main concepts scientists sections
SimplePopularAdvanced

Why Thunderstorms Travel in Packs

Original: "Space-time extremes of severe US thunderstorm environments"
· Jonathan Koh, Erwan Koch, Anthony C. Davison
arXiv:2201.05102 · 2022-01-13 · CC BY 4.0 · 1 min
Applied Statistics stat.ME
A warm ocean current makes thunderstorms in the U.S. cover more ground.

A thunderstorm front is like a rubber band. If you leave the band alone, it lies bunched up, and storms form in isolated spots. If you stretch the band, the band of bad weather widens and covers several states at once. Scientists wanted to understand what exactly stretches this band in the atmosphere.

They collected data on stormy weather in the U.S. over 37 years. For each month, they picked the strongest values of three indicators: how much heat and moisture the air holds, how strongly the wind spins up a storm, and an overall measure of their combined effect. Then they built a model that shows at what distance thunderstorm centers are still linked to each other. They called this distance the range.

Range works like the length of a rubber band. In summer it is short: storms are localized. In winter and spring it is long: severe weather is synchronized over a large area.

It turned out that the warm El Niño current in the Pacific Ocean stretches the band even more. In northern states, the range increases by 20–30%, and in the northeast, for the overall indicator, it can grow by almost a factor of two. Previously, it was thought that the connection between storms in different places was always the same. Now it is clear that it changes with the seasons and depends on ocean currents. This knowledge helps us prepare in advance for the possibility that one storm will cover an entire region rather than a single city.

Whole graph →

The range parameter for PROD in the northeastern US during a strong El Niño can increase by nearly a factor of two compared to the neutral phase, meaning that thunderstorm systems can cover an area comparable to several states.

In Kim Stanley Robinson's novel 'Forty Signs of Rain', climate models predict extreme weather events, and the accuracy of such predictions depends on accounting for global oscillations like El Niño.

\gamma(\mathbf{s}) = (\|\mathbf{A}\mathbf{s}\|/\rho)^\alpha
Defines spatial correlation: the larger the distance, the weaker the dependence; parameters ρ (range), α (smoothness), and matrix A (anisotropy) control the shape.
\theta(\mathbf{s}_1,\mathbf{s}_2) = 2\Phi\left(\sqrt{\gamma(\mathbf{s}_2-\mathbf{s}_1)/2}\right)
Relates the probability of simultaneous threshold exceedance to the semivariogram; a value of 2 means independence, 1 means complete dependence.
CLIC_b = \frac{1}{B}\sum_{b=1}^B \left[ 2\log L_C\left(\hat{\psi}(\hat{\lambda}); t_{\hat{\lambda}}(\mathbf{Y})\right) - 4\log L_C\left(\hat{\psi}_b^*(\hat{\lambda}_b^*); t_{\hat{\lambda}}(\mathbf{Y})\right) \right]
Bootstrap estimate of the unstandardized composite Kullback–Leibler divergence; the smaller the value, the better the model.
Concepts
ENSO spring predictability barrier Jin-Timmermann model Multivariate ENSO Index El Niño-Southern Oscillation El Niño index La Niña Southern Oscillation Westerly wind burst recharge oscillator model space weather NOAA CPC tropical cyclone natural variability TAO/TRITON moorings ERSST seasonal cycle seasonal forcing forecast interannual variability