Space & signals
Scientists flew into storms to test whether they could change a hurricane
Stormfury tried to weaken hurricanes through cloud seeding. Later observations showed how easily a natural change could be mistaken for an experimental success.
A storm as an experiment
Project Stormfury explored whether cloud seeding could alter a hurricane's structure and reduce its strongest winds. A 1975 NOAA technical memorandum by Robert Sheets and Samuel Pearce examines two cloudline seeding events. It belongs to an experimental programme, not a demonstration that storms can be directed at chosen destinations.
The proposed mechanism was specific. Stimulating convection outside the existing eyewall might encourage a larger replacement eyewall, spreading the circulation and weakening peak winds. NOAA's retrospective account describes an ambitious research effort whose initially promising observations later became much harder to interpret.
The missing comparison
Suppose a storm weakens after an intervention. Two events have happened in sequence, but the sequence alone does not establish causation. The critical question is what the storm would have done without the intervention. Hurricanes change continually, so a before-and-after comparison is unusually vulnerable to a misleading answer.
NOAA's review identifies natural eyewall replacement as a major problem for the original interpretation. Unmodified hurricanes could exhibit changes resembling the desired response. Observations also found too much naturally occurring ice and too little supercooled water for the proposed seeding mechanism to work as expected. These were challenges to both the statistical comparison and the physical premise.
An experiment can fail usefully
The distinction between changing weather and understanding weather matters here. A programme may fail to establish a reliable intervention while contributing observations, instruments and questions that improve scientific knowledge. Its existence proves that researchers investigated a possibility; it does not prove that the possibility became a controllable technology.
The 1975 case study and the later NOAA account should therefore be read together. A contemporary report records what researchers could infer from particular events. A retrospective assessment can incorporate observations from other storms and revise that inference. Keeping both prevents the archive from freezing the story at the most optimistic stage.
A bounded claim is more interesting
Nothing in these sources establishes that modern hurricanes are manufactured or steered. Those are different claims requiring direct evidence of a mechanism, intervention and effect. Nor should one historical programme be treated as a final judgment on every conceivable form of weather modification.
Stormfury's distinctive lesson is methodological. A good explanation must connect an intervention to an observable mechanism and distinguish its effect from the system's ordinary behaviour. The more dynamic the system, the harder that task becomes. Here the storm itself supplied a convincing imitation of success. The eventual recognition of that problem is not an incidental footnote; it is the part of the story that tells us what the experiment actually taught.
Sources and further reading
- A Case Study of Two Stormfury Cloudline Seeding Events ↗
NOAA Technical Memorandum ERL WMPO 21, 1975; repository record and linked case-study file
- NOAA Hurricane Research Division, Project Stormfury ↗
Programme description and two reasons seeding interpretations were reassessed
- Willoughby et al., Project Stormfury: A Scientific Chronicle (1985) ↗
Developments since 1977: cloud physics and vortex dynamics