Western Massachusetts probably isn’t the first place that comes to mind when the topic is severe thunderstorms and tornadoes, but that’s where I lived as a kid when two significant severe weather events crossed my path. Two weeks after my family moved into a new home in the city of Ludlow, our house was hit by lightning, blasting charred paneling and insulation across my sister’s bedroom and exposing glowing electrical wires. A few years later, on Oct. 3, 1979, a rare (for New England) supercell thunderstorm spawned a short-lived but very strong tornado as it rolled northward up the Connecticut River Valley and into Feeding Hills, Mass., which was our home at the time. To this day, the tornado spawned by the so-called Windsor Locks, Conn., storm ranks as the ninth most destructive tornado in history. These events left a mark on me and certainly helped influence my career choice as a severe storms meteorology researcher.
Most tornadoes are (thankfully) short-lived and weak, but a very small fraction of tornadoes are long-track EF5 tornadoes, the most devastating. EF5 is the strongest class of tornado on the Enhanced Fujita scale, indicating sustained winds exceeding 200 miles per hour. A long-track EF5 leaves a damage path dozens of miles long. In the past five years, long-track EF5 tornadoes have devastated cities and towns and destroyed lives in Iowa, Missouri, Oklahoma, Mississippi and Alabama. In all cases, the tornadoes formed within supercell thunderstorms, which are long-lived and characterized by a rotating updraft.
While meteorologists have vastly improved their skill in forecasting the likelihood of severe weather occurring over a broad area, days in advance, they are currently unable to predict whether a newly formed thunderstorm will produce a tornado, much less whether a specific storm will produce a long-track EF5. Our inability to forecast the behavior of individual thunderstorms motivates my own research on supercell thunderstorms. I am interested in understanding the internal workings of supercell thunderstorms that produce long-track EF5 tornadoes, and, utilizing the “Blue Waters” supercomputer at the National Center for Supercomputing Applications (NCSA), my collaborators (Robert Wilhelmson of the University of Illinois, Louis Wicker, National Severe Storms Laboratory and Bruce Lee and Catherine Finley of WindLogics) and I have recently made significant progress in this area.
Utilizing Blue Waters, we initialized the CM1 cloud model with the atmospheric conditions of the May 24, 2011, supercell that spawned a long-track EF5 tornado near El Reno, Okla. Within this environment, a cloud was initiated by analytically forcing an updraft. This cloud grew into a supercell, and, more importantly, spawned an EF5 tornado that, over the course of its 65-mile path, produced winds near the ground of up to 300 miles per hour, which is also the strongest wind ever measured by Doppler weather radar in an observed tornado.
This simulation is a milestone in severe storms meteorology: Never before has a storm of this strength been modeled and visualized with such fidelity. To achieve this result the team had to meet both meteorological and computational challenges. Getting the model to produce the “storm we wanted” proved to be the biggest challenge, and it took dozens of attempts before we were able to simulate a supercell that produced a long-track EF5. On the computational side, the challenges primarily involved I/O. The amount of data produced by a simulation like this can be staggering – our simulation produced nearly 100 TB of data. I spent a significant amount of time trying different I/O approaches to maximize throughput during writes to Blue Water’s Lustre® file system.
Once we chose and coded up a strategy, we developed middleware to provide an easy-to-use interface to the model data, which was spread out among dozens of directories and thousands of HDF5 files. Finally, visualization, analysis and conversion tools were interfaced to the middleware. You can view visualizations of the supercell and the EF5 tornado it produces, along with recent scientific presentations given on this storm, here.
This simulation is the first of many that will explore the violent nature of long-track EF5 tornadoes and the supercells that spawn them. It will be the subject of much study, but now that we know this kind of simulation is possible, the door is wide open to numerically explore the conditions that make these devastating storms possible. Only by simulating dozens of storms in different environments, some of which produce long-track EF5s and some of which do not, will we be able to tease apart the important physical factors that help predict when storms produce this kind of tornado. If such factors can be observed with weather radar in advance of tornado formation, forecasters could provide more timely and accurate forecasts during severe weather outbreaks, ultimately saving lives.