BBC_Knowledge_Asia_Edition_-_May_2016_

PETER MOORE IS A HISTORIAN, LECTURER

AND AUTHOR OF THE WEATHER EXPERIMENT.

forecast. In his Weather Prediction

By Numerical Process, Richardson

said that 64,000 human computers

would be needed. A more recent

calculation has put the number

closer to a quarter of a million.

Richardson nursed his ideas while

serving as an ambulance driver

behind the trenches during the

Great War. With data from

Bjerknes, but no 64,000 clerks to

assist him, he set off on the heroic

endeavour of calculating the

required results himself. He

snatched every moment he could,

essentially establishing his own

computer ised forecast model as the

shots f lew around him.

Richardson’s ideas cleaved a new

path for forecasting in the 20th

Century. With the arrival of the

computer age in the 1950s, the sort

of numerical weather prediction that

he had imag ined became viable.

Models that utilised Newton’s

second law of motion, as wel l as the

laws of thermodynamics and applied

physics, were developed, processing

data from an increasing number of

land and airborne stations. Likely

scenarios were projected onto more

detailed grids and maps –

descendants of those conceived by

FitzRoy and Loomis long before.

Nowadays, all major forecasters

rely on numerical weather

prediction (NWP) models, which

are processed by supercomputers.

The Met Office got its first

computer in the 1950s – it

apparently bor rowed one from J

Lyons, the tea and cake company –

and over time these have become

increasingly powerful, able to

handle more complex models and

higher numbers of calculations. The

latest Met Office supercomputer was

announced in 2014 at a cost of

$137m. It can perform more than

16,000 trillion calculations per

second and weighs the same as 11

double-decker buses.

But for all this, weather

forecasting can never be an exact

science. One of the great discoveries

of the 20th Century was Edward

Lorenz’s chaos theory: his realisation

in 1961 that the smallest variation in

input can have an enormous effect

on output. The triumph of chaos

theory gave forecasters yet another

challenge to face. Weather

forecasting is, just as those at the

Royal Society in 1663 suspected, the

most difficult of scientific arts. ß

BBC weather

reporter Michael

Fish’s blunder

surrounding the

1987 hurricane

led to major

improvements in

forecasting

The familiar isobars

we see on today’s

weather maps show

regions of equal

pressure

EXPL AIN IT TO A FRIEND

COLD FRONT

The leading edge of a cool mass

of air that displaces warm air from

the ground. It is typically

associated with rain.

ENSEMBLE FORECASTING

The practice of running a specific

weather forecast numerous times

with var ying initial conditions to test

the predictability of the atmosphere

at a given moment.

NOWCASTING

A term coined by Met Office

scientist Prof Keith Browning to

describe short-term forecasts that

are updated frequently.

SYNOPTIC CHART

A map that simultaneously shows

various attributes of the

atmosphere at a given moment.

TROPOSPHERE

The lowest layer of Earth’s

atmosphere and the space in which

nearly all organic life lives and

almost all weather occurs.

PHOTO: BBC, SCIENCE PHOTO LIBRARY