Skip to main content
  1. News
  2. Weather & Climate

Scientists juggle the idea of long-range tornado predictions

See also

Top meteorological minds have already begun to wrestle with the idea of long-range tornado season predictions similar to those that currently circulate for future hurricane seasons. The proposal began as a white paper in 2012 lead by the NOAA’s Scott Weaver, but a better understanding of the global climate will enable forecasters to more accurately predict the odds of an outbreak.

Accuracy in Prediction

Weather prediction in general relies on the comparison of observed atmospheric conditions to established standards: probability represents a statistic that could at any point fail. Beyond this, tornado prediction is inherently difficult: many of the conditions that contribute to their creation or destruction evolve quickly and constantly, and only recently has modern technology increased tornado warning times to around 15 minutes. Our long-range hurricane predictions aren’t much more accurate, and we can only pin down individual hurricanes eight days in advance (with failure). According to a Washington Post article by Rick Grow, data shows how even our best forecasters can succumb to unseen variables. In terms of predicting individual hurricanes or tornadoes, long-range forecasting isn’t even close to that precise yet.

Large-Scale Influences

Tornadoes are directly produced by columns of rotating air, but indirect causes can be as catastrophic as hurricanes or as global as ENSO. When Hurricane Rita hit the border of Louisiana and Texas in 2005, it created an outbreak of 101 tornadoes that would twist across Louisiana, Arkansas, Mississippi, Alabama, Tennessee over a two day period. Researchers have also shown that ENSO shifting from its cool phase to its warm phase during springtime can mean double the odds for major tornado outbreaks in the Midwest and South. For more information about how global weather factors into tornadoes, read the UCAR’s article on the details of long-range tornado prediction.