It must be clearly understood that these remarks refer only to observations at a fixed station. If a fall of barometer is observed in travelling, it may be due either to a change in the state of the atmosphere or to a change in the traveller’s height above sea-level. This is the reason why it is absolutely essential, in making barometric estimates of height (or boiling-point determinations), to have simultaneous observations going on at a base-station, or preferably at a series of intermediate stations.
The ordinary prognostics of the approach of rain or bad weather differ in different localities, and require a considerable amount of local knowledge before they can be utilised. The peculiar absorption band in the solar spectrum due to the water vapour of the Earth’s atmosphere, and called the rain-band, is a valuable guide to an experienced observer with a spectroscope in predicting rain. The only instrument, however, likely to be useful to the ordinary traveller is the wet and dry bulb thermometer. When the two thermometers have the same reading, indicating saturation of water vapour, or when they approach at temperatures above 60° F. within two degrees or so, rain may be expected, or possibly mist. The appearance of low clouds clinging to the hillsides is an indication that the temperature at the place where they are is below the dew-point. The appearance of the upper clouds, taken in conjunction with the readings of the barometer, is a valuable indication of forthcoming weather changes. The increase of cirrus clouds in a clear sky with a falling barometer, or the appearance of a solar or lunar halo, may be taken as a sure sign of an approaching cyclone, the intensity of which may be foreseen by the rate at which the barometer is falling.
While the weather of places on the west coasts of temperate continents exposed to the prevailing sea-wind is usually made up of a succession of cyclones of different degrees of intensity, and of the anticyclonic intervals between them, over the greater part of the Earth’s surface the climate is much more uniform, and the seasonal changes are the principal cause of changes of weather. To understand these general conditions it is necessary to consider the elements of climatology.
3. Outlines of Climatology.—The air is in constant movement on account of the unequal way in which the heat of the sun falls on different parts of the Earth’s surface, and at different seasons of the year. All the conditions of the atmosphere show a certain diurnal periodicity which is most marked in the regions of steady climate between and near the tropics. Thus, as a rule, the minimum temperature of the air occurs just before sunrise, the maximum temperature from two to three hours after noon. The amount of difference between the maximum and minimum temperature of the day (daily range) is least near the sea or in wet regions (a maritime climate) and greatest in the interior of the continents, especially where the rainfall is slight (a continental climate). Over the sea itself the daily range of air temperature averages only 3 Fahrenheit degrees; but in the heart of a continent, especially in a desert, it may exceed 60 Fahrenheit degrees.
Diurnal changes of pressure are proportionally much smaller in amount than changes of temperature, and are to be observed as a regular phenomenon only in the tropics, or elsewhere during very settled weather. There are usually two maxima daily, about 10 A.M. and 10 P.M., and two minima occurring about 4 A.M. and 4 P.M. It is only in rare cases that the total barometric range exceeds 0.10 inch, very frequently it is not greater than 0.04 inch. Still it is convenient to remember in the tropics that a fall of the barometer not greater than 0.10 inch between 10 A.M. and 4 P.M. is to be expected, and does not indicate either the approach of a storm (if the observer is at rest) or the ascent of 100 feet (if he is on the march).
Associated with the diurnal changes of temperature in settled weather are changes of wind due to local configuration of the ground. The wind, for example, usually blows up a mountain side, or up a steep valley, during the day, and down a mountain, or down a steep valley, during the night. So, too, the regular land and sea breezes found on the borders of the sea or of great lakes blow from water to land in the day time and from land to water at night. Here the determining cause is the fact that land is warmed and cooled by radiation, and in turn heats or chills the air much more than water does. In the settled climates of high tropical plateaus a regular diurnal change of wind direction has been observed, the wind blowing successively from all points of the compass.
A similar diurnal periodicity occurs in the amount of cloud, in the moisture of the air, the fall of rain, the occurrence of thunderstorms, etc. It is also to be noticed in the flow of rivers in mountainous regions where the streams take their rise from glaciers or snow, the rapid melting of which by the heat of the sun causes the volume of water to increase greatly in the afternoon, while the cessation or reduction of the rate of melting at night diminishes the volume of the river in the morning and forenoon.
Periodic changes of greater amount but similar in kind are produced by the alternation of the seasons, the difference between the mean values of the months in which the phenomena are at a maximum and minimum respectively being termed the annual range. With regard to temperature, very moderate changes occur in the tropical zones where the altitude of the noon-day sun is always great, and the length of day and night varies little with the season (for the most part less than 5 Fahrenheit degrees); but in the temperate and frigid zones there are strongly marked annual changes. As in the case of daily range, proximity to the sea is a controlling factor in the annual range of temperature. To take a very characteristic instance, the annual range between the mean temperature of July and January is about 23 Fahrenheit degrees in the Lofoten islands on the margin of the Atlantic, while it is 120 Fahrenheit degrees at Verkhoyansk in the same latitude, but in the centre of the Asiatic continent.
The extreme months for air temperature are January and July in almost every part of the world, the maximum occurring north of the equatorial belt in July and south of it in January.