Weekly atmospheric feature
One weather event each week, investigated in public
The featured location can be anywhere in the world, and the event can range from haze, contrails, and unusual cloud development to wildfire smoke, severe weather, or a documented atmospheric intervention. Every story uses the same standard: preserve the observations, compare competing explanations, show the data sources, and state what the evidence cannot yet establish.
This week's feature story
North America · Washington, DC · opened July 16, 2026Is aviation-related cloudiness being undercounted in the DC haze story?
We are testing whether persistent contrails, spreading contrail cirrus, and humid atmospheric conditions explain more of the regional veil than wildfire-smoke reporting alone acknowledges.
A global beat with one consistent evidence standard
Current field observation
Privacy-reviewed field observation · Washington, DC area · June 8, 2026Stored linear-trail observation
Review the preserved visual record, stored location precision, and evidence limits. The record was not GPS-verified and does not establish attribution to any aircraft.
Research archive
Long-form evidence audits remain available beneath the rotating weekly feature.
A contrail that would not persist
A controlled research flight produced short-lived trails at several altitudes. The absence of persistence is evidence, not a failed observation.
NASA SUCCESS · persistent-trail analysisHow a narrow line became a spreading ice cloud
Satellite, in-situ, and modeling evidence connect high ice supersaturation and wind shear to growth over 15–180 minutes.
The audit template
Define the claim
“A trail appeared” is different from “this aircraft made it,” “it persisted,” or “it changed cloud cover.” Each requires different evidence.
Build a synchronized timeline
Convert every source to UTC, document time resolution, and distinguish observation time from data retrieval time.
Test necessary conditions
Formation needs the right exhaust-mixing threshold. Persistence needs ambient air at or above ice saturation. Neither condition alone identifies a specific flight.
Seek an alternative
Compare natural cirrus, multiple overlapping trails, incomplete flight data, and model error against the preferred explanation.
Grade the conclusion
Use “observed,” “supported,” “consistent,” “uncertain,” or “not supported” according to the available record.
Why older cases can still be high-value: the underlying thermodynamics have not changed, and the unusually rich SUCCESS instrument record lets readers examine a complete evidence chain. Modern operational products improve coverage, but they do not eliminate upper-tropospheric humidity uncertainty.