Green Fireballs: Visualisation · Signals from the Periphery

← Intro 中文

Event

FIREBALL

Solved altitude

—

Path slope

—

Sound at ground

SILENT

GREEN FIREBALLS · THE TRIANGULATION BENCH
New Mexico · 1948-1951 · the LaPaz network
two falls, a bearings machine, a colour strip
level at 8-10 miles · constant speed · silent
the meteor baseline: steep, slowing, loud
▶ replays the event · ⊕ fires a new one

Station

…

…

Evidence

fact ⇄ what kind of claim · live

Two sight lines decide a path; more stations shrink the error

triangulation is geometry, and needs no theory of the object

MEASURED

The class: level at 8-10 miles, constant speed, ~2 s, instant on

LaPaz to the conference · DOE-UAP-D004

REPORTED

Only three real paths were ever solved: 12 Dec, 20 Dec, 30 Jan

the whole wave, reduced to three lines of geometry

REPORTED

No sound was ever confirmed, though it was canvassed hard

Teller's bound: a silent body must be near a cubic centimetre

REPORTED

Cu 510-522 nm · Mg 517-518 nm · [O I] 557.7 · Na 589

laboratory wavelengths; copper burns green, and so does magnesium

MEASURED

LaPaz's 5,200 Å is a naked-eye estimate; no spectrogram exists

the eye cannot split copper green from magnesium green

REPORTED

209 sightings logged around Sandia, 1948-50, in 116 pages

with copper-powder residue investigations · DOW-UAP-D017

REPORTED

Project Twinkle closed with no solution and no spectrogram

the instruments arrived after the peak · unresolved since

REPORTED

View

The event

…

▶ The sky view replays a scripted event; the sound front runs 12× faster than real sound so you don't wait 45 seconds. The delays printed are the true ones.

The network · triangulation view

Stations reporting2

One station is a direction, not a place. Two decide a path. LaPaz's answer to the Godman trap was a network.

Bearing error± 0.50°

Patrol inspectors with compasses and stopwatches, not theodolites. Crank it and watch the solve wander.

Show true paththe line nobody had in 1948

The candidates · colour view

Copper · 510-522 nmrare in meteorites: the anomalists' green

Magnesium · 517-518 nmabundant in meteorites: the sceptics' green

Oxygen · 557.7 nmthe air's own auroral green

Sodium · 589 nmthe yellow control

Iron · 490-545 nmthe whitish background picket

LaPaz's 5,200 Åa naked-eye estimate, drawn dashed

Eye band± 12 nm · what an estimate can resolve

Display

Labels

Glow

Readout · live from the solve

Event in force

…

Stations reporting

2

Bearing error

± 0.50°

Solved altitude

—

Solved slope

—

Implied speed

—

Fit vs truth

—

Selected line λ

—

…

What is measured vs modelled here

Measured: the triangulation geometry (a least-squares intersection of sight lines), the sound-delay arithmetic, and the laboratory emission wavelengths. Reported: every clause of the fireball's behaviour (level at 8-10 miles, constant speed, ~2 s, instant onset, silence, green near 5,200 Å), the three solved paths, the 209-sighting Sandia log, the copper-powder investigations, Twinkle's non-result. Modelled: the station layout, the event paths this bench fires, the bearing-error sizes, the meteor baseline's specific numbers, and the ± 12 nm eye band. Reading: whether the solved paths describe unusual meteors or something else. LaPaz and Bradbury split on it in the same minutes, and this bench keeps both cards.

The triangulation model

Each station reports azimuth and elevation for the event's first and last point; each report defines a sight line from the station; the solved endpoint is the point minimising the summed squared distance to all the lines (the 3×3 normal equations, solved exactly). Bearings carry Gaussian error on the dial you set. Altitude is read at the midpoint, slope from the endpoint drop over the ground run, speed from the solved length over the stopwatch duration. With one station the solve is refused: every range along the ray fits the same bearings, which is the whole reason networks exist.

Where this drawing takes liberties

The map is stylised New Mexico, not a survey; the AESS posts sit at round coordinates; real reports came from moving patrol cars and airline cockpits, not five fixed pylons. The sky view compresses the sound front twelve-fold and scripts both events. The colour strip renders line positions truthfully but line brightness schematically, and the ± 12 nm eye band is an honest guess at an honest witness, tagged MODELLED. The wavelengths, the geometry, and the quotations take no liberties.

…1 / 11

…

…

drag = orbit · scroll = zoom · ▶ replays the event