The General Circulation of the Atmosphere

The July chart of mean pressure shows no fundamental differ ence from the conditions which prevail in January, so far as the southern hemisphere is concerned. In the northern hemisphere, however, the distribution of pressure has changed fundamentally. The Asiatic anticyclone has disappeared, and its function as the chief controller of conditions over Asia has been taken by a de pression centred to the north of India. There is a continuous increase of pressure southward from the centre of the depression as far as the sub-tropical anticyclonic belt of the southern hemi sphere. As a result, there is over the whole Indian ocean a broad current of air which blows into the southern edge of the depres sion as a south-west wind, known as the south-west monsoon of India (Simpson, "The South West Monsoon," Q.J.R. Met. Soc.). This current, having passed over warm ocean during a journey some thousands of miles in length, reaches India as a warm and very damp current. On reaching India it is forced by the con figuration of the land to rise over the coastal ranges of moun tains, and so gives rise to copious rainfall, which is known as the monsoon rainfall. It may be noted in passing that the word mon soon denotes seasonal, so that it might be applied to any seasonal changes of wind. The winds which blow around the south-eastern edge of the Asiatic anticyclone in winter are known as the north east monsoon. An examination of charts of mean pressure month by month shows that the Asiatic anticyclone begins to diminish rapidly in intensity in April, and by early June the depression is formed. The monsoon winds, with the associated rainfall, develop during Jane, and usually persist until late September.

No verbal description of the distribution of temperature, pres sure and wind over the earth's surface can take the place of an examination of the charts. The reader who desires fuller in formation is referred to Sir Napier Shaw, Manual of Meteorology, vol. ii., where maps are given showing the distribution of these elements, and some others, for each month of the year.

Rainfall.

In many respects rainfall (which for our present purposes will be taken to include all forms of precipitation, snow, hail, etc.) is the most important of all the meteorological ele ments. It is at the same time the most difficult of these elements to treat in a satisfactory manner, on account of its extreme vari ability with place, and the fact that an hour of torrential rain associated with a thunderstorm may affect in a marked manner the representation of both the mean diurnal and mean seasonal variations. What is known as the "normal" seasonal variation of any element is the average form of the annual cycle of that ele ment taken over a number of years, but it is only over very restricted areas of the globe that the variation of rainfall in any one year will resemble at all closely the "normal" variation. The outstanding case where this is possible is that of the Indian peninsula, whose heavy rainfall in June to September is associ ated with the monsoon winds. The maximum rainfall in any year will occur within the period of the monsoons, and never out side it. Even in the British Isles averages taken over many years point to a definite maximum of rainfall in a particular month of the year. This month is October at Greenwich, December at Aberdeen and July at Edinburgh, but it would be extremely hazardous to apply this result to forecast the rainfall of any particular year.

Within the tropics rainfall usually has a well marked maximum and occasionally two maxima.

World Weather.

Under this title may be grouped a number of very extensive statistical investigations of the correlation be tween meteorological factors over different parts of the world.

Among the most notable of these are two papers by G. T. Walker in the Indian Meteorological Memoirs (1923 and 1924) one by Walker and Bliss in the Memoirs of the Royal Met. Soc. vol. ii. No. 17, several papers by Exner in the Sitzungsber. Akad. Wiss. Wien (1913, 1924 and 1926) ; and a valuable general discussion of the work of other writers by Defant in the Meteorologische Zeit schrift (1926). Results of considerable value have been derived (in the northern hemisphere). In the lowest layers, next to the ground, the wind speed increases rapidly, while the direction changes slowly, but at greater heights the increase of wind be comes slower, while the direction changes rapidly. The variations of wind in the lowest 2,000 ft. are nearly always in the sense in dicated above. At greater heights the nature of the changes of wind depends upon the horizontal distribution of temperature, and upon the wind direction.

The variation of wind with height in the lowest layers can be represented with considerable accuracy by a formula V =a+b log (h+c), where h is height and a, b and c are constants (E. H. Chapman, M.O. Professional Notes No. 6). Fig. 3 (Dobson Q.J.R. Met. Soc., 1920), shows the variation of wind with height in the upper part of the tropo sphere and the lower stratosphere. The heights are relative to the base of the stratosphere, and the data are classified according to the wind velocity in the highest two kilometres of the troposphere, the groupings being (a) less than 13 metres per sec., (b) 13 to 19 metres per sec., (c) greater than 19 metres per second.

The direction of the winds considered above showed no marked changes within the troposphere apart from the usual changes at the ground, and no appreciable change of direction was usually noted on passing into the stratosphere. The most noteworthy feature of fig. 3 is the steady increase of winds within the tropo sphere and the equally steady and rapid decrease within the lower stratosphere. The rapid falling off of wind in the strato sphere denotes that the stratosphere is warmer over low than over high pressure.

The general distribution of wind at different heights can be most readily summarized by reference to the distribution of pressure at different heights. Shaw (Manual of Meteorology, vol. ii., figs. 164-172) gives diagrams of isobars at the surface, and at heights of 4 km. and 8 km.

Further details of the variation of wind with height in different regions will be found in Manual of Meteorology vol. ii., 6. See also: van Bemmelen Proc. K. Akad. Wet. Amst., (1918), Har wood, "The Free Atmosphere in India," Memoirs Indian. Met. Dept. (1924) ; Arbeiten Preuss. Obs. Lindenberg, especially vol. xiii. ; Cave, Structure of the Atmosphere in Clear Weather (1912) ; Alfred de Quervain, Schweizerische Gronlande Expedi tion 1912-13 (Zurich, 1920) ; E. Gold, "Barometric Gradient and Wind Force," M.O. 190, and "The International Kite and Balloon Ascents," Geophysical Memoirs, No. 5, M.O. 210.

by these writers, and an idea of the type of result obtained may be gathered from the following extract from a paper by Walker in the Indian Meteorological Memoirs, vol. xxiv. part 4: "We can best sum up the situation by saying that there is a swaying of pressure on a big scale backwards and forwards between the Pacific ocean and the Indian ocean, and that there are swayings, on a much smaller scale, between the Azores and Iceland, and between the areas of high and low pressure in the north Pacific: further, there is a marked tendency for the highs of the last two swayings to be accentuated when pressure in the Pacific is raised and that in the Indian ocean lowered."