CHAPTER XII THE METEOROLOGICAL LABORATORY. THE STUDY OF THE ATMOSPHERIC HEAT-ENGINE AND THE CYCLE OF PHYSICAL CHANGES IN THE GENERAL CIRCULATION THE ENERGY OF THE ATMOSPHERE, THERMAL AND ELECTRICAL We have reserved for a separate chapter the enumeration of the special instruments for exploring the energy-changes in the atmosphere. These changes are developed in the most conspicuous degree at surfaces of discontinuity, as for example between air and earth, or air and sea, on the one hand, or between air and water-drops or between layers of different composition and density on the other. The most notable manifestations are solar and terrestrial radiation and the thermal energy associated therewith and the energy of electrical and electro-magnetic discharges. These are naturally associated together and become merged one with another as being related to the all-pervading ether according to an older theory, or to wavemotion of varied wave-lengths according to the more modern view. It is the more desirable in these days to recognise the association of thermal energy and electrical energy as fully as possible because the atmospherics that disturb wireless instruments tuned to definite wave-lengths have been associated with the electrical energy of thunderstorms from the earliest days of radiotelegraphy, and have more recently been associated with rainfall which is one of the chief evidences of changes of thermal energy in the atmosphere. No doubt, in order to be strictly logical we ought to have brought the measurement of rainfall and sunshine into this chapter with solar radiation and atmospheric electricity because sooner or later we shall have to regard the complete cycle: the expenditure of the solar energy in the evaporation of water, with or without an accompanying rise of temperature of the water, the development of instability1 by the saturation of the free air, the consequent convexion and condensation, the development of vast electrical charges by the breaking up of falling drops, the electrical discharges, and the consequent atmospherics, and the return of the condensed water to the earth again after its energetic excursion in the free air. We notice the defect in the arrangement without regret if the attention of meteorologists is thereby drawn to the fact that solar radiation and rainfall may be regarded as almost convertible terms for the expression of the energy of changes in the atmospheric structure. We thereby call attention to the circumstance that rainfall has been regarded as an indispensable meteorological measurement from time immemorial while its counterpart, solar radiation, is not measured at all except at a few special observatories. The reason for this circumstance is sufficiently evident—the measurement of rainfall can be carried out effectively with an instrument of the simplest construction by following a few easy rules, but the exploration of the various aspects of solar or terrestrial radiation or of the manifestations of electrical or electro-magnetic energy in the atmosphere requires the resources of a laboratory and the skill of an experimenter. The apparatus needs adjustment and readjustment, not merely observation or the reading of a permanently adjusted instrument like a clock or a barometer. We have therefore brought together these and other like methods under review in a chapter with the comprehensive heading of the Meteorological Laboratory. INSTRUMENTS FOR THE MEASUREMENT OF THERMAL RADIATION, SOLAR AND TERRESTRIAL In the heading of this chapter we have enumerated in the briefest possible manner the preliminary steps in the progress towards the effective measurement of the changes of energy which are associated with radiation. We have not thought it necessary to recapitulate the steps of that progress in detail. An excellent summary of the instruments for measuring radiation is given in R. S. Whipple's paper before the Optical Society2, and a still more recent 1 1 An ingenuous reader of the proof of this chapter has marked the word "instability" with a note of interrogation. I share his doubts. The word instability is over-worked in the literature of the mechanics of the atmosphere. Stability is used normally to denote a body or system whether at rest or in motion which is so arranged that it recovers its original condition automatically if it is displaced; but here instability means that the column of air is not in statical equilibrium. I crave permission to use a new word "sistible" to describe a condition of possible statical equilibrium, derived from sistere just as stable is derived from stare. A stork standing on one leg is obviously sistible and by use of its muscles is also stable, a drunken man is really sistible but not stable, so is a wireless-mast in the absence of its guys, so also is a layer of atmosphere in convective equilibrium. A vortex-ring on the other hand is stable but not sistible, it would lose its distinctive pressure-distribution if it lost its rotation. A cyclonic depression is regarded by some meteorologists as having a similar property. A column of atmosphere would be sistible if the density of the air diminished continually with height but not otherwise. So I would satisfy the reader's justifiable curiosity by writing unsistibility for instability. In view of the necessity of separate words for separate connotations the demand for a new word is to me irresistible. 2 Trans. of the Optical Society, London, Session 1914-15, pp. 1-63. summary is given in the Dictionary of Applied Physics1. An exhaustive survey of the position of the subject at the end of the nineteenth century is given by Jules Violle in a report to the International Meteorological Committee at St Petersburg, 18992. We simply enumerate the instruments here, referring the reader to the sources mentioned for descriptions and explanations. Solar maximum and grass minimum thermometers For a long series of years meteorological stations have included in their equipment a solar maximum thermometer which consists of a maximum thermometer with a blackened bulb mounted in a glass vacuum-tube. The whole is exposed freely to the sky and the maximum temperature is registered day by day. It has produced some interesting facts, as for example that the solar maximum registered in the Antarctic in R. F. Scott's first expedition (1901-04), in a very cold environment, agreed very closely with that registered in Madras3, but it has not added much to our knowledge of the energy-cycle of the atmosphere. Objection is often taken to it on the ground of want of agreement between different instruments of the same or slightly different pattern, but the basic fact about the measurements is that a single reading of the intensity of sunshine in the day does not give us a measure which can be incorporated with the physical processes of the day. Much more is wanted before the effect of solar radiation can be traced. The same may be said of the grass minimum thermometer, another climatological instrument. It indicates the lowest temperature reached during the night; terrestrial radiation reduces the temperature of the air close to the ground below that registered within the screen, often by as much as 5 t. Such a reading is of considerable value for climatological purposes because frost close to the ground is often of real significance for the plants similarly exposed. But it is a very imperfect indication because the temperature attained depends upon the method and manner of exposure. Poynting has shown that very low temperatures can be obtained by providing suitable thermal insulation of the thermometer and protection from air-currents. S. Skinner has suggested a Dewar vacuum vessel with a cup-shaped top exposed to the sky as an instrument for collecting dew with the name Drosometer; that might perhaps be effective as an integrator of terrestrial radiation but it has not been calibrated and the instrument for calibrating it is what is really required. As a contribution to the physics of the atmosphere we ought to have a measure of the loss of thermal energy due to radiation to the sky, and that the grass minimum does not give. 1 Dictionary of Applied Physics, Macmillan and Co., Ltd., 1923, vol. III, s.v. E. A. Griffiths 'Radiant Heat and its Spectrum Distribution, Instruments for the Measurement of.' 2 Meteorological Office publication, No. 148, London, H.M. Stationery Office, 1900, p. 42. 3 National Antarctic Expedition 1901-04, Meteorology, Part 1, p. 514, London, Royal Society, 1908. 'The Drosometer, or Measurer of Dew,' by Sidney Skinner, Q. J. Roy. Meteor. Soc., vol. XXXVIII, 1912, p. 131. Pyrheliometers The instruments which are in use at the present time for the absolute measure of the thermal energy received by a blackened surface in a limited beam of direct sunlight are the compensating pyrheliometer of Ångström which has been approved as a standard instrument by various international authorities, the silver disk radiometer of Abbot and the actinometer of Michelson. The first, Ångström's instrument, depends upon the comparison of the electrical resistance of two blackened strips of thin metal, one exposed to the sun and the other receiving a compensating amount of heat from an electric current. Abbot's instrument measures the amount of heat absorbed by a water-reservoir from the solarisation of the blackened surface of a small silver disk screened from everything but direct sunlight, and Michelson's instrument is a bimetallic couple, blackened and suitably exposed to direct sunlight, the intensity of the radiation being indicated by the deformation of the couple. Meteorologists will welcome the introduction by Messrs Richard Frères of a simple and effective self-recording pyrheliometer, designed by L. Gorczynski, which is described in the Procès-Verbaux de la réunion à Madrid de la Section météorologique de l'Union géodésique et géophysique internationale. The recording is by the deviation of a galvanometer-needle through the effect of radiation on a thermopile of the pattern designed by Moll. All these instruments require calibrating or standardising in order to obtain the amount of radiation in absolute measure. It has been customary for a long series of years to express the results in gramme-calories per square centimetre per minute, but such a unit is only tolerable when radiationmeasurements are regarded as belonging to a separate physical compartment and the transformations of energy in the atmosphere which are the natural result of radiation are ignored, although the comprehension of those transformations is the very purpose of the measurement of solar and terrestrial radiation so far as meteorology is concerned. We prefer therefore, and shall endeavour to be consistent in the use of, a dynamical unit based upon the C.G.S. system for the expression of the thermal energy of solar and terrestrial radiation. The unit which we have found to be most convenient in practice is the kilowatt per square dekametre, which is indeed the same as a milliwatt per square centimetre; its relation to the customary unit is expressed by the equations: = I gramme-calorie per square centimetre per minute 69.7 kilowatts per square dekametre. I square dekametre is 1076-4 or approximately 1000 square feet. I 1 kilowatt per square dekametre centimetre per minute. = 0143 gramme-calorie per square The actual reading of solar radiation obtained from a pyrheliometer expresses the conditions arrived at when the temperature of the receiving surface has become steady. When that state is reached the thermal energy received by the solarised surface is passed on to its environment as fast as it comes and, as in every other case of radiation, what is registered is the balance of a complicated system of exchanges. The sun's rays are the immediate cause of the rise of temperature of the receiving surface, but the receiving surface itself is also radiating outwards. Some radiation comes from the atmosphere itself in line with the sun and with that part of the sky which is included within the exposure, and the atmosphere itself acts to some extent as an absorbing screen, so much so that the measured intensity of direct solar radiation falls off quite notably as the altitude of the sun diminishes from its maximum, and a greater thickness of the atmosphere is traversed. Hence if we wish to bring the measurements of solar radiation into account in seeking an explanation of meteorological processes we want to understand clearly what we are dealing with. Two ways of approaching the subject are open; the first is to analyse the whole complicated process into its separate parts, to estimate the original rate of radiation from the sun and the rate of simultaneous radiation from the earth and atmosphere, and work out a balance of receipts and disbursements. Since the starting-point of the balance-sheet is the rate of issue of energy from the sun beyond the confines of the atmosphere, which goes by the name of the solar constant, about 135 kilowatts per square dekametre, or 1.93 gramme-calories per square centimetre per minute, and since we are at the moment unable to make direct observations of the solar radiation beyond the confines of the atmosphere we have by some means or other to correct the reading of the pyrheliometer for the losses due to the other items of the balance-sheet. To this object a great deal of effort has been directed by the Smithsonian Institution of Washington, first under S. P. Langley and now under Abbot, Fowle and Aldrich. Various efforts have been made by these investigators, by choosing sites at which the amount of atmospheric absorption is small, to obtain an effective measure of the residual correction. Many estimates of the value of the solar constant have been evaluated which, when corrections have been most carefully made, show residual variations of the "constant" of about 10 per cent. between 125 and 139 kilowatts per square dekametre. Radiometers The other way of approaching the meteorological problem is different. It takes as its fundamental observation the thermal condition of a horizontal black surface exposed to the radiation of the whole of the sky, or of a part of it. That, at any rate, helps us to define what happens at the earth's surface. We have then to make out some sort of estimate as to the effect of changes in the aspect or condition of the surface. What will be the result if instead of a black solid there is white earth, green grass, snow, water or cloud. These things have to be thought of, but the fundamental consideration is how much there is to dispose of. The instruments designed for that purpose are called radiometers and among those in use are Callendar's radiometer consisting of a |