WEATHER SCIENCE
HIGH STRATUS AND CIRRO-STRATUS CLOUDS. (Photo by Capt. Wilson Barker, F.R.S.E., F.R.Met.Soc,, &c.)
Frontispiece,
WEATHER SCIi :E
AN ELEMENTARY INTRODUCTION
TO METEOROLOGY
BV
F. W. HENKEL, B.A.
FELLOW OF THE ROYAt. ASTRONOMICAL SOCIETY, MEMBER OF THE BRITISH ASTRONOMICAL ASSOCIATION, LATE DIRECTOR OF MARKREB O8SKRVATORY, ETC., ETC
T. FISHER UNWIN
LONDON LEIr
ADELPHI TERRACE IN> ASSE so
HIGH STRAITS AVU C»HK« (Photo by t>pt, Wilson Barter V
WEATHER SCIENCE
AN ELEMENTARY INTRODUCTION
TO
METEOROLOGY
BY
F. W. HENKEL, B.A.
FELLOW OF THE ROYAL ASTRONOMICAL SOCIETY, MEMBER OF THE BRITISH ASTRONOMICAL ASSOCIATION, LATE DIRECTOR OF MARKREE OBSERVATORY, ETC., ETC.
T. FISHER UNWIN
LONDON LEIPSIC
ADELPHI TERRACE INSELSTRASSE 20
1911
[All rights reserved.]
PREFACE
IN the preparation of this work I must acknow- ledge my indebtedness to previous writers on the subject, and to the admirable Text-books, both popular and scientific, that have appeared from time to time. Amongst these I may mention the volumes by Dr R. H. Scott and Mr Abercromby in the International Scientific Series, Dr Waldo's " Modern Meteorology," Dr Hann's " Meteorologie," my friend Mr Inwards' " Weather Lore," Flammarion's "I/ Atmosphere," Dr Archibald's and Mr G. F. Chambers' popular works, as well as the transactions and journals of Meteorological Societies which have been consulted and referred to in due course. My indebtedness is also great to the late Commander M. F. Maury's "Physical Geography of the Sea," to the Text-books of Professor S. P. Thompson and Dr R. W. Stewart on Electricity, to the late Dr C. A. Young's " General Astronomy," etc. For the illustrations I am indebted to the
kindness of Captain Wilson Barker, of H.M.S.
5
261609
6. ••••'.: Preface
training ship Worcester, who sent me seven photographs of typical clouds, taken by himself, one of which appears as the frontispiece ; whilst Mr Inwards has permitted the reproduction of his plate of cloud forms, as well as the illustra- tions of Greenwich and Kew Observatories, and the Temple of the Winds, Athens ; to Dr H. R. Mill, head of the British Rainfall Organisa- tion, I owe the Rainfall Map that appears in chapter iv. Messrs Casella, Pillischer and Short and Mason have also kindly supplied illustrations of some of the instruments, barometers, ther- mometers, etc., described in the chapters on instruments. To all of these gentlemen for their help in these and other ways, as well as to others, my thanks are due.
In addition 1 wish to express my gratitude in especial to Dr H. R. Mill and Mr Inwards for their kindly advice and recommendations during the course of publication.
F. W. HENKEL.
CONTENTS
CHAPTER I
PAGE
Introductory — Ancient ideas — The barometer as a ' ( weather glass " — Dawn of modern meteorology — Cyclones and anticyclones — The earth : its size, shape, and motions — The atmosphere 13
CHAPTER II
The barometer — Mercurial barometer — Glycerine barometer — Fortin's barometer — Adjustments — Syphon barometer — Fishery barometer — Aneroid barometer — Self-recording barometers — Corrections to reading of barometer — Weather variations with barometer — Oscillations of pressure, diurnal and annual — Extreme range — Dove's rules — Regions of high and low pressure — Pressure at various altitudes above the surface . 46
CHAPTER III
The thermometer — Scales : Fahrenheit, Centigrade, Reaumur, De L'Isle — The hypsometer — " Wet and dry bulb " hygrometers — De Saussure's hair hygrometer — Maximum and minimum thermometers — Self- recording ther- mometers (thermographs) — The Stevenson screen — Glaisher screen — Black bulb thermometers — Ranges in temperature . . . . . . . . .71
7
8 Contents
CHAPTER IV
PAQB
Rain gauges — Sunshine recorders — Wind instruments —
Anemometers — Distribution of rainfall . . .89
CHAPTER V
\l Weather forecasts — Types of circulation — The seven funda- mental forms of isobars — Cyclones — Secondaries — V-shaped depressions — Cols — Straight isobars — Anti- cyclones— Wedges — Lower and upper winds . . .103
CHAPTER VI
More detailed description of course of circulation — Weather in (1) cyclones, (2) anticyclones — Special varieties — Whirlwinds — Tornadoes . . . . . .114
CHAPTER VII
Clouds — Howard's nomenclature — Cirrus — Cirro-cumulus- Cumulus — Stratus — Cirro - stratus — Cumulo - stratus — Nimbus — Heights of different kinds of clouds — Motions of clouds and air currents — Fogs, mists, etc. — Dust particles or other nuclei necessary — Cloud " prognostics " of weather changes 131
CHAPTER VIII
The winds — Direction — Compass and true bearings — Relation to isobars — Beaufort's scale — Buy Ballot's law — Trade winds — General circulation — Special winds : Fohn, scirocco, bora, mistral, land and sea breezes — [Note on wind velocities and Beaufort's scale] . . . .148
CHAPTER IX
Miscellaneous phenomena — Snow — Hail — Dove's theory — Volta's theory — Dew — Fog — Mist — Rainbows — Halos — Coronae — The mirage — Blue of the sky — Tints of sunrise and sunset — Dust . . . . . . . .161
Contents 9
CHAPTER x
PAGE
Methods of heat transference — Ocean currents — The Gulf Stream — Influence on temperature and other weather conditions — Theories as to its origin, etc. — The Humboldt or Peruvian Current — Cooling effect — China Current — " Kuro Siwo '' — Currents of the Indian Ocean — Agulhas Current — Polar Currents — Ideas as to their causes . .181
CHAPTER XI
Atmospheric electricity — Instruments — Quadrant electro- meter — Portable electrometer — Thunderstorms — Light- ning conductors — St Elmo's fire — Electric "Hum" — The Aurora — Magnetic storms — Magnetic and true bearing 199
CHAPTER XII
Seasonal variations of weather — Recurrent types — Buchan's hot and cold periods — Indian summer — St Luke's summer — St Martin's summer — May and November cold spells, etc. — Southerly, northerly, westerly, and easterly types ( Abercromby) — Southerly type — Weather conditions — Westerly type — Northerly type — Easterly type — Distribution of pressure over the globe — Polar cyclones —The Doldrums 225
CHAPTER XIII
Observatories — First order, second order, and third order stations — Greenwich Observatory — Self-recording instru- ments— Kew Observatory — Second order stations — Their work — Bi-daily observations — Instruments — Observations — Morning series — Evening series — Forecasts of weather — Storm warnings — List of districts — Daily and weekly weather reports — [Note on metric and " English " measures] -Temple of the Winds, Athens . . . 239
CHAPTER XIV
Weather signs and portents of coming changes — Sun — Moon — Clouds — Cirrus — Cirro-stratus — Cirro-cumulus —
io Contents
PAGE
Stratus — Cumulus — Cumulo-stratus — Colours of the sky — Animals — Birds 261
CHAPTER XV
Cycles of weather — Bruckner's investigation — Rainfall — Wet and dry periods — Oscillations of mean pressure and temperature — Sun-spots and the weather — Sun-spot " period": its irregular length — The moon . . . 272
CHAPTER XVI
The seasons — Heating action of the sun — Zones — Climate affected by proximity to sea — Isothermal lines : their course across the globe — Extremes of temperature — Isabnormals — Sea temperatures — Influence of Gulf Stream, etc. — Minimum temperature of sea 285
CHAPTER XVII
Upper air — Balloon ascents — Kites — Isothermal layer —
Division of air into two regions, lower and upper . .311
CHAPTER XVIII
Practical applications — " Signs" of rain — The leech - Seasonal changes — Migration of birds — Plant " Phen- ology"— Flowering of plants — Fall of leaves — Seasonal prevalence of diseases— Salubrity of particular regions —Ozone .' . .318
INDEX 331
LIST OF ILLUSTRATIONS
HIGH STRATUS, AND CIRRO-STRATUS CLOUDS Frontispiece
STANDARD BAROMETER. (FORTIN'S PATTERN) . page 50
ANEROID . „ 54
BAROGRAPH ... . „ 56
WET AND DRY BULB HYGROMETER . . . ,, 76
MAXIMUM THERMOMETER ) _Q
f » /y
MINIMUM THERMOMETER >
STEVENSON SCREEN ,,83
SOLAR RADIATION, BLACK BULB THERMOMETER
84
GRASS MINIMUM THERMOMETER
THE HYGRODEIK . . . . . . „ 88
RAIN GAUGE AND MEASURING GLASS . „ 90
SUNSHINE RECORDER ,,93
ANEMOMETER WITH WIND VANE . . . • „ 95 ROBINSON'S ANEMOMETER ... . „ 96
MEAN ANNUAL RAINFALL MAP OF THE BRITISH
ISLES to face page 102
SELF-RECORDING RAIN GAUGE .... page 102
CLOUD FORMS, BY COL. H. M. SAUNDERS . tofacepage 131
CUMULUS AND HIGH STRATUS )
} . „ 134
CIRRUS •
)
SQUALL CUMULUS CUMULUS
11
12 List of Illustrations
CIRRO-CUMULUS ^
> to face page 140
GROUND FOG J
QUADRANT ELECTROMETER . ... page 202
GREENWICH OBSERVATORY (METEOROLOGICAL
SECTION) to face page 239
KEW OBSERVATORY (NATIONAL PHYSICAL
LABORATORY) .... „ 243
TEMPLE OF THE WINDS, ATHENS . ,,260
MEAN ANNUAL ISOTHERMS FOR THE BRITISH ISLES
(AFTER BUCHAN AND MILL) .... page 286
WEATHER SCIENCE CHAPTER I
INTRODUCTORY ANCIENT IDEAS — THE BAROMETER AS A " WEATHER
GLASS" DAWN OF MODERN METEOROLOGY CYCLONES AND
ANTICYCLONES THE EARTH I ITS SIZE, SHAPE, AND MOTIONS
THE ATMOSPHERE.
EVERY one thinks that he or she knows some- thing about the weather, and its vagaries, real or assumed, are an unfailing subject of conver- sation, affording ample opportunities for the indulgence in the Englishman's privilege of grumbling. Yet, notwithstanding the multi- plicity of weather saws and sayings, the enor- mous mass of statistical details collected by the industry of countless observers in our own and other countries and published in ponderous tomes, whose covers no one ever opens, by Govern- mental departments and learned societies, it must be confessed that the science of the weather is as yet in its infancy. Very little has been done in the way of utilising " the dry masses of figures which accumulate year after year, and lie abso- lutely idle in yearly volumes, unread and un- studied," but still the game of accumulation goes
13 A
14 Weather Science
on as merrily as ever. However, of recent years some attempts have been made to arrive at a better state of things ; our knowledge, not merely of the bottom of the "ocean of air" in which we live, but also of portions of its upper regions, has been increased by the judicious flying of kites with self - recording instruments of pre- cision, and much important information as to the distribution of temperature, moisture, etc., in those hitherto unknown heights has been obtained. The application of dynamical methods by Bezold and others, the development of theories of atmospheric circulation, the improvement of instruments and the establishment of regular meteorological observatories, not only in the lowlands but also in mountain regions, these and other things have all helped in throwing light upon some of the obscurer problems of the science. The complexity of the subject, how- ever, seems to give little hope that the ideal of deducing all weather changes from the knowledge of the variations in the amount and direction of the solar radiation, in their effect upon the mixture of air and water vapour of which our atmosphere consists, will be soon realised.
For a long time to come Meteorology must be regarded as a science of pure observation, and our knowledge of it confined to generalisa- tions obtained with more or less exactness as the result of these observations. In early days, Meteorology (from the Greek ra /uerewpa, things
Aristotle's Views 15
above), included not only the study of the atmosphere, its clouds and weather changes, but also such things as comets and shooting stars (or meteors, as they are distinctively called), which, being now known to be extra-terrestrial phenomena, are removed to the domain of Astronomy. Accordingly we find in Ptolemy's "Almagest," the greatest work of the ancient Greek astronomy, no mention of comets at all, the latter being regarded as mere temporary exhalations or vapours from the ground drawn upwards, and burning when they reached the " region of fire." Aristotle, perhaps the most universal genius who ever lived, whose works and ideas on almost every subject were held in the utmost reverence till about two centuries ago, considered that the atmosphere is divided into three regions. The first region is that in which animals and plants live, supposed to be immovable like the earth on which it rests ; the second is an intermediate region, intensely cold ; the third region, contiguous to the region of fire, or the heavens, partakes of the diurnal motion of the latter. Vapours arising from the earth ascend to this region and are heated, engendering igneous meteors and comets. A characteristic feature of these ideas is the slender basis of fact upon which they rest.
Where experiment or observations were few, or wanting altogether, the ancient writers seem to have found no difficulty in supplying the
1 6 Weather Science
deficiency by means of vivid imagination. A writer of authority made a definite statement, and his word passed for ages unimpeached, no attempt being made to ascertain whether the results of experiment agreed with the predic- tions of theory or not. In this way there have arisen fancies as to the connection between the " changes " of the moon and weather change, still widely believed in, though absolutely without foundation in fact ; others, after the spirit of the old astrologers, have brought in the planets, the infinitesimal variations in the sun's light and heat accompanying the greater or less spottedness of his surface, etc., as efficient causes. All such speculations, by attributing specific and different actions in different regions of the earth, bring their own refutation. Whatever influence, for instance, a sun spot may have upon terrestrial conditions, it is difficult to believe that it can cause extra heat in one region, deficient rain- fall in another, and specially fine weather in a third. So far we may be satisfied to look to nearer causes for such variations, and it must be confessed that the complexity of the subject is such that we can hope only for slow progress in our knowledge in this respect.
The science of weather predictions, and the results obtained, will be alluded to in the course of this work in further detail. So far, however, greater success has been obtained in average results than in furnishing information which is
Napoleon and Laplace 17
available for any specific short interval, such as the total rainfall on a given day, or the actual temperature at a given place.
An instance of the uselessness of average results as to mean temperature and other weather con- ditions in enabling predictions to be made for any specific future occasion is given by Mr Abercromby in his well - known book on the " Weather." Nearly a hundred years ago, Napoleon, on the eve of his invasion of Russia, requested Laplace to calculate when the cold set in severely over that country. The latter found that on the average it did not set in hard until January. " Napoleon made his plans accordingly, a sharp spell of cold came on in December, and the army was lost." In a similar manner we have found by long continued series of observations the mean height of the baro- meter for every day of the year and almost for every hour of the day at most inhabited places on the earth's surface, but this gives no informa- tion whatever as to what will be the actual height at any particular moment, in fact this is almost bound to be either higher or lower than the mean value.
The old weather prognostics, many of which we inherit from classical days, and some from yet earlier periods, have been handed down into all European languages, and are still of great value. Most of the saws and sayings relating to the approach of rain, perhaps the most
1 8 Weather Science
important point which concerns all, young and old, rich and poor alike, are well known, and their general accuracy has been often testified to, though it is only within recent years that some reasons for their verification as well as explanations of their occasional failure have been forthcoming.
The invention of the barometer nearly three centuries ago, and the consequent knowledge of the variations in atmospheric pressure, led to the discovery that, on the whole, the mercury fell for rain and windy wreather, and rose for fine, being generally low when the conditions were unfavourable, and high when fine settled weather prevailed. Hence arose the designa- tion of this instrument as a "weather glass," and the conventional but incorrect notation still to be met with even now, some makers alleging that the public will not buy barometers without these indications. Yet a very short series of observations will suffice to show that rain some- times falls with a high and rising barometer, and there are frequent occasions of good weather when the "glass" is low. Of more scientific value are the instructions given in the well- known Fitzroy or " fishery " form of barometer, these embody the results of careful observations.
In our latitudes the barometer usually falls with S.E., S., or S.W. winds, and rises with W., N.W., and N. ones. Thus most rain coming with southerly and south - westerly
Synoptic Charts 19
winds, owing to their having blown over the ocean before reaching us, this explains the ordinary statement, but since rain sometimes comes with a northerly wind also, the rise of the barometer often deceives those who expect fair weather from the latter circumstance.
The method of synoptic charts, giving the conditions of temperature and pressure, direction and force of the winds, etc., over large areas of the earth's surface (as illustrated by the daily diagrams published by the Meteorological offices of this and other countries which are reproduced in the newspapers), constituted a distinct advance in the science of weather predicting. Tele- graphic information from a number of stations is transferred to an outline map. Lines are drawn through all places having the same barometric height, these isobars? as they are called, being usually marked at intervals of Y1^ inch or 2 mm. (5 mm. = *2 inch very nearly) of the mercurial barometer, and from the nature and present arrangement of these lines important information as to impending changes in weather may be easily gathered.
The direction and force of the wind at various places is marked, as also the temperature, at a given time, lines drawn through places having the same temperature being known as isothermal lines, or simply "isotherms." A general rule
1 All places in the neighbourhood where the pressure is for the moment the same are on the same isobar (uros, equal ; fiapvs, heavy).
2O Weather Science
governing the relation between the direction of the wind and the position of the isobars was enunciated by Buys Ballot in 1860, and is accordingly known by his name. "If one stands with his back to the wind, the baro- meter will be lower on his left hand than on his right." Thus when the barometer is higher to the north than to the south, the wind will be east, when it is higher to the east than to the west the wind will be southerly, and so on. It is evident, however, that this knowledge can only be possessed by an observer who knows the conditions at other stations than his own. The movement of air being due to the difference of pressure at different regions, it follows that where the "gradient" is small, or the isobars are far apart, the wind will be in general slight ; when, on the other hand, the isobars are close together, there will be more wind, and a "serious storm may be expected when there is a difference of half an inch of pressure between two neighbouring stations." Speaking generally, it may be said that the force of the wind will not exceed that of a "strong" or "fresh" breeze, unless the gradient exceeds 0*02 inch for a distance of 15 geo- graphical miles, approximately equal to 2 mm. per degree of latitude (Abercromby).
Seven different forms of isobars are usually distinguished, of which two, the " cyclone " and the "anticyclone," are the most familiar to
Cyclones and Anticyclones 21
ordinary readers. The former is a region of low pressure, surrounded by more or less nearly circular isobars, hence the name (from the Greek KVK\O$, a circle). These, however, usually have the form of somewhat elongated ovals, and are by no means concentric. At or near the centre of the innermost oval is the point of lowest pressure, spoken of as the centre, or "eye," of the cyclone. Generally speaking, cyclones are in fairly rapid motion, in this country usually from S.W. towards N.E., and the wind rotates round the centre in the positive or " counter- clockwise" direction (that contrary to the direction in which the hands of a clock move), its intensity depending on the closeness of the isobars, being nearly proportional thereto. Occasionally cyclones are stationary, break up or move in an unexpected direction — hence one cause of the failure of predictions founded on the supposition that they will follow their usual course. The cyclones which cause the much dreaded tropical storms are of the same nature, and differ principally from those we experience in these latitudes, in the much greater varia- tions of pressure within a short distance and the consequent greater violence of the wind. In our country the velocity of a cyclone centre may vary from 10 to 70 miles per hour, and, as we have said, its motion is usually eastward, though a few move in the westerly direction, whilst others are stationary. The air to the front
22 Weather Science
is usually warmer than that in the rear, and thus gives rise to the well-known close, "muggy" sensation, that in the rear is characterised by a brisk, exhilarating feeling. The anticyclone, on the other hand, is a region of high pressure, surrounded by isobars usually much further apart than those of a cyclone, and consequently covering a larger area. Light winds circulating in the " clockwise " direction ; or calms, prevail, and sometimes for days or even weeks there is practically no motion of the system at all, until it breaks up or moves slowly on. The ordinary features of weather are a clear sky and dry air, varied sometimes by a few light showers, in summer a hot sun, in winter fog and frost. The other forms of isobars will be dealt with more in detail later on in the course of this work (chap. v.).
The earth, " our common mother," the stand- point from which we view the rest of the universe, is one of a number of more or less spherical bodies of varying sizes moving round a centre, known as the planets of the solar system. It is the third in order of distance of the planets moving round the central body, and its motions may be roughly divided into two principal ones, the diurnal rotation on its axis, causing the succession of day and night, and the annual motion round the sun, causing (in our latitudes) the phenomena of the seasons. In shape the earth is not strictly of
Size and Shape of Earth 23
the form of any regular geometrical figure, but approximates to an oblate spheroid of revolu- tion, i.e., a figure like a sphere (the distance of every point of whose surface from the centre is the same), but with a greatest and least diameter, round the latter of which it turns once in twenty-four (sidereal) hours. The least diameter (polar) is as nearly as possible 7,899 miles, the greatest (equatorial) diameter is 7,925 miles, so that the deviation from exact sphericity, or the polar flattening, as it is sometimes called, is not great. Neither the elevation of mountains, nor the depression of the sea bottom, nor even the polar flattening, bear any considerable pro- portion to the whole size of the earth, the highest mountain never rising more than 6 miles above the general surface, whilst the deepest seas have less than that distance from surface to bottom, and the difference between equatorial and polar diameters (25 miles) is but a small fraction of either, so that a model made accurately in wood would be undistinguishable by the eye from an exact sphere.
The diurnal rotation on its axis causes a point on the Equator (circumference 7,925 x 3-1416 miles = 25,000 miles nearly) to move over more than 1,000 miles per hour, points in higher latitudes having slower motions, till at the Poles there is no rotation at all.
In the latitude of London (51 J° N.) the speed of rotation is about 600 miles per hour, this
24 Weather Science
varying directly as the cosine of the latitude, being at its maximum on latitude 0° (equator) and zero at latitude 90° (poles, where cos 90° = 0). The rotation of the earth on its axis, besides causing the succession of day and night, rising and setting of the stars, etc.. has a most important meteorological effect in causing the phenomena of the "trade" and "anti-trade" winds, the deviation of air currents produced during their movement from places moving more quickly to others of slower rotation, and vice versa, which will be dealt with later on in the course of this work.
In addition to this motion of rotation the whole earth has a progressive motion round the sun, completing one circuit in the course of a year, moving in a path which is approximately circular (in reality slightly oval), and which lies in one plane (the ecliptic), which makes an angle of about 23|° with the perpendicular to the axis of rotation (the Equator) ; or in other words, the inclination of the Equator to the ecliptic is this angle, whose mean value on 1st January 1910 is given in the Nautical Almanac as 23° 27' 3'58", diminishing at the rate of 0*468" per annum. These planes inter- sect in two points, called technically the first point of Aries and the first point of Libra, the sun appearing from the earth to be in the former on 21st March, when his longitude is said to be zero, and in the latter on 23rd September
Annual Motion 25
(approximately, varying slightly through the calendar arrangement of leap year, and the slightly unequal apparent motion of the sun), or perhaps, since it is the earth that moves and not the sun, it would be more correct to say that the earth is in "Libra" on 21st March and in "Aries" on 23rd September.1
Since, after all, all motion is in reality relative, it is not only convenient, but, so far as we are concerned, correct, to use the current language. Thus between 21st March and 23rd September the sun's position is on one side (north) of the equinox, whilst for the rest of the year he is on the other side (south) of that plane. When in the celestial Equator, which we may regard as the indefinite prolongation of the plane of the terrestrial Equator to the heavens, the sun is everywhere above the horizon for twelve hours, and for an equal time of day below it, and these times are consequently called the equinoxes (Latin : equus, equal ; nox, night), day and night being equal. After this, from 21st March to 21st June, the sun's distance from the Equator northwards becomes greater and greater; the "days" in our latitude, or times
1 The signs Aries and Libra are now distinct from the constella- tions of those names, and are now approximately situated in the preceding constellations Pisces and Virgo. The " first point of Aries" is not very near any bright star, but not very far from the fourth magnitude star w Piscium ; the " first point of Libra " is between the double star p Virginis and the star 77 of that con- stellation, both of the third magnitude. Both these points, how- ever, are slowly changing in position owing to " Precession of the Equinoxes."
26 Weather Science
during which he is above the horizon, become longer and longer; the "nights," or periods when he is below, becoming correspondingly shorter. After this, reaching his greatest dis- tance north of the plane of the Equator ("first point of Cancer "), his distance from that plane gradually diminishes, till on 23rd September he is again in the celestial Equator. The days and nights are once more equal, the former having been gradually growing shorter and the latter longer from 21st June to 23rd September. After this time the sun's apparent motion carries him south of the Equator; the days are now shorter than the nights, the former continue de- creasing, the latter increasing, till the extreme values, both for the sun's angular distance (south) and the length of these periods, are attained on 21st December (shortest day in our hemisphere). The varied phenomena of the seasons and the different altitudes of the sun, relative length of day and night in different parts of the globe, are fully explained in most geographical and astronomical works, so that we need not go much into detail here. It will suffice to say that since the days are longer, and the altitude of the sun greater in summer than in winter, we have here the primary cause of the differ- ence of temperature between those seasons ; whilst in spring and autumn both the length of the day and the altitude of the sun have intermediate values, and, as is well known, these
Climates and Zones 27
seasons are colder than summer, and (in general) warmer than winter with us. Thus the first and most general cause of the difference in temperature at different times of the year in our country is roughly indicated, whilst the differ- ence between the temperature in our latitudes and that experienced in " hot countries " nearer the Equator is primarily due to the more nearly vertical position of the sun in the latter regions. As one writer puts it, the same amount of heat is spread over a larger area, and so each spot receives less.
At all places between latitudes 23^° N. and 23^° S., which zone, bisected by the terrestrial equator, is known as the Torrid Zone, the limiting latitudes, 23^° N. and 23^° S. being known as the Tropics of Cancer and Capricorn respectively, the sun is in the zenith or vertical at some time or other during the year, and its altitude, when highest, at noon is never less than 90° — 47° = 43°, whereas its greatest altitude above the horizon in the latitude of London (51£°), though 62° on 21st June, is less than this on every other day of the year, being only 15° on 21st December. Everywhere throughout the Torrid Zone the length of the time the sun is above the horizon never varies greatly from twelve hours, the night being of about the same length. On the Equator itself the days and nights are always (theoretically) of equal length (neglecting the effect of refraction
28 Weather Science
by the atmosphere, which is always to slightly extend the former at the expense of the latter, and of which we shall soon have to speak), but further north and south there is a slight variation.
At the Tropic of Cancer (23^°), for example, the sun is vertical on 21st June, when he is said technically to enter the sign of that name, and the length of the day has then its greatest value, thirteen and a half hours, whilst the night is only ten and a half hours long. Conversely, on 21st December the sun has his least altitude at noon, only 43°, and the day is only ten and a half hours long, the night being then thirteen and a half hours long. Between latitudes 23 -J-° N. and S. and latitudes 66^° N. and S. respec- tively, the sun's greatest altitude varies from 90° (when he is in the zenith), to nothing (when he is on the horizon), and the length of the day under- goes corresponding variation. Everywhere within these zones, the " Temperate Zones," he rises and sets at least once every day, though, theoreti- cally, at the summer solstice he barely touches the horizon at midnight, for latitude 66^°, whilst at the winter solstice he scarcely rises at all.
Between latitudes 66^° N. and S. and the respective Poles — the " Frigid Zones " — the sun has never an altitude greater than 47°, about half-way between the horizon and the point overhead, so that his rays always fall more or less obliquely; and whilst at one time of the
Phenomena at the Poles 29
year (during summer of each hemisphere respec- tively) he is visible near the horizon at the time that would be midnight in lower lati- tudes, and the phenomenon of the "Midnight Sun" is seen, at another time he does not rise at all, and perpetual night, lasting for a longer or shorter period (during the winter of each hemisphere), arises. At the Poles the sun is alternately visible and invisible for about six months at a time (but see " Refraction " by the atmosphere, infra), but his greatest altitude never exceeds 23^°, or its value at noon on 8th February in the latitude of London, so that his rays at no time fall more than very obliquely, though his long continuance above the horizon will help to raise the temperature above what might otherwise be expected. The phenomena at either Pole will be somewhat different from anywhere else on the earth, so that the measurement of the sun's apparent altitude will afford a certain test as to whether the observer is, or is not, in latitude 90°. If, now the sun's altitude be carefully measured when he is in one part of the sky, and again twelve hours later, the corrected readings, after allowance for refraction and instrumental errors, differ only by the amount of the sun's change of declina- tion during that interval, as given by the Nautical Almanac, then the observer may be confident that he is very close indeed to the position of the terrestrial Pole.
30 Weather Science
Though, as we have just stated, the primary cause of the different temperatures prevailing in different parts of the earth is to be looked for in the unequal amount of heat directly received from the sun, yet the relative distribution of land and water cause the actual values to be very different from what would otherwise be the case. Water having a greater specific heat than almost any other substance (that is, more heat is required to raise its temperature by a given amount than is the case for land surfaces, and more heat is given out by it in cooling through the same degrees of temperature than do land masses), the heat of summer is mitigated and the cold of winter moderated by the presence of large masses of water, for the lands adjoin- ing them — hence arises the difference between "insular" and "continental" climates. Water covering ^ths of the earth's surface, whilst of the ^ths surface uncovered by water the greater part lies in the Northern Hemisphere, we find very considerable differences in temperature between points in corresponding latitudes, which receive almost exactly the same amount of sun heat, directly. As an equally important cause may be mentioned the existence of warm and cold currents, such as the Gulf Stream flowing from the Central Atlantic towards Western Europe, a warm current conveying some of the heat received by the tropical regions to the latter, the " Humboldt current " of the Southern
The Atmosphere 31
Pacific and "Polar currents" helping to cool the lands towards which they flow.
But the earth is not merely a mass of land and water, otherwise neither man himself nor the science of meteorology could exist. Above and around; it, to an unknown height, extends the atmosphere or " ocean of air," a gaseous envelope. The main constituents of this atmos- phere are the gases oxygen, nitrogen, argon and carbon dioxide, with a variable amount of water vapour, and small traces of ammonia, nitric acid, etc. The density of this atmosphere is greatest at the earth's surface, and rapidly decreases as we go upward, but no certain limits can be assigned beyond which we can positively assert that there is no air at all, though at a height of above 100 miles from the surface the quantity of air must be very small indeed. Being a gas, or rather a mechanical mixture of a number of gases whose relative proportions vary very slightly at different times and places, the atmosphere possesses elasticity, exerts pressure, and is easily affected by changes of temperature. As a fluid never in equilibrium, it is in constant motion, and the consideration of its motions under various influences forms the main part of our study. Were the whole atmosphere of uniform density, equal to that which it has at the surface of the ground, its height would be only about 5 miles, and this height (20,000 feet) is sometimes called
32 Weather Science
the height of the " homogeneous atmosphere " ; this would produce a pressure equal to that actually existing at the surface of the earth, which pressure on every square inch is about equal to that produced by the weight of a column of mercury 30 inches high and 1 square inch in cross section. In terms of the C.G.S. system (in which the centimetre is the unit of length, the gramme the unit of mass, the second the unit of time) the air pressure on each square centimetre of the ground is about equal to that of a column of mercury 76 cm. high and 1 sq. cm. in cross section, upon an area of 1 sq. cm. at its base.
This pressure is conveniently known as atmos- phere of pressure ; in English measure this equals about 147 Ibs. per square inch. The pressure higher up is less, and decreases gradually as we ascend. It is commonly and conveniently measured by the barometer (fidpvs, heavy ; perpov, a measure), perhaps the most important of all meteorological instruments.
Though, as we have stated, the mean pressure is as above, yet its actual value is always vary- ing, and the study of these changes involves much of our science, sudden changes of pressure almost always accompanying changes of weather, whilst fairly settled or slowly changing conditions invarably indicate settled weather.
Owing to the presence of the atmosphere, we do not see external (heavenly) bodies exactly in
Refraction Effects 33
their true positions. Rays of light coming from the sun, moon, or a star have their directions changed by this action, which is called " refrac- tion," just as we observe the image of a straight stick immersed partly in water to be apparently bent at the surface, the part below being seen raised above its true position. In a similar manner the sun and stars are seen above the position they would otherwise occupy in the sky, by a variable amount depending on their altitude above the horizon, and (in a less degree) on the temperature and other conditions of the air.
There is an important difference, however, between refraction by the atmosphere and that produced by a fluid such as water. The latter being homogeneous, a ray of light entering it in any direction not perpendicular to its surface, is bent in a definite direction, making an angle with its original course (^ sine of angles of in- cidence and refraction in a constant ratio to one another"), and pursues its new course without further change so long as it remains in the water, whilst a ray of light entering our atmos- phere from outside undergoes a gradual and increasing deviation as it enters more and more dense air, whereby it is made to traverse a slightly curved path, and the celestial body is seen in the direction of the tangent to the point of the curve which reaches the eye of the observer. The density of the lower atmosphere
34 Weather Science
being greater than that of the upper air, the refraction is greater for the former. As a result of refraction the sun and moon are seen above the horizon at a time when they are really below it. The time of sunrise in our latitudes is accelerated, and that of sunset is retarded, and so the day is lengthened by from five to eight minutes. In more northern latitudes this difference is yet greater, whilst in the Arctic Regions the result is that the length of "per- petual day " is increased by several " days," and that of "perpetual night" is shortened by a corresponding amount.
When the sun and moon are near the horizon, we notice that they are distorted from a circular form into ovals, and are usually of a reddish colour. The former effect is due to the fact that the amount of refraction changes rapidly near the horizon, being greater below than above. Thus the lower edges of these bodies are raised more than the upper ones, so that the vertical diameter is diminished. The horizontal diameter is not affected, and so the sun and moon appear to be oval and not circular. Their reddish colour is due to absorption. White light is composed of all the " colours of the rainbow " — red, yellow, green, blue, violet, etc., — and of these, the green, blue, and violet (the shorter "waves") are more readily absorbed by the atmosphere, or rather its vapours (for pure dry air exercises scarcely any absorption), than the red and
Twilight 35
yellowish rays (which are longer). Thus this process of selective absorption results in the loss of a greater amount of the blue and violet rays, and so the sun and moon, when seen low down near the horizon, appear reddish or yellowish rather than white, an effect seen also when the sun shines through a " November fog." Twilight is also a phenomenon due to the presence of our atmosphere. For a short time after the sun sets (and before it rises) some of its light falling in the upper regions clouds, etc., is reflected downwards, and thus gives illumination to the lower atmosphere and ground. It is not certain whether this reflect- ing power is due to the gases of the atmosphere or to the presence of dust particles, water vapour, and possible ice crystals existing in it. So long as the sun is not more than 18° below the horizon of any place, some light will thus be received by reflection from the upper atmos- phere, and thus the length of the day will be appreciably increased. It thus happens that from the latter part of May till the middle of July, though the sun is below our horizon at London for more than seven hours out of the twenty-four, there is no real night during that period, though at midnight the amount of light received from the sun is at times very small. The red tints of sunset are due to a similar cause to that from which arises the red colour of the rising and setting sun.
36 Weather Science
The generally diffused light of the sky in the daytime is also due to the presence of the atmosphere. If there were no atmosphere the stars would be visible by day as well as by night. The scattering and reflection of light by the air, its vapours and dust particles, give rise to a general illumination, strong enough to render the more feeble illumination of the stars invisible, except under special circumstances. The blue colour of the sky is also due to the scattering of light by small particles in the air, the red, as we have just seen, being transmitted ; though much of the other radiation forming white light is absorbed, a portion appears to be reflected. It is by no means certain what substances are most effective in this. Lord Rayleigh thinks that fine salt particles floating in the air, or even the oxygen itself may be the cause. The presence of metallic meteoric dust has also been suggested as an important factor in producing this colora- tion. Thus, apart from its essential character as a supporter of life and every kind of com- bustion, the other services rendered by our atmosphere are many and important. To its presence is due much of the varied play and coloration of the inorganic world, the brilliancy of the clouds, and the wondrous tints of sunrise and sunset ; the blue of the sky, so different from the inky blackness that would prevail were it absent, the gradual coming on and fading away of illumination in the morning and evening, instead
Atmospheric Absorption 37
of the instantaneous appearance of day and night respectively, are all results of its presence.
As a storehouse of the solar heat after it has reached the earth, the atmosphere is a most important meteorological agent. The sun's rays pass fairly freely through the air, only a small portion being absorbed by the vapour of water and traces of carbon dioxide (oxygen and nitrogen, the main constituents being almost perfectly diathermanous or " transparent " to radiant heat), and reach the surface of the ground or the waters. Most of these rays are then reflected, but in the act of reflection they are changed somewhat in character, and are then largely absorbed by the water vapour and the clouds instead of going back into outer space again. Part of these absorbed and reflected rays are again reflected downwards by the clouds, and so in large measure the heat is kept in, and remains as a permanent gain to the earth. On a cloudy night the air never becomes so cold as on a clear one, when there is less vapour in the atmosphere. In the latter case the heat reflected from the ground not being stopped passes more or less completely out, and as none comes in from outside, the temperature rapidly falls. Theories of dew and hoar frost formation depend upon an action of this kind, as we shall see in a later chapter.
Until quite recently all observations were confined to the denser layer of atmosphere in immediate proximity to the ground, and nothing
38 Weather Science
was known as to the conditions prevailing in the upper regions. At a few mountain observatories some information had been obtained, occasional balloon ascents were made, and inferences drawn from the behaviour of the upper clouds, which revealed the presence of currents whose velocity much exceeded anything known to occur near the surface of the ground. Of late years, however, by means of kites sent up with self-recording instruments, much important information as to these hitherto unknown regions has been obtained, some account of which will appear in a later chapter of this work.
It has been already stated that there is no definite knowledge as to how far the atmos- phere extends upwards. The density, however, decreases very rapidly, and there can be very little air at a greater height than about 50 miles. Observations of the duration of twilight indicate about this limit, but meteorites visible only to us by their heating to incandescence by friction against a resisting medium, have been seen at a height of 100 to 150 miles, so that a very small amount of air must exist at a greater height than is shown by the twilight observa- tions. The aurora, a phenomenon probably due to electric discharges in the rarefied upper air, is not often known to exist at a greater height than about 40 to 50 miles. Though the extension of the atmosphere upwards is to a small degree un- certain, yet from a knowledge of its specific
Weight of the Atmosphere 39
gravity and pressure at the surface, its total weight may be inferred.
One cubic foot of dry air, measured at the temperature of freezing water (32° F.) and under the barometric pressure equal to 30 inches of mercury, weighs 1*3 ozs. or 565 grains, more exactly, the exact experiments of Jolly giving for the weight of 1 litre of air at 0° C. (82° F.), and under the pressure of 760 mm. (of mercury) values varying between 1*30493 and 1*30575 grammes. The total weight of the atmosphere of the earth resting on its surface was given by Sir John Herschel as llf millions of pounds or 5*3 trillions of kilogrammes (5'3x 1021 grammes). The composition of this air is remarkably con- stant. Though a mechanical mixture, and not a chemical compound, the proportions of its ingredients vary very slightly, wherever the air may be collected. Of the four principal con- stituents, oxygen, nitrogen, argon, and water vapour, only the last varies to any appreciable degree. By volume, the oxygen forms 20*9 per cent., nitrogen 76*0 per cent., argon 1 per cent., whilst the water vapour varies from 1 per cent, to 4 per cent., when the air is very damp. The carbon dioxide, whose amount in pure country air is less than 0*03 per cent., sometimes may exceed 0*07 or 0*08 per cent, in the air of a town, whilst in badly ventilated crowded halls it may reach to 0*20 per cent., or even more. Though these different gases differ considerably
4O Weather Science
in density, water vapour being the lightest, and carbon dioxide the densest ; owing to diffusion, the proportion of these ingredients does not sensibly vary at different altitudes. The heavy carbon dioxide is carried upwards, and the lighter water vapour downwards by the action of the same principle. Nevertheless, there is a tendency for the lighter gases to spread more rapidly up- wards, and the heavier to remain below, so that it has been supposed that the absence of hydrogen, the lightest of all gases, from our atmosphere in a free state, is due to the upward diffusion whereby it has " got beyond the power of recall by gravitation." But all such speculations de- pending as they do upon the uncertain indica- tions of the " kinetic theory " of gases, are to be deprecated, and we should rather wait for further experimental knowledge than attempt to supply its place by premature hypotheses. Within the last few years several new gases previously un- known have been discovered (argon, xenon, crypton, etc.), and hydrogen has been detected in the Jumerolles or jets of steam of Tuscany, and also in the human breath under certain conditions (Thorpe).
The following quotation from the works of the late Dr Buist of Bombay may fitly be inserted in this place: —
" The atmosphere is a spherical shell which surrounds our planet to a depth which is unknown to us, by reason of its growing tenuity, as it is
Buist on the Atmosphere 41
released from the pressure of its own superin- cumbent mass. Its upper surface cannot be nearer to us than 50, and can scarcely be more remote than 500 miles. It surrounds us on all sides, yet we see it not ; it presses on us with a load of 15 Ibs. on every square inch of surface of our bodies, or from 70 to 100 tons on us in all, yet we do not so much as feel its weight. Softer than the softest down — more impalpable than the finest gossamer — it leaves the cobweb undisturbed, and scarcely stirs the lightest flower that feeds on the dew it supplies ; yet it bears the fleets of nations on its wings around the world,1 and crushes the most refractory substances with its weight. When in motion its force is sufficient to level the most stately forests and stable buildings with the earth — to raise the waters of the ocean into ridges like mountains, and dash the strongest ships to pieces like toys. It warms and cools by turns the earth and the living creatures that inhabit it. It draws up vapours from the sea and land, retains them dissolved in itself or suspended in cisterns of clouds, and throws them down again as rain or dew when they are required. It bends the rays of the sun from their path to give us the twilight of evening and of dawn ; it disperses and refracts their various tints to beautify the approach and the retreat of the orb of day. But for the atmosphere sunshine would burst on us and fail us at once, and at once remove us from
1 This was written before the days when the atmosphere is polluted by the reckless waste of the stored up accumulation of ages ; the consumption of thousands of tons of coal, to satisfy the cravings of a few would-be ' ' record breakers/' to save a few hours of their worthless time, regardless of all other circumstances.
42 Weather Science
midnight darkness to the blaze of noon. We should have no twilight to soften and beautify the landscape, no clouds to shade us from the scorching heat, but the bald earth, as it revolved on its axis, would turn its tanned arid weakened front to the full and unmitigated rays of the lord of day. It affords the gas which vivifies and warms our frames, and receives into itself that which has been polluted by use, and is thrown off as noxious. It feeds the flame of life exactly as it does that of the fire — it is in both cases consumed, and affords the food of consumption ; in both cases it becomes combined with charcoal, which requires it for combustion, and is removed by it when this is over."
The temperature of the atmosphere is perhaps of more immediate interest to our physical well- being than almost any other phenomenon, for, as it has been well put by an esteemed friend of the writer, "the struggle for existence is essentially, after all said and done, a struggle with cold." Warmth is more than food to the body, and in fact the need for the latter is the want or otherwise of the former, food being the fuel of the living " steam-engine at work." Many eminent meteorologists, too, consider the determination of temperature as the most im- portant of all observations in the science, for it is to the unequal heating of different parts of the atmosphere, the sea, and the land, mainly by the sun's radiation, that almost all circulation and other movements are primarily due. From
Tidal Action, Sunspots, etc. 43
these arise the equatorial warm currents, both serial and aqueous, the polar colder streams which take their places, the differences of pressure in different regions (modified somewhat by the earth's rotation), the trade winds and the anti-trades, the various regions of high and low pressure (anticyclones and cyclones), etc. Of minor import, if indeed the action is certainly detectable, is such tidal effect as is produced by the unequal attraction of the moon and sun upon different parts of the atmosphere. Changes in the number and intensity of sun spots, and other signs of solar activity have relations more or less direct with the phenomena of terrestrial magnetism, the Aurora (Borealis and Australis), magnetic storms, etc., and have been suspected by some to have a yet more intimate connec- tion with weather conditions. Upon this point prejudice has hitherto perhaps come more into play than sober deductions from ascertained facts, one party rejecting as preposterous the notion that there can be any such connection between the greater or less development of sun spots and local weather conditions, the other party as unhesitatingly accepting it. Attempts have even been made to show that such things as the recurrence of famines, floods, excessive or deficient harvests, commercial "booms" and panics, are all intimately related to the sun- spot period.
Such evidence as we have so far is of such a
44 Weather Science
nature that altogether contradictory results have been arrived at by different investigators, but some short account of these will be dealt with in a later chapter of this work (chap. xv.). In the "Dark Ages" the immediate influence of not only the sun and moon, but of all the heavenly bodies, upon terrestrial phenomena, was generally believed in. The phenomenon of dew was considered a product of the stars ; one planet " produced " rain, another gave fine weather, and so on. Even within comparatively recent times the " wild fancy " of Bishop Whiston attributed the Noachian deluge to the near approach of a "watery comet," which later on is to destroy the world by fire! This was offered to the world in Newton's day !
Till little more than two centuries ago comets and "shooting stars," still called meteors, were considered as terrestrial phenomena, and were accordingly not considered by astronomers, but as they are now known to be extra-terrestrial objects, they no longer fall within the scope of our science, and will not be further referred to here. A science of celestial meteorology, or the probable weather conditions existing in other worlds than ours, can scarcely be said to exist, though speculations have been often indulged in as to the habitability of the nearer planets. The phenomena of belts on the major planets Jupiter and Saturn have been likened to the trade -wind zones of our own earth,
Objects of the Science 45
etc. With all such matters, notwithstanding its etymology (TO, fjieTeupa, things above), meteorology does not deal, leaving to astronomy whatever may be known or imagined as to the physical and climatic conditions of the other bodies of the universe ; confining itself to the study of the physical state of our own atmosphere, its dis- tribution and temperature ; changes in these elements from day to day and throughout the year, their bearing on climate and habitability ; and, lastly, the possibility of inferring future changes from present conditions.
CHAPTER II
THE BAROMETER — MERCURIAL BAROMETER — GLYCERINE BAROMETER — FORTIN'S BAROMETER — ADJUSTMENTS — SYPHON BAROMETER —
FISHERY BAROMETER ANEROID BAROMETER SELF-RECORDING
BAROMETERS CORRECTIONS TO READING OF BAROMETER —
WEATHER VARIATIONS WITH BAROMETER OSCILLATIONS OF
PRESSURE, DIURNAL AND ANNUAL EXTREME RANGE — DOVfi's
RULES — REGIONS OF HIGH AND LOW PRESSURE — PRESSURE AT VARIOUS ALTITUDES ABOVE THE SURFACE.
THERE is probably no instrument to which the progress of modern meteorology is due so much as the barometer, the measurer of the weight, or rather pressure, of the atmosphere. The name is derived from the two Greek words, /3apv$, heavy, and MeVpoi/, a measure. It is an often-told tale that some Florentine workmen, finding that they were unable to raise the water in a pump to a greater height than 32 feet above its level in the well, came to Galileo to enquire the cause. The rise of water in a pump having been attributed to nature's abhorrence of a vacuum, he remarked upon this that it was evident that this abhorrence did not extend beyond 32 feet ! Torricelli, one of his favourite pupils, some time after Galileo's death, con- sidered that this water column was supported by
46
The First Barometer 47
the pressure of the external air upon its surface, this forcing the fluid up to a height varying with the pressure and the density of the matter raised. A heavier liquid than water he thought would rise to a less height. Taking a glass tube closed at one end, and filling it with mercury, whose density is thirteen and a half times as great as water at the same temperature, he inverted this, holding it upright, over a larger vessel partly filled with mercury also. The liquid in the tube fell somewhat, part of it pass- ing into the vessel, and on measuring the height of the top of the liquid above the surface of the mercury in the vessel, he found this to be about 30 inches. Thus was formed the first mercurial barometer. The space above the top of the mercurial column, between this and the top of the tube, was empty, and is known as the Torricellian vacuum. It is a nearly perfect vacuum, containing only a trace of mercury vapour. The height of the mercurial column was found to vary somewhat from time to time at the same place, and very considerably, when the apparatus was moved from place to place. Thus Pascal, the famous self-taught geometer, afterwards not less celebrated as a theologian, found that on the top of the Puy-de-D6me, one of the extinct volcanoes in the Auvergne region of France, the height of the mercurial column above the level of the cistern was only about 25 inches as compared with 30 inches in the
48 Weather Science
town of Clermont. Thus it was shown that the smaller amount of air over the instrument on the mountain supported a shorter column than in the valley below. The earliest form of mercurial barometer differs but little from those at present in use ; the Accademia del Cimento employed a nearly closed cistern containing mercury into which dipped an upright glass tube graduated along its length and supported by the neck of the cistern. The most ordinary fluid for barometers is mercury, on account of its being much denser than any other known liquid, and giving off but little vapour at ordinary temperatures ; but other liquids have been also used.
There is a glycerine barometer at the Geo- logical Museum in Jermyn Street, London, and water barometers have also been occasionally employed, but though the great length of these instruments enables minute variations in atmos- pheric pressure to be more easily seen, yet their very length is itself a disadvantage, and the amount of vapour given off thereby producing a depression of the barometric column vary- ing with temperature and other conditions, necessitates the use of troublesome corrections before the true pressure can be obtained. In addition, there is the very considerable effect of capillary "attraction." On the other hand, mercury is a liquid which may be easily obtained reasonably pure, and its rate of expansion for
A Simple Barometer 49
different temperatures is well known and very regular within the ordinary range employed. Of the various forms in use we propose to describe only one or two of the most ordinary. The simplest kind of barometer, as already mentioned, consists of a glass tube supported in an upright position over a cistern nearly full of mercury, its lower end dipping into the latter. The barometer tube is filled nearly full with mercury, and is then heated till the fluid boils. Thus any air or moisture contained in the tube is expelled. It is thus completely filled with mercury, and inverted in the cistern. The atmospheric pressure at any moment is measured by the height of the top of the mercury column above the surface of the liquid in the cistern. A scale whose zero point is at the level of the latter surface, divided into inches and fractions of an inch (or into millimetres), affords the means of doing this. As, however, the mercury in the tube is supported by the pressure of the atmosphere, and the latter varies continuously, the length of the column must also vary accord- ingly ; consequently the level of the mercury in the cistern must also change, for when the fluid rises in the tube its level must fall in the cistern, and vice versa ; thus the position of the surface in the latter must change, falling when the height in the tube increases, and rising when mercury flows from the tube into the cistern. Thus, to allow for this change of level, an arrangement,
Weather Science
such as that adopted for the well-known Fortin form, must be adopted. This instrument is a cistern barometer, containing mercury as the fluid, with a scale whose lower end is a fixed ivory pencil. Before read- ing the instrument, in most cases the point will either be below the surface of the mercury in the cistern or else slightly above, but by means of a screw at the bottom of the cistern the latter may be raised or lowered till the point of the ivory pencil is in contact with the mercury surface, which is shown by the apparent contact of the point and its image reflected from the liquid metal.
The reading of the scale at the top of the upper surface will then give the apparent barometric height. Another adjustment is necessary, however, before this reading can be ascertained. It will be noticed that the surface of the top of the mercury column in the tube is not flat but curved, slightly convex upwards, being
standard Barometer, highest in the centre. The read-
(Fortin's pattern.) ing of the scale ^ ^ highest
point must be taken, the eye must be as nearly
The Scale and Vernier 51
as possible on the same level, and most instru- ments are provided with a vernier, whereby the reading may be made with a greater degree of accuracy. In barometers in ordinary use in this country the scale is divided into inches, lOths and 20ths of an inch. The vernier is a small movable scale divided into a number of equal divisions, such that a fixed number of these are equal in length to one more or one less than a number of divisions on the fixed scale. Very frequently, as we have stated, the scale of the barometer is divided into 20ths of an inch, and twenty-four of these divisions correspond to twenty-five spaces on the vernier, whereby •£$ * •£$ = -gfo inch may be read off, or y^^j inch by estimation. The bottom of the vernier being brought exactly on a level with the convex top of the mercury column, the reading is made. The line on the scale next below this point is noted, and this gives the nearest ^ inch ('05), the smaller fractions being taken from the vernier. Thus suppose that the line on the main scale next below the bottom of the vernier is 29 '550 inches, and that the next line above the 3 on the vernier coincides with one on the main scale, when the instrument is set each long line on the scale corresponds to •1 inch, and the intermediate short lines '05 inch, every long numbered line on the vernier gives the -01 (lOOths), and the short lines '002 (2,000ths of an inch.) Then the reading is 29 -550 + -03 +
52 Weather Science
•002 = 29,582 inches. The attached thermometer is to give the temperature at the time of the reading, for in reducing the reading of the barometer the correction for temperature is a very important one. This thermometer should be read before the barometer. Of other forms the best known are the "syphon" barometers, the " fishery " barometer, and " aneroid " barometers.
The syphon barometer, as its name indicates, has a tube bent up into a U-shaped form, a short open end about 6 to 8 inches long, and a longer closed end, the whole supported in a vertical position, the two ends upwards. There is no need, therefore, for a cistern in this form of the instrument, for when mercury is introduced into the open end it runs down and gradually rises in the other limb, the atmospheric pressure on the open end supporting a column in the closed part, whose height varies according to the variations in the former.
The distance between the levels of the mercury in the open and closed ends respectively gives the height of the barometer, and scales are pro- vided to enable the necessary measurements to be made. If properly made with a correct scale and uniform tube, no correction for capillarity will be required for this instrument, the only corrections being those for temperature and altitude above (or depression below) mean sea- level. This form of instrument, however, is not so much in use in this country as on the
Fishery and Aneroid Barometers 53
Continent. A thermometer is mounted to the same framework to give the temperature of the instrument at the time of reading.
The fishery barometer was designed by the late Admiral Fitzroy, some time Director of the Meteorological Office. It is a mercurial barometer fastened to a box-wood frame, the mercury column dipping into a box-wood cistern which has a flexible sheep - skin base and is provided with a lifting screw. There are two verniers, reading to hundredths of an inch, one being placed on each side of the tube, and a large attached thermometer. Many of these instruments are erected at exposed positions on the coasts, at coastguard stations for the " use of fishermen, sea-faring persons, and the public generally" (whence the name). The lettering on the instrument is the result of considerable experience, and is more accurate and reliable than that on the ordinary "weather glass." These instructions, from the name of their designer, are known as Admiral Fitzroy's rules.
A common and portable form of barometer is the well-known aneroid, usually made in the form of a watch or chronometer. In this instru- ment the pressure of the atmosphere is measured by means of its effect in altering the shape of a metallic box from whose interior the air has been partly removed, the upper surface of the box being corrugated to make it yield more easily to external pressure. At the centre of
54
Weather Science
the top of the box there is a pillar connected with a powerful spring, to keep the box from collapsing. The top of the box rises or falls with the variations of pressure, and these move- ments are transferred by means of levers and
Aneroid.
springs to a hand which moves on a dial like a clock face. The instrument has to be graduated by comparison with a mercurial barometer ; it is very quick in indicating changes of pressure, and may be made small enough to go in the waist- coat pocket, and is thus very convenient for mountain observations, etc.
It has the drawback of being affected by
Wheel Barometer 55
changes of temperature to an uncertain degree, each instrument acting differently in this respect, and, moreover, it is liable to gradual changes which necessitate its frequent comparison with a standard mercurial barometer, so that for exact scientific work there are serious objections to its use.
The wheel barometer consists of a mercurial syphon barometer, whose two branches have usually the same diameter. On the surface of the mercury in the open branch there floats a piece of metal or glass suspended by a thread, to the other end of which is fixed a pulley on which the thread is partly rolled. Another thread, rolled parallel to the first, supports a weight which balances the float. A needle moving on a dial is fixed on the axis of the pulley, and the float moving with the rise or fall of the mercury, the pulley turns and the needle with it, thus recording the variations of pressure. The instrument has the disadvantage resulting from the friction arising from this additional apparatus, and is consequently slow in indicating changes.
Self - recording mercurial and aneroid baro- meters or " barographs " are in use at large fixed public observatories, and some of the cheaper forms are used also by private observers.
They usually need comparison with standard mercurial barometers, and when this is regularly done their records, as giving the continuous
56 Weather Science
variations of pressure for every moment, are of great value in detecting minute irregularities and variations of short period, which would otherwise escape notice. The photographic method in use at Kew and Greenwich consists in recording photographically the varying position of the top of a mercurial barometer. Sensitized paper
Barograph.
is wrapped round a cylinder, which is turned with uniform motion. Usually one complete revolution of the cylinder is performed in a day. Light from a lamp behind the barometer falls upon the paper, but is intercepted by the mercurial column to a greater or less degree, the line of division between the part acted upon by the light and the unaffected part varies according to the varying height of the column, every part of the cylinder, except that on which
Corrections to the Barometer Readings 57
the spot of light falls, being covered with a case of blackened metal. Thus there is traced upon the paper a boundary curve whose ordinates (heights) are proportional to the movement of the barometer, and on developing the image the trace becomes visible to the eye.
The readings of every barometer, to bring the indications of different instruments into harmony, need various corrections before they can be applied for comparison, or other purposes. They are usually given as five (or six) in number.
I. Correction for Index Error. (II. „ „ Capacity).
III. „ „ Capillarity.
IV. „ „ Temperature.
V. „ „ Altitude above Sea-level.
VI. „ to reduce to mean latitude 45°.
Of these the first three corrections vary for different instruments, and are often previously obtained by comparison with standard instru- ments (II., does not apply to Fortin's baro- meter), whilst the corrections IV. and V. are conveniently applied by means of tables, since they are the same for all instruments under similar conditions.
The Correction for Index Error is the amount to be applied to the readings owing to the fact that the scale (inches or millimetres, etc.) is not quite accurate, so that a given reading, say 29 inches on the standard barometer, corresponds
58 Weather Science
to a somewhat lower or higher reading on the instrument under comparison.
It is stated that some instruments have been found to read half an inch too high, others as much too low (Scott). Others again, which are correct in one part of the scale, are found to be several tenths of an inch wrong in other parts. The usual practice at Kew is to test them at every half inch from 27*5 to 31 inches (of mercury), and the corrections thus found are registered in a table to be used with that particular instrument. Any barometer whose index error is greater than 0-010 inch is to be rejected (Kew, 1875).
The Correction for Capacity. — In barometers with closed cistern there is a certain height for the mercury at which the column is correctly measured by the scale. Below this, since when the mercury sinks in the tube it rises in the cistern, the height read off must be too great ; and, conversely, the level in the cistern falling as the mercury rises in the tube, readings will be too low when this is the case. When the ratio between the capacity of the tube and the cistern is known we may allow for this as follows : Suppose the capacity of the tube to that of the cistern to be Tfoj, and we know the height at which the reading of the scale is correct, this is called the " neutral point." Then by adding a T^jth part of the difference between the height read off and that of the neutral point, we get the
Capillarity and Temperature 59
true reading when the column is higher whilst we subtract the difference when the column is lower. It will be remembered that Fortin's barometer having a movable cistern, the zero of the scale is always at the level of its surface, and so no such correction is wanted for instruments of that pattern.
The Correction for Capillarity arises from the fact that there is what is called a capillary repulsion between mercury and glass. The surface of the former in a glass vessel is never horizontal, but is lower at the edges than at the centre, and, moreover, this depression is greater in narrow tubes than in broad ones, being nearly inversely proportional to the diameter of these latter. It has also been found to be greater in tubes in which the mercury has not been boiled than in those where it has been so treated.
But the more important of the corrections to be applied are those for temperature and altitude, to be next described.
The Correction jor Temperature, or Reduction to Freezing Point (32° F. or 0° C.).— The attached thermometer gives the temperature at the time of observation. Since all bodies are affected by change of temperature, usually expanding with its increase and contracting with decrease, the length both of the mercury column and of the scale itself will vary with variations in temperature, apart from the rise or fall of the mercury due to pressure changes. Mercury
60 Weather Science
being a liquid, is more expansible by heat than either glass or the scale, and consequently the level of the barometer will rise with any rise in temperature and fall with a fall of temperature, even though the air pressure remain constant. All readings of the instrument are accordingly reduced to the temperature of the freezing point (32° F.), and tables are given which enable the correction to be applied to the observed length at other temperatures to reduce them to the value they would have at that standard temperature, the temperature of the instrument at the time of observation being given, of course, by the attached thermometer.
Correction jor Reduction to Sea-level. — The height of the barometric column, being a measure of the pressure of the atmosphere, is greater or less, according as the latter varies, and so it is evident that if we ascend above the level of the surface of the ground, there being less air as we go up, the pressure will be less, and consequently the barometer will fall, whilst, conversely, if we go down into a mine, the air pressure will be greater than that at the surface. For moderate elevations above the surface the barometer falls about ^ of an inch for each 100 feet of altitude, but this amount varies not only in different places, but also with variations of temperature. We may determine the height of a mountain or other elevated station by readings taken at the sea-level and the station
Reduction to Mean Latitude 61
respectively, and the barometric determination of heights depends upon this comparison of readings. Tables of more or less exactness are given to enable the correction to sea-level to be made, and for the British Isles the mean sea -level at Liverpool (Ordnance Survey) or "high water at London Bridge," is taken as the zero of altitude.
The last "correction," Reduction to Mean Gravity at Latitude 45°. — To reduce the read- ings of the barometer to a standard of com- parison available all over the globe, it has been proposed to take as the standard the value of g, the acceleration due to gravity, at latitude 45°, its value varying slightly at different latitudes from 32-086 in foot second units (or 977*99 C.G.S. units at the Equator to 983*21 at the Pole) at the Equator to 32*258 at the Pole, and multiply the barometric readings by the proper factor, ratio of gravity at the station to that at latitude 45°. For latitude 50° this is nearly equivalent to a correction (addition) of +*014 inches to the barometer at 30 inches, and about +'013 inches (barometer 29 inches). This correction, however, is not often applied, and, of course, is only for mercurial barometers and not for aneroids. In practice the corrections for index error, capacity (if required), and capillarity are contained in one, the "Kew correction," when the instrument has been tested at that institution, and after this is made, the corrections for temperature and
D
62 Weather Science
altitude are next applied. As an example of the way in which these corrections are made, we may take the following, given by Mr Scott :—
f Attached thermometer 68° F.
Uncorrected reading . . 29 '946 in. i Altitude of cistern, 105 feet
^ above sea-level.
Kew correction . . . -HH* '
29-960
(Deduct) Temperature cor- 1 .-,«/. rectionfor68° \ " 106
29-854 = (Reading at 32° F.). Altitude of 105 feet at tern- ") perature 50° and approxi- > + -116 mate pressure 30 inches J — - (Readi corrected and re-
29'970 i duced to 32' F. and M. S. L. ).
Note. — In applying the correction for altitude we must know the air temperature at the time, for the difference of pressure between the sea-level and the place of observation is for the height of the vertical column between the two positions, and the weight of this column varies with its temperature, being less for hotter air than for colder.
The old and familiar name of " weather glass " for this instrument has already been alluded to as well as the notation engraved on the scales of many barometers.
This usually appears in accordance with the idea of high barometer for fine weather and low barometer for bad, something as follows : —
31 inches = very dry. 30-5 „ = set fair. 30 = fair.
29 inches = rain.
28 !5 „ =much rain,
28 „ = stormy.
29 '5 „ a= change.
Apart from the fact that these terms are of
Diurnal Variations of Barometer 63
very limited application, being only roughly true for places at mean sea-level and under normal conditions, it often happens that rain falls with a high and rising barometer, and we frequently get fine weather when the latter is low, so that this notation is practically worthless. The fishery or Fitzroy barometers, instead of the above mis- leading and imperfect notation, have instructions engraved on their scales applicable to the usual course of the weather in these islands.
In a general way the barometer in these latitudes varies most irregularly, but in tropical regions, apart from occasional storms, there is a regular daily rise and fall. It generally rises from a minimum at 4 A.M. to 10 A.M., when it attains its first maximum, then falling again to 4 P.M. second minimum, rising again to a second maximum at 10 P.M. So regular is this oscilla- tion that it is sometimes stated that the hour of day can be ascertained by the height of the baro- meter. Any irregularity in this daily oscilla- tion is an unfailing sign of a storm. The morning maximum and the afternoon minimum (10 A.M. and 4 P.M.) differ more from the mean than the other oscillations. At Calcutta the diurnal range, or difference between the extreme values, is about 0*12 inch. This range is very much smaller at stations in higher latitudes, but has been detected everywhere where the alternation of day and night exists (Herschel), and is evidently due (partly) to the action of
64 Weather Science
the sun, "though the causes are not yet thoroughly worked out." There is a prob- ability, however, that the moon, too, has an action here, and that, like the ocean tides, both sun and moon are instrumental in producing an "atmospheric tide." On this point letters by Mr Dines and Mr Langdon in Symons' Meteorological Magazine, April 1910, may be consulted, under the heading of "Atmospheric Tides."
"It is fairly certain that the natural period of oscillation of the atmosphere as a whole is about twelve hours. Thus the tide-producing power of the sun must inevitably produce some such tide as we see in the double daily baro- metric oscillation." — DINES.
It appears, from examination of various series of long - continued observations, that there is also an annual period as well as a daily one, but this variation is small for our islands, though for continental regions it is sufficiently well marked. The characteristic high barometer over central Siberia during the winter, as com- pared with the much lower readings during the summer in that country, are well known. The same thing is generally true to a less degree of the mean pressure over North America during the same seasons.
The extreme range at sea-level in our latitudes may be roughly taken as about 3 inches, from 28*0 to 31*0 inches of mercury ; its mean height
High and Low Readings 65
being rather under 30 inches, " one atmosphere " being conveniently expressed as the pressure equal to this value. More exactly, the value of the standard pressure is defined as such that it will support a column of mercury 76 cm. high, at latitude 45° and at the sea-level, the temperature of the mercury being 0° C. (32° F.) ( Watson's " Physics "). For our latitude (51° N.) the standard pressure usually adopted is 29*905 inches of mercury, under which pressure the boiling-point of water is 212° on the Fahrenheit scale, which is slightly smaller than the value 76 cm. for the latitude of Paris, corresponding to 29-922 inches.
It is not often that readings of the barometer under 28 -5 inches or over 30 -5 are recorded at stations on the sea-level. Occasionally, during storms, readings under 28 inches are taken, as at Ochtertyre in Scotland, during the great storm of 26th January 1884, when a reading of 27*332 inches was recorded, and throughout the day readings under 28 inches were taken in the north of Ireland and Scotland. As examples of high barometric readings, I have myself on several occasions observed the barometer standing slightly over 31 '0 inches at Markree, at the end of January 1902. Much higher readings have been observed during the long continuance of the winter anticyclone over Siberia, 31'62 inches having been recorded at Barnaoul in December 1877.
66 Weather Science
Professor Dove's practical rules on the relation of barometric changes and wind direction may be thus condensed (for Europe) :—
The barometer falls with east, south-east and south-west winds. It rises with west, north-west and north winds. It ceases to fall with a south- west wind and begins to rise ; with a north-east wind it ceases to rise and begins to fall. These rules were originally given for the whole Northern Hemisphere, but it has been shown that for Asia and the extreme northern regions of America and Greenland, they are not correct (Scott). Similar rules for the Southern Hemisphere by interchanging of north and south, were pro- pounded by the same authority.
Dr Schreiber of Leipzig has, however, shown that these relations are not necessary, that wind direction and barometric pressure are probably more complex functions of the general weather conditions. The warmest currents are usually the dampest, and the relative humidity and probability of rain are roughly inversely as barometric pressure, but have also a minor rela- tion to wind direction. Thus we have the usual relation between a higher rising barometer and fine weather and falling barometer with rain and wind ; the fall being greater for high winds than for heavy rains. Most generally a rise of the barometer is accompanied by a fall of the thermometer and vice versa, but whenever a simultaneous rise of both instruments occurs it
"High" or "Low" Barometer 67
may be taken as "a sure sign of steady fine weather." The contrast between the weather conditions implied by the slow or rapid move- ments of the barometer is well expressed in the following couplet: —
" Long foretold, long last ; Short notice, soon past."
A rapid rise or alternate rising and falling indicates unsettled weather conditions, whilst a gradual rise foretells fine weather, and a gradual fall during fine weather is a promise of a spell of continued bad weather. With regard to the amount of rise or fall of the barometer in a given time it may safely be said that a fall of ^ inch in an hour, or f3^ inch within three or four hours, is a sure indication of a coming storm (Chambers).
Professor Mohn in his " Meteorologie " sum- marises the general reasons for a high or low baro- meter (apart from storm and violent changes) as follows. The barometer is high when the air is very cold, the lower strata being more dense than when it is warm, and more contracted, thus causing the upper strata to sink down and bring a greater mass of air, thus increasing the pressure at the base (recorded by the barometer). When the air is dry it is denser than if partly composed of moisture (the vapour of water having only •62 the density of air at the same temperature and pressure), and thus the barometer is high.
68 Weather Science
Whenever an upper current sets in towards a given area, compressing the lower strata, this also causes a high barometric pressure. On the other hand, the barometer is low when the lower strata are heated, the isobaric surfaces (surfaces of equal pressure) are elevated, and a barometric slope causing motion results, so that the mass of air pressing on the lower regions is reduced. When the air is damp its density is reduced, and the more so the greater the amount of this moisture, since, as we have already stated, water vapour at the same temperature and pressure has only f ths the density of dry air. Whenever the air has an upward tendency, its pressure is lessened, so the barometer reads lower.
The presence of deep snow is another favour- able condition for a high barometer, for this tends to prevent changes (rises) of temperature by in- terposing a non - conducting layer between the air and the ground.
We accordingly find that the regions of highest barometrical readings are situated in the interior of continents and in high latitudes at the time when the temperature is lowest, as, for instance, in Eastern Siberia during the winter, where a constant anticyclone prevails for many months. The much greater proportion of water to land in the Southern than in the Northern Hemisphere, whilst it tends on the whole to raise slightly the mean temperature of places in southern latitudes over those of corresponding regions north of the
Difference of Hemispheres 69
Equator, similarly tends to lower the pressure over the former. Between latitude 40° S. and the South Pole there is little land beyond a few scattered islands, with the exception of the supposed Antarctic continent, which in any case is probably much smaller than the land round the North Pole. Thus the atmosphere over a large portion of the Southern Hemisphere is more heated than at equal latitudes of the northern half of the globe, so that the " isobaric surfaces " are raised and the differences of pressure between the Equator and higher latitudes diminished. This difference of pressure being the force tending to produce a flow from the Equator towards the polar regions, a smaller mass will ascend, and consequently the pressure (due to the weight of the superincumbent atmos- phere) will be correspondingly less. The presence of any considerable mass of land near the South Pole must act as a " centre of indraught " for the upper currents from the warmer water surfaces around it, thus tending to still further reduce the pressure. To this defect of pressure is also attributed the exceptional force and frequency of the " brave westerly winds " which blow so persistently over the Southern Ocean.
[Note.— The variations of barometric pressure at different heights above the sea-level have been determined, and a table prepared by Dr Sprung gives the annual mean pressures, their mean values for January and July, and also for altitudes of 2,000 and 4,000 metres (6,700 feet and 13,000 feet), above the surface for 5° intensity of latitude from 80° N. to 70° S. From this table we find a minimum annual mean of 758 -2 mm. at latitude 65° N., increasing thence to
yo Weather Science
a maximum 762*4 mm. at 35° N., again decreasing to 757*9 mm. at latitude 10° N. From this to southern latitudes the mean annual pressure increases to 763 '5 mm. at latitude 30° S., falling off again at higher latitudes (738-0 at latitude 70° S., the most extreme southern latitude given). At an altitude of 2,000 metres we have for latitude 80° N. 582 '0 mm., increasing at first somewhat more rapidly to 600-9 mm. (latitude 30° N.), then more slowly, till in latitude 20°S. the maximum 602'7 is reached, after which there is a fall, at first slow, then more rapid, to 569*9 in latitude 70° S. The pressures at altitude 4,000 metres increase from 445*2 (latitude 80° N.) to 471*1 (latitude 10° to 20° S.), and then decrease to 437 '2 at latitude 70° S. Thus for the Northern Hemisphere from this table we get a maximum pressure at latitude 30° N., a slight minimum at the Equator, and a more pronounced minimum at latitude 65° N., with a slight increase northwards, onwards towards the Pole. At elevations 2,000 metres the maximum pressure is nearer the Equator, and the same is the case with the observations for 4,000 metres altitude, there being only one minimum for each towards the polar regions, no increase of pressure in the highest latitudes being evidently such as the observations on the surface indicate. Sprung's numbers are given on the metric units ; they may be reduced to British statute feet, and inches of pressure by remembering that 39 -37 inches = 1 metre.]
Invention of the Thermometer 71
CHAPTER III
THE THERMOMETER '. SCALES, FAHRENHEIT, CENTIGRADE, REAUMUR, DE I/ISLE THE HYPSOMETER Cf WET AND DRY BULB " HYGRO- METERS— DE SAUSSURE'S HAIR HYGROMETER — MAXIMUM AND
MINIMUM THERMOMETERS SELF-RECORDING THERMOMETERS
(THERMOGRAPHS) — THE STEVENSON SCREEN — GLAISHER SCREEN — BLACK BULB THERMOMETERS RANGES IN TEMPERATURE.
THE thermometer seems to have first made its appearance about the end of the sixteenth century, and its invention has often been attributed to Galileo, though there is reason to believe that some kind of instrument for measuring temperatures was in use before his time. Galileo, however, used at Padua a glass bulb with a narrow open tube attached to it. This bulb was heated, and the open end of the tube placed under water in a vessel. The water rose to a certain extent in the tube, and as the air in the bulb cooled it rose still further, falling again when the bulb was warmed, exactly as the modern differential thermometer. About 1630 a French physician suggested inverting the Galilean thermometer, nearly fill- ing it with a liquid, and observing the rise and fall of the liquid with changes of temperature.
72 Weather Science
The "Florentine thermometer," invented by pupils of Galileo, consisted of a bulb with a fine tube above, and the fluid employed was alcohol ; the tube divided arbitrarily by means of small glass particles attached to its side, the zero being at about the point to which the liquid fell in a freezing mixture of salt and water. The substitution of mercury for alcohol, first suggested by Halley, has such obvious advantages that the former is now always used for exact observations, except for such temperatures as are occasionally met with in very cold countries, where mercury freezes in the winter. Mercury is a substance which can be obtained with considerable purity, and is liquid throughout the ordinary range of meteoro- logical observations. In addition it has a low specific heat and great conductivity, so that it is quickly affected by changes of temperature, and soon indicates these. The first use of the freezing- and boiling-points of water as fixed temperatures is variously attributed to Newton and Fahrenheit, the latter of whom also divided the interval between these temperatures into 180 parts or degrees. Finding that the temperature of a mixture of snow and salt was 32 of these degrees below the freezing-point of water, he called this latter the zero temperature (perhaps supposing no lower temperature to be possible), and the freezing-point being thus 32, the boiling-point became 212. This scale is in
Scales in General Use 73
almost universal use in this country, though originally "made in Germany," and has the advantage that in expressing observed tempera- tures it is very rarely necessary to use the negative sign (-) for temperatures below zero, whilst the small size of its degrees is also useful.
The scale, however, which is in general use throughout Europe is that known as the Centigrade. In this, the interval between the freezing- and boiling-points of water is divided into 100 degrees, the freezing-point being 0° (or zero) and the boiling-point 100°. It is commonly stated that the invention of this scale was due to Celsius, but it appears that the credit should be given to Linnaeus (Scott, " Meteorology," p. 19). The former called the boiling-point 0°, and the freezing-point 100°, the degrees running downwards, whilst the suggestion of Linnaeus that the freezing-point should be called 0°, and the boiling-point 100° agrees with the present universal practice. A third scale, proposed by Reaumur, is still sometimes used in Germany and Russia. On this the freezing-point is 0°, but the boiling-point is 80°, the degrees being thus larger than on the centigrade scale. A fourth system, proposed by Dr De L'Isle in 1733, divides the interval between the freezing- and boiling-points of water into 150 degrees (but this is nowhere in use at present so far as we are aware), reckoning backwards from the
74 Weather Science
boiling - point. Rules are sometimes given whereby temperatures on one scale may be con- verted into corresponding readings on any or all of the others, but being a matter of pure arithmetic and of very little general utility, we need not waste space to do that which any reader can do for himself. It is to be hoped that in a few years' time all save one scale will be mere matters of antiquarian interest. At present, however, it is necessary for the English reader to acquaint himself with the Fahrenheit and Centigrade scales, the former because it is in general use in this country, the latter because almost all continental writers use it, and it is even gaining ground here also, being generally adopted by chemists and physicists, if not by meteorologists.
Under ordinary conditions we may look upon the freezing-point as a constant temperature, since it is only slightly affected by variations in atmospheric pressure, being only lowered 0-0075° C. = 0-0135° F. by one "atmosphere" increase of pressure, and for the ordinary varia- tions the depression or elevation is smaller than the probable error of the thermometer. On the other hand, the boiling-point of water is greatly affected by variations in the intensity of the superincumbent atmospheric pressure, and a difference of 1 inch above or below the standard (usually taken at 30 inches, or more exactly 29*905 inches of mercury) raises or
The Hypsometer — The Hygrometer 75
lowers the boiling-point by 17° F. (on the metric system an increase of 27 mm. in pressure corresponds to 1° C. rise in boiling-point, and conversely). On this variation in the boiling- point depends the hypsometer, an instrument for the measurement of height (Greek ^09, height ; /merpov, a measure) : which is nothing but a boiling-point thermometer with a large scale whereby the temperature at which water boils on a mountain may be easily ascertained by performing the experiment. Then from the tables the barometric pressure corresponding to the ascertained boiling-point is known, and, with less exactness, the height of the mountain may be inferred. As, however, we know that the barometric pressure at any point on the surface varies within a range of 2 or 3 inches, the results of hypsometrical observation are liable to considerable uncertainty.
The simplest form of hygrometer, or instru- ment for measuring the amount of vapour present in the atmosphere at any given time, consists of the " wet and dry bulb " arrange- ment. This is merely two similar thermometers mounted side by side on a frame (often inside a Stevenson's or other screen), the bulb of one being wrapped in muslin or other soft material, whilst the other thermometer has its bulb uncovered. The muslin is tied round the bulb by means of a few threads of worsted or cotton, the other ends of which dip into
76 Weather Science
a small vessel of water, which is kept fairly full. The two thermometers being compared, it will be found that the " wet bulb " instrument will in general read somewhat lower than the
other. So long as the air is not saturated with moisture, some of the water carried up by the threads to the muslin will be continually evap- orating, and this evapora- tion causing an absorp- tion of " latent heat," the temperature of the instru- ment will be loVered thereby. When the air is saturated at any given temperature no more moisture can evaporate, and so the wet bulb and dry bulb thermometers will give the same read- ing. Thus we may see that the difference be-
Wet and Dry Bulb Hygrometer.
a measure of the amount of vapour present in the air at any given time. The theory of this instrument, sometimes called the " psychro- meter " (V^X/00'^ cold ; perpov, measure), has been investigated by Apjohn, August, and others, and tables have been formed by Glaisher, as the
The "Hair Hygrometer" 77
result of experiments carried on at Greenwich and elsewhere, whereby from the observed read- ings of the dry and wet bulb thermometers the amount and pressure of the aqueous vapour of the atmosphere may be obtained, approximately at least for our latitudes.
Regnault's and De Saussure's hygrometers are but little used in this country, though the latter instrument was recommended by the Vienna Congress of Meteorologists. This, the " Hair Hygrometer," consists essentially of a human hair, which is fixed at one end and stretched by a small weight at the other, the cord of connection passing over a block to which is attached a pointer, which moves over a graduated arc. The hair stretches as it grows damp, and contracts as it dries, the pointer at the same time moving either forwards or back- wards, its position varying with the length of the hair. Other hygroscopic substances, such as seaweed, which grows damp in wet weather and dries when the weather is fine, wild oats, whose life-like movements under the influence of moisture are well known, catgut, etc., are sometimes employed, but rather as "hygroscopes" (or indicators of moisture) than hygrometers (measurers}.
Whilst the reading of the thermometer gives the temperature at the moment of observation, it is for many purposes desirable to know not only what the temperature is now, but what has
E
78 Weather Science
been the highest and the lowest temperature during the interval since the last observation, the greatest heat of day and the lowest tempera- ture at night, etc. There are many varieties of instruments whereby this information may be obtained, but of course only a few of the more commonly used forms can be here mentioned. Instruments for recording the highest tempera- ture during a given interval (e.g., the course of a day) are called maximum thermometers, those giving the lowest temperature are consequently known as minimum thermometers.
Of maximum thermometers perhaps the most generally used form is the Phillips', invented by the well-known geologist, though that of Negretti and Zambra is as good and quite as often recom- mended. This is a mercurial thermometer having a tube whose bore is narrowed close to the bulb. The mercury on expanding rises past this obstruc- tion, but when the temperature falls again, the liquid cannot flow back again with ease. Thus if the thermometer be set on any day, say in the morning, and examined after the temperature has risen and fallen again, the column of mercury in the tube above the constriction will represent the amount of liquid forced past it when the temperature was highest, and the end furthest from the bend or constriction will indicate the maximum temperature. Phillips' thermometer has a small air bubble separating a detached portion of mercury from the main column.
Maximum and Minimum
79
When the latter expands, it pushes the bubble and short column before it, but when the tem- perature falls, though the main column contracts and leaves the air bubble behind, the detached column also contracting, but to a minute degree,
Maximum Thermometer.
its end furthest from the bulb indicates the highest temperature since the last reading. To set the instrument it is inclined gently, bulb downwards, when the air bubble becomes contracted to its smallest dimensions, the detached column com- pressing it, and coming close to the main portion.
The minimum thermometer in most common use is that known as Rutherford's. In this instrument the liquid is coloured spirit (usually
Minimum Thermometer.
alcohol) having a metal index. This index is allowed to run to the end of the column by
8o Weather Science
inclining the thermometer with its bulb slightly upwards. At the same time any detached portion of the spirit is brought back to the main column by gently warming, or sometimes this may be done by swinging the instrument to and fro, bulb downwards. When this is done, the index being set at the top of the column, the thermometer is left in a horizontal position. If now the temperature rises, the spirit will flow past the index without moving it, but if the temperature falls, the index will be drawn back with the contracting fluid, for the spirit will not easily admit of its surface being broken by the index getting outside, " capillary attraction " keeping the index with its upper end just below the top of the liquid column. Thus the top of the index will mark the lowest temperature, since it recedes with a fall, but remains stationary with a rise of temperature. Care must be taken that the index is within the spirit, and that no detached column remains in the upper end of the tube.
Though a spirit thermometer can record lower temperatures than a mercurial one (in this country of course so low a temperature as the freezing-point of mercury, —40° is never naturally reached), yet its expansion and contraction are less regular than is the case with the liquid metal, errors are frequent from the presence of air in the tube, from evaporation of part of the liquid column, etc., so that it needs frequent
Self-recording Thermometer 81
comparison with a standard mercurial thermom- eter, and even then its readings are often not very reliable.
Self-recording thermometers are in use in various large observatories. A common method is to obtain a continuous automatic record by means of photography, in a manner similar to that adopted for the barometer. A sheet of sensitised paper is mounted on a vertical cylinder placed behind an upright thermometer, and is shielded from light by a blackened metallic cover, with the exception of a narrow strip just behind the position of the mercury. A beam of light from a lamp is thrown upon the apparatus, being stopped by the mercury in the tube; so that it only falls on the part of the paper above the liquid column. The cylinder revolving slowly (usually once in twenty-four hours, or sometimes a shorter interval), if the temperature remain the same, the boundary between the unaffected and blackened parts of the paper will be horizontal ; if the temperature change, which it will do con- tinuously, there will be seen a wavy line of separation, and thus the position of this will be a measure of the temperature at each moment.
The photographic thermograph affords another method of obtaining a continuous record. Here there is a bubble of air on the top of the mercurial column which moves up and down with the temperature. A lamp is placed in front of the instrument and a photograph of
82 Weather Science
the position of the air bubble is taken on sensitised paper, which is rolled on a drum revolving uniformly.
For meteorological purposes various forms of screens, of which Glaisher's and Stevenson's are the most generally used, are adopted by different observers.
Whilst, on the one hand, it is necessary to obtain free exposure to the air, on the other hand, for instruments intended to give the air temperature, care must be taken to guard against direct radiation from the sun or neighbouring hot bodies, whereby the temperature of the thermom- eter may be unduly raised, whilst we must also see that it may not be lowered by radiation to colder bodies. As far as possible, the ordinary wet and dry bulb, as well as the maximum and minimum thermometers, must be surrounded by freely circulating air and be not too near other objects.
The Stevenson screen is made of wood, " double louvred," the louvres sloping in opposite direc- tions, so that the air may freely circulate, but direct radiation, rain and snow are unable to enter. It is erected on four legs, each about 4 feet high, and should stand on open ground or over grass, and not be within 10 feet of any wall or other building if possible, nor near trees. Its dimensions are usually about 2 feet (length), 14 inches (breadth), and 18 inches (height). Inside it are placed the dry and
Stevenson Screen
wet bulb instruments, as well as the maximum and minimum thermometers, the two former being hung vertically side by side, the two latter suspended horizontally, one over the other. Below and slightly to one side of the wet bulb instru- ment is placed the vessel of water (cup or small glass), into which the ends of the strands tied round the muslin sur- rounding the ther- mometer bulb, dip. In the Glaisher pattern there is a vertical board on which the thermo- meters hang, their bulbs being freely exposed to the air. This is attached to a sloping screen supported by a tripod. As the direction of the sun's rays changes during the day, the whole structure requires shifting once or twice daily, whilst in time of rain with northerly wind the instruments cannot be kept dry, as on that side they are merely sheltered on top and on one
Stevenson Screen.
84
Weather Science
Solar Radiation
Black Bulb Thermometer.
side. Thus it is in some respects less convenient than the Stevenson form of screen, whose double louvres prevent the entrance of any rain or snow. Special forms of instru- ment, such as the black bulb thermo- meter, which is a maximum thermo- meter of a particular kind, are used for measuring the intensity of radiation, its readings always being much higher than those of the other instruments. To the late Sir John Herschel, the well-known astronomer, is due the suggestion of its construction. It consists of a mercurial maximum thermometer, whose bulb and lower part of the stem are coated with lamp black. This apparatus is enclosed in a larger glass tube, one end of which is blown out into a large bulb, and the thermo- meter bulb rests in the centre of this. This cover has been previously ex- hausted of air as far as possible, by means of an air pump or otherwise, and is a partial vacuum. The whole instrument,
Grass
Minimum
Thermometer. J
Black Bulb Thermometer 85
cover and all, is hung horizontally on a wooden structure at about the same height above the ground as that at which the other thermom- eters are placed (i.e., about 4 feet) in the open air, and at a distance from walls of buildings, trees, or other obstructions. The bulb is exposed to the full rays of the sun, and protrudes usually some slight distance beyond the sup- ports ; it is usually directed south or south- eastwards.
As a 'rough rule for measuring the maximum radiation from the sun, it is suggested to take the difference between the maximum reading given by this instrument and the maximum as given by the ordinary maximum thermom- eter in the screen, but it is not very easy to attach a meaning to the number thus obtained, which will, moreover, differ with different ther- mometers, according to differences in " the coat- ing of the bulb, in the glass of the outer jacket, or finally, in the perfection of the vacuum" (Scott). The reading of the black bulb ther- mometer giving, perhaps, the highest tempera- ture in the sun at any moment during the day without reference to its duration, can afford but an imperfect idea of the amount of solar radia- tion, so that we may regard this instrument rather as a curiosity than as of scientific value. Its high readings, sometimes even in winter, may excite our admiration, but, just as the phenomen- ally low temperatures occasionally recorded by
86 Weather Science
grass minimum spirit thermometers, they are practically meaningless.
The daily range of temperature is least in winter, usually not exceeding about 5° or so, and greatest in summer, when it amounts (in this country) to about 15° or 16°, its value for spring and autumn being intermediate between those for summer and winter. Though the sun's altitude and the total amount of sunlight re- ceived are greatest on 21st June and least on 21st December, the hottest and coldest times of the year follow about one month after these dates. In accordance with the old proverb, " As the days begin to lengthen, so the cold begins to strengthen," it is when the days are getting sensibly longer, about the end of January or the beginning of February, that we experience the coldest weather in this country. Similarly it is in August that the highest temperatures are experienced, though the day is considerably shorter than during June. The coldest time of day is usually just before sunrise, and the hottest time about 2 P.M. The general and fairly evident explanation of all these phenomena is essentially the same. During the day heat is being received in increasing amount from sunrise to noon, after which a smaller quantity comes till sunset. All the while heat is being radiated away from the earth, but at a slightly less rapid rate, and thus during the morning the temperature increases
Variations in Temperature 87
even after the amount of heat received begins to diminish. Thus the hottest time of the day is after noon. During the night no heat is received from the sun, and so the temperature continually falls, radiation going on all the while till sunrise. In a similar manner the amount of heat received each day from the sun during spring continually increases up to 21st June, but the amount radiated away, though greater as time goes on, does not quite keep pace with the increased amount received. Thus the temperature continues to increase even after the maximum daily amount of heat has been re- ceived, until at about the end of July the amount radiated equals that received daily, and the temperature is a maximum. After this, the daily amount received continuing to diminish, the radiation by night exceeds the daily receipt, and so the temperature falls.
The mean daily temperature in this country will not differ much from the mean of the temperatures at 8 A.M. and 8 P.M., or 9 A.M. and 9 P.M., or if three observations are taken, the mean of the 9 A.M., 3 P.M., and 9 P.M. temperatures will give a very close approxima- tion to the truth.
The mean of the maximum and minimum thermometers will not usually give quite so accurate a result, the maximum day tempera- ture being generally slightly more above the mean than the minimum is below, so that the
88
Weather Science
mean of the readings of the two instruments will be somewhat higher than the true mean temperature.
In instruments even of the best class occasional changes of zero, sometimes as much as 1° F., take place, necessitating examination of thermometers, and though of recent years great improvement in the construction of standard thermometers has been attained, it must be confessed that few instruments can be trusted within 0*2° or so, and this accuracy is perhaps more than is to be usually expected.
The Hygrodeik.
Measurement of Rainfall 89
CHAPTER IV
RAIN GAUGES — SUNSHINE RECORDERS WIND INSTRUMENTS
ANEMOMETERS — DISTRIBUTION OF RAINFALL.
THE measurement of the rainfall of any district is a matter in which perhaps more interest is taken than almost any other branch of meteorological science, and accordingly we find a very large number of stations throughout the British Islands where a rain gauge is installed and regular observations are taken daily, and reported to a central authority. The British Rainfall Organisation, founded by the late G. J. Symons, and now under the directorship of Dr H. R. Mill, publishes an annual volume, British Rainfall, giving the results of observa- tions from more than three thousand five hundred stations in our islands, illustrated by maps and diagrams, with articles upon various branches of rainfall work, whilst Symons1 Meteorological Magazine, a monthly magazine under the same editorship, contains records of rainfall and temperature at a number of stations each month, articles on meteorological topics, remarkable storms, a climatological table for the British Empire, and occasional contributions
Weather Science
on other kindred subjects, as well as letters on remarkable meteorological phenomena, or points of interest to meteorologists generally.
There are many forms of rain gauge in use, from a simple upper funnel dipping into a glass bottle, to the most elaborate automatic self- recording gauge, or "pluviometer," as it is called. The commonest form, and one giving quite satisfactory results, is merely an improvement of Howard's rain gauge, so named from the eminent meteorologist whose careful study of cloud phenomena led him to a classification and nomencla- ture for these, now uni- versally adopted. A copper funnel, whose upper diameter is 5 inches, dips at its lower end into a glass vessel — a wine bottle or jar with a neck will serve very well for this purpose. Sometimes the funnel and bottle are enclosed within a cylindrical copper vessel, on whose bottom the bottle receiving the rain rests, and to whose top the rim of the funnel is fitted. At the top, and forming part of the funnel, is a "vertical rim" about 6 inches in depth, and the inner surface of this rim fits over the metal cylinder. The object of the rim is to prevent
Rain Gauge and Measuring Glass.
Rain Gauge 91
the "plashing" of raindrops and to catch any snow that may fall. Care should be taken that the mouth of the funnel is not dented, otherwise the full amount of rain will not be collected. It has been found that the diameter of 5 inches is the most convenient one for ordinary use, and though at some stations a diameter of 8 inches is used, and a consequent greater amount of rainfall may be collected, the results of experiments on gauges "from 3 inches to 24 inches in diameter show that the difference in indications is very small, hardly exceeding one per cent" (Scott). The gauge is ordinarily examined every day at 9 A.M., and the rain fallen in it during the previous twenty - four hours (if any) is poured into a measuring glass and measured, the amount being entered in the register as having fallen on the previous day. The measurement of snow or hail is usually effected by thawing this by means of the addition of a measured quantity of warm water, and subtracting the amount of this latter from the total resulting volume of liquid. Of course the snow may be melted by thawing it in a warm room, but this process is a slow one. It is sometimes said that by measuring the depth of the snow in a place where it has not drifted, and reckoning ^th the depth in inches, i.e., one foot of snow for an inch of rain, we get the amount of rain that would correspond to the given snowfall, but apart from the fact that the snow is not likely
92 Weather Science
to be of uniform density there is an uncertainty arising from evaporation, so that the result will only be a rough approximation.
The height of the rim above the ground for a " ground gauge " should be about 1 foot ; the latter should be perfectly level, and to keep it so, as well as to prevent the whole apparatus being blown over by the wind the lower part of the cylinder should be embedded for several inches in the ground, leaving little more than the funnel and rim above the surface. A frequently used contrivance to diminish the effect of the wind as far as possible consists of a shield of wire-netting encircling the mouth of the gauge, this was proposed by Nipher in 1878 ; the effects of evaporation are reduced to a minimum by making the aperture of the funnel which dips into the receiver as small as possible, and placing the latter in a vessel so that it is shielded from the sunlight (as in the outer copper vessel). It was found that rain gauges placed at an elevation above the ground catch considerably less rain than those near its surface, an effect now known to be due to the action of the wind, as it is stated that a Nipher shielded gauge will catch about the same amount on the roof of a house or other building as on the ground (Waldo).
Of self-recording instruments for the measure- ment of rainfall a simple form is now on sale by various makers, and a short description of
Sunshine Recorder
93
one is given in our chapter on Observatories (chap. xii.).
The use of the sunshine recorder has been greatly extended of recent years, and now most seaside and watering places possess one or more of these instruments, and publish the results of these records as an attraction to possible future visitors. The most ordinary form of this instru-
Sunshine Recorder.
ment is that known as the Campbell Stokes recorder. This consists of a spherical lens of glass whose principal focus lies at a short distance outside its surface. It is mounted on a pedestal, and attached to this is a curved metal frame- work with three grooves, at a distance from the lens equal to its focal length, the whole resting on a slab whose surface must be horizontal ; and this slab should be cemented on to the surface of a wall of other position where it will be freely
94 Weather Science
exposed to the sunlight, and kept in a fixed position. The sun's rays passing through the lens are focussed at some point or other on the surface of the curved metal framework, according to the time of day and the season of the year, and a sheet of cardboard is fitted into one of the grooves. As the sun's position changes the position of its image changes also, and thus burns a trace upon the card, the latter being unburnt when the sun is covered by cloud. Owing to the sun's varying declination (and consequent altitude) at different times of the year, its image will fall higher or lower upon the framework, and three sets of cards of different sizes and shapes are provided, the smaller card for winter fitting into the upper groove, the intermediate size for spring and autumn fitting into the middle groove, and the longer fitting into the lowest groove, for summer. The instru- ment must be so set that the central position must be just opposite to that of the sun at noon, and its image falls upon each card at the point marked noon on the graduated surface of the latter. When the instrument is in adjust- ment, on a clear, sunshiny day, there will be formed a broad horizontal track from sunrise to sunset, the position of this track changing from day to day during the course of the year, as stated above, but the middle point of each day's trace will always coincide with the time of apparent noon, On a cloudy day the trace
The Anemometer
95
will be fainter and often broken. It has been suggested to expose a photographic plate towards the north at night-time, and thus obtain a trace of the pole star on clear nights ; the record, of course, being broken or absent in cloudy weather, thus supplementing the sunshine recorder for the night hours.
Instruments for wind measurements may be divided into two classes, those which measure
Anemometer with Wind Vane.
its pressure upon any surface, and those which record its velocity. The direction is indicated by a vane or weathercock, whose construction
g6 Weather Science
and manipulation needs little description, though even in this simple instrument errors occasion- ally arise from the confusion between true and magnetic bearings, the difference in this country amounting from 15° to 20°.
One of the most commonly used forms of " anemometers " for measuring the velocity of the wind is that invented by the late Dr Robinson
Robinson's Anemometer.
of the Armagh Observatory, and known by his name. It consists essentially of four hemi- spherical cups which are fixed at the ends of two horizontal rods forming a cross. These are fastened to a vertical axis, the latter passing through the point of intersection of the rods. The wind causes these rods or arms to revolve, and they are so constructed that a definite rate of revolution corresponds to a definite wind
Different Forms of Anemometer 97
velocity. The motion is registered by a wheel- work arrangement at the base of the apparatus, being communicated to a system of dials which record the number of revolutions made, whence the speed of the wind may be inferred. There is sometimes an arrangement for marking the numbers of miles of wind on a paper strip fastened to a drum moving regularly by clock- work.
As an instrument for recording pressure, Lind's anemometer may be mentioned. This consists essentially of a glass syphon, one end closed, the other end bent and open to the wind. The limbs of the syphon are vertical, the open end is horizontal. Water is poured in up to a given height, standing at the same level in both legs when the air is calm. When the wind blows, the level will be more or less depressed in the one, and rises correspondingly in the other to a greater or less degree according to its force.
Dines' pressure anemometer is a more exact instrument, but depends on a similar principle.
Osier's and Gator's anemometers are more complicated instruments. In these the pressure against a plate of given area is measured, whilst a resistance is afforded by springs or by weights.
Wild's pressure gauge consists of a rectangular plate, hung on hinges to a horizontal axis. The angle made by this with the vertical measures the force of the wind. Its indications, though sufficiently accurate for light winds, do not show
98 Weather Science
the differences between those of strong force (Scott).
The distribution of rainfall throughout the world is a subject of considerable importance from the economic point of view as well as the meteorological one, for the habitability of any district is largely dependent upon this factor. Even in our own islands the local variations are very considerable, there being a mean annual rainfall of over 160 inches in the lake district of Cumberland, whilst in Lincolnshire the annual precipitation does not usually much exceed 20 inches. Similarly in India, Cherrapunji, on the Khasia hills, has an annual rainfall of 500 inches, whilst in the Deccan there are frequent droughts, and the total amount is far below the average. As a very rough approximation it is sometimes stated that the annual rainfall at any place within the Torrid Zone is about 100 inches, the Temperate Zone has 30 inches, and the Frigid Zone 15 inches. Nevertheless, though it is probably true that as a general rule more rain falls over the Torrid Zone than else- where, there are also to be found over the same zone the driest regions of the globe, the Sahara and Arabian deserts, and the great desert of Gobi or Shamo. In South Africa the driest region is probably the Kalahari desert, whilst in America there are the " nitrate " regions of Peru and Northern Chili, and the great Salt Lake district of the United States. The rainfall of
Agencies affecting Rainfall 99
Australia is in general small, and throughout the greater part of the interior of the island continent there falls considerably less than 10 inches per annum.
There are three principal agencies which are most efficacious in bringing about a fall of rain: — (1) mixture of masses of air at different temperatures, (2) ascending currents of air, (3) contact of warm air with the cooler surface of the ground. Of these three agencies, probably the second — ascending currents of air, which by expanding fall in temperature — is the most commonly acting, at least within the tropics. It has been calculated that the temperature of a mass of dry air will fall about 1° C. for every 100 metres of ascent (Hann). If, however, the air, as is always the case, contains more or less moisture, the rate of cooling will be consider- ably less, since condensation, converting some of this vapour into the liquid form, sets free " latent heat," which tends to warm the sur- rounding air. Nevertheless, the fall of tempera- ture produced by a rise from the surface of the ground to the top of a mountain is very con- siderable. The heated warm air over the surface of the ocean always containing much vapour in suspension, rising to a greater or less height, sooner or later loses most of its moisture, which falls in the form of rain. Near the Equator, generally speaking, there is an almost continual precipitation in the form of heavy
too Weather Science
showers, and the fall is greatest when the weather is hottest.
The contact of warm damp air with the colder surface of the ground is the main cause of the greater rainfall on the western part of these islands (Great Britain and Ireland) as compared with the eastern. The prevalent wind being south-westerly, which has blown over more or less of the Atlantic before reaching us, arrives, nearly saturated with moisture, first upon the hilly western portions of the country. The air, being forced to ascend, is cooled by contact with the hilltops, and thus we get the heavy rainfall of our western counties, which, though they are thus wetter, have also a milder tem- perature than the inland and eastern parts of the country, (partly) through the " latent heat " set free during the process of condensation. The third method of condensation — mixture of masses of air at different temperatures — though considered by some authorities as important, is estimated by Dr Hann as of minor value as a cause of precipitation.
The distribution of rainfall throughout the globe, though in general, as already stated, de- creasing from the Equator polewards, is subject to great local variation. In tropical regions the year is usually divided into the " dry " and " rainy " seasons respectively. Thus at Panama the rainy reason is from May to November, the dry season from November to May. In regions
"Dry1' and "Rainy" Seasons'' 'ioi'
where the trade winds blow there is but little rain during the time of their occurrence, whilst the descent of the return trades brings abundance of rain. Since these regions, approximately 30° to 40° N. and S. latitudes, have their rain when the sun is lowest, they may be called the regions of the " winter rains," in contra- distinction to those regions where most rain falls when the sun is at its highest.
Most of the countries round the Mediterranean, California in America, Cape Colony, etc., come under the former category. On the other hand, in Natal, the Argentine Republic, China, and the Eastern United States, more rain falls during the summer than during the winter periods of the year. In our own islands, though the total amount of rainfall is less than in tropical and sub-tropical regions generally, there is not the same seasonal distinction, though there is, perhaps, a greater amount of precipitation during the winter than the summer months ; yet the difference is not a large fraction of the whole amount, so that we may say that we have " rain at all seasons."
It is a general rule that the heaviest rain- falls occur in the western counties in Ireland, Scotland, and England alike, partly owing to the fact that the most rain-bearing winds coming from the south-west reach these regions earlier, and partly owing to the mountainous nature of the western districts, the cooling of the
102
Weather Science
air currents by passing over them producing a deposition of most of the moisture. Maps showing the total rainfall at several thousand stations throughout the United Kingdom, are published annually by the British Rainfall Organisation, founded by the late G. J. Symons, and now under the direction of Dr H. R. Mill.
Note. — By the term annual rainfall, so many inches, is meant the depth of water that would be obtained if all the rain which falls there in a year were collected into one horizontal sheet, and none were lost by evaporation or absorption into the soil. A 5-inch gauge, the area of whose surface consequently
=i T d2 = £ (3-14]6)(25) = 19'63 square inches
thus collects nearly twenty times as much rain as falls upon each square inch in its vicinity, and so enables measurement of rainfall to the nearest hundredth of an inch (or even less) to be made. One inch of rain means that if all the rain falling over any given area were spread out uniformly, it would form a layer having that depth.
Self-Recording Rain Gauge.
60
oo
MEAN ANNUAL RAINFALL MAP OF THE
BRITISH ISLES. (By kind permission of Dr. H. R. Mill.)
Annual Rainfall.
Below 25 in Blank
,, 25-30 ,, ... ... ... Dots
„ 30-40 ,, ... ... ... Faint shading
„ 40-60 „ Darker shading
Above 60 Black
To face p. 102.
Weather Forecasts 103
CHAPTER V
WEATHER FORECASTS TYPES OP CIRCULATION THE SEVEN FUNDA- MENTAL FORMS OF ISOBARS CYCLONES SECONDARIES V-SHAPED
DEPRESSIONS COLS — STRAIGHT ISOBARS — ANTICYCLONES — WEDGES
LOWER AND UPPER WINDS.
FROM the earliest times inferences as to the probable future course of the weather have been drawn with greater or less certainty from the state of cloudiness, direction of winds, halos, and other optical phenomena, and many " prog- nostics " are well known and widely prevalent. A very complete collection of these " saws and sayings" is given by Mr Inwards in his " Weather Lore," and though many are little warranted by the actual facts of the case, on the other hand many are of great value, and " in conjunction with other aids to weather fore- casting, prognostics will never be entirely superseded, especially for use on board ship" (Abercromby).
The invention and use of the barometer has led to a fresh set of prognostics, and there are still many persons who have great faith in the legends "fine," "rain," "set fair," etc., still to be found on the " weather glass," though these
IO4 Weather Science
terms are only correct in the broadest and most general sense, and are often without meaning. All such indications and prognostics, however, are of a very uncertain character, and their frequent failures are a subject of common remark. Even as it is now, with the immense mass of statistical information, records of temperature pressure, rainfall, etc., all over the habitable globe, we are in little better position to issue reliable forecasts. The average temperature is thoroughly well known, but no one can say what the actual temperature at any hour throughout the year will be; the total amount of rainfall has been measured at some stations for two centuries, but no human being can say with certainty what amount will fall at any given place four days hence.
Occasional attempts to predict the general course of the weather for months ahead may be one and all set down as premature, nay, in some cases they are little better than the char- latanry of the fortune-tellers and astrologers. Such knowledge as we have yet gained is of a purely general and statistical character, lead- ing us to mean and average results, but in no case to inferences for individual future times. The weather forecasts issued by our own and foreign meteorological offices are often remark- ably accurate, but are necessarily of the most general character, and can only be issued a day or so ahead, though in times of fairly settled
Fundamental Forms of Isobars 105
weather a good idea may be gained of its course for the next three or four days.
This forecasting by means of " synoptic charts," and the information as to the conditions pre- vailing in surrounding districts, obtained and telegraphed to a central office, has already been alluded to, but we propose here to give a some- what more detailed account of this branch of experimental meteorology, a science (if it may be called so) as yet only in its infancy. The seven fundamental forms of isobars, already referred to, are the cyclones, anticyclones, secondary cyclones, V-shaped depressions, cols, wedges, and straight isobars. The cyclones, secondaries, and V's enclose areas of lower pressure, the centres being at a pressure more or less below the average ; the anticyclones, wedges, and cols enclose regions of higher pressure. It has been found that as a rule cyclones, secondaries, and wedges move eastwards and north-eastwards in our latitudes, whilst anticyclones are often stationary for days or weeks, and occasionally even for months at a time, breaking up rather than moving on.
A cyclone may be defined as a "large disc of nearly horizontally moving air circulating spirally round a central area over which the barometric pressure varies from one-fifth to as much as three inches (of mercury) below that at its border" (Archibald). The diameter of a cyclone varies from 20 to 2,000 or even 3,000
io6 Weather Science
miles, and the form is more often oval than circular, the isobars being usually not quite concentric. Two cyclones differing from one another in the greater or less closeness of the isobars, the general character of the weather experienced during their passage will be the same, but the wind will be stronger in the former than in the latter case. The main difference between the tropical cyclones causing storms and the cyclones of temperate regions consists in this point.
Tropical cyclones are usually small and move forward at the rate of from 2 to 10 miles per hour (Abercromby), but the wind round them moves at a very high speed, perhaps 100 miles per hour. Those of temperate regions, on the other hand, have quicker translatory movement, perhaps from 20 to 50 miles per hour, but their rotatory speed (and consequent wind) is much slower. Thus we may consider the " rotatory " phenomena as due to the circulation, the " trans - lational " to the forward movement of a cyclone. In Europe and North America the usual course of a cyclone is towards the east, but in the northern tropics hurricanes move towards the west. Occasionally, even in our latitudes, a cyclone has a westerly motion, and then the usual prognostics and weather signs are said to " fail."
The usual symptoms of the approach of a cyclone in our latitude are as follows (apart from the fall of the barometer, usually accom-
Signs of a Cyclone 107
panied by the rise of the thermometer) : — The air has a close, "muggy" feel, drains begin to smell, persons subject to rheumatism complain of their pains, cirro-stratus and cirrus clouds gradually cover the sky, the sun and moon when seen low down are pale and "watery," frequently surrounded by halos. As the centre approaches, light showers or drizzling rain begins to fall, the wind blows in gusts, and usually from the south-east or south - west. After a while the barometer ceases to fall, and a short bright interval, followed by squally showers, succeeds. The centre is now passed, and the air assumes a brisk, exhilarating "feel." A hard sky with detached masses of clouds of the cumulus type characterises the rear of the cyclone, just as the front is characterised by cirrus or cirro-stratus. The north-west wind "improves men's tempers as opposed to the neuralgic and rheumatic sensations of the front." Secondary cyclones are small depressions usually associated with larger cyclones, though also found at the edges of anticyclones. The wind usually blows in angry gusts, not in the steady, regular manner of " cyclone wind." The motion of a secondary is generally parallel to that of the primary. Secondaries are frequent indications of rain without much wind. Their sudden formation frequently produces the falsi- fication of previously made forecasts. Contrary to the general rule of rain with a falling barom-
io8 Weather Science
eter, heavy rain, at first in gusts and then more steadily, sometimes for several hours continually, with steady or even slightly rising barometer, accompanies the passage of a secondary.
Allied to secondaries, we have the form of isobars known as V-shaped depressions, the isobars enclosing an area of low pressure, taking a shape like the letter V. These often occur along the southern prolongation of a cyclone, and the point of the V is usually directed south- wards in our hemisphere, or else lies in the "col" or region of low pressure between two adjacent anticyclones. Two distinct types of V's are usually enumerated. In the first and most common kind "a narrow strip of cloud pre- cedes an area of rain, followed by detached clouds and blue sky." In the second kind we have the front cloudy, and "half a crescent- shaped area of rain in the rear." The trough of the depression marks off the front of this area, but the rear is ill defined. An example of this kind of circulation bringing with it a "line squall" caused the capsizing of H.M.S. Eurydice off the Isle of Wight in 1878. The usual sequence of weather, as this form passes over a station, is from blue sky to cloud and wind from the south-west, with falling barometer. Then a heavy bank of cloud comes from the north-west, passing over with a squall; the wind suddenly changes to north-west, and the barometer rises. After the squall, driving rain
Cols, Straight Isobars 109
continues for some time, gradually ceasing, and the sky clearing once more. Many of the "southerly bursters" of Australia are said to belong to the class of V's in which rain falls in rear of the trough ; the point of the V is here turned northwards, whilst the wind is north-east in front, and south-west or south in the rear.
The col, as just mentioned, is a neck of low pressure lying between two anticyclones. In the middle there is usually no wind, whilst on the edges the wind blows according to the usual rule of isobars. The weather is generally dull and gloomy, and frequently violent thunder- storms occur (in summer) during the prevalence of these conditions. The col does not itself move, and no regular sequence of weather can be assigned to it.
Straight isobars are sometimes found near the northern edges of anticyclones. They rarely persist for long, and are soon followed by a cyclonic depression. They thus usually indicate cloudy, unsettled weather, with some wind, soon to be succeeded by more or less rain, though the air is drier than in the case of cyclones. "Visibility," "the sun drawing water," and "audibility" frequently occur as accompanying this form of isobars.
In every respect contrary to the cyclone is the next most frequent type of circulation, called, from its general opposition in properties, the
"anticyclone."
G
no Weather Science
In this the isobars are usually much wider apart, than is the case with those composing the cyclone, and the gradient of pressure is upwards towards the centre, diminishing gradually towards the outer portions. Whilst the cyclone is usually in fairly rapid motion, the motion of an anti- cyclone is very slow, and sometimes the system remains almost stationary for days or weeks at a time, finally breaking up and being replaced by a cyclone. The circulation round the centre is "clockwise," but the wind is much less in force than for cyclonic disturbances, as is shown by the greater distance apart of the isobars. During the prevalence of an anticyclone we have in summer, blue sky, hot sun, and little wind; in winter, frost and fog, and sometimes biting east winds, with gloomy black sky. In the centre of the anticyclone there is a dead calm, whilst the circumferential winds are usually slight and centrifugal, or slightly curved outwards ; the wind in a cyclone being centri- petal or incurved. Extreme dryness, accom- panied by heat in summer (though sometimes slight showers fall), and cold, frosty (sometimes also foggy) weather in winter, are the constant characteristics of this type of circulation. Anti- cyclones are the most persistent types of atmos- pheric circulation.
Wedge-shaped isobars are projecting areas of high pressure moving between two cyclones, and may be regarded as the converse of V
Wedges 1 1 1
depressions, just as anticyclones are the con- verse of cyclones. The wedge may point in any direction, but most commonly is directed towards the north. On the front or eastern side the weather is bright and the sky clear, the wind being of moderate force, and blows round according to the general law of gradients (from the north-west on the east side) ; in the centre there is a calm, and on the west side the wind is south-west, the sky becomes overcast, usually with cirro-stratus clouds, and next comes rain from the following cyclone.
Thus the sequence of weather, when the wedge travels eastwards, is fine, with north-west wind and rising barometer ; then calm and mist or fog ; then halo and gloomy sky, falling barometer and rain, with south-west wind from the new cyclone. Thus the prognostics of halos, strips of cirrus, known popularly as " Noah's Ark," etc.
It is to be remarked that all cyclones are not preceded by a wedge, but "only those which roll, as it were, along the northern edge of large stationary anticyclones " ( Abercromby). Appearances of the sky characterising the front of a wedge are signs of coming rain, of the very opposite kind to those of the rain prog- nostics of a cyclone.
Thus in a very broad, general sense we may consider cyclones, V-shaped depressions, straight isobars, and cols as indicative of unsettled condi- tions ; anticyclones and, to a less degree, wedges,
H2 Weather Science
high-pressure areas of fine weather, the former of a more or less permanent character, the latter transient and " too fine to last." We may dis- tinguish between the rain of a cyclone, which is heralded by great dampness of the air, the rain following wedges, due also to the succession of a cyclone, though the air in the wedge is itself dry, and the slight showers associated with straight isobars.
We have just given some account of the course of circulation, direction of winds, etc., in these various types of isobars, and must now proceed to the consideration of the upper currents associated with these.
The surface winds of the cyclone may be described as moving in an " ingoing spiral," more incurved in the right front than elsewhere, and less incurved as we approach the centre, than in the regions outwards.
The upper currents, on the other hand, blow outwards in front (in the fore part of the cyclone), their direction makes a considerable angle with that of the lower currents, but at the rear they are more nearly parallel to the latter. For the anticyclone also, the surface winds and the upper currents are even more opposed in direction than is the case for cyclones; the former blow spirally outwards in the clockwise direction, the latter inwards. In every case the upper currents move more quickly than those near the surface; this is, no doubt, partly due
Surface and Upper Winds 113
to the decrease of friction, just as winds over the surface of the sea blow more strongly than on the neighbouring lands. The increase of speed upwards, from observations made first by Stevenson, of Edinburgh, later by Professor Archibald, and more recently by the observers at Blue Hill and elsewhere, appears to be very considerable. Whilst the surface winds experi- ence the full effects of friction with the ground, at a very small altitude this effect rapidly diminishes. Professor Archibald found that the average velocity at 1,600 feet is just double that at 100 feet, and still continues to increase for greater heights. Clayton's observations at Blue Hill give for the level of the stratus clouds (1,670 feet), an average speed of 19 miles per hour, increasing to 24 at 5,326 feet (height of the cumulus), and velocities of 71 and 78 miles per hour respec- tively, for altitudes of 22,000 feet (cirro-cumulus), and 29,000 feet (cirrus). In winter the speeds are greater than for summer, and speeds of the upper cirrus clouds amounting to as much as 96 miles per hour are recorded. The speeds noted in European localities for the upper air seem to have usually somewhat less than these values. (See also chap vii.)
CHAPTER VI
MORE DETAILED DESCRIPTION OF COURSE OP CIRCULATION - WEATHER IN (1) CYCLONES, (2) ANTICYCLONES - SPECIAL VARIETIES — WHIRL- WINDS - TORNADOES.
THE name " Cyclone " (from the Greek /c a circle), though popularly applied only to the violent storms of the tropics (the air in which moves round and inwards towards a central region, hence the name), is applied by meteor- ologists to any portion of air moving in such a manner, the pressure at the centre being lower than that at the borders ; the difference between one cyclone and another being (1) a difference in size ; (2) difference in shape, and closeness of the isobaric lines. It is upon this latter that the force of the wind depends, as we have already seen ; the ordinary cyclones of (usually unsettled) weather, and those which cause violent storms, differ mainly in this respect ; but the latter are also usually much smaller in size than the former.
In our own country, in addition to this circu- latory or spiral motion of cyclones, they have usually a motion of translation from south-west towards north-east, though they are sometimes
114
Weather Signs of Cyclone 115
stationary ; at other times they move in the opposite direction, and occasionally fill up with- out moving on.
The weather signs at the front of a cyclone are usually those already mentioned, gloomy, close, and muggy weather, drains smelling offen- sively, the sun and moon seen dimly surrounded by halos, and the cloud formations of a cirro- stratiform type. If now the cyclone remains stationary, or dies out, or moves in a different direction, then these prognostics of coming bad weather would be said to fail. During the course of a cyclone travelling more or less centrally across any district, we find the follow- ing accompaniment of weather, cloud, etc., and various popular sayings are associated with different phases. The first signs of approach of a cyclone, "the front," is often heralded by halos, seen round the sun and moon, commonly when they are low down in the south-west. Next follows denser clouds, giving rise to the " watery sun," seen dimly through their greater thickness. After this follows rain, at first slight, but soon more continuous and heavy, accom- panied by driving wind. All this time the barometer is falling continuously with greater or less rapidity. The wind at first (assuming the the usual direction of a cyclone in this country, from south-west towards north-east, roughly speaking), is from the south-east, and moderate in force, but changes in direction and increases
u6 Weather Science
in force, as the depression advances; veering towards the south, and then becoming south- westerly, perhaps increasing in force to a gale. Now the barometer will begin to rise again, the centre line or trough having passed, and the wind suddenly veers round, or rather "jumps" towards the west or west-north-west, its force being much greater than has hitherto been the case, rain coming on more heavily than ever. This ceases after a while, the barometer con- tinuing to rise, and patches of blue appear in the sky with " rocky " cumulus cloud and moderate wind.
Lastly, after a few "clearing" showers, the wind falls to a gentle breeze, the heavens becoming clear and cloudless, and the cyclone has passed. When the observer is situated ex- actly on the path of the cyclone centre the wind "jumps " or changes when the centre passes with- out " veering " or " backing " through the inter- mediate directions, from south-west to north- east; if he, as is usually the case, is to the south of the centre, the wind " veers," very rarely for this country, when the cyclone centre is to his south- ward the wind " backs." If the wind changes with the " sun," i.e.9 from east by south to west, or in the same direction as the apparent motion of the heavenly bodies in the sky, this is called "veering"; if its change is in the opposite direction, i.e., from westwards by south to east, or from east by north to west, it is said to " back " or change
"Veering" and "Backing" of Wind 117
against the sun. A change of direction, such as that from south-west to north-east, without either veering or backing, such as occurs after the centre of a cyclone has passed directly over the observer's position, is conveniently called a "jump." Mr Abercromby quotes the following prognostic with reference to the " backing " of the wind : —
" When the wind veers against the sun Trust it not, for back 'twill run."
This he explains by the usual sequence of dis- turbance. When in Northern Europe a cyclone passes to the south (a rare phenomenon), thus producing the backing of the wind against the sun referred to, it is almost always followed by another cyclone to the north, which brings more bad weather and fresh changes of the wind.
From the general indications given by clouds, wind, and barometer as outlined above, the observer can easily ascertain his position with regard to the disturbance, and its general course. The various directions of the wind in the different portions of a cyclonic disturbance are related to the pressure (in different parts) by the "law" of Buy Ballot, given more fully in our chapter on the Winds (viii), sometimes enunciated thus : —
"Stand with your hands stretched out on either side and your back to the wind, then, in the Northern Hemisphere the centre of the cyclone will be to your left hand ; if you are in
ii8 Weather Science
the Southern Hemisphere the centre will be to your right."
This form, however, ignores the incurvature of the wind, which, instead of blowing directly along the isobars, makes a considerable angle therewith ; this varies with position on the surface of the globe, being also different for storms on land and on sea. The " inclination " was found to be 62° for the Philippine Islands (latitude 14°) ; in Bengal (latitude 20°) it was 57° ; over the Atlantic 30° ; in England only about 20° ; thus apparently greater for equatorial regions than for stations further northwards. From this it follows that the old law of storms, which supposed the wind to " blow in a circle " directly along the isobaric lines, is quite unreliable for tropical latitudes, where the deviation is often considerable. Thus it sometimes happened that by following this rule, in the old sailing-ship days, a captain might sometimes run his ship directly into danger when seeking to avoid it. Whilst cyclones form, perhaps, the most pro- minent feature of the atmospheric circulation in our latitudes, they are less common in the tropics, though of a more dangerous type, as has been mentioned, smaller but with much greater and more violent wind motions, and it is from these that the word "cyclone" has acquired its popular meaning of a fearful storm.
The researches of Piddington, Redfield, and
Meldrum on Cyclones 119
Dove on Indian, American, and European cyclones respectively, established the true form of the movements of wind, that the latter circu- lates in a spiral curve round the centre or point of lowest pressure.
The late Dr Meldrum, for many years Director of the Royal Alfred Observatory and Government Meteorologist at Mauritius (whose researches on cyclones have recently been utilised by Mr Maunder of Greenwich to point out an unexpected relation between their sequence in tropical latitudes and the sun's rotation period), by his work largely helped to develop a more satisfactory set of rules than those afforded by the " circular theory." Even so, all that can be done is to give rough general rules, for, as pointed out by Dr Meldrum, in some cases the wind, instead of blowing at right angles to the radius, blew directly towards the centre (Scott). The modern rules now advise the mariner " to avoid" running before the wind, to lie to on the starboard tack (i.e., with the wind on his right) in the Northern Hemisphere, or on the port tack (wind on left) in the Southern.
Tropical cyclones occur more frequently in September and October for the Northern Hemi- sphere; in February and March, the "cyclone seasons," for the Southern ; or more generally in the summer and autumn of both hemispheres. Off the Indian coasts they are most common and dangerous at the changes of the monsoons
I2O Weather Science
during May and October respectively. At times there is a difference of not less than 2 inches of pressure (mercurial barometer) between the centre and outer circumference of a cyclone. During one storm the barometer at Marie- Galante in Guadaloupe fell from 29-646 inches to 27*953 inches in the course of seventy minutes, between 6.30 and 7.40 A.M. on 6th September 1865 (Buchan).
It sometimes happens that such a cyclone will travel from the tropics northwards towards Europe and more temperate regions, where it will become a much milder phenomenon. This is another proof that the ordinary storms of Europe are phenomena of the same nature as hurricanes, though differing from them not only in intensity, but probably in the shape of the storm area, and also apparently in the unequal development of the wind from the different points of the compass (Scott).
Calculations have been made by the curious as to the amount of energy developed in some of these disturbances. For instance, Professor Reye of America calculated that the Cuban cyclone of 5th October 1844 used up 473,000,000 units of horse-power in three days.
As a theory of the cause and movement of cyclones, a brief outline of Ferrel's views may be of interest. Ferrel, sometimes called the " Newton " or the " Father of Modern Meteor- ology," originally an elementary school teacher,
Fen-el's Theory 121
who by his original researches and other work has so largely added to our knowledge, con- structed a consistent theory of these phenomena.
By assuming an inflow towards and an upflow over a given area, he showed that (neglecting friction) the air would tend to rotate round a central area, at the inner portions of which the pressure would be very low, but this would gradually increase from the centre outwards, the whole moving in a direct or counter clockwise direction. Outside of this area there would be another area moving in the opposite or clock- wise direction. The interior region would thus be the cyclone ; the outer a kind of anticyclone or pericyclone. The effect of friction would be to somewhat modify this ideal state of affairs ; the pressure near the centre would be low, but not quite so low as would otherwise be the case, the (interior) motion of the inner portions being somewhat more "centripetal" (towards the centre), that of the outer part "centri- fugal" (outwards).
The clear, calm region usually only a few miles in extent, at the centre of a tropical cyclone, sometimes called the "eye" of the storm, was explained by him.
Some modern "specialists" consider that extra tropical cyclones, at least in part, are due to somewhat different causes from those of the tropics. It is well known that stormy con- ditions, heavy rain, etc., are, on the whole, more
122 Weather Science
prevalent in winter than in summer, and it has been thought that they are due to eddies or whirls in the upper "return currents" flowing over from the Equator, which are crowded into the " narrower latitudes." These eddies cause the lower air nearest the surface to ascend ; this air forms clouds, whence sometimes rain falls, but the movements are less violent near the ground than in the upper regions.
The general direction of motion of cyclones eastward in these latitudes has already been alluded to. They appear to be governed chiefly by the prevalent west wind, both upper and under, and are carried along like eddies by the current. They often exhibit a remarkable tendency to follow the same course, several successive depressions quickly succeeding one another. Thus during weather of the "westerly" type, in Great Britain, when the depressions are so far south as to cross that island, the centres have a decided tendency to traverse either the line of the Caledonian Canal in Scotland, or the low-lying ground which separates the valleys of the Forth and Clyde (Abercromby.) Other cyclones coming in from the Atlantic often hug the coast of Norway instead of going north-east.
Mountain chains also powerfully influence their direction. For instance, the Alps forms a natural boundary between the Mediterranean weather and that prevailing in the more northern
Bebber on Cyclones 123
parts of Europe ; the Himalayas, to the north of India, also, even more powerfully influence the weather conditions of that great country. This tendency of cyclones to follow certain definite directions is, of course, of great im- portance in connection with forecasts as to their probable course and duration. The influence of heat in determining their course has also been considered by Dr Bebber and others. The former has enunciated the following relations, especially for Central Europe : —
"When the distribution of air pressure and temperature are in the same sense, then the depression is propagated nearly in a direction perpendicular to the temperature and pressure gradients ; if they are distributed in the opposite sense the motion of the depression ceases or is checked."
He considers pressure to be the more im- portant factor in determining cyclone motion in winter, whilst in summer, temperature difference is the predominant influence. With regard to the "sense" of temperature and pressure gradients, the following definition is given : —
" If the highest pressure and highest tempera- ture are both to the north, or both to the south of a cyclone, they are said to be in the same sense (and the depression will move at right angles to both). But suppose pressure was highest to north, and temperature highest to south, then these two elements are said to be
124 Weather Science
distributed in the opposite sense, and the cyclone would probably be arrested in its usual eastward course" (Abercromby).
Cyclones frequently exhibit a tendency to move round anticyclones usually in such a manner as to keep the anticyclones on their right in this hemisphere, whilst occasionally they appear to move round one another, or again the phenomenon of two cyclones moving round a common centre is presented.
As already stated, whilst the surface currents in a cyclone move after the manner of an ingoing spiral, but with their direction less incurved towards the centre, in the upper parts of the disturbance, at heights above 10,000 feet from the earth, the wind blows in a more irregular spiral outwards, being in front very much inclined outwards, but in the rear nearly parallel to the lower surface currents. At intermediate heights the course of the wind is considered to be nearly parallel to the isobars, or moving almost in a circular, or more strictly oval, manner, since few cyclones are even approximately circular, the late Professor Loomis stating that in the United States a circular "cyclone" does not occur more than once during the course of the year, the average ratio between the longest and shortest diameters being about 1-94 to 1, or roughly, rather less than 2 to 1. We have already detailed the general distribution of
Size of Anticyclones 125
cloudiness in a cyclone, but the actual circum- stances vary somewhat in different cases. The most common occurrence of rain with low pressure does not, however, prevent occasional instances of an exception to the general rule being sometimes perceived. In some cases well- marked and large depressions have been formed, but the barometric reading at the centre was only slightly less than on the outside, perhaps not lower than 297 inches (of mercury), the gradient is consequently very small, with light winds, small and slow fluctuations of pressure ; unaccompanied by rainfall, or at most very slight showers have fallen.
We next come to consider the large often stationary areas of high pressure from which the air often flows outwards, to feed the cyclones, whose characteristics are in so many respects opposite to those of the latter that the name of anticyclone has been universally applied to them. Anticyclones are generally of much larger size than cyclones, and sometimes cover a whole continent, or extend half over the ocean, and are very persistent phenomena. The pressure is highest at the centre and the isobars are more nearly circular than those surrounding the cyclone centre, and much wider apart ; in consequence there is usually but little wind any- where. There is practically a calm in the central portion, but on the outer parts of the system the wind blows round the centre in the clock-
H
126 Weather Science
wise direction. Like the wind of cyclones the surface wind direction is not along or parallel to the isobars, but spirally outwards. Just as with depressions the surface and the upper winds are opposed in direction, the upper currents blowing spirally inwards, their direction being more inclined to the isobars than the lower. The whole system, though often at rest and persisting till it breaks up, occasionally moves slowly along, usually in a direction east to west, or sometimes north-west to south-east.
During the summer-time this type favours dry, hot weather, in winter east wind and overcast conditions. Often the morning is somewhat hazy or foggy, but this mistiness is usually dispersed by the power of the sun's rays ; the broad features of anticyclone weather being put as blue sky, dry cold air, hot sun, and hazy horizon, with very little wind, the type called by Abercromby "radiation weather." In summer, mist in the morning and evening, with fine, hot, cloudless day ; in winter similar conditions, but more pronounced fog, and often instead of a clear sky, overcast, gloomy, rainless weather, sometimes accompanied by the well-known east wind whose presence is so unwelcome: —
" When the wind is in the east, "Tis neither good for man nor beast."
Many indications and popular sayings are very familiar, some being common to all parts of the world. The far flight of birds, wild animals
Descending Current of Air 127
disporting themselves in the open air, and the exhilarating effects of fine bright weather upon human beings as well as the animal creation generally, are well - known accompaniments. The prevalence of morning and evening mists during hot weather, of fogs, white or black, as the case may be, and frost during winter (the necessary accompaniment of the absence of wind during anticyclonic conditions), as also the veering of the light breezes with the sun " in by day, and out by night," the land and sea breezes due to unequal heating of sea and land, experienced at maritime stations, are all more or less common phenomena accompanying this form of isobars.
Whilst in cyclones it is considered that we have to deal with an ascending current, for anti- cyclones the current is a descending one, and, as stated, sometimes serves as a " feeder " to the former. The ascending damp air of the cyclone more or less nearly saturated with moisture favours precipitation and consequent rainy con- ditions, whilst the air being drier, and such moisture as is present far from the tempera- ture of saturation, in the anticyclone, it in general favours fine and settled conditions of weather.
It has been already stated that both forms of circulation in general depend upon one another, and that cyclones frequently travel round anti- cyclones, near whose edges they have been
128 Weather Science
formed, though it sometimes happens that they will travel far from these latter, occasionally moving not only across the Atlantic into Europe, but even passing thence to Asia.
In addition to the more common forms referred to, special groups of local disturbances of smaller size, variously known by the names of whirlwinds, tornadoes on land, and water-spouts on the sea, the simoon or dust cloud of the desert, the hurricanes of the West Indies, and the typhoons of the China Seas, are all related to cyclones, though "the motion of the wind in storms of the eddy type is probably more truly spiral, in- curving towards the centre, than circular " (Scott).
The whirlwind or tornado may be described as a mass of air in rotation round an axis usually nearly vertical. It may be as much as 200 feet in height, but its breadth does not usually exceed 10 feet. The harmless "dust whirl " of the roadside and the terrible tornado of America are extreme examples, whilst the "simoon" of the desert is a whirlwind carrying sand and dust. When the whirl is large and the air moist we sometimes get a thunderstorm, such as accompanies the "pampero" of the Argentine Republic, which is described as a south-west wind, ushered in by a sudden squall, with rain and thunder and a typical form of cloud wreath. In northern latitudes a similar kind of disturbance is known as a " line squall " (Abercromby).
Tornadoes and Waterspouts 129
The most characteristic feature of the tornado is its funnel or spout, the cylinder of air in rotation, whilst the system moves forward, usually in a north-east direction at a rate of about 80 miles per hour. The rotation is of a somewhat complex character ; in addition to the counter-clockwise or cyclonic motion there is usually a violent upward current, also a rising and falling motion, the end of the spout some- times rising from the ground and then descending again, whilst the axis is seldom upright, but sways slightly to and fro. Rain and thunder usually accompany this manifestation. Great damage is done by its destructive violence — houses, trees, fences, churches, etc., being blown down or carried up into the air for a great distance. It has been estimated that the velocity of the wind rotating near the centre of a tornado may reach as much as 500 miles per hour; whilst the upward velocity may sometimes attain to over 100 miles per hour. The central column of rarefied air being cooled by expansion, any vapour within it is condensed. Thus some- times a "water-spout" is produced. That this water is not drawn up from the sea is shown by the fact that even when a water-spout passes over the sea, the water in it is quite fresh, and not at all salt. The funnel shape of the water- spout or tornado clouds is considered to be due to the increased pressure of the air near the surface. Above, the absence of friction and the
130 Weather Science
pressure from below causes the central rarefied area to extend somewhat, but lower down there is increased internal pressure, and the rotating air is confined to a narrow space.
Tornadoes appear to be more common in the spring and early summer, whilst in autumn and winter they are of very rare occurrence. They usually occur on sultry days, and either in the south-east, or right front of cyclones, or in front of the trough of V depressions ( Abercromby).
The destruction done by these fearful, though fortunately short-lasting manifestations, which "break the monotony of a tropical calm," has been so often described in glowing terms by different writers that it seems hardly necessary to repeat the well-worn tale. Those who are fond of complaining of the badness of our own weather may at least be reminded that we have much to be thankful for in our exemption from such catastrophes ; though besides this purely negative benefit, the existence of many positive advantages alluded to in the course of this work may with more justice cause our own climate (or at least that of the more maritime western regions), to be regarded as one of the best in the whole world, though perhaps that of certain stations in the Southern Hemisphere is more salubrious.
Howard's Nomenclature 131
CHAPTER VII
CLOUDS HOWARD S NOMENCLATURE CIRRUS CIRRO-CUMULUS
CUMULUS — STRATUS CIRRO-STRATUS CUMULO-STRATUS NIMBUS
HEIGHTS OF DIFFERENT KINDS OF CLOUDS — MOTIONS OF CLOUDS AND AIR CURRENTS FOGS, MISTS, ETC. DUST PARTICLES OR OTHER
IT has been often said that the study of clouds is one of the most important elements necessary for successful forecasting of coming weather. More than a century ago, Howard, in his essay on the " Modifications of Clouds," proposed the nomenclature for the different kinds which is now universally adopted. " Clouds," says he, "are subject to certain distinct modifications, produced by the general causes which effect all the variations of the atmosphere ; they are commonly as good visible indications of the operation of these causes, as is the countenance of the state of a person's mind or body." He first discriminates the three simple kinds, cirrus, cumulus, and stratus, the names being derived from the Latin, and nearly correspond to their general appearance. He defines cirrus cloud as " parallel, flexuous, or diverging fibres, extensible in all or on all directions " (nubes cirratd). The cumulus (nubes cumulata) cloud is a "convex
132 Weather Science
or conical heap increasing upwards from a horizontal base." The third form, stratus cloud (nubes strata), is " a widely extended, continuous, horizontal sheet, increasing from below." Of these three forms the cirrus clouds are the finest and most lofty ; cumulus, more dense and formed in lower regions of the atmosphere ; stratus, the lowest form, usually forming in the evening and dissipating towards the next morning. Howard distinguishes four intermediate forms, the cirro- cumulus, the cirro-stratus, the cumulo-stratus, and the cumulo-cirro-stratus or nimbus. Other forms are sometimes given, but these seven kinds will describe 90 per cent, of all skies (Abercromby). Howard's definition of these intermediate forms we take from his famous essay, mentioned above. The " cirro-cumulus " consists of" small, well-defined, roundish masses," in close horizontal arrangement. The "cirro- stratus," " horizontal or slightly inclined masses, attenuated towards a part or the whole of their circumference, bent downward, or undulated, separate, or in groups consisting of small clouds having these characters." Next comes the " cumulo-stratus," " the ' cirro-stratus ' blended with the cumulus and either appearing inter- mixed with the heaps of the latter or super- adding a widespread structure to its base. The seventh form, nimbus or cumulo - cirro - stratus, 'the rain cloud.' A cloud, or system of clouds, from which rain is falling." " It is a horizontal
Cirrus Clouds 133
sheet above which the cirrus spreads, while the cumulus enters it laterally and from beneath."
In addition to the above seven well-marked forms some writers distinguish between cirro- stratus and strata - cirrus, cirro - cumulus and cumulo- cirrus, according as one or other character predominates, Howard's forms cirro-stratus and cirro-cumulus being each subdivided into two. Captain Wilson Barker * proposes a very simple classification, considering all clouds as belonging to one or other of two types : — (1) The Cumulus or heap type; (2) the Stratus or large sheet type, including all forms save Howard's cumulus under the second heading. Of all these forms the cirrus clouds are those which have the greatest elevation and variety of extent with least density. They are the earliest indications of change after a period of fair, settled weather, signs that this is to be followed by less favour- able conditions.
At first there appears a few threads " pencilled, as it were, on the sky." These increase in length, and new ones are added laterally. Often these first threads serve as stems to support numerous branches, and from these in their turn other branches spring.
Owing to their great height, cirrus clouds though often in fairly rapid motion, seem more slowly moving than clouds of other forms. It is fairly certain that they consist of small ice
1 Essay on " Clouds and Weather/' 1895.
134 Weather Science
crystals, whence their effects in causing halos and other optical phenomena. Investigations by Hildebrandson in Sweden and the Rev. Clement Ley in this country have added considerably to our knowledge of atmospheric movements in the upper regions of the air. Cirrus clouds seem to be found in all parts of the earth, but those seen in tropical regions are probably at a greater altitude above its surface than the polar ones. With regard to their duration, Howard remarks that this varies from a few minutes to many hours, being long when they appear alone at great heights, and shorter when they are formed lower and in the vicinity of other clouds. In fair weather the sky is seldom quite free from small groups of oblique cirrus, whilst continued wet weather is attended by horizontal sheets, which subside quickly and pass to the cirro- stratus form. Before storms they appear lower and denser, and usually in the quarter opposite to that from which the storm arises. This form of cloud is often called "mare's tails," more especially when it is curved in form. In observing these clouds, it should be noted whether they are developed in any particular region of the sky rather than another, as well as the relation between their longitudinal extension and the direction in which they are moving.
The cirro-cumulus is also a lofty cloud, though usually less high than the cirrus. It differs from the latter in being of a more rounded form,
CUMULUS AND HIGH STRATUS.
CIRRUS. (Photos by Capt. Wilson Barker, F.R.S.E., F.R.Met.Soc., &c.)
To face p. 134.
SQUALL CUMULUS.
CUMULUS. (Photos by Capt. Wilson Barker, F.R.S.E., F.R.Met.Soc., &c.)
To face p. 135,
Cirro-Cumulus — Cumulus 1 35
consisting usually of small, detached masses, sometimes compared to a flock of sheep lying down, or the markings on the back of mackerel, whence the name " mackerel sky." It seems to be formed from a cirrus or from a number of small separate cirri, by the fibres collapsing and becoming small, roundish masses, the change taking place either through the whole mass at once, or gradually from one end to the other. This form is frequent in summer during warm, dry weather, and is more sparingly seen in intervals between showers, or before thunder- storms, when very dense and compact masses in close contact are often visible.
The cumulus cloud, commonly called the " wool pack," is formed by an ascending current of air whose vapour is rapidly condensed. It is the densest kind of cloud formed in the lower atmosphere. Its lower surface is roughly plane, whilst its upper rises into conical or hemispherical heaps, which sometimes " continue nearly of the same bulk or rapidly rise to mountains." If remaining unchanged in size they are usually numerous and near together, when swelling they are few and far apart ; in either case their bases always lie nearly in one horizontal plane. These horizontal bases are, of course, evidence of the existence of strata of air of different temperatures. If the lower stratum be at a higher temperature than the upper it can contain more moisture, and thus will dissolve any portions of the cloud
136 Weather Science
which descend into it, so that the cumulus cloud appears to stand on the surface of separa- tion between these two layers of different temperature. In fair weather the variations of these clouds are often periodical during the course of the day. They begin to form some hours after sunrise, arrive at their maximum size during the afternoon, at the hottest time of the day, and then diminish, totally disappearing towards sunset. If, on the contrary, they increase rapidly in size, sink downwards, and do not dis- appear in the evening, rain may be expected. The formation of large cumuli to " leeward " in a strong wind indicates the approach of a calm with rain ; if they do not subside, but continue to rise towards sunset, thunder is to be expected in the night. In winter time the appearance of cumulus in the south after a fine day often indicates approaching snow.
The stratus is a cloud lying in horizontal layers or strata, whence the name. Howard applied this term also to ground mists and fogs, but this is now discarded by meteorologists. It may be considered as the " cloud of night," since it owes its origin to the evening mists and grows denser during the night, dissipating again towards morning. The air being tolerably still and radiation from the ground going on, the general mass of the atmosphere above it cooling gradually, some stratum arrives at the dew point temperature and its moisture is condensed into
Cirro-Stratus and Cumulo-Stratus 137
cloud. The pure stratus cloud is an accompani- ment of fine weather, and if stratus at night be followed by diffuse fog in the morning we have generally settled atmospheric conditions.
The cirro-stratus, according to Howard, appears to result from the subsidence of the fibres of the cirrus to a horizontal position ; at the same time they approach one another laterally. The form and relative position often suggests "shoals of fish." The structure is always thickest in the middle and thinner towards the edges. This form of cloud often precedes wind and rain, the nearer or more distant approach of which may sometimes be inferred from the greater or less permanence of these clouds. Owing to the great extent but little perpendicular depth of this form, the sun or moon may often be seen shining through it surrounded by a halo, so that the appearance of these phenomena is often regarded as a sign of approaching foul weather. It seems fairly certain also from these phenomena that the cirro-stratus clouds are largely composed of frozen particles of vapour.
The cumulo-stratus cloud is a compound of cirro-stratus resting either on the top of a cumulus or crossing an isolated patch of the latter. It may be regarded as the cumulus cloud passing gradually into the nimbus form. It is usually a forerunner of rain or snow, according to the season of the year.
The nimbus cloud is a name loosely given to
138 Weather Science
any kind of cloud from which rain falls, so that the term may be applied rather vaguely to different species. Abercromby distinguishes two kinds — the cumulo-nimbus, the rocky, cumulus cloud from which rain falls in squalls or in showers; and pure nimbus, a flatter cloud more like heavy strato-cumulus, that forms from or under cirro - stratus. The name cumulo - cirro - stratus, suggesting its mode of origin, has been already alluded to. " The reason for making nimbus a class of its own comes from the fact that a sudden striking change comes over the look of the upper surface of a cloud the moment rain begins to fall " (Abercromby). This change is possibly associated with the discharge of electricity at the moment of precipitation.
In addition to these principal varieties a number of minor forms, some of which are of importance in judging coming weather, may be distinguished. Sometimes before the approach of a cyclone a blue sky becomes white, then grey, and drizzling rain falls without the formation of any true cloud form. For this Mr Ley has given the name " cirrus haze " or " cirro-nebula." Small detached clouds seen in rapid motion under any mass of cloud just before the pre- cipitation of rain are frequently called " scud," or sometimes fracto-cumulus. Before the advent of squalls and thunderstorms there is sometimes seen a long roll of narrow black cloud in rapid
Height and Motion of Clouds 139
motion, and this form goes by the name of " cloud wreaths."
With regard to the height above the surface of the ground at which the various cloud forms exist, we may class cirrus, cirro - cumulus, and cirro-stratus as high clouds, since they often exist at an altitude of 20,000 to 30,000 feet ; cumulo- and strato-cirrus forms are found at intermediate altitudes, whilst cumulus, stratus, and nimbus are low, many being below 2,500 feet. The levels for Upsala of the principal varieties in summer have been given as: — cirrus forms, 20,000 to 27,000 feet ; middle forms, 12,000 to 15,000 feet ; cumulus, nimbus, and stratus, below 6,000 feet. Though these altitudes vary with latitude and season of the year they serve to illustrate the principle, " that clouds everywhere tend to form at a few definite levels, widely separated from each other."
Motion of clouds. — Low-lying clouds usually move in the same direction as that of the wind felt at the surface of the ground, though in mountainous regions they are subject to local variations. Their motion is, however, always more rapid than that of the air current close to the ground. Clouds at great altitudes have been carefully studied, especially in recent years, by the observers at Blue Hill, near Boston, U.S.A., and elsewhere, and it has been shown that at a height of about 5 miles the movement is practically three times as fast in summer and
140 Weather Science
six times as fast in winter as the currents at the earth's surface.1 As a general rule, the greater the altitude the faster the movement. Their motion, too, is often in a different direction from that of the surface currents, and sometimes we find several layers of clouds floating at different heights, each moving in different directions.
Fogs and mists are closely related to clouds. Aqueous vapour rising by evaporation from the ground is itself invisible, but it becomes con- densed in the form of minute droplets, with the liberation of its latent heat. At any definite tem- perature air has the power of maintaining a cer- tain quantity of aqueous vapour in the gaseous condition, this being greater as the temperature is higher. When the air contains the maximum quantity of vapour for a given temperature it is said to be saturated, and the smallest diminution of temperature causes a portion of the vapour to leave the gaseous condition and become "precipitated," as it were. Two masses of air of different temperatures mixing together, each saturated with moisture, producing a mixture of an intermediate temperature, a portion of their moisture is thus condensed and a fog is formed. The condensed particles having a tendency to form on solid matter floating in the air, such as dust and soot, etc., we get the black fogs of London and other large towns. No fogs can
1 Inwards, presidential address to Royal Meteorological Society ' ' On Some Phenomena of the Upper Air."
CIRRO CUMULUS.
\
GROUND FOG. (Photos by Capt. Wilson Barker, F.R.S.E., F.R.Met.Soc., &c.)
To face p. 140.
Fogs and Mists 141
be produced in an atmosphere perfectly free from foreign particles by the condensation of aqueous vapour alone ; this has been shown by Aitken and Tyndall. In addition to the cooling produced by the mixture of two masses of air of different temperatures, fogs may be produced by the passage of a warm damp current over a cold surface, and also by the passage of saturated air over a warm water surface. The water, being warmer than the air, gives off more vapour than the latter can contain, and thus a fog is pro- duced. Of course no fog can be formed in windy weather, since it is dissipated by the least motion of the air. Mists are similar in character to fogs, but the particles are larger, and they feel wetter. They are more commonly found on parts of hills covered with trees, and near the banks of rivers and marshy places, than elsewhere. A cloud is in reality only an extensive fog or mist existing at a greater altitude, and its forma- tion, as at a mountain top, is due to the con- densation of moisture from the warm lower air in passing over it. This has its temperature lowered and is thus forced to deposit part of its contained moisture.
In making observations of meteorological phenomena it is customary to estimate the amount of cloud visible in the sky at the time of observation. The ordinary estimation is by tenths, 0 being recorded for a clear blue sky, 10 for one altogether covered by cloud. Varia-
142 Weather Science
tions in estimation as to the amount of cloudi- ness naturally prevail to a considerable extent, and the same cloud appears of a very different size near the horizon than when higher up in the sky. With regard to the speed of clouds generally much information has been recently gained. The Blue Hill (U.S.A.) Observatory results give an average speed of about 20 miles per hour for stratus clouds at an altitude of about 2,000 feet, whilst for cirro-cumulus at 4 miles high the speed increases to 70 miles per hour, and for very lofty cirrus (5 to 6 miles above the ground) this may amount to as much as 80 miles per hour. It has been asserted that speeds of 250 miles per hour have been observed ! (Archibald). For every 1,000 feet of ascent add on about 2 miles an hour to the velocity of motion (Archibald). The speeds are greater in winter, but the average height of clouds are greater in summer than in winter. The observa- tions at Upsala and other places in Europe give less rapid but still large speeds for the upper clouds. Speeds of 19 miles per second for clouds at 4,300 feet, and 38 miles per hour for clouds at 38,000 feet were noticed. These rapid movements of the upper air may eventually be utilised in " flying machines " travelling with the wind. The successive cloud layers, which, as we have already said, tend to form at about certain definite levels, coincide with air streams differing from one another in point of velocity,
Formation of Clouds 143
temperature, and humidity, and these must exercise a marked influence on the weather con- ditions below.
We shall next devote a few lines to the question of the formation of clouds. Every cloud is the visible top of a column of invisible water vapour, sometimes stretching from the ground upwards, and becoming condensed on reaching a colder stratum of air. At every temperature a certain proportion of water vapour may be maintained in the gaseous state, but when this amount is exceeded, or the air is "saturated," the excess is usually condensed, though it has been found that perfectly clear dust-free air may be supersaturated ; but when fine particles of solid matter are present con- densation takes place round these as nuclei. It has been supposed that sudden rainfalls are due to this condition of vapour saturation, but Mr Aitken has found that the presence of dust is absolutely necessary to the formation of rain. The amount of water in the air rapidly decreases as we go upwards, though cirrus clouds at a height of 50,000 feet have been occasionally noticed, these latter being probably composed of small ice particles or "needles." The shallow stratus, cirro - cumulus, and cirro - stratus clouds are supposed to be due to the mixture of a layer of warm air with an underlying colder stratum. When one current crosses another it raises waves in the latter; the "mackerel
144 Weather Science
sky," and the long rolls of dark cloud following one another at the rear of a storm, with showers and brighter intervals, being examples of such aerial waves. The clouds around mountain-tops are due to the cooling produced by these latter causing condensation of part of the moisture in the air which is rising upwards.
The "table-cloth" over Table Mountain and other hills is formed by the passage of a warm moist current of air over the cold hilltop, whose action condenses part of this moisture. When it passes beyond the mountain the cloud mixes with warmer air, and is once more dissipated, but as fresh air is continually rising, a cloud is almost constantly formed over the mountain, but consists of constantly changing particles of water. The clouds occurring in connection with cyclones are due to the ascent of damp air rising and mixing with the drier atmosphere above.
The literature of cloud prognostics is very extensive, and from all ages the portents and signs of coming weather changes have been discussed with more or less ingenuity by many writers. The remarks of the late Admiral Fitzroy, though given elsewhere, may be here quoted : —
" After fine weather the first signs in the sky of a coming change are usually light streaks,
Cloud Prognostics 145
curls, wisps, or mottled patches of white distant clouds, which increase, and are followed by an overcasting of murky vapour that grows into cloudiness. The appearance, more or less oily or watery, as wind or rain may prevail, is an infallible sign. Usually the higher and more distant such clouds seem to be, the more gradual but general the coming change of weather will prove."
Further remarks by the same author with reference to the motions, colours, etc., of cloud forms, are quoted in Mr Inward's valuable " Weather Lore." Some of the descriptions of these phenomena to be found in the Bible show the results of careful observation on the part of the scriptural writers. Many references to cloud phenomena are especially to be found in the book of Job. In Aristophanes' comedy the " Clouds," written as a satire on Socrates and his teaching, "the clouds" are supposed to be the new, hitherto unknown deities introduced by that philosopher to the Grecian world, to replace the gods and goddesses of Olympus. Shakespeare's description of some cloud appear- ances and changes may be worth quoting here : —
" Sometimes we see a cloud that's dragonish A vapour sometimes like a bear or lion, A towered citadel, a pendant rock, A forked mountain, a blue promontory
146 Weather Science
With trees upon't that nod unto the world And mock our eyes with air. That which is now a horse, even with a thought The rack dislimns and makes it indistinct As water is in water."
Of a coming storm he says in the Tempest : —
" And another storm brewing ; I hear it sing i' the wind Yond' same black cloud, Yond' huge one, Looks like a foul lumbard That would shed his liquor . . . Yond' same cloud cannot chuse But fall by pailfuls."
The shepherd of Banbury is full of weather wisdom from the appearances of clouds. When they increase, we are told: "If the sky from being clear becomes fretted or spotted all over with bunches of clouds, rain will soon fall." Again, with regard to wind, he says : " If you see a cloud rise against the wind, when that cloud comes up the wind will blow the same way," etc. As signs of a coming storm : " In summer, when wind has been south for two or three days, and it grows very hot, you see clouds rise with great white tops like towers, there will be thunder and rain suddenly. If two such clouds arise, one on either hand, it is time to make haste to shelter." The appear- ance of cloud hats or "caps" on distant hills, especially when these lie to the south or south- west of the observer, is looked upon as a sure
Cloud "Hats" 147
sign of approaching rain. Many such sayings are common in Scotland —
" When Largo Law puts on his hat. Let Kellie Law beware of that ; When Kellie Law gets on his cap, Largo Law may laugh at that,"
the latter being to the south - west of the former.
CHAPTER VIII
THE WINDS DIRECTION COMPASS AND TRUE BEARINGS RELATION
TO ISOBARS — BEAUFORT'S SCALE — BUY BALLOT'S LAW — TRADE
WINDS GENERAL CIRCULATION SPECIAL WINDS I FOHN, SCIROCCO,
BORA,, MISTRAL, LAND AND SEA RREEZES NOTE ON WIND
VELOCITIES AND BEAUFORT'S SCALE.
THE wind is a body of air in motion, and this motion being produced by differences of baro- metric pressure in different directions, it is found that as a general rule the "force" or velocity of its movement is roughly proportional to the closeness of the isobaric curves, being consider- able when they are near together, and small when they are wide apart. The direction of the wind is recorded according to the point of the compass from which it blows, and it is per- haps advisable in this connection to recollect that the north - seeking end of the magnetic needle does not point exactly towards the true or geographical north, but at present (1911) makes an angle of about 15° to the west with it, for the neighbourhood of London. This " varia- tion " is slowly changing from year to year, at
148
Wind Directions 149
present diminishing, and in about fifty years' time true and magnetic north will coincide. It is customary to divide the whole circum- ference of the compass card into thirty-two parts called points, each point making thus an angle of 360° -4- 32 = 11 J°, with the neighbouring points on either side of it. In practice it is usually sufficient in estimating the direction of the wind to use eight principal points only, N., N.E., E., S.E., S., S.W., W. and N.W. In estimating the closeness of any two isobars, on which the force of the wind depends, we take the slope of the barometric gradient, measured at right angles to the isobar curves, this being the shortest line which can be drawn between them. Gradients are measured by the number of millimetres of barometric pressure difference in one geographical degree, or their equivalents in English measure, 1 mm. = 0*04 inch, and 60 nautical miles = length of 1°. According to its intensity the wind is variously designed as light, moderate, fresh, strong, gale, storm, or hurricane. The well- known Beaufort's scale is still in very general use. On this system an arbitrary series of numbers, from 0 to 12, "calm" to "hurricane" is taken to estimate approximately the various intensities of wind corresponding to velocities from 0 to 100 miles per hour.
The following numbers give the approxi- mately equivalent velocities of the wind as
Weather Science
determined by the British Meteorological Office :—
Velocity in miles Wind. per hour.
7 Moderate gale . 40
8 Fresh „ . 48
9 Strong „ . 56
10 Whole „ . 65
11 Storm . . 75
12 Hurricane . 90
and upwards.
|
V( Wind. 0 Calm 1 Light air . 2 Slight breeze 3 Gentle „ 4 Moderate „ 5 Fresh „ 6 Strong „ |
slocity HI miles per hour. . 3 . 8 . 13 . 18 . 22 . 28 . 34 |
[See also note at end.]
Though, as we have said, the force of the wind varies directly with the closeness of the baro- metric gradients, yet for any given gradient in this country, winds from north and east are stronger than those from the south and west points, by at least one-third of their whole amount. In certain cases of tropical and sub- tropical winds, such as the " northers " of New Mexico and the " nortes " of Panama, the force of the wind is quite disproportionate to the gradient. There are also some few winds distinctively called "non isobaric," whose origin does not appear to be due to differences of barometric pressure. Since in general, whenever there is a difference of pressure, the air must flow from a region of high pressure towards one where it is low, we see at once that there must be a relation between wind and air pressure. The law express- ing this relation is known by the name of Buy Ballot, a Dutch professor who drew attention to
Buy Ballot's Law 151
its importance. It is commonly enunciated in the following terms: —
" If you stand with your back to the wind in the Northern Hemisphere, the barometer will be lower on your left hand than on your right. In the Southern Hemisphere, standing with your back to the wind, you will have a lower baro- meter on your right hand than on your left."
Thus (in our latitudes) if the barometer is higher to the north than to the south, the wind will be east, southerly if the pressure is higher to the east than the west, and so on. Thus in every case in the Northern Hemisphere " whenever we find an area of low readings, the wind moves round it against watch hands, and whenever we find an area of high readings, the wind moves round it with watch hands " (Scott). The converse is the case for the Southern Hemisphere.
Though in our own country we are accustomed to regard the direction of the wind as a symbol for all that is most variable, yet in other regions there is more regularity in the phenomena of its motion. Two main currents may be distin- guished, the equatorial warm current northwards and southwards towards either Pole, and the polar cold current from either Pole towards the Equator. The cause of these great atmospheric currents is of course the difference of temperature between the equatorial and polar regions. Air
152 Weather Science
heated in the former regions, becoming lighter, ascends, and its place is taken by colder air flow- ing in from other parts. Hadley, nearly two hundred years ago, pointed out that a mass of air moving polewards will be deviated towards the east by the effect of the earth's rotation, since it is coming from a region of quicker rotation towards one of slower movement, and conversely the polar currents will be deviated in the opposite direction (or lag behind). His theory was, however, imperfect in that he assumed that only currents moving along the meridian (i.e., due north or due south) are thus affected, but it was shown by Poisson in 1837, that the effect of the earth's rotation on a freely moving mass near its surface, is to cause a devia- tion to the right in the Northern Hemisphere, and to the left in the Southern Hemisphere, inde- pendently of the direction in which the mass may be moving. Ferrel during 1858 and 1859, developed a theory of atmospheric motions, based on this theorem of Poisson, which he independently discovered. He gave briefly the theory of cyclones, tornadoes, etc., and showed why the law of Buy Ballot is true.
We have said that air currents from the Equator into higher latitudes are deflected by the earth's rotation, and thus in the Northern Hemi- sphere the south wind becomes a south-west one, and the polar current flowing equatorwards be-
Trade Wind Zones 153
comes a north-east one. The latter is known as the north-east trade wind, and blows with great persistency over the Northern Atlantic and North Pacific Oceans. In the Southern Hemi- sphere we have the south-east trade winds. Near the Equator, we have a calm belt occasionally broken by violent storms. The trade-wind zones as well as the intervening belt of calm, shift their position somewhat during the course of the year, being about 10° lower in March than in September. Thus the north-east trade wind blows between latitudes 25° and 3° N. in the spring, but in September its position is between about 35° N. and 10° N. The "calm" and " south-east " trade zones undergo corresponding changes. In Southern Asia and over the Indian Ocean, we find the well-known "monsoons" blowing for one half the year in one direction, and for the other half in the opposite direction. The south-west monsoon blows between May and October over the Northern Indian Ocean, whilst the north-east trade wind blows during the rest of the year.
South of the Equator the south-east trade blows from May to October, and the north- west monsoon "formed from the north-east trade wind drawn across the Equator," blows from October to May.
The following is in a few words the most general account of the circulation of the air, and
154 Weather Science
variation in mean barometric pressure through- out the globe. We have first a zone over the Equator at which the pressure is about 29*8 inches (of mercury), on either side of which there is a belt of higher pressure (about latitude 30 c N. and S.), reaching to 30'2 in. Within this area the trade winds blow throughout the year, except over the north part of the Indian Ocean, where in July they blow inwards towards an area of low pressure and high temperature, the " south- west monsoons," whilst in January, the " north- easterly " winds prevail over this region. Further north and south (polewards) of these regions of higher pressure, the winds in general blow towards the poles.
Throughout Europe the most frequent wind is the south-western, whilst in Asia and Eastern North America the north-west wind is perhaps more prevalent. On the whole, from Hann's investigations, it may be stated that the warmest winds, the southerly and westerly, produce a mean elevation of temperature in Central Europe of from 2° to 6° above the value it would other- wise have, whilst on the other hand, the northerly and easterly winds produce a lowering of from 5° to 7°, the north-east causing the greater depression. The conditions are otherwise in Asia, the most frequent wind, the north-west, lowering the temperature as much as 4*5°, though on the other hand, the south wind raises it by 10*4°, but is a very rare phenomenon
Wind Variation 155
(Scott). This predominance of the south-westerly winds over Europe is not a general phenomenon, but, speaking generally, " in winter the air flows off the land on to the sea, and in summer it flows off the sea on to the land " (Scott). The latter are the " rain bringers," the former usually dry winds. Thus in our country we find the south and south-west winds are the rainy ones (and warm), the north-east wind is cold and