Wed 30 Mar 2011
Forecast Offices Then and Now
Changes over 50 years

By Philip Eden

Weather forecasting has changed completely in the last 50 years or so, not just in the accuracy of the predictions, but also in the working environment of the meteorologists.

A typical forecasting office in 1950s was peopled by two 'castes' - forecasters and assistants. The job of an assistant was to carry out the hourly weather observations, and to plot all the data from Europe and the north Atlantic on charts. Most would become extremely proficient at chart plotting, using a double-nibbed pen with black and red inks, completing each station report in under five seconds. Thus a chart with over 500 stations plotted could be completed in about an hour. It was then passed to the forecaster whose first job was to analyse the chart, drawing isobars and fronts, and highlighting zones of significant weather.

The forecaster's main tools were a soft lead pencil, an eraser, and a selection of crayons. He would examine the latest charts and, using rules and guidelines developed by his predecessors, leavened by his own experience, he would construct a sequence of forecast charts covering the next 48 hours. The office would be noisy with a continual clatter of teleprinters, and there would be paper everywhere.

An assistant in 1971 could well have ended up a senior forecaster by 2011. His office is very different. Assistants have vanished, and there are fewer forecasters too. They work at ordinary desks with two computer terminals, one for displaying data and one for creating forecasts: this really is a paperless office. It is also a quiet one with a barely audible hum of hard disks and air-conditioning units.

Nearly all the raw forecast material is now produced at the nation's central forecasting office, and comprises computer output overseen and occasionally adjusted by a human forecaster. At the outstation the forecaster's job is simply to interpret that output, incorporate local knowledge, and present his forecasts according to the clients' requirements. More and more of this is being automated, and the need for the human forecaster on site is diminishing. Nowadays, airline pilots, mariners, and offshore oil-industry personnel, are all encouraged to 'self brief'.

There is one further big change from the 1960s. Fifty years ago practically every weather forecaster in the UK was a civil servant; now 10-20 per cent of them work in the private sector.