FitzRoy's Meteorological Department

The Office Nobody Meant to Create

In 1854, the Board of Trade in London set up a small statistics-gathering unit and gave it a name that told you almost nothing: the Meteorological Department. Its job was bookkeeping — collecting wind and weather logs from ships, reducing them to tables, and making oceanic trade routes a little more predictable. The man appointed to run it, Vice-Admiral Robert FitzRoy, had commanded HMS Beagle on its second surveying voyage and had a reputation as a careful, exacting observer. He was not hired to invent anything.

He invented everything anyway.

The Department occupied a modest room in Parliament Street and employed a handful of clerks. FitzRoy began exactly as instructed: designing standardised log forms, corresponding with ship captains, persuading the Admiralty to enforce the new recordkeeping. The underlying theory — that enough systematic data, gathered everywhere at once, would eventually reveal how the atmosphere behaved — had been stated clearly by the American oceanographer Matthew Fontaine Maury, whose Physical Geography of the Sea FitzRoy admired. But Maury's interest was in the sea lanes; FitzRoy's attention kept drifting upward, to the weather itself.

The telegraph changed the question. By the early 1850s, lines ran between most British towns, and a message that once took days on horseback now arrived in minutes. FitzRoy saw immediately what this made possible: a weather observation recorded simultaneously in Aberdeen, Valentia and Plymouth could be sent to London, plotted on a map, and read as a picture of the atmosphere at that moment. The operative word, coined in France at almost exactly the same moment, was synoptic — literally, seen together. Before the telegraph, "seen together" was physically impossible. After it, the technology was there. Someone only had to build the system.

From Records to Warnings

FitzRoy built the system. By the late 1850s he had a network of land-based stations — fifteen at first, then more — each equipped with standardised instruments and instructed to wire their readings to London each morning. His design for a portable barometer, robust enough for use aboard fishing vessels, went into production; the Board of Trade distributed them to harbours around the coast. Every storm warning that was eventually hoisted — the black canvas cone at the harbour mouth, raised or lowered depending on the direction of approach — depended on readings from those instruments reaching London in time to matter.

None of this had been authorised. The Meteorological Department's founding charter said nothing about public warnings. FitzRoy began issuing them in February 1861 on his own initiative, using a word he appears to have invented for the purpose: forecast. The forecasts went to the Board of Trade, which passed them to harbour authorities and, from August 1861, to the press. The Times began printing them; other papers followed. For the first time in history, ordinary people woke to a printed prediction of tomorrow's weather.

The scientific establishment was not uniformly pleased. The Royal Meteorological Society ↗ valued FitzRoy's observational work and found his forecasting irregular. Francis Galton, conducting one of the earliest systematic reviews of forecast accuracy, found the results mixed. The fundamental criticism was methodological: FitzRoy's prognostications rested on pattern-recognition and experience — on the behaviour of the barometer needle, the look of the sky, the track of a low-pressure centre — rather than on any deductive chain of reasoning. He was, the critics implied, guessing.

The criticism was not wrong, exactly. What FitzRoy lacked, and what nobody in 1861 yet possessed, was the physics. Vilhelm Bjerknes would not state the forecasting problem in mathematical terms until 1904. Lewis Fry Richardson would not attempt to solve it by arithmetic until the 1910s. FitzRoy was working a generation before the theoretical framework existed, using observation and hard-won instinct to do what would later require supercomputers. That the method was imperfect is less remarkable than that it worked at all.

What the Office Became

FitzRoy died in 1865, and within months the Board of Trade suspended the storm warnings and forecasts, bowing to scientific criticism and to the cost. The observational network continued. The international exchange of surface data — which the World Meteorological Organization ↗ would eventually formalise into a global system — had already taken root in the infrastructure FitzRoy built. British ships carried his log forms. His barometers hung in harbour-master's offices from Lerwick to Penzance.

Forecasting resumed in 1879, this time with the sanction of the Royal Society and on terms that acknowledged the scientific objections. The Meteorological Department was reorganised, eventually becoming the Meteorological Office — the Met Office — which moved over the following century from London to Bracknell and then to its current home in Exeter. The institutional line from FitzRoy's back room in Parliament Street to the supercomputing centre in Devon is real and unbroken, even if the science would have been unrecognisable to him.

What the history of the Department reveals, more than anything, is how bureaucratic intent and individual energy can diverge so completely that the institution that emerges is nothing like the one commissioned. The Board of Trade wanted tables of shipping data. It got, through the impatience and curiosity of one man, the prototype of every national weather service that followed. The network of coastal observing stations FitzRoy assembled was the kind of infrastructure that, once built, proved impossible to dismantle — because the fishermen and harbourmasters who had come to rely on the warnings made their discomfort audible when forecasts stopped.

The forecasts were imperfect. FitzRoy knew it, wrote about it, and understood that improving them would require better theory than anyone then possessed. He was right on all counts. What he could not have foreseen was how long it would take — how many decades of physics, how many iterations of the observational network, how much arithmetic — before the science caught up with the intuition he had acted on in 1861. By then, his office was a ministry, his system was global, and the word he had coined was in every dictionary.