A grass fire in eastern Idaho became an unexpected laboratory when it generated its own thunderstorm. In an October 2 account, NASA described how researchers chasing fire-driven clouds this summer caught a plume from the Wildhorse fire and gathered measurements that could help explain one of wildfire’s most dangerous behaviors.

The campaign, INSPYRE, focuses on pyrocumulonimbus clouds, often shortened to pyroCbs. These towering storms can loft smoke into the stratosphere, connecting events on the ground to changes high above it. The National Center for Atmospheric Research’s project record lists the 2026 deployment as July 27 through September 10 and frames three central questions: Which fires generate these storms, what controls their smoke injection, and how does that smoke alter atmospheric composition and radiation?

Lightning is another part of the puzzle. The Naval Research Laboratory, a campaign partner, described airborne measurements designed to investigate why some fire-generated storms produce substantial lightning while others produce little. Its iSTORM instrument is designed to measure gamma-ray flashes associated with lightning. That is a research capability, not evidence that the team has already solved the forecasting problem.

On August 26, researchers aboard the NCAR Gulfstream learned that Wildhorse was unexpectedly intense. They diverted and spent about three hours sampling the cloud plume and its drifting smoke. NASA says the encounter provided observations roughly an hour after the storm first erupted.

The two aircraft have complementary jobs. NCAR’s Gulfstream emphasizes direct sampling and radiation measurements, while NASA’s ER-2 observes from high altitude. Ground instruments, satellites and models add other perspectives. Combining those views should help separate the role of fire intensity from the atmospheric conditions that allow a storm to develop.

The practical stakes are substantial: fire-generated downdrafts and wind shifts can endanger crews, while high-altitude smoke can travel far beyond the original burn. Researchers hope better understanding will eventually support more useful warnings and simulations.

For now, analysis is the next step. The University of Iowa’s campaign data page cautions that its lidar products remain preliminary pending final calibration. Collecting the observations was the summer’s achievement; establishing what they mean will take more work.