t we just use one or the other? Because they feed off one another. This is so important to realize: observations help to verify and improve our modeling abilities. There is only so much we can learn (albeit, a lot) from our idealized models. On the other hand, modeling helps improve our observational research. How would we know what data we need to collect without seeing the whole system at work?

We have gone through many years of this process: collect data, create model, test model, discover data holes, collect more data, repeat. The discoveries made from the original VORTEX project as well as the recent VORTEX2 project have been invaluable in improving our models.

I could write on this topic all day, as I truly believe both components of research are irreplaceable. While I hope this tragedy leads to more calculated chasing (my own crew has certainly been too close), I shudder to think of the implications of limiting field research entirely.

Leave a Reply

Your email address will not be published. Required fields are marked *