Weak lightning during thunderstorm development can trigger deadly wildfires, study finds

Lightning-ignited wildfires are among the most unpredictable natural disasters affecting forest ecosystems. They are rare, fast-moving, and often occur in remote, high-altitude regions where detection is difficult. Despite decades of research, scientists still lack a clear understanding of what types of lightning are most likely to ignite fires.

Researchers from the University of Science and Technology of China and the Institute of Atmospheric Physics at the Chinese Academy of Sciences investigated a catastrophic fire in MuLi County, Sichuan Province. The fire was confirmed to have been triggered by lightning during a developing thunderstorm, a period previously thought to pose low ignition risk.

Fire investigators found a lightning strike that penetrated a tree from top to bottom, confirming it as the ignition source. The incident, which occurred at 09:07–09:10 UTC, started in a dense coniferous forest at an elevation of 3 549 m (11 644 feet).

The blaze spread rapidly due to erratic winds and steep terrain, resulting in 31 fatalities, including firefighters and residents. The event shocked China and prompted a detailed scientific inquiry into how such weak lightning could start such an intense blaze.

Reconstructing the fire from satellites and lightning networks

To understand what made this event possible, scientists combined multiple global datasets. They used the Himawari-8 satellite to monitor cloud-top brightness temperature, known as TBB, to track the storm’s evolution. Simultaneously, lightning activity was analyzed using two major global detection networks: the World-Wide Lightning Location Network (WWLLN) and the Global Lightning Dataset 360 (GLD360).

WWLLN has a detection rate of about 18% and measures lightning energy in joules, while GLD360 achieves around 70% efficiency and records peak current in kiloamperes, along with polarity information. By combining both datasets, the researchers could cross-check each lightning event, minimizing uncertainty.

They also used ERA5 reanalysis data from the European Centre for Medium-Range Weather Forecasts to assess surface meteorological factors, including wind speed, humidity, temperature, precipitation, and air pressure. This comprehensive reconstruction made it possible to pinpoint the exact atmospheric conditions that coincided with ignition.

MuLi County, located at the boundary of the Tibetan and Yunnan–Guizhou Plateaus, has an average elevation above 3 800 meters (12 467 feet). Evergreen coniferous forests cover about 65 percent of the area, while grasslands make up roughly 29 percent. These dense forests, rich in resin, are highly flammable and difficult to access, making fire suppression operations particularly challenging.

Together, the datasets revealed that ignition occurred not at the height of lightning activity but during the developing stage of the thunderstorm when lightning was weaker and less frequent.

When weak lightning meets the perfect storm

During the fire’s ignition period, only three to six lightning discharges occurred between 09:07 and 09:10 UTC. Compared with the later mature stage of the storm, this was a quiet period for lightning activity. However, surface conditions during this window were exceptionally favorable for fire ignition.

Near-surface humidity fell to around 30%, while precipitation remained near zero. Wind speeds were moderate but fluctuating, creating erratic air movement that helped flames spread once ignition began. Satellite data showed that the fire started near the outer edge of a developing thundercloud, where descending dry air created a localized low-humidity zone at ground level.

As the storm developed, cloud-top brightness remained relatively high, indicating incomplete convection. By 11:00 UTC, roughly two hours after ignition, convection intensified and lightning frequency increased sharply. However, by then the fire was already burning out of control.

This mismatch between lightning frequency and ignition timing highlights a key finding of the study: the most dangerous lightning is not always the strongest or most frequent but the one that strikes in dry, windy conditions before rain begins to fall. Surface pressure analysis from ERA5 data revealed minimal change before and after ignition, suggesting that the fire’s spread was driven primarily by local moisture and wind rather than atmospheric pressure dynamics.

Negative lightning: the unexpected culprit

The polarity of lightning plays a critical role in whether it can ignite fires. In most North American cases, positive cloud-to-ground lightning is blamed for wildfires because it delivers higher peak current and longer continuing current. But in MuLi County, researchers found that almost all lightning on the day of the fire was of negative polarity.

Negative cloud-to-ground lightning can produce multiple return strokes, each delivering short bursts of current that sustain heating of the ground and nearby vegetation. Despite being weaker, these repeated discharges can ignite surface fuels under dry conditions.

Data from both WWLLN and GLD360 confirmed this dominance of negative lightning. Only one weak positive stroke was recorded near the ignition point, and it occurred outside the confirmed time window. The rest were clustered around the fire’s ignition zone, strengthening the conclusion that negative lightning was responsible.

Researchers also noted that the local forest composition, rich in resinous conifers, may enhance the flammability threshold, making ignition easier even with smaller energy input from lightning. This explains why regions like southwestern China may exhibit different ignition behavior compared with other parts of the world.

A second case confirms the pattern

To verify whether the MuLi event was unique, the team analyzed another lightning-ignited fire that occurred on June 8, 2019, in Tangyang Township, about 70 km (43 miles) north of the original site. Fire department reports confirmed that a lightning strike between 08:10–08:15 UTC on June 7 caused ignition.

In this case, the parent lightning was also predominantly negative. Both WWLLN and GLD360 data showed a cluster of negative discharges close to the fire site, while positive lightning occurred farther away. Environmental conditions mirrored those of the March event: low humidity, near-zero precipitation, and moderate winds.

Together, the two case studies suggest a consistent mechanism linking negative lightning to wildfire ignition in this part of China. The pattern appears to depend on regional topography, local vegetation, and atmospheric dryness rather than solely on lightning energy.

Researchers caution that the results are based on limited cases but emphasize that the evidence points toward a significant, previously overlooked fire hazard associated with weak negative lightning in dry, developing storms.

Why this discovery changes wildfire science

For decades, fire monitoring systems have treated the mature phase of thunderstorms as the highest-risk period for lightning-ignited fires. Most models rely on lightning density or energy rather than considering storm evolution or surface meteorological context.

This study overturns that assumption. It shows that weak lightning in developing storms can be more dangerous than intense lightning later on, provided the environment is dry, windy, and rain-free. This is especially true in high-elevation, mountainous regions where lightning strikes are frequent but rainfall is sporadic.

According to Professor Yong Xue, the study’s corresponding author, lightning in developing storms “may pose a higher risk of igniting combustible materials because of distinct atmospheric conditions such as low precipitation, low humidity, and strong winds.”

The findings call for a reevaluation of lightning-fire early warning systems. Integrating polarity data, humidity, and storm stage could help create more accurate models for predicting lightning ignition potential. This could ultimately save lives by allowing faster evacuation and more targeted firefighting responses.

The path forward for wildfire prediction

The implications of this study extend well beyond southwestern China. Similar dry, mountainous ecosystems exist in the western United States, the Mediterranean Basin, and parts of Australia, where lightning is already a major cause of wildfires.

Future fire-risk models could combine lightning polarity information with high-resolution humidity and temperature data to identify ignition-prone storms before the first strike occurs. This would mark a shift from simple lightning counting toward true ignition probability modeling.

The study also emphasizes the need for global lightning monitoring systems that can distinguish between positive and negative discharges. Improved detection and data-sharing could help identify emerging hotspots in regions where weak lightning events have been historically overlooked.

Finally, by focusing on the storm’s developmental phase, meteorologists could gain precious minutes to hours of lead time to issue alerts or deploy resources, especially in remote, high-elevation regions where firefighting access is limited.

References:

¹ Weak lightning in developing thunderstorms triggers deadly wildfire – Institute of Atmospheric Physics, Chinese Academy of Sciences – October 14, 2025

² On the possible meteorological factors contributing to lightning-ignited wildfires in West Sichuan: A case study of MuLi wildfire in March 2019 – Atmospheric and Oceanic Science Letters – Zhengyang Qu et al. – https://doi.org/10.1016/j.aosl.2025.100714 – September 6, 2025

Reet Kaur

I’m a science journalist and researcher at The Watchers, contributing to the Epicenter edition, where I cover peer-reviewed scientific research and emerging discoveries across Earth and space sciences. With a background in astronomy and a passion for environmental science, I’ve worked in shark and coral conservation in Fiji, conducting reef and shark-behavior research, contributing to mangrove restoration, and earning PADI Open Water and Coral Reef Certifications. I bring a blend of scientific rigor and storytelling to illuminate the discoveries shaping our planet and beyond.

This author does not have any more posts.