Handbook of Meteorology/Distribution of Pressure
THE AIR: THE DISTRIBUTION OF PRESSURE
Measurement of Atmospheric Pressure.—Practically all the movements of the air are due to differences in its temperature; but, inasmuch as differences in temperature result in differences in the density of the air, it is convenient to express such differences in terms denoting the force with which the air presses upon the earth at sea level. It is also more convenient in weather science to express this pressure in terms of the length of a column of mercury which the air balances—that is, a barometer.[1]
Thus, a column of air 1 square inch in cross-section weighs at sea level about 14.7 lbs., or 1 atmosphere. It balances a column of mercury of the same sectional area, 29.92 inches in length. In metric terms, the weight of a column of air 1 square centimeter in sectional area is 1033.3 grams, and it balances a column of mercury 760 millimeters in length.[2] The length of a column of water which balances a column of air of the same sectional area is about 34 feet.
Distribution of Pressure.—The movements of the air caused by heating, cooling, expansion and contraction include the general or planetary movements, as well as the massing of the air in one locality and the counterbalancing depressions formed in another. The expansion of the air by heating has been determined many times. If 1000 parts of air at 32° F be heated to 33° F, its volume will be increased 2.035 parts; at 50° F the increase will be 36.63 parts; at 130° F, a temperature very common under a summer sun, the 1000 parts will become 1199.43
Pressure and winds in January.
parts. In centigrade terms, the increment is 0.00037 for each degree, measured from the absolute zero. Small as seems the rate of expansion, the actual increment over a continent, even for a rise of temperature of a few degrees, must be measured in terms of cubic miles, and its aggregate weight in millions of tons.
Observations show that the mean pressure over the earth varies from season to season; at any given locality existing pressure varies from hour to hour. It also varies according to latitude; but the variation for latitude is not regular. The maps on pages 40 and 42 show the several regions of high pressure. They show also that the pressure in these regions in January is slightly lower than in July. The regions of summer high pressure are situated in latitude 30° to 35° north and south.
The southerly regions cover the ocean and, for the greater part, are far removed from human activities. The northerly regions are of great importance from the fact that they modify the climate, the one of North America, the other of Europe and Asia.
The North Pacific high covers ocean waters in July, and it tends to carry cool air to the adjacent coast. In January it covers the western part of Canada and at times pours an enormous volume of cold air over the greater part of the United States.
The North Atlantic high covers western Asia in January; in July it covers the ocean east of the United States. At times, the area of maximum pressure, 30.30 inches (1026 mb) or more, is close enough to the continental coast to retard the easterly flow of air and cause a pretty general stagnation of air over the eastern part of the United States. It is therefore a feature in the formation of hot spells over that region.
The area covered by the North Pacific has a January pressure of 30.20 inches and a July pressure of 30.30 inches (1026 mb). The Bermuda, or Atlantic high, is about 30.25 inches (1024 mb); over Siberia, however, the winter high is not far from 30.50 inches (1033 mb). It is thought that the two zones of high pressure in mid latitudes are due to the descent of the upper currents that constituted the updraught in equatorial regions. This is denied by many meteorologists, however.
Pressure and winds in July.
The areas of low pressure are much larger in extent than those of high pressure and, as a rule, they are not so well defined. The area of lowest pressure is in the south polar region; it is inclosed by the sixtieth parallel; its mean pressure is 29.40 inches (996 mb). A low pressure area in the North Atlantic lies east of Greenland; a similar low pressure area in the North Pacific covers Bering Sea.
The zone of ascending air currents in equatorial regions is a region of low pressure. Its mean, summer and winter, does not vary much from 29.80 inches (1009 mb). Indeed, the changes in pressure from month to month throughout tropical regions are very slight. The ascending currents in equatorial regions and the descending currents near the tropics are assumed to exist. Their existence, established circumstantially rather than positively, explains satisfactorily the position of the constant highs and lows.
The great differences between summer temperature and winter temperature explain the apparent shifting of each summer high from a position over the ocean in summer to one over the nearby continent in winter. Cold air is heavier than warm air; and, in the latitude of the constant highs, the temperature of the air over the land in winter is much lower than over the ocean; in the summer, on the other hand, the temperature is lower over the ocean. In each case the high forms in the region of lower temperature.
Mean Pressure over the Earth.—It is customary to reduce all pressure observations used for comparison to a sea level basis and to a temperature of 32° F (0° C). The maps, pages 40 and 42, show the marked variations in pressure that are seasonal. From these pressures the mean pressure over the earth has been calculated by meteorologists to be between 29.90 and 29.85 inches. W. M. Davis has calculated the mean pressure over the northern and the southern hemispheres, for the summer and the winter months, deducing the following values: 29.99 inches (1016 mb) for January and 29.87 inches (1012 mb) for July in the northern hemisphere; 2991 inches (1013 mb) in July (midwinter) and 29.79 inches (1009 mb) in January (midsummer) in the southern hemisphere.
The determination of mean pressure over each of the two hemispheres is more important than that of the mean pressure of the earth as a whole. The fact that the pressure over the southern hemisphere is lowest at the time when it is highest over the northern hemisphere, and vice versa, indicates the shifting of an enormous volume and weight of air from one hemisphere to the other, twice a year. Davis estimates that the weight of air thus moved is equivalent to a pressure of 0.12 inch (4.1 mb), or between 30 and 35 million tons in weight.
Density.—It is evident that a close relation exists between the temperature, pressure and density of the air. With lowering temperature, the volume of air contracts, and air flows in to equalize the loss. The cold air contains a greater number of molecules per given volume and therefore its density is increased; because it contains more matter per given volume its weight, and therefore its pressure is increased. Density varies directly with pressure and inversely with temperature. It also varies inversely with the moisture content of the air.
For the greater part, human activities are carried on at the plane of contact, where earth and air meet. Weather science, however, includes a study of density and pressure at all observable altitudes from sea level upward. Density of the air at different heights also affects air flight and the flight of projectiles. Hence, a knowledge of the density of the air at different altitudes is necessary.
The changes in the density of the air are most marked at the earth’s surface. The daily range in density may be as much as 10 per cent, and the extreme range in a year has exceeded 20 per cent. The range is greatest in temperate latitudes at or near sea level. The changes in density due to temperature variations explain the high midwinter pressure over inland regions and the high midsummer pressure in oceanic regions; the rock envelope of the earth radiates its heat more rapidly than does the water. Very low temperatures in winter increase the density of the air. High temperatures in summer decrease the density, with the result that oceanic regions are cooler in summer and warmer in winter than far-inland continental regions.
Diurnal and Semi-Diurnal Changes in Pressure.—At stations of considerable elevation a maximum daily pressure at the warmest part of the day is observable. It is attributed to the heating of the air, thereby causing an accumulation which 
The barogram of a quiet week in equatorial regions, showing the semi-diurnal osillations in pressure. The maximum at 12 o’clock is due to clock time instead of solar time.
practically forms the crest of the wave of greatest warmth. During the coldest hours of the day a reverse movement takes place and forms a corresponding trough of pressure. This diurnal maximum and minimum of pressure is practically a raising and lowering of the center of mass. As a result, a greater mass means greater pressure, and vice versa.
The semi-diurnal maximum and minimum is very regular and obtains in every part of the earth. It is best studied from the barogram, a strip of paper attached to the revolving drum of a recording barometer. The line drawn by the barograph pen shows a slight rise above mean pressure at 10 o’clock, morning and night, followed by a depression at 4 o’clock, morning and afternoon. These oscillations in pressure are probably due to the waves of temperature which ceaselessly follow the sun with the rotation of the earth.
The semi-diurnal maxima and minima are greatest in equatorial latitudes; they decrease in higher atitudes. According to the observations of General Greely, the oscillations of the barograph pen were scarcely noticeable in polar regions. The amplitude of oscillation is greater by day than by night; it is greater at the equinoxes than at the solstices. The day amplitude is greater over the continents than over the sea; the night amplitude is the reverse. According to Humphreys, the whole atmospheric shell vibrates in waves which happen to be in 12-hour wave lengths. Records made by P. R. Jameson on the East African Coast near the equator show a maximum of about 0.025 inch above and the same minimum below normal pressure.[3] In tropical regions the irregular variations in pressure are infrequent; the semi-diurnal oscillations, on the other hand, are very regular. The claim that an observer can tell the time of day by the barometric pressure is not without foundation.
Other Variations in Pressure.—Pressure ranges exceeding 1.5 inches during a week—the record of a barograph sheet—are not uncommon. Weekly and monthly ranges are usually much greater in winter than in summer and much greater in mid-latitudes than in low or high latitudes. Professor Mohn has summarized as follows:[4] The barometer is high when the air is cold, when it is dry, and when an upper current flows into a given area. It is low when the lower air is heated, when it is damp, and when it has an upward movement.
The variations in pressure with which weather science is chiefly concerned are the daily highs and lows which cross the continents in mid-latitudes from west to east and, for the greater part, are lost in mid-ocean. These great billows of the atmosphere are comparable to the billows of the sea; but, as is shown in Chapter XIII, the lows are usually storm centers and the highs are rapidly moving masses of cold air. The former are the cyclones of the forecaster; the latter, the anticyclones. When accompanied by rain or by snow they are the “storms” of popular tradition. Local disturbances, such as tornadoes, water spouts and thunder-storms, usually affect pressure, and leave each its record on the sheet of the recording barometer. Observers learn quickly to interpret these records.
Isobars and Gradients.—The distribution of pressure is best shown by means of lines drawn on a map through adjacent points having the same pressure. These lines are isobars. The maps, pp. 40 and 42, show midwinter and midsummer isobars. For the sake of comparison, the figures are reduced to the basis of sea level and temperature of 32° F (0° C). The isobars on the daily weather map show the respective positions of highs and lows, and from them the daily forecasts are made.
On the daily weather map the conditions of pressure are interpreted by a study of the relative positions of the isobars, which constitute a contour map of the air. If the isobars of a high, or of a low, are close to one another, the slope of the air wave is steep; if far apart, the slope is gentle. Therefore, in either case, they show the gradient of pressure, and from the gradient of pressure an approximate velocity of the wind may be indicated.
About half a century ago, Whipple, of Kew Observatory, prepared an empiric table based upon isobars 15 nautical miles apart—that is, a 15-mile gradient. If, for instance, the gradient is 0.1 inch, the indicated velocity of the wind will be approximately 9 miles per hour; if 0.2 inch, it will be about 17 miles per hour, etc. The study of the pressure gradient, therefore, is a fairly accurate indication of wind velocity. It also enables the observer to make a reasonably accurate forecast of wind velocity from twenty-four to thirty-six hours in advance.
Actual and Recorded Pressure.—Pressure decreases with altitude, at a varying rate. If the lowest of a pile of ten books is lifted, the weight of the nine books above it must be overcome; but if the fifth book is lifted, the weight of only four books must be overcome. The same principle applies to the atmosphere. At sea level a column of air I square inch in cross-section, presses with a weight of 14.7 lbs., but at a height of 19,000 feet the weight of the column is only half as great. For the first few hundred feet above sea level the pressure decreases at the rate of 0.1 inch for each 90 feet of ascent; at an altitude of 3000 feet it is at the rate of 0.1 inch for 100 feet of ascent. The greater number of weather stations in the United States are 1000 feet or more above sea level, and many of those west of the Denver meridian are more than 5000 feet above sea level.
For purposes of comparison in the preparation of daily weather maps, all pressure observations are reduced to sea level basis. For this purpose such a reduction is necessary, and all reduced pressures within an altitude of a few hundred feet of sea level are sufficiently correct for practical purposes. For altitudes materially greater than 1000 feet, the results when applied to mean pressure, are erroneous. Thus, at Mount Washington, the mean recorded pressure for January, reduced to sea level, is greater than that for July. As a matter of fact, the actual mean pressure is less in January than in July. The following illustration will explain:
A compress 12 feet in height is filled with loose cotton. The pressure of its weight at the bottom is, say, 16 lbs. per square foot. Half way to the top, at the 6-foot level, the pressure is half as much. Now let us assume that the cotton is compressed so that its depth is only 9 feet. The pressure at the bottom remains the same; but at the 6-foot level, there is only half as much cotton as before compression; hence the pressure is half as great. The center of mass has been lowered in the process of compression.
The same principle applies in the case of measurements of the atmosphere. Thus, at Mount Washington, and at other stations of considerable altitude, expansion due to temperature-increase raises the center of mass in summer; mean pressure, therefore, is raised. In winter, low temperature causes contraction, lowering the center of mass and therefore the pressure. In other words, while sea level and also the observer’s station are at fixed altitudes, the center of mass is a varying altitude; it is raised by increasing temperature and lowered by decreasing temperature. Hence its effect on mean pressure.
- ↑ From two Greek words meaning “measure of weight.”
- ↑ British observers usually express the pressure of the atmosphere in millibars: 1 inch = 33.864 mb; 1 mb = 0.02953 inch; therefore at 29.53 inches, the barometer reading is 1000 mb. At 29.92 inches, or 1 atmosphere, the barometer reading is 1013.2 mb. In certain computations, the millibar scale of the barometer possesses many conveniences.
- ↑ At Mount Vernon, N. Y., Lat 40° 54', the values are approximately 0.022 inch—possibly less.
- ↑ Grundzüge der Meteorologie.