Search

The instrument described gives a continuous record on photographic paper of the change in weight of a naturally exposed grass surface, 330 cm2 in area. The sensitivity is 2 mg cm−1 per cm of record and the time scale is 0.5 hr/cm. Observations suggest that on very calm nights the condensation of vapour reaching the surface by diffusion from the atmosphere is only a small fraction of the total ‘dew.’ The maximum rate of deposition that has been observed on short grass is 0.03 mm hr−1.

Rainfall from non-freezing clouds has been widely reported but detailed observations are sparse. During June 1952, ten days of observations were made during trade-wind weather conditions on the island of Oahu. Rainfall was observed during each of the 10 days, and in no instance did cloud temperatures below 7°C occur. Aircraft were used to obtain temperature and humidity soundings supplementary to the Honolulu radiosonde and to measure the dimensions of the clouds. Raindrop sizes were recorded on a continuously moving dye-treated paper tape at two locations. Data on rainfall were obtained from 5 recording raingauges on the crest of the Koolau Mountain Range and from 32 standard raingauges in an area of 130 m2 Pilot-balloon ascents were made daily on the windward coast of Oahu to supplement the Honolulu Weather Bureau wind data. A continuous record of cloud conditions and motions was made by lapse-time motion-picture photography and supplemented with photographs from aircraft. Rainfall intensities exceeding 3 mm/hr were observed from clouds with bases at 2,000 ft (18.6°C) and tops at 7,500 ft (10.2°C). Raindrop sizes as large as 2 mm in diameter were observed from clouds 6,000 ft deep. Light-intensity rains from small cumulus clouds frequently consisted of raindrops which were relatively large (2 mm diameter) but low in number.

WALLACE AND CLUM (1938) have recently published a paper on leaf temperatures in which they present many interpretations and conclusions which appear to me as incorrect and unsound. Since the writers have published other papers that are thoroughly sound and since many readers may read the summary only or may fail to recognize the fallacies in the interpretations given-for several of them are easily overlooked-and since several of these interpretations and conclusions directly contradict those I have reported, I feel rather obliged to write the present paper in order to clarify certain points relating to factors influencing the temperatures of leaves, especially the influence of transpiration and radiation. Wallace and his colleagues have devised ingenious equipment by which one can obtain continuous records of various conditions such as radiation intensities, temperature, etc., which can be recorded by electrometric methods. By the use of this equipment, Wallace and Clum have attempted to obtain continuous records of leaf and air temperatures as influenced by various factors. From records obtained in these experiments, they concluded, among other things, that transpiration commonly cools leaves by amounts ranging from 6? to 90C.; that "there seems to be no doubt . . . . that transpiration is necessary at times to prevent leaves or leaf parts normal to the sun from being injured"; that a leaf of Heliopsis in full sunlight "normally wilts every day and that the greatest cooling of 7?C. below air coincides with the period of most severe wilting"; that a leaf even if covered by vaseline may transpire rapidly enough to be cooled below air temperature by as much as 50C.; that "'at night leaves are seldom, if ever, below air temperature due to loss (by radiation) to the sky ."; that " vaselining failed to prevent transpiration in leaves," while " waxing leaves with a mixture of beeswax and rosin prevented transpiration and that a leaf covered with this wax becomes heated to killing temperature in the sun solely because the wax prevents cooling by transpiration. An examination of the evidence and reasoning that has led to these conclusions, however, indicates that each of these conclusions as well as other conclusions of Wallace and Clum are entirely unjustified. One of the major sources of error in their work is that the thermopile which they assumed. was at air temperature was probably 20 to 70C. above air temperature at least a part of the time, and thus their comparisons between leaf and air temperatures at these times were incorrect. That the thermopile was at times not at air temperature is clearly indicated by several independent sets of their own data, as well as by evidence from other sources. For example, in

The wind velocities given in this paper were obtained by two anemometers at heights of 12·7 m. and 94·5 m. above the ground. Though the diurnal variation of wind velocity at different heights, has been worked out by numerous observers, there are not many results from anemometers as high as 95 m.; for this reason the ordinary diurnal variation at this height in summer and winter is given below, together with that at 13 m. for comparison. The diurnal variation at the two heights in strong, moderate and light winds is then considered, somewhat on the lines of Hell-mann's paper.1 The vertical gradient of temperature up to 87 m. is also. recorded. Wind gradient must depend largely on temperature gradient, and in the next part of the paper the relation between the difference in wind velocity and the difference in temperature over approximately the same height interval, is worked out for various wind strengths. The various factors controlling the wind gradient are investigated, and the results are found to agree with Taylor's, theory of turbulence.2 The gustiness of the wind is also studied in its relation to vertical temperature gradient and to wind direction. Finally some individual charts, illustrating the various points, are reproduced.

Hitherto, for synoptic and climatic purpose, means of temperature at the surface, and in the upper air, have been obtained without regard to the different classes of air masses, the means being for all masses combined. With the object of obtaining a greater insight into the structure of the atmosphere, it is pertinent to inquire what are the difference between (i) the vertical distribution of temperature, and (ii) the temperature near the surface of the earth, in the different classes of air which cross the British Isles. This investigation gives, for the winter season (December, January, and February combined), for 12 different classes of air over the British Isles: (i) the average and extreme temperatures, the lapse rates and the degree of thermal stability at a number of levels in the atmosphere, and the heights of the 32°F. and 8°F. isotherms; (ii) the daily maximum and minimum temperatures, the mean temperature, and daily range of temperature at the surface at Kew during the 24-hour period, 0-24 h., on days when no important change of air mass occurred during this period. The levels in the atmosphere for which values have been obtained are 1,000, 2,000, 3,000, 5,000, 10,000, 15,000, and 20,000 feet. Over 600 upper air ascents, chosen for the most part during the winters of 1930-31 to 1943-44, and 1,000 surface maximum and minimum temperatures at Kew during the winters 1924-25 to 1943-44 have been used. For comparison, and to obtain an estimate of the changes of temperature and stability, means of temperature at the same levels in the atmosphere were obtained over western Russia in Siberian air, and over the Icelandic and Azores regions in direct polar air and in tropical air. The reality of the difference of temperature between the various classes of air chosen has been examined. A discussion of the physical problems raised in this paper is reserved for another paper.

Many attempts have been made to develop simple, accurate methods of measuring open water evaporation or evapotranspiration, or to devise formulas which assign satisfactory weight to each of the various factors involved in such atmospheric water movements. Very little appears to have been accomplished on the development of a general hypothesis of the relation of water movemeiits to the climates of the earth as a whole. Most previous approaches to the problem have worked from the bases of climatology and physics with the utilization of measurements from watersheds, fresh-water lakes, irrigated fields or lysimeters for obtaining correlations between physical factors and the movement of water from vegetated or non-vegetated surfaces. This paper presents a hypothetical, planet-wide pattern of atmospheric water movements based on natural vegetation, which latter supposedly reflects environmental conditions accurately. In the development of a diagram with logarithmically spaced value lines of temperature and precipitation to determine world plant formations or life-zones, Holdridge (1947) found unexpectedly that one of the three requisite sets of parallel boundary guides for the hexagons was potential evapotranspiration ratio lines. This made possible the extrapolation from the diagram of a value of annual potential evapotranspiration for any station which had been located on the chart by the plotting of mean annual data of biotemperature and precipitation. Subsequently, Holdridge (1959) (1960) found from his plant formation diagram that the same value for potential evapotranspiration could be determined with greater ease and more precisely by simply multiplying the mean annual biotemperature in degrees C by the factor 58.93. Further consideration of the interesting direct relation between biotemperature and potential evapotranspiration, coupled with observations of vegetation over a wide range of conditions in the western hemisphere, has led on to the derivation of the nomogram for atmospheric water movements from areas of climatic vegetation associations, as presented in Figure 1, and of formulas for other types of vegetation associations given in the text of this paper. However, before proceeding with the presentation of the nomogram and formulas, it appears desirable to define the terms utilized and to discuss the concept of potential evapotranspiration, which is of special importance because it serves as a basic value to which other atmospheric water movements may be related.

This paper describes the three dimensional airflow and precipitation within small ‘polar’ depressions which sometimes cross Britain during northerly outbreaks and which in winter can be responsible for heavy snowfalls. The data used in the study consist of Doppler and conventional radar information, together with routine synoptic data and sequential radiosonde ascents from the radar station. Three dimensional airflow was derived from the radiosonde data assuming that wet bulb potential temperature was conserved. Horizontal and vertical air velocities were also derived from the Doppler radar measurements. Previous knowledge of polar lows is meagre; they are generally thought to be shallow features resulting from enhanced convection within cold air flowing over a warm sea. However, the well-developed polar low which is the main subject of this paper is shown to have been an essentially baroclinic disturbance. Although enhanced small-scale convection occurred in one sector, the main area of widespread precipitation associated with the polar low was produced not by small-scale convective overturning but rather by slantwise convection within a narrow tongue of air ascending steadily at about 10 cm s−1. The speed of travel and short wavelength (900 km) of the polar low in this study are consistent with its having formed in a region of enhanced baroclinicity within the polar air below 850 mb rather than in the major baroclinic zone bounding the polar air mass. Considerable low-level baroclinicity within the polar air is also shown to have been present during the formation of other intense polar lows.

The paper describes a relatively simple model of a non-precipitating cumulus cloud that grows by the ascent of successive spherical thermals through the residues of their predecessors, mixing with the surroundings being determined by the relative velocity and the radius of the thermal. The model appears to account for several important features of the cloud structure and the droplet-size spectrum that are not produced in models of a single parcel or plume. In particular, it produces modest clouds in 30-40 min in which the vertical velocity, cloud depth and liquid-water content agree quite well with observations. The computed droplet spectra closely resemble measured spectra and reproduce the bimodal structure observed by Warner. In model maritime clouds containing small concentrations of droplets, the spectra broaden quite rapidly and produce droplets of r = 25 μm by condensation on nuclei of m = 10−11g in concentrations of order 100 m−3 within half an hour, beyond which size they may continue to grow rapidly by coalescence to precipitation size. However, in continental clouds containing droplet concentrations of a higher order, it is difficult to produce a significant number of droplets of even 20μm radius.

In a paper by Davies and Oakes (1962) concerning the problem of formulating a model of the general atmospheric circulation, the boundary condition used at the surface of the earth involved the assumption that the surface meridional, zonal, and vertical velocities are zero. In this note the effect on the model is evaluated by using as boundary condition at the earth's surface the more realistic assumption that the Reynolds stress is proportional to the first power of the surface wind speed.

Much micrometeorological work to date has assumed, or tried to ensure, that conditions of horizontal homogeneity exist at and near the earth's surface. The present paper reports an observational study of the variation in the meteorological elements with distance and height downwind of a pronounced discontinuity from dry to wet surface conditions. In particular the temperature and humidity fields were observed together with the changes in the radiative-balance components and in the wind structure. It was found that temperature and humidity changes at a height of 5 cm within 16 m of the discontinuity were commonly as much as 5$C and 5 g m−3 respectively. Theoretical treatments of the problem are outlined, and a slight modification of Philip's (1959) analysis is used to compare observed and predicted changes in temperature and humidity with distance downwind up to a height of 150 cm. It is found that the observed humidity changes are in good agreement with expectations, but that the measured temperature changes are about half those predicted. Changes in the evaporation rate in various distance intervals downwind from the leading edge have been computed. These are compared with an estimate of the rate which would have occurred under the same conditions from a wet area of infinite extent. The discrepancies between theory and observation are attributed to the fact that the available theory fails to take into account the (demonstrable) differences in the aerodynamic properties of the adjacent surfaces. There is scope for further theoretical and experimental work on this matter.

Several models expressing the decrease of solar radiation in vegetation canopies have been proposed. The models are similar in (1) including a method of projecting the shadow of one more leaves onto a horizontal plane and (2) including a method of relating the projected shadow to the decrease of solar radiation. The models differ in the equations used and in the assumptions made. The models used the binomial, the Poisson, or some modification of the binomial to relate the projected shadow to the radiation at different levels. When the models were tested against the percentage of sunlit area at different levels in the canopy, the best fitting models was one proposed in this paper in which the shadow of the leaf is projected as a shadow of a cone and the binomial is used to express the decrease of radiation. Negative exponential models usually overestimated the percentage of sun. Including the shadow cast by twigs and branches in the prediction improved the agreement between the observed and the expected pe...

This paper describes a representation of the distribution of sensible and latent heat from the surface through the atmospheric boundary layer which has been formulated for use in a 10-level primitive equation model atmosphere. The transfer process is represented in two parts : (i) the transfer of energy across the Earth's surface into the lowermost 100 mb layer of the model atmosphere; and (ii) the subsequent redistribution of this energy through two or more such layers by small-scale convection. The fluxes of energy across the surface are calculated using empirical ‘bulk aerodynamic’ relationships. In land regions consideration of the energy balance at the surface is also necessary, and diurnal variations of radiation are taken into account. The redistribution of energy by small-scale convection is represented by convective adjustments which ensure that a certain neutral lapse rate of temperature is never exceeded. Some results of the incorporation of these effects into the 10-level model are described.

Observations of hypolimnetic currents in Lake Mendota, intcgratcd over entire seiche periods, indicate that steady motions exist below the thermocline. Seiche currents and turbulcncc are shown to bc insufficient to explain the observed currents. Hutchinson ( 1957, p, 287, p, 341) has suggested that deep water currents, at least in medium-sized stratified lakes, are primarily due to internal seiches. He also quotes Mortimer ( 1953) to the same effect. On the basis of the small amount of hublished data on the observed currents below the thermocline and the fact that both the usual “surface” seiche and the internal seiche must involve readjustment of mass distribution throughout the entire lake, such a conclusion might be reached on theoretical grounds. It is the purpose of this paper to examine the role of the seiche in producing hypolimnetic currents utilizing statistical data on such currents observed in Lake Mendota. THE OBSERVED CURRENTS IN LAKE MENDOTA During the past 10 years, about 2,000 current observations have been made in Lake Mendota. All were made by the “free drag” method, i.e., by observing the displacement of a vane or drag suspended at the desired depth and allowed to drift with the current. This method, which may be used to obtain currents integrated over any time interval desired, is more sensitive to small current velocities than the finest propeller type current meters. The bulk of the current measurements made in Lake Mendota1 are summarized in Figure 1. Most of these data were obtained with “drags” half a meter square, and so indicate the mean current in a layer 0.5 m thick centered at the indicated depth from the surface. In the 34 cases indicated for the 1 Data from the entire lake arc included, though most are from the southwestern portion. layer O-O.25 m, drags both 25 cm and 10 cm high were used. As might be expected, the mean and modal velocities at the surface of the lake are larger than at a depth of 1 m as indicated when the data from the O-O.25 m and O-O.5 m layers are compared to those at 1 m. (Table 1.) Surprisingly, the currents at 2 m do, not fit the general pattern of decreasing mean with increasing depth. Furthermore, the number of currents in the interval O-O.99 cm/set might be expected to increase with depth, and generally does except for the striking anomaly of the 2-m depth. Thcsc features suggest some cause other than the usual wind drift explanation of the nearsurface currents, possibly a larger than expected role of that level in the seiche circulation. Most of these current measurements were made in midsummer when the thermocline is generally at a depth of 5 to 8 m and generally less than 10 m. Thus, the data from 10 m are largely for currents within the hypolimnion, while for greater depths there TABLE 1. Mean and modal velocities for currents in Lake Mendota Depth MC%11 Mode Number cm/see cm/set of c.?ses O-O.25 m 9.36 6-7 (10-11) 34 O-O.5 m 5.93 4-5 107 0.75-1.25 m 4.48 1-2 583 1.75-2.25 m 4.71 3-4 ( l-2) 367 4.75-5.25 m 3.23 l-2 255 9.75-10.25 m 2.78 2-3 248 14.75-15.25 m 1.67 o-1 26 19.75-20.25 m 7.7 not dcfincd 10

The paper contains a study of the day-to-day, seasonal and latitudinal distribution of ozone over the southern hemisphere in comparison with those observed in the northern hemisphere. It has been observed that there is more ozone over the middle latitudes of the southern hemisphere in summer, autumn and early winter, than in the northern hemisphere, whereas the spring maximum values are more or less the same in both hemispheres, suggesting lesser destruction of ozone from spring to autumn in the southern than in the northern hemisphere. The distribution of ozone with height computed from umkehr observations over Aspendale by method B has been compared with that over Tateno in Japan. The results are discussed in relation to the idea of a poleward flow of air from the lower stratosphere and the neighbourhood of the equatorial tropopause in the southern hemisphere. It is concluded that the mean meridional circulation in the lower stratosphere of the southern hemisphere is weaker than, or at any rate only as vigorous as that in the northern hemisphere in spring.

There exists at the present time no adequate method of determining either the rate of respiration or that of photosynthesis of a plant under entirely normal conditions. It is necessary to confine a plant or a portion of a plant, such as a leaf, in a relatively small volume. The gaseous exchange of the plant can then be determined either by continuous displacement of the air surrounding the plant or by determination of the change of composition of the atmosphere when the plant is kept in a confined space. The principle applies not only to land plants in air, but also to aquatic plants surrounded by water. Although most of the evidence permits the conclusion that under such experimental conditions the photosynthetic and respiratory activities are the same as under normal conditions in nature, recent experience has shown that great care must be exercised in interpreting the results of the rates of gas exchange obtained under such experimental conditions. The purpose of this paper is to demonstrate the effect of changes in the CO2 concentration of the air on the apparent rate of respiration of excised leaves. In measuring the rate of respiration or photosynthesis under different conditions of carbon-dioxid content of the air about the leaf, the correct interpretation of results depends upon an exact knowledge of the manner in which the plant or leaf responds to a change in these external conditions. In determining the rate of photosynthesis on the basis of the CO2 fixed, it is necessary also to determine the rate of respiration. It has been very generally assumed that the latter, determined in the dark, remains the same during illumination. A correct estimation of the rate of photosynthesis depends, therefore, largely upon an exact determination of the respiratory value. Moreover, for practical experimental purposes it is often desirable that the photosynthesis determinations be made in an atmosphere enriched in CO2. The rate of respiration is very often determined by measuring the CO2 emitted in a definite period of time when a stream of air free [The journal for July (II: 4I7-492) was issued Aug. I4, I924.; 493

The meteorograph records of the almost daily ascents over England during the months of April, 1939, and September, 1930, have been evaluated according to a method described in an earlier paper. (Q.J.70, 1944, p. 293). The humidities so obtained indicate (1) that supersaturation with respect to ice is very frequently found in the upper troposphere; (2) that, within the stratosphere, dry air of a humidity mixing ratio of about 10−5 sometimes does occur, which can only be explained as tropical stratosphere air; (3) that, on one occasion, very dry air from the stratosphere appears to have been drawn into the uppermost regions of the troposphere.

This paper discusses the thermodynamic transports of heat, liquid water and (briefly) water vapour by non-precipitating cumulus convection. It is shown that because of the irreversible mixing between cloud and environment, there is a downward transport of enthalphy in the cumulus layer. A lapse-rate adjustment model relates stratification to the life-cycle of a model cloud parcel. A sub-cloud layer model specifies the lower boundary of the lapse-rate model, and the convective transports through cloud-base. Budget equations together with the lapse-rate model, and its time dependent boundary conditions, predict the time development of the cumulus layer, and show the dependence on large-scale mean vertical motion, cloud-base variations, and the surface sensible heat flux.

The downwelling velocities associated with Langmuir circulations in Lake Ontario correlate well with wind speed under surface cooling. With surface heating the correlations with wind speed are poor and calculations of a stabihty parameter indicate a different proportionality. Observations of surface phenomena indicate that surface slicks and capillary wave damping arc in operation at the point of downwelling. Since Langmuir (1938) first observed surface windrows and suggested that roll vortices existed in the surface waters of the ocean, observations of these phenomena have been made in various regions (Assaf et al. 1971; Faller and Woodcock 1964; Myer 1969; Sutcliffe et al. 1971). Enough evidence has now been collected to show that Langmuir cells are perhaps the most important mechanism for the vertical transport of heat and matter in the surface waters of lakes and oceans. Despite their importance there is as yet no agreement as to the mechanism of formation and maintenance of these circulations, Scott et al. ( 1969) discuss six possible mechanisms and also suggest that the generating mechanism may vary from site to site. There appear to be few or no data published pertaining to Langmuir circulations in the Great Lakes. This paper presents the results of a l-month study on Lake Ontario and discusses the various circulation generating mechanisms in the light of observed data. The primary purpose of the study was to determine the speed of downwelling and to relate to this the vertical motions of phytoplankton populations. We gratefully acknowledge the assistance given by the many members of the Canada C,enter for Inland Waters who provided facilities, equipment, and assistance. Particular thanks to M. Donelan and F. M. Boyce, M. Mawhinney, and D. Hore. Thanks also to our students and technicians : C. D. Turner, S. Humby, and R. Smith. 1 This work was supported by grants from the National Research Council of Canada. METHODS The observations were carried out from a moored barge and tower installation 6 km NW of Niagara-on-the-Lake (Ontario) in 12 m of water. The barge and towers were positioned and serviced by the Canada Center for Inland Waters, Burlington, Measurcments of wind speed at 4 m, solar radiation ( total shortwave), air temperature (wet and dry bulb), and surface water temperature were made from the barge. The downwelling beneath the streak was determined by measuring the downward motion of neutrally buoyant drogues consisting of a rectangular metal sheet attached to a central float (suggested by Langmuir 1938). The drogues were calibrated for zero buoyancy ( or nearly so) in a 1.5-m-deep tank filled with surface water. The sinking rates in the streaks exceeded the sinking rates in still water by up to 10 times, indicating clearly that some downwelling existed. On many occasions the drogues drifted many meters sideways on reaching a depth of 5 to 6 m, indicating a possible roll circulation, The surface streak position was identified by the use of computer cards strewn on the surface from a small launch. Over 70 measurements of the rate of the downwelling jet were made, as well as observations of streak separation and surface wave

In this paper are presented some solutions of analytic models in which large-scale velocity fields act on initially large-scale distributions of temperature. Phenomena very similar to atmospheric fronts both surface and upper tropospheric are described. At the surface front there is a tendency to form discontinuities in velocity and temperature in a finite time. The upper tropospheric front and its jet stream are associated with the descent of a tongue of stratospheric air. The main assumption underlying the models is the justifiable one of geostrophic balance across but not along the front.

The theory of scattering and attenuation by rain, snow and cloud is reviewed and theoretical results are presented in the form of equations, tables and graphs, so that the radar response to meteorological particles can be calculated at six wavelengths (10, 5.7, 3.2, 1.8, 1.24 and 0.9 cm) and various temperatures. Particular emphasis is placed on developments since Ryde's comprehensive paper in 1946. Published experimental results are compared with the theory. All results computed from the theory are contained in Tables 4 and 5. The attenuation by water vapour and oxygen is given in an Appendix.

A result from a previous paper (Priestley 1947) indicates the existence of a broad upper limit to the size of pressure systems, and a lower limit to the distance separating the centres of two significant closed pressure systems. The pressure field has thus an inherent scale of pattern which varies with latitude, imposed by the essential geometry and dynamics of atmospheric flow. Orders of magnitude deduced theoretically agree well with experience. The implications contribute to an understanding of the mechanism of the mean global circulation.

This paper describes some features of the climatology of the free atmosphere over India which are related to the onset and withdrawal of the monsoon.

During 1955, 46 measurements of frost point and temperature were made at varying altitudes up to about 50,000 ft using a Dobson-Brewer pressurized frost-point hygrometer and a flat-plate thermometer mounted on the Canberra aircraft of the Meteorological Research Flight. The general character of these ascents in the lower stratosphere confirms the conclusions of earlier papers (Murgatroyd, Goldsmith and Hollings 1955). The mean frost point at 48,000 ft for the ascents made in 1954 and 1955 is -117°F and at 46,000 ft -115°F, with a standard deviation of about 3°F in each case.

Being well aware of the profound importance of evaporation as a factor affecting plants and animals, biologists are careful to regulate or observe it in experimental work. Evaporation is a dynamic process and represents the net difference between the number. of water molecules leaving a surface and the number returning to the surface during some unit of time. It is frequently difficult or impossible to control the number of water molecules leaving the surface of living tissues, but it is often possible for us to control the rate at which water molecules return to these surfaces from the surrounding atmosphere. This is usually attempted by regulating the relative humidity of the atmosphere. Biologists sometimes fail to realize that identical relative humidity values do not indicate identical atmospheric moisture conditions unless the temperature is also the same. If identical atmospheric moisture conditions are desired at different temperatures it is necessary to maintain different relative humidities. Although this fact has been pointed out earlier by a number of different workers (cf. Bolas, '26; Braun-Blanquet, '32; Day, '17; Imms, '31) it seems worth emphasizing here again. It is the purpose of this paper to show graphically some of the significant relations between relative humidity values and atmospheric moisture conditions and to point out some of the advantages of using vapor, pressure deficits rather than relative humidities in measuring the effect of atmospheric moisture conditions upon living organisms. It is important to realize that, when considered independently of other factors, the actual amount of water vapor present in the air has little, if any influence upon evaporation (Day, '17). The dryness or wetness of a climate is not correlated with the amount of water vapor in the atmosphere. The very " dry" atmosphere of Death Valley, California, probably the most arid region in the United States, contains on the average in July almost exactly the same amount of water vapor per unit of volume as does the " moist " atmosphere of Duluth, Minnesota, at the same time of the year. BraunBlanquet ('32) cites an equally striking illustration of the lack of correlation between the " dryness " of the atmosphere and its moisture content by stating that the arid deserts of Morocco have a relative humidity of 90 per cent in the summer months. Very obviously the importance of atmospheric moisture as a factor affecting the rate of evaporation lies not in the absolute quantity of moisture present but in the relation between the amount present and the amount that could exist under the same conditions without condensation. 277

Two equations are presented giving the vertical eddy diffusivity coefficient K, in the thermocline of a well-stratified lake as a function of its lake mean rate of downward displacement and intensity. Both equations are derived by transforming the heat transport equation onto a quasi-Lagrangian coordinate system, the origin of which moves down with the lake mean thermocline depth. Application of the equations to Lake Ontario yields summer mean values of K, in the thermocline of 0.02 to 0.07 cm”/sec. During an exceptionally quiet period in July and August 1907, K, approached the value of molecular diffusivity. The eddy diffusivity coefficient can bc determined by direct methods, such as dye releases or shear measurements, or by indirect methods making use of changes in the temporal and spatial distribution patterns of naturally occurring parameters. In layers with strong vertical density gradients, such as thermoclines, special problems arise that complicate determination of the eddy diffusivity coefficient by direct measurement; indirect methods, however, may still be applicable. Using thermal data from Lake Ontario, I have developed a quasi-Lagrangian equation to estimate a mean value for the vertical coefficient of eddy diffusivity in the thermocline of a stratified lake. In contrast to the more traditional Eulerian formulations, the quasi-Lagrangian diffusivity equation is written with reference to a moving coordinate system. The origin of this system is fixed with respect to the thermocline rather than the surface, and thus moves up and down with internal waves, seiches, and seasonal changes in thermocline depth. The quasi-Lagrangian l I am indebted to Dr. D. B. Rao who, through many stimulating discussions, greatly contributed to the work described in this paper. I am aIso grateful to the Marine Sciences Branch for their kind permission to let me complete the work started during my earlier assignment with the Inland Waters Branch. equation can take one of two different forms, depending on the number of simplifying assumptions that can be made. In a large, deep lake KZ can be expressed as a function of the lake mean rate of downward movement and the intensity of the thermocline. In shallower lakes an additional heat-flux term is introduced. Derivations of the Eulcrian and both quasi-Lagrangian equations for the diffusivity coefficient are presented below. Lake Ontario data have been interpreted using all three equations; the quasi-Lagrangian equations, within the narrow limits of their applicability, yield more meaningful results than the Eulerian method.

This paper describes the role of sonic disturbances in redistributing the mass of the atmosphere when heating or cooling is applied to a part of it. It is shown that the effect, for meteorological purposes, of heating is a local reduction of density and surface pressure. The pressure field thus set up causes an advective disturbance which later modifies the field by transport of mass with the wind. It is concluded that no disturbance of meteorological significance ever travels from one place to another otherwise than by advection.

The paper shows the so-called ‘agitation’ of the atmospheric electrical elements in relation to the atmospheric turbulence, and demonstrates in a series of examples the behaviour of the agitation and its application to the analysis of atmospheric electrical processes. We can consider it as a further element in the coupling between atmospheric electricity and meteorology. It appears that the central significance of the coupling can be seen in the various atmospheric exchange processes and their different electrical effects.

Wind blowing across a water surface will transmit momentum to the water and cause a surface current. The ratio of water velocity to wind velocity is called the wind factor. This paper presents observed values of the wind factor obtained as median values and by regression analysis of wind velocity vs. water velocity. The data show that the wind factor is a discontinuous function at a critical wind speed. Water velocity in the surface layers increases with wind velocity until a critical wind speed is reached, and then it decreases. This observation is in agreement with Munk’s (1946) theory of a critical wind speed for air-sea boundary processes, which yields air-sea boundary instability for winds exceeding 6.5 m/sec. The observations taken in Lake Mendota yield a critical wind speed of 5.7–6.1 m/sec. The present results were arrived at after a study of 356 observations.

Several workers have reported enhanced radar echoes, from regions well above the freezing level, in layer-type rain clouds. Usually, there is a decrease with time in the height of these ‘upper bands,’ as Bowen has called such enhanced echoes. This paper describes an investigation into upper bands which has been made at Cambridge, using a 3-cm radar with a vertically directed beam. The mean downward velocity of upper bands at a height 9,000 ft above the freezing level is found to be about 20 ft/sec, the velocity generally decreasing systematically with time. Such velocities are much greater than the rate of fall of the ice particles which are shown to be responsible for the upper-band echoes, which, it is suggested, are produced by precipitation streaks in a region of vertical wind shear. A simple theory is worked out which gives the apparent rate of fall of an upper band in terms of vertical wind shear, the horizontal wind speed near the cloud top, and the terminal velocity of the ice particles forming the precipitation streak. Measured values of the fall speed of upper bands agree reasonably well with those predicted by this theory. A previous theory of upper bands (Bowen 1951) is shown to be untenable, though it is possible that the upper bands observed by Bowen in Australia have a different origin from those observed at Cambridge. It is proposed to make further observations to settle this question.

The paper describes an experimental programme to study the characteristics of momentum and heat transport in the first 22-6 m of the atmosphere. Sonic anemometers and fine platinum wire thermometers were used for flux computations by the eddy correlation technique. Two drag plates were used to measure the surface stress. The results indicate that the fluxes are constant with height to within ± 20 per cent provided a long enough averaging period is chosen. Shorter averaging periods (15 min or less) show considerable variation in the fluxes and in their vertical gradients especially during unstable conditions. It appears that much of this variability can be attributed to the strong effect on the fluxes of submesoscale circulations. The momentum flux gradient seems to be particularly sensitive to these circulations. The most consistent estimates of u* were obtained from the drag plates. They show much less scatter when used in similarity relationships than estimates obtained from the eddy correlation technique at higher levels. However, because of inherent difficulties in installing the drag plates in such a rough surface, their absolute reading was obtained by calibration against the eddy correlation measurement. The correlation γUW, between the longitudinal and vertical components of the wind, is generally much smaller than γWT, the correlation between the vertical component and temperature. With increasing instability γWT increases to a value of about 0-6 while γUW drops to negligible levels.

The rǒle of mixing in the production of fogs and clouds has been the subject of frequent discussions, and many authors contend that turbulent mixing is an important factor in the formation of fog. This result is derived from the well-known Bezold effect and supported by observations showing that fogs sometime occur together with winds of considerable strength. The writer (1939, 1940, 1941) has maintained that turbulent mixing tends to prevent fogs from forming and dispel existing fogs. The aim of this paper is to evaluate the amounts of condensation caused by mixing under various atmospheric conditions. Throughout it will be assumed that the air contains a sufficient amount of condensation nuclei, with the result that no appreciable supersaturation can be maintained.

This paper contains:— (a) A description of the use of the wet-bulb potential temperature for determining changes in level of air in situations when all levels of the original air column have not the same trajectory; (b) A derivation of the amount of subsidence which takes place in air moving off the ridge of a high, amounting at times to 1.7 km. in 24 hours from the 3 km. level, and a demonstration of the subsidence of an inversion in the centre of a high of 300 m. in 24 hours; (c) A demonstration of the overrunning and uplift of a stream of warm air over another stream of colder air; (d) Some remarks on the divergence of the subsiding air as it moves off the high pressure region; (e) A discussion of the changes in moisture content which took place in the masses of air considered.

The paper contains a study of the mean meridional distributions of ozone in different seasons. The vertical distributions of ozone at a number of stations at different latitudes in the Northern Hemisphere were worked out from the intensities of light scattered from the zenith sky on individual clear days using method B, i.e., the method of curve fitting, with suitable correction for secondary scattering. Diagrams of vertical distribution corresponding (1) to high ozone amounts (frequent in winter and spring) and (2) to low ozone amounts (frequent in summer and autumn) are shown. Mean distribution diagrams are also given for the months of March, July and November. An important feature of the distribution is the strong increase in ozone amount in March below 18 km when the latitude increases beyond 30° . The results are discussed in relation to the idea of a poleward flow of air from the lower stratosphere and the neighbourhood of the equatorial tropopause, as suggested by Dobson and Brewer from frost-point measurements over the United Kingdom. The existence of a stratospheric warm air pool over the middle latitude in winter and spring as revealed by sounding-balloon ascents is confirmatory evidence of this. The influence of the pool of cold air which exists in winter above 20 km in the stratosphere of the dark polar night is also considered. It is suggested that in winter and spring there is an ozone-regenerating cycle due to meridional circulation in the stratosphere of tropical and middle latitudes, which carries air from the lower equatorial stratosphere to lower levels in the stratosphere of middle latitudes and puts back some of this accumulated air to higher levels in the equatorial stratosphere above 25 km. The more vigorous this circulation, the greater will be the rate of ozone storage in the lower stratosphere of extra-tropical latitudes.

The paper discusses relationships observed between ozone and the upper-air measurements made at Brisbane, Aspendale and Macquarie Island. The correlation coefficients between the short-term fluctuations of ozone and the temperatures at 100, 200 and 300 mb levels at these places are presented. In general, high ozone was observed to be associated with the sinking of the tropopause, descending of stratospheric air, warming of the lower stratosphere and a southerly flow of air in the lower stratosphere. At Macquarie Island, an instance of the ozone fluctuations in the baroclinic waves of the polar night westerly vortex suggested that the middle stratospheric waves contributed to the unexplained long term variance in total ozone. The meteorological parameters at the 200 mb level did not reveal the type of oscillation shown by the spring maximum level of ozone with a periodicity of 24 months. From the study of the 60 mb temperatures, it is concluded that the middle stratospheric circulation is playing an important role in deciding the spring level of ozone in middle latitudes of the Southern Hemisphere.

Studies of the vertical distributions of CO2 fluctuation in a cornfield were made in the 4-to 0.25-cycle/min fre- quency range. Amplitude of fluctuations decreased with height above the ground. Frequency in this range ap- peared rather constant, however. Sources and sinks for CO, within the cornfield contribute to the fluctuations; however, eddy structure originating inside and/or outside the cornfield plays an important role too. I T has long been known that the CO 2 content of the air near the earth's surface undergoes temporal and spatial fluctuations arising from inhomogeneous dis- tribution of sources and sinks of carbon dioxide as well as the degree of air mixing by meteorological processes (Huber, 1952). The fluctuations can pre- sent serious problems when sampling to obtain repre- sentative values of the carbon dioxide content of the air. We have found this especially true in our studies of CO2 exchange above active plant communities. We have been well aware of the shorter time fluc- tuations of IA to 2 cycles per minute, and as a conse- quence have attempted to obtain integrated samples for 10- to 20-min sampling periods. However, previ- ously no attempts were made to study the short-time fluctuations themselves. This paper will discuss such a study in a field of full-grown corn (Zea mays L.). Although the site has been described previously (Lemon et al., 1963), certain features of the location are pertinent to this study. The corn was relatively isolated as a frost-sensitive crop in an area of mixed vegetation of frost-hardy species. Also, the wind fetch over the corn to the sampling site was approximately 100 m. The significance of these two points will be discussed later.

It is well known that if special types of days are considered separately, systematic diurnal temperature variations may be found to exist at polar stations even during the dark season. In the present paper it is shown that, in all probability, this phenomenon is of trivial nature and that a perfectly simple explanation of its origin is possible.

Precise numerical calculations of the flux divergence due to narrow regions of the ozone 9.6 micron band were made, using the relative strengths and positions of the spectral lines calculated by Kaplan, Migeotte and Neven (1956). The absolute line strength was estimated on the basis of Walshaw's (1957) laboratory measurements. Spectral models or Curtis's approximation were not used, but Lorentz shapes and uniform widths were assumed for all lines. Calculations up to a height of 33 km were made for two ozone ascents of Brewer and Milford (1960). The atmospheric heating rates obtained for narrow regions of the band were then combined to estimate the heating rate by the whole band with results in qualitative agreement with, though nearly twice as great as those of Plass (1956a). The effect of the radiative temperature of the atmosphere below the ozone layer (whether of ground or cloud top) appears to be great. The radiative heating by ozone in the 10-20 km region is 0·2 or 0·3 degrees C per day for a ground temperature of 10°C, but may disappear altogether, if cloud is present just below the tropopause. Using a larger computer, the procedure could be extended to cover the whole band, thereby avoiding the approximate extrapolation employed in this paper. The procedure could be applied to any absorption band, and is flexible to permit the use of any desired pressure and temperature dependence of the shape, width and strength of the spectral lines; results from it might serve as a standard against which conclusions reached by approximate methods can be checked.

Meteors falling into the earth's atmosphere produce, by a process which was described briefly, a trail of ionised air. By studying the reflection of radio waves from these trails (‘meteor echoes’) the atmosphere at heights near 95 km can be sampled several hundred times per hour throughout the day, simply and at low cost. The paper summarized results obtained by many workers at Jodrell Bank since 1946 (see references). The intensities and heights of appearance of meteor echoes lead directly to estimates of air density (about 3 × 10−9 g/cm3 at 95 km) and of scale height (about 6·5 km at 95 km), while the durations of meteor echoes afford a means of measuring the atmospheric diffusion coefficient. The measurements were compared briefly with independent estimates. By a special technique, movements of meteor trails of the order of 1 m are detected, and hence winds measured at heights of the order of 95 km. Although ‘winds’ obtained by ionospheric methods may represent travelling disturbances, the ‘meteor’ winds are true mass movements of the air. It is found that there is usually a prevailing wind of the order of 10 m/sec, upon which is superimposed a semi-diurnal variation which may have an amplitude of 30 m/sec. The phase and amplitude of the semi-diurnal component show seasonal changes. The winds increase with height, the shear having a value of about 2 × 10−3 sec−1, a value comparable with that found in the troposphere, although there is no change in wind direction with increasing height. There is no detectable vertical component of wind.