Atmospheric sciences
Probes, satellites, and data transmission
Kites equipped with meteorgraphs were used as atmospheric probes in the late 1890s, and in 1907 the U.S. Weather Bureau recorded the ascent of a kite to 7,044 metres above Mount Weather, Virginia.
In the 1920s the radio replaced the telegraph and telephone as the principal instrument for transmitting weather data. By 1936 the radio meteorgraph (radiosonde) was developed, with capabilities of sending signals on relative humidity, temperature, and barometric pressure from unmanned balloons. Experimentation with balloons up to altitudes of about 31 kilometres showed that columns of warm air may rise more than 1.6 kilometres above Earth’s surface and that the lower atmosphere is often stratified, with winds in the different layers blowing in different directions. During the 1930s airplanes began to be used for observations of the weather, and the years since 1945 have seen the development of rockets and weather satellites. TIROS (Television Infra-Red Observation Satellite), the world’s first all-weather satellite, was launched in 1960, and in 1964 the Nimbus Satellite of the United States National Aeronautics and Space Administration (NASA) was rocketed into near-polar orbit.
There are two types of weather satellites: polar and geostationary. Polar satellites, like Nimbus, orbit Earth at low altitudes of a few hundred kilometres, and, because of their progressive drift, they produce a photographic coverage of the entire Earth every 24 hours. Geostationary satellites, first sent up in 1966, are situated over the Equator at altitudes of about 35,000 kilometres and transmit data at regular intervals. Much information can be derived from the data collected by satellites. For example, wind speed and direction are measured from cloud trajectories, while temperature and moisture profiles of the atmosphere are calculated from infrared data.
Weather forecasting
Efforts at incorporating numerical data on weather into mathematical formulas that could then be used for forecasting were initiated early in the century at the Norwegian Geophysical Institute. Vilhelm Bjerknes and his associates at Bergen succeeded in devising equations relating the measurable components of weather, but their complexity precluded the rapid solutions needed for forecasting. Out of their efforts, however, came the polar front theory for the origin of cyclones and the now-familiar names of cold front, warm front, and stationary front for the leading edges of air masses (see climate: Atmospheric pressure and wind).
In 1922 the British mathematician Lewis Fry Richardson demonstrated that the complex equations of the Norwegian school could be reduced to long series of simple arithmetic operations. With no more than the desk calculators and slide rules then available, however, the solution of a problem in procedure only raised a new one in manpower. In 1946 the mathematician John von Neumann and his fellow workers at the Institute for Advanced Study, in Princeton, N.J., began work on an electronic device to do the computation faster than the weather developed. Four years later the von Neumann group could claim that, given adequate data, their computer could forecast the weather as well as a weatherman. Present-day numerical weather forecasting is achieved with the help of advanced computer analysis (see weather forecasting).
Cloud physics
Studies of cloud physics have shown that the nuclei around which water condenses vary widely in their degree of concentration and areal distribution, ranging from six per cubic centimetre over the oceans to more than 4 million per cubic centimetre in the polluted air of some cities. The droplets that condense on these foreign particles may be as small as 0.001 centimetre in diameter. Raindrops apparently may form directly from the coalescence of these droplets, as in the case of tropical rains, or in the temperate zones through the intermediary of ice crystals. According to the theory of Tor Bergson and Walter Findeisen, vapour freezing on ice crystals in the clouds enlarges the crystals until they fall. What finally hits the ground depends on the temperature of air below the cloud—if below freezing, snow; if above, rain.
Properties and structure of the atmosphere
Less than a year after the space age began with the launching of the Soviet Sputnik I in 1957, the U.S. satellite Explorer I was sent into orbit with a Geiger counter for measuring the intensity of cosmic radiation at different levels above the ground. At altitudes around 1,000 kilometres this instrument ceased to function due to saturation by charged particles. This and subsequent investigations showed that a zone of radiation encircles the world between about latitude 75° N and 75° S, with maximum intensities at 5,000 and 16,000 kilometres. Named after the American physicist James Van Allen, a leading investigator of this portion of Earth’s magnetosphere, these zones are responsive to events taking place on the Sun. The solar wind, a stream of atomic particles emanating from the Sun in all directions, seems to be responsible for the electrons entrapped in the Van Allen region as well as for the teardrop shape of the magnetosphere as a whole, with its tail pointing always away from the Sun.
In 1898 Teisserenc de Bort, studying variations of temperature at high altitudes with the aid of balloons, discovered that at elevations of about 11 kilometres the figure for average decrease of temperature with height (about 5.5 °C per 1,000 metres of ascent) dropped and the value remained nearly constant at around −55 °C. He named the atmospheric zones below and above this temperature boundary the troposphere and the stratosphere.
Toward the end of World War II the B-29 Superfortress came into use as the first large aircraft to cruise regularly at 10,000 metres. Heading westward from bases in the Pacific, these planes sometimes encountered unexpected head winds that slowed their flight by as much as 300 kilometres per hour. The jet streams, as these high-altitude winds were named, have been found to encircle Earth following wavy courses and moving from west to east at velocities ranging upward to 500 kilometres per hour. Aircraft have also proved useful in studies of the structure and dynamics of tropical hurricanes. Following the destruction wrought to the Atlantic Coast of the United States in 1955 by hurricanes Connie and Diane, a national centre was established in Florida with the missions of locating and tracking and, it is hoped, of learning how to predict the paths of hurricanes and to dissipate their energy.