Complete index

Everything on the site

Grouped by section. Nothing here sits behind a filter or a search box.

By section

Arithmetic5

FitzRoy4

Instruments7

The room7

Where it fails6

01Arithmetic

A forecast you calculate rather than guess.

Six Weeks of Arithmetic for Six Hours of WeatherLongRichardson calculated a single six-hour forecast by hand in 1922.
Bjerknes states the problemLongBefore anyone could calculate a forecast somebody had to say precisely what calculating one would mean: a set of equations, a set of initial measurements, and no meteorology in between.
Why it came out wrongLongThe arithmetic was sound; the initial data was not smoothed, so the calculation amplified noise that was never in the atmosphere.
Sixty-four thousand people with slide rulesMediumHis estimate of what it would take to keep pace with the weather, arranged in a hall like an orchestra, is the clearest picture of what a model does.
What a Forecast Actually IsShortA probability dressed as certainty, derived from a snapshot that no one quite gets right.

The man who coined the word, and was attacked for presuming.

FitzRoy, and the word forecastLongHe built the first storm warning service, coined the term, and was attacked for presuming that the weather could be predicted at all.
FitzRoy's Meteorological DepartmentLongA government office set up to collect weather observations from ships, which quietly became the first national forecasting service without anyone authorising it to be one.
The first storm warningsLongTelegraph made it possible to know about weather before it arrived, which turned observation into warning and warning into responsibility.
A cone hoisted at the harbourShortThe warning itself was a black canvas shape raised on a mast where crews could see it — a forecast delivered as an object rather than a text.

A thermometer in a louvred box, and everything after it.

The Stevenson ScreenLongLouvred wood, white paint, a specified height: the point is to measure air rather than sunshine, and the design has barely changed.
The radiosonde ascentLongA balloon rises for around ninety minutes until the latex gives way at roughly thirty kilometres, transmitting all the way up.
Anemometers, and Robinson's cupsMediumFour cups on a spindle turn at a rate related to wind speed — a relationship that turned out to be less simple than its inventor believed, and had to be calibrated rather than derived.
The Buoys Nobody VisitsMediumMost of the planet is ocean, so most observation is unattended and has to survive years without maintenance.
What satellites cannot seeMediumA satellite measures radiance, not temperature, and turning one into the other is a calculation with assumptions in it.
The Station PlotMediumEight or nine numbers, one small diagram: how every dot on a synoptic chart carries the whole atmosphere on its back.
The Aircraft That ReportShortThe busiest instrument network in the atmosphere is also the one nobody designed.

Richardson's hall is now a room of screens.

ENIAC, and the first machine forecastLongIn 1950 a machine reproduced in about a day the kind of calculation Richardson had done by hand in six weeks, and the result was good enough to prove the method rather than the weather.
How a model is actually runLongRunning a numerical weather model is not pressing a button and waiting.
Somebody still draws the isobarsMediumThe synoptic chart looks like it predates the computer. It does.
The People on ShiftMediumA forecast office runs continuously, and the handover between shifts is where judgement is actually transferred from one person to the next.
Getting the observations inMediumBefore the model can run, it needs to know where the atmosphere is right now — and that is harder than it sounds.
Fronts as a Drawing ConventionShortThe triangles and semicircles on a weather map are not decorations — they are a theory drawn in ink.
Isobars, and what they are forMediumPressure drawn as lines turns a page of measurements into something a forecaster can read at a glance — and the distance between the lines says how hard the wind will blow.

And why the failure has a shape.

Where it fails, and why the failure has a shapeLongForecast error is not random: it grows fastest where the atmosphere is least stable, which is why some situations are predictable for days and others for hours.
The butterfly, in practiceMediumSensitivity to initial conditions is a measured property with a timescale attached, not a metaphor.
The Grid, and What Falls BetweenLongA model divides the atmosphere into boxes, and anything smaller than a box — a shower, a hill — has to be represented rather than resolved.
EnsemblesMediumRunning the model many times from slightly different starting points converts a single answer into a spread, which is the honest output.
VerificationShortA forecast is only as good as the record of how it did, and the scoring is its own small discipline.
ParameterisationLongThe grid stops at a certain size. Everything smaller than the grid box is represented by a formula, and those formulas — not the equations of motion — are where models diverge most from one another.