Art. III.—Australasian Weather-charts and New Zealand Storms.

Transactions and Proceedings of the Royal Society of New Zealand, Volume 29, 1896, Page 61

Art. III.—Australasian Weather-charts and New Zealand Storms. By Major-General Schaw, R.E., C.B. [Read before the Wellington Philosophical Society, 9th September, 1896.] Plate V. There are three great forces which produce movements in our atmosphere—heat, gravitation, and the rotation of the earth on its axis. Heat primarily causes motion through the heating of the equatorial zone of the earth by the direct rays of the sun. This heat is communicated to the layer of air in contact with it, which expands, and, being thus lighter than the air not so heated, rises, and the cooler air flows in below. The lighter air flows north and south towards the poles, whence cooler and heavier air flows in to take its place. This is the first and simplest cause of motion in the atmosphere, and it is the result of sun-heat and the attraction of gravitation. But the rotation of the earth causes the stream of air flowing north and south to diverge from meridian lines. Popularly we see why when we consider that air at the equator is moving eastwards with the earth's circumference—which there is nearly twenty-four thousand miles, or at the rate of a thousand miles an hour—while at the poles there is no eastward motion, and at all intermediate circles of latitude the eastward rate of motion is graduated according to the distance from the equator, that at 60° being just half the equatorial rate. Hence all currents of air flowing towards the poles are moving also more and more quickly than the surface of the earth towards the east, and they seem to be from the west of north or south, while all currents of air flowing from the poles towards the equator are also apparently moving in the reverse direction, from the east of north or south; because, as they advance equatorwards, the surface of the earth is moving more and more rapidly to the east, and the currents of air lag behind. But the mathematical law is more complete than this—viz., that in the Southern Hemisphere, between the equator and the poles, any body freely moving over the surface of the earth in any direction appears to be deflected to the left, and, in the Northern Hemisphere, to the right, the amount of deflection varying directly as the sine of the angle of latitude. Hence the deflection is much greater at the antarctic circle than it is, say, in our latitude. This is important for us to bear in mind when we come to

consider the probable causes of the motions of the atmosphere in cyclones (or “lows”) and anti-cyclones (or “highs”). Heat, however, has an additional influence on the motions of our atmosphere, on account of the vapour of water which is present so largely in the atmosphere. Vapour of water contains a considerable amount of heat in an imperceptible or latent condition. This heat has been abstracted from the wet surface where the evaporation takes place (a fact well known to us by the chill experienced as our skin dries when wetted), and it is given up again when the vapour is reconverted into water by reduced temperature and pressure. Thus the apparent anomaly results that the cool air saturated with vapour which the westerly winds bring over our Southern Alps loses there its moisture—condensed into rain and snow—and passes on as the warm, or even hot, dry wind experienced in the plains of Canterbury. Under suitable conditions this setting-free of the latent heat in water-vapour by condensation gives rise to ascending currents in temperate or cold regions like those produced directly by the sun's heat in the tropics, or in localities where the character and configuration of the surface of the ground causes an abnormal heating of a portion of the surface of the ground, and so of the air in contact with it. This latter condition of things is believed to be the cause of most of the comparatively small, but very violent, tornadoes experienced in tropical regions—in America, and Asia, and other places exposed to great heat from the sun's rays. The former is thought to be one of the main causes of those more widely-extended but generally less violent storms experienced in the temperate zones. We see, then, two or three classes of circulation in the earth's atmosphere. First, the constant great exchange between the equatorial and polar districts, which, modified by the rotation of the earth, causes the trade winds north and south of the equatorial belt of calms, the two belts of comparative calms and high barometer in the vicinity of the 35th parallel of latitude north and south, and the counter-trades on the polar sides of these belts. Secondly, the constantly-recurring phenomenon of circular storms moving eastwards in these regions of the counter-trades. And thirdly, the occasional tornadoes or hurricanes which are met with in the tropics. It is with the second class of disturbances that I propose to deal this evening, by the aid of the charts of Australasian weather prepared day by day at Brisbane under the superintendence of Mr. Wragge, the Government Meteorologist there, and sent regularly to Sir James Hector, who has kindly placed them at my disposal. I believe that, it is mainly owing to the energy, zeal, and scientific knowledge of Sir James Hector himself that we are

indebted for the initiation of this system of weather observations and weather-charts, although it so happens that the Governments of Queensland and South Australia, and more especially the former, have been most wisely liberal in providing sufficient funds for maintaining an efficient meteorological department of State. Before proceeding to discuss these weather-charts, however, I wish to draw your attention, first, to the schematic representation of the general circulation of the atmosphere given in the diagram (Pl. V.), and to the general theory of cyclones and anti-cyclones, premising that on both subjects we are far from having attained to accurate knowledge, or to a general consensus of opinion. Starting at the equatorial belt of calms, which, you will observe, lies north of the equator, we see the upper current flowing over towards the poles, while a cooler under current in the opposite direction flows towards the equator to replace the other. This cool current, deflected to the left, as before mentioned, causes the south-east trade winds in this hemisphere. At about 35° latitude the upper current flowing pole-wards is met by an opposite upper current flowing towards the equator, and in this region lies the belt of comparative calms, and high pressure with anti-cyclonic circulation (i.e., against the hands of a watch in this hemisphere), and the currents reappear near the surface of the earth, flowing out from the calm belt north and south. That flowing north, in this hemisphere, or towards the equator, we have already noticed as the south-east trade winds; that flowing towards the southern pole, deflected to the left or eastwards, produces the “brave west winds,” “the roaring forties,” or the “counter-trades” so well known to navigators. In the higher atmosphere above these westerly winds a reverse current exists flowing northwards towards the equator: and here our fairly-certain knowledge ends; but there are strong grounds for believing that the general circulation is completed as shown in the diagram, an upward circulation taking place at about the 60th parallel of latitude, with a very low barometer when the opposing low-level currents meet, just as the downward circulation with a very high barometer occurs at the meeting-place of the high-level currents near the 35th parallel of latitude. Beyond this is pure conjecture at present; but reason leads us to believe that the high currents from all sides meet in the region of the pole, causing, anti-cyclonic calms and high pressure there, and a vigorous outflow on all sides at a low level. This appears to be the vertical or meridional system of general circulation. As regards the two contrary and apparently supplementary systems of circulation in a horizontal, or, rather, gyratory,

manner observed in the anti-cyclones and cyclones, no thoroughly exhaustive theory has yet been established to account for them. Yet these two systems are practically invariable in each hemisphere, and in reverse directions in the two hemispheres. It is clear, therefore, that they depend on fixed laws, however complicated those laws may be owing to the extreme variability of the behaviour of the atmosphere under different conditions of pressure, of temperature, and of moisture; and, also, it is evident that the rotation of the earth rules the directions of their rotation. In a popular way we may perhaps regard the phenomena of the anti-cyclones in this way. The total mass of air between the equatorial belt and the calm belt near 30° latitude is about equal to the total mass of air between that belt and the pole, hence the opposing currents must meet in that zone. As they form the upper strata of the atmosphere, with very little pressure from superincumbent air, that current which is in any degree lighter than the other rises and rides over it, compressing and forcing down the other; and this process of one stratum gliding over another is being constantly renewed by fresh air pressing on behind from the opposite sides. But this sandwiching of opposing currents and their piling up above one another must reduce their onward velocity and increase their tendency to deflect to the left, while at the same time they are being pressed downwards; it follows that, as the fronts of these successive layers of air, after passing over or under the front edges of the opposing layers, are turned to the left and downwards, a downward screw-like circulation to the left (or against the hands of a clock) is set up and maintained. The level at which the outflow of air takes place, and the principal direction of that outflow, depend on the pressures in the vicinity, the principal outflow being towards the centre of lowest pressure. Thus in some measure cyclones are supplementary to anti-cyclones, forming, as it were, drainage-basins for their outgoing air. Yet there is no permanent connection between the two circulations. Rather, apparently, does the cyclone attack the anti-cyclone, which offers a more or less passive resistance to its onward movement. The Hon. Ralph Abercromby distinguishes two classes of cyclones which reach England—those born at the edge of the anti-cyclone tropical belt and moving north-east, and those born in the arctic regions and moving south-east. Ours seem to correspond with the latter. The theory of the cyclonic circulation—of the birth and of the progress of cyclones such as we experience in these latitudes—is still in its infancy. Yet, as the direction of the circulation is invariable (with the hands of a watch in this

hemisphere), and as the motion of translation is uniformly to the eastwards and northwards, although frequently obstructed, or even reversed, for a time by opposing anti-cyclones, it is clear not only that the circulation is according to law, but also that the maintenance of the circulation, and also of the onward march, it may be, are due to some force developed within the cyclone itself. The origin of a whirlwind in a hot region is simple. Some locality under special conditions gets excessively heated by the sun, and an upward current is produced. From all sides cooler air flows inwards and upwards, and so a spiral upward motion is produced, either right-handed in this hemisphere or left-handed in the Northern Hemisphere. But what is the cause of our cyclonic storms, which reach us from the west and apparently from the south? They cannot be due to any such superheating of the ground, for all is sea until we reach antarctic ice-clad land. I put forward the following suggestion, partly derived from the ideas of various writers on the subject and partly from my own ideas. I conceive that our cyclones have their birth on the confines of the antarctic region—at the meeting-place of the north-west counter trade-winds with the low-level outflow from the south polar anti-cyclone. Where these opposing currents meet they cannot glide one over the other as in the case of the high-level currents which form anti-cyclones, because of the presence of the superincumbent atmosphere; therefore they form a calm neutral mass between them, and by the constant pressure both are forced upwards, each being deflected strongly in this high latitude to the left, outside the calm column of air, and so producing the right-handed upward spiral, in which motion the central column itself may partake. But as the air ascends temperature and pressure decrease, until the water-vapour is condensed into rain or snow or hail, and the latent heat is given up to the surrounding air, which expands and rises, and causes a stronger upward draught. This condensation and liberation of latent heat would be greatly accelerated by the presence of snow-clad land at the meeting-place of the winds, and it may be that in some favourable spots on the coast-line of Antarctica high cliffs and glaciers meet the force of the vapour-laden west winds and act as the primary determining influence by forcing the west winds upwards, and, by condensation of the vapour and consequent warming of the air, increasing the upward current. At the level of the surface of the land the force of the opposite current would be felt, causing first gyration and then northerly motion of the whole gyrating system, which would be borne eastwards and northwards by the prevailing westerly winds—northwards because the southern edge of the storm-whirl is circulating against the west wind while the northern edge is circulating

with it, hence the pressure is greater on the south than on the north, and the whole system is forced northwards as well as eastwards. As this upward whirl moves over the ocean it is constantly sucking up water-vapour with the air which enters it below. This is condensed in its turn as it rises into regions of lower pressure and lower temperature, forming clouds, rain, hail, or snow, and liberating more heat, and so giving fresh force continually to the ascending current. The ascending currents must be strongest on the north and east, which are fed by warm air, and less strong on the south and west, which are fed by cold air; and this, probably, is the chief cause of easterly and northerly motion. Our circular storms act as immense irrigating and warming machines, pouring out over sea and land the water taken up as vapour, and diffusing the heat in the upper air originally communicated by the sun to the upper layers of the ocean, while at the same time they mix and purify the air. These rotary storms are frequently met with by vessels navigating the ocean south of the belt of anti-cyclones and high pressure. Captain Maury collected an immense number of statistics on the subject, which place it beyond doubt that they occur frequently at least as far south as 50° or 55° S. latitude, where, although west winds prevail, east winds sometimes occur. In our latitudes they generally first become apparent to our observation in the neighbourhood of Cape Leuwin, whence they advance eastwards along the border of the belt of anti-cyclones, sometimes as rapidly as twelve hundred miles in the twenty-four hours, or fifty miles an hour, sometimes much more slowly. Sometimes they are stopped or even recede for a time, apparently being resisted by the anti-cyclones. Sometimes they partially break through the barrier northwards, but never completely, as far as I have been able to ascertain. Most frequently the storms pass south of New Zealand; but very frequently they extend as far as Cook Strait, through which they pass. More rarely they extend to the north of the North Island, in which case they generally present the normal type of a completely closed circulation, while those that pass south of Wellington usually present the appearance of a partial circulation open to the south. Probably this is frequently due to our having no observatories farther south to record the variations of barometric pressure and of the direction of the wind; but sometimes the circuit may be open to the south, as is the case occasionally in North America towards the north. Always it is to be observed that our great mountain-ranges largely influence the form and progress of the storm circulation, and it would seem also that over the ocean the progress and extension northwards of the depres-

sion is much more notable than over the land. This last observation, if established, would seem, to favour the view that the inherent force of the storm is due to the vapour of water drawn up from the sea, which gives out its latent heat when condensed in the upper strata of the atmosphere. Occasionally the first appearance of one of these depressions is near Tasmania. I have not yet found in the records more than one case where it first appeared at New Zealand. This leads one to suppose that there is some special nursery for storms on the shores of Antarctica, whence one after another is constantly being launched on their ocean voyage. How far a storm can travel before it ceases to maintain its rotary character is yet, I think, uncertain, but a few cases are recorded where the same storm seems to have been traced halfway round the globe. In this Southern Hemisphere circular storms seem to be more regular and persistent than in the Northern Hemisphere, which probably is due to the smaller interference of land in this hemisphere. The trade winds and counter-trades are more powerful in the Southern than in the Northern Hemisphere. The whole air-circulation is freer and more active, and probably this is the main cause of the displacement northwards of the equatorial belt of calms. Cyclones are generally now called “lows,” because the barometer shows a low pressure of the atmosphere towards the centre of a cyclone circulation, and the anti-cyclones are called “highs” for the contrary reason. Now, we have every reason for believing that when a cyclone borders on an anti-cyclone the surplus air from the anti-cyclone pours down into the cyclone, and yet that the circulation within the anti-cyclone is screwing downwards and that in the cyclone is screwing upwards from below. We conceive, therefore, that, as air-motion depends on gravity (apart from the motion due to the rotation of the earth), the altitude of the plane where the air leaves the anticyclone must be above the level of the plane where it enters the cyclone. The latter appears to be at the level of the earth's surface, and for an unknown height above it. Evidently, if the cyclone is not filled up, and so quenched, the air pouring in must escape upwards and outwards, and this we believe it does, while the anti-cyclone is replenished by air pouring in from above. Now, if the level of the atmosphere above the anti-cyclone be above the average level of the atmosphere, what force causes the surrounding air to rise up and flow down into it? To this I can find no answer in books, although some German meteorologists attribute it to the rotation of the earth. Hence it appears possible that really the true height of the column of air near the 35th

parallel of latitude is below the average, and that the high barometer, or greater atmospheric pressure, there is due to the downward motion of the air rather than to the greater height of the column of air. Similarly it may be that the comparatively low barometer near the equator, and the very low barometer near the antarctic and arctic circles and in the centres of rotating storms, may be due to ascending currents of air there rather than to a lower level of the surface of the atmosphere. Should this be as I suppose—namely, that the general level of the surface of the atmosphere is higher at the equator and towards the poles than it is near the 35th parallel of latitude—the circulation would take place, as we know it does, according to the laws of gravity. The velocity acquired by the air sliding down two inclined planes from the equator and polar regions and meeting about the 35th parallel would produce the downward movement there, made gyratory by the rotation of the earth. The friction, of one stratum of air gliding over or beside another stratum of air has been found to be infinitesimally small, so that the velocities of two such bodies of air moving from north and south, and meeting about the 35th parallel, would be the final velocities attained by the accelerating force of gravity at this point, and the collision would cause the masses to fall with the momentum produced by the vertical components of their onward and downward motions. Is not the high pressure observed in an anti-cyclone due—in part at least—to this downward momentum? And similarly, is not the low pressure observed in a cyclone due—in part at least—to the upward motion produced primarily by collision of oppositely-flowing currents of air under the influence of gravity, and sustained by the latent heat set free as the vapour of water in the subtropical inflowing current of moist air is condensed by the opposing polar cold current? I suggest the question, but I am unable to assert that the facts are as I suggest. I do not think that any good instrument has yet been devised for registering the upward or downward movements of the wind. Such instruments installed at high levels and at low levels would give us, I believe, much valuable information on this very obscure subject. As regards the interaction of cyclones and anti-cyclones upon one another in such cases as we observe here, it will be evident that, whatever be the full causes of the high and low barometers in the two systems, they tend to destroy one another, the high filling up the low and the low eating up the high, and both being thus reduced towards the normal pressure. But it is a question of supply and demand. When an anti-cyclone is bordered on both the north and south sides

by cyclones it generally is rapidly diminished in extent and pressure, and the lows advance towards each other; but if this advance of the lows be across Australia they soon shallow and die out—as I conceive; because their inherent force, the constant condensation of vapour of water, is lost in that arid land. Sometimes a large cyclone with very low pressure will rapidly diminish an anti-cyclone with which it collides; sometimes the regular supply to the anti-cyclone from above seems to exceed the drain upon it by the cyclone, and it either is undiminished or increases. If the high be north of the low—the most common case—the effect of the outflow from the high into the law will be to increase the motion eastwards of the latter, owing to the eastward deflection due to the rotation of the earth. If the high be east of the low the tendency will be to retard the motion eastwards and to deflect the low to the south. If the high be south of the low it will retard its progress; if west of the low it will accelerate its progress and deflect it northwards. These theoretical results seem to correspond generally with the facts shown by the weather-charts. In considering the two series of Australasian weather charts which exhibit types of winter storms and summer storms, we observe, first, the belt of calms and anti-cyclones to extend generally farther north in winter (10° to 45°) than in summer (25° to 45°), and thus to be wider in winter than in summer; secondly, that it is broken up into a series of anti-cyclonic circulations, all circulating from north by west to south and from south by east to north; thirdly, that these anti-cyclones seem to be mobile and pliable, yet without any decided tendency to move east or west along the belt, while they appear to offer great resistance to the movements of cyclones.* It is to be noted that ordinarily, as each successive cyclonic disturbance appears on the south-west coast of Australia, it is accompanied by an anti-cyclone following it on its north-west side, which travels across Australia as the cyclone progresses eastward along the south coast, until the Tasman Sea is reached. There a comparatively permanent anti-cyclone exists which is attacked by each successive cyclone, and with its remains the following anti-cyclone blends and restores it to its normal condition as the cyclone passes on eastward. The cyclones generally seem to lose their force when-passing over land, and to increase in force over water. They have a powerful energy of motion in an easterly and northerly direction, which, when unopposed, may carry them twelve hundred miles in twenty-four hours; but they are often checked, or even for a time forced back, by opposing anti-cyclones. They rapidly alter their forms, throwing out great arms of depression where the resistance is diminished by

straits or by the absence of anti-cyclones, or highs. These arms sometimes appear to break off and form separate cyclones; and sometimes a cyclone stopped by a high in front of it may be overtaken by a succeeding cyclone, which coalesces with it, and the increased energy thus obtained forces a passage for, the combined system. In summer the cyclones appear generally on the south coast of Australia, farther to the east than in winter; they are less frequent, and travel somewhat slower. In winter they appear usually as a depression off Cape Leuwin, and they travel to New Zealand—about three thousand miles—in from six to twelve days—in one instance three days—or at the rate of about five hundred to two hundred and fifty miles a day including stoppages, and they succeed each other at the rate of about one in every six days. In summer they travel at about four hundred to two hundred miles a day, and they do not usually occur more frequently than once in twenty days. We can see a reason for this in the greater difference of temperature during winter than in summer between the equatorial and the polar regions, and the consequently more rapid and vigorous circulation of air between them in winter than in summer. To one other point I would draw attention, in conclusion. We have seen that the great circulating systems of depression appear to travel from west to east, and sometimes with great velocity. But the force of the wind does not appear to be influenced by this eastward movement of the systems; the force of the wind depends apparently on the velocity of circulation, not on the rate of translation. We must think, therefore, that the eastward movement of translation, and the corresponding movements north or south of these depressions, are of the nature of wave-motions, not of horizontal motions of the particles of the atmosphere. Such wave-motions are not easily conceived, and they must be very complicated. Fresh particles of the atmosphere are constantly being drawn into the vortex and whirled upwards and outwards as the storm moves on, and the force of the wind depends upon the vigour of this circulation, not on the rate of its propagation. The whole subject is extremely difficult. The enormous forces influencing winds and weather have been in operation for thousands of thousands of years performing the beneficent designs of the Creator, and men hitherto have known little or nothing of the how. Within late years careful systematic observation has given us some little insight into the modes of operation, and I have no doubt that by degrees we shall learn more perfectly the laws governing our weather, and be able more accurately to forecast the immediate future; and that we shall understand better than we do now where and how

our storms arise, what is their motive-force, and what circumstances govern their direction and rate of progress. But in this, as in all true science, our knowledge must be based on long-continued, careful observation and comparison of facts. Some of these facts I have endeavoured to bring before you this evening. My own study of them has given me great interest and pleasure, and I hope that in some small degree I may have been able to communicate to you some of that interest and pleasure. The charts exhibited were those extending from the 23rd December, 1895, to the 9th January, 1896, showing a typical summer storm, which first appeared in the Great Australian Bight on the 23rd December, and the progress eastward of which was traced day by day until it reached New Zealand on the 3rd January, when for four days a north-westerly gale prevailed, followed by a day of moderating west-by-north wind; then, on the 8th, southerly squalls; and on the 9th the storm had passed, and fine weather, with southerly breeze, prevailed. The charts showing the rise, progress westward, and termination of a tropical storm between New Caledonia and North Queensland were also exhibited. New Zealand winter storms were illustrated by a series of charts extending from the 13th June to the 13th July, during which period five storms passed over New Zealand, some of them south of the South Island, some through Cook Strait, and one in a completely-closed circuit passing north of New Zealand; all having been traced from their first appearance near Cape Leuwin. In some cases succeeding storms overtook and became blended with the preceding storms. Examples were also shown of various other storms in which the southern portion of the circuit, with its easterly winds, was more or less completely exhibited near New Zealand.