Robert FitzRoy created the weather forecast and established a field of study no less relevant today than that popularized by his more famous associate, Charles Darwin.

After working for years to protect the lives of others, Robert FitzRoy’s ended his own life 150 years ago. He was troubled by recurrent depression and by the critical public reception of his new innovation: the weather forecast.
Today, this vice admiral, former governor of New Zealand, and founder of a still-vital institution, is perhaps most often remembered in the context of his first command: the HMS Beagle. The Beagle surveyed the coasts of Patagonia, Tierra del Fuego and the Straits of Magellan during the 1830s. The  expedition later became famous for carrying Charles Darwin to the Galapagos Islands, where finches provided the inspiration for his theory of evolution.
Consequently, in the annals of scientific history, FitzRoy is frequently relegated to a sidekick role, mentioned briefly as Darwin’s companion and captain or maybe in reference to his rejection of his friend’s evolutionary theory upon its publication. Yet FitzRoy was a man of science in his own right and established a field of study no less relevant today than Darwin’s own.
Robert FitzRoy began his career as a sailor. Even in his later years as a purely dry-land official, the dangers of the water remained a strong motivating force in his efforts. FitzRoy viewed many lives lost at sea as tragedies that could have been prevented by improved prediction of oncoming weather.
In 1854, FitzRoy worked with the British government to establish the Meteorological Department of the Board of Trade, today known as the Met Office,
of which he became the first director. With a staff of three, the department’s stated purpose was to “reduce sailing times with better wind charts,” but FitzRoy saw additional opportunities for weather predictions.  Making use of information conveyed to London via telegraphs and a barometer of his own design, FitzRoy began using synoptic charts and canvas signals to issue storm warnings. He also issued general weather reports for the public.
Today, the Met Office employs a staff of more than 1,900 people in 60 offices across the globe, and includes the Public Weather Service, the National Severe Weather Service, and research on climate change modeling.
Dr. Dave Jones from the Met Office’s Department of Applied Weather Science spoke with Earthzine about ways in which the Met Office has changed since FitzRoy’s time and ways in which its guiding motives have remained the same.
In FitzRoy’s time, protection of human life and protection of property were the driving forces behind a desire to improve weather projections. FitzRoy also released public weather reports in the newspaper as a side product of his research, and the public used these reports for everything from planning picnics to betting on horse races.
As Jones explains, this multiple use approach to weather information has only become more prominent over time.
“The protection of life and property is still at the core of what National Met Service is today … [but] I think what’s come out of the recent skill of our forecast predictions is the ability to do additional things over and above protection of life and limb.”
Those additional services are quite varied. Customers of weather data range from renewable energy producers to national defense and aviation services.
“An example would be a wind production forecast for commercial customers … giving them an idea what the wind speed will be so they can assess the amount of power for the wind farm,” Jones said. “Another example would be in the road area: providing route forecasts to enable local authorities to treat the roads for snow and ice. In the defense area, it can be helping people make better use of things like radar systems and electro-optical systems for operations.”
Translating weather data into useable products is such an important service that the Met Office has devoted an entire department to the task. Demand for these products relies heavily upon trust for the Met Office’s ability to turn observations and measurements into projections. Public acceptance of this concept has come a long way since the 1850s.
In FitzRoy’s time, the idea of predicting weather by via scientific rationale was a novelty. Previous methods of weather prediction had been largely tied to myth, superstition, or local anecdote. The very word “forecast” was coined by FitzRoy in an attempt to indicate that his methods were driven by logic rather than the mystic skills associated with “predictions.”  However, lacking a vocabulary  and clear knowledge of the degree of uncertainty in weather reports, FitzRoy struggled to establish public confidence in his methods. Sometimes lauded for success, he also suffered ridicule for failed forecasts, and criticism weighed heavily on him.
Although weather forecasts are still far from perfect today, Jones feels that public understanding of the scientific skill involved in forecasting has considerably improved since FitzRoy’s time.
“In a fully deterministic sense, we’ll never be able to say from today exactly what the weather is going to do in two days time,” Jones said. “But we can have a very good attempt at forecasting it, and we can be fairly clear about how confident we are about it. FitzRoy basically had to prove the point that was that there was any level of skill at all.”
FitzRoy’s attempt to demonstrate skill in forecasting faced considerable challenges. At the time, limited knowledge existed about how the atmosphere functioned, and many believed that it was completely chaotic, following no clear patterns or rules. FitzRoy’s tools were limited as well: he largely had to rely on thermometers, barometers (one of which he designed himself), and wind reports from the coast.
FitzRoy could not foresee the future success that his newly established office would have, but his work in creating a methodical approach to weather forecasting laid important foundations for later weather services in the U.K. and across the world.
“To actually be bold enough to produce a forecast and actually publish it, I think, that was considerable foresight on his part,” Jones says.
Communicating confidence and uncertainty still remain challenges for the Met Office, but as scientific understanding has improved, so has public confidence in forecasting. Today’s technologies offer far greater details of information than those of the 19th century and are able draw observations from a greater portion of the globe: from land surface measurements to measurements taken from under the surface of the ocean, and data gathered from high in the atmosphere. The tools used to collect this data include weather balloons, aircrafts, and satellite imagery.
“We’re observation hungry,” Jones says, laughing.
Even processing such a large volume of data is only possible because of recent technological developments. The data are fed into computer modeling systems that then generate projections, either on a fine scale (~1.5 km resolution) for weather projections, or on a coarser scale (~60 km resolution) for decadal climate projections.
By running multiple models with slightly different inputs, comparisons can be made among weather projections, and a range of uncertainty can be established. Being able to quantify uncertainty is important because the more effectively uncertainty can be communicated to users of weather information, the more valuable the weather information provided to them is likely to be.  Public use is, after all, the overarching objective behind the collection of all this information in the first place.
“There’s a great degree of scientific interest, and there’s a huge beauty associated with the weather,” says Jones, “but we have to make it useful for people, we have to make it accessible, we have to help people make useful decisions with it.”
This desire for public service provides a clear link between today’s bustling, international Met Office and the tiny forecast office established by Fitzgerald more than a century and a half ago. As the next century unfolds, the Met Office has a range of challenges to address, including communicating uncertainty in long-term climate models and seeking accurate ways to incorporate the interplay between Earth systems into a computer model. Who knows what the technologies and discussions of the 21st century may bring?