Thunderstorms may eventually become more predictable

As I’ve mentioned in previous articles, there are over 40,000 thunderstorms forming every day across the globe. That’s nearly 1,700 per hour. For an entire year, it’s estimated that our planet will receive approximately 14.6 million thunderstorms.

With thunderstorms comes lightning, which is still one of the most mysterious meteorological phenomena. They are known to form mostly on warm and humid days, especially when the air is unstable. However, predicting where exactly a thunderstorm will develop and when it will occur has often been extremely difficult.

According to the World Meteorological Organization, thunderstorms have likely caused around 30,000 deaths globally between 2010 and 2019. Without warning, the storms will often produce heavy rainfall very suddenly and locally, that will often lead to flash flooding. It’s estimated that thunderstorms have cost billions annually in economic losses.

New research that was published in Nature states that scientists believe they’ve discovered a system to help predict where a thunderstorm is expected to hit, especially on hot days.

Developing thunderstorms will contain powerful, localized vertical updrafts of air. These rising currents can be detected by ground-based weather radars and are valuable for real-time weather forecasting and severe thunderstorm warning systems. By combining satellite observations of soil moisture with measurements of vertical wind shear, which are the changes in wind speed and direction with height, scientists say they can identify the most intense thunderstorms within roughly 5 to 25 miles and 15–60 minutes before they form. This important discovery is based on 21 years of observational data.

The research revealed that 68% more explosive storms occurred in areas with favorable soil moisture patterns, which describe how water is distributed across the soil over space and time. A new technological approach allowed scientists to collect high-resolution satellite data, providing detailed daily images of soil wetness that were not previously available.

To conduct the research, scientists analyzed satellite observations of atmospheric conditions of about 2.2 million storms from 2004 to 2024. The researchers believe that the underlying principle they discovered will also apply to thunderstorm development in other parts of the world. The study focused mainly on sub-Saharan Africa, a region that frequently experiences intense thunderstorms. Improving forecasting in this part of the world is especially important because flash flooding can have severe impacts on large and rapidly growing urban populations.

In addition, the timing and nature of thunderstorm formation may influence whether individual storms grow into large clusters known as mesoscale convective systems. These systems can last for many hours and extend across hundreds of miles. In the future, the work could be expanded to help predict the entire life cycle of severe storms, from their initial formation to their eventual dissipation.

Thunderstorms are often thought to appear suddenly and unpredictably. However, this research shows that the locations where storms begin seemed to be more predictable than previously believed. The new study found that clouds are most likely to grow rapidly in areas where soil moisture patterns align with favorable wind shear conditions. This relationship provides meteorologists with clues about where thunderstorms are likely to form later in the day, especially during the afternoon and evening hours when conditions are often the most favorable. With the help of artificial intelligence, researchers are looking to incorporate this new knowledge into improved computer models that can better predict where storms will develop in this region and eventually in other parts of the world. This information would be very valuable for sectors including transportation, energy, agriculture, and storm-water management.

Here in North Idaho, thunderstorm season is approaching and the average number of days with thunderstorms, which include thunder, lightning, and rain, across the lower elevations is 14 (one in April, two in May, five in June, two in July, two in August, one in September and October). In Spokane, the average number of days with thunderstorms is about 10-15, with the majority occurring during the warmer months. The normal number of severe weather days in the Inland Northwest for an entire year is slightly less than one. With the lingering effects of the weak, cooler than normal sea-surface temperature event, La Nina, the number of thunderstorms in the region are expected to be close to normal levels with the highest numbers in April and into the middle of May. 

In terms of our near-term local weather, more moisture is expected across the Inland Northwest through at least the end of the month. The long-range computer models continue to indicate that some of these Pacific storm systems may be cold enough to produce measurable snowfall in the lower elevations, but much of it is likely to fall as rain. As of the weekend, about 20 inches of snow has been measured at Cliff Harris’s station. The average snowfall for March in Coeur d’Alene is about 6 inches. Based on the current weather patterns, we believe that snowfall totals in March may be close to normal.

• • • 

Contact Randy Mann at [email protected].