B. Haurwitz

COMMENTS ON THE SEA-BREEZE CIRCULATION

Abstract Since the sea breeze is caused by the temperature difference between the air over land and that over water, its intensity might be expected not only to increase while the temperature difference increases to its maximum but also to continue increasing until the difference decreases to zero. It is shown that in a model taking friction into account the intensity of the sea breeze begins to decrease considerably earlier, in better agreement with the observations. The diurnal rotation of the sea breeze can be explained as an effect of the Coriolis force. The observations of the diurnal variations of the sea-breeze direction made at Boston agree reasonably well with the theory, especially insofar as the modifying effects of a superimposed general wind are concerned.

The diurnal and semidiurnal barometric oscillations global distribution and annual variation

SummaryThe global distributions of the annual and seasonal means of the diurnal (S1) and semidiurnal (S2) surface pressure oscillations are investigated by spherical harmonic analysis. The main waves are,S11 (with wave number 1) forS1 andS22 forS2.S11 is much less predominant among the waves ofS1 thanS22 among those ofS2. As in the case of the lunar semidiurnal barometric tideL2 the pressure maxima occur earlier in the Southern than in the Northern Hemisphere. In the case ofS2 the standing waveS20 and the waveS23 are also of interest besidesS22. Although the present analysis extends only from 60°N to 60°S, whileS20 is largest at polar latitudes, its results show thatS20 should be smaller at high southerly than at high northerly latitudes, as has been observed. Thus this observed asymmetrical distribution ofS20 may be due to causes outside the polar regions rather than to their geographical differences. The best approximation to the observed distribution ofS20 is obtained by including a mode representing an oscillation independent of longitude and latitude indicating a small semidiurnal variation of the mean global surface presure, which is an unlikely result on physical grounds.The seasonal variation ofS11 expressed in percent of the annual mean is smaller than that ofS22, and both are less than the unexplained seasonal variation ofL22.The main wavesS11 andS22 are expressed not only by associated Legendre functions, but also by Hough functions.

INSOLATION IN RELATION TO CLOUDINESS AND CLOUD DENSITY

Abstract From records of insolation and from observations of cloudiness and cloud density obtained at Blue Hill Observatory relations have been derived which give the amount of insolation, T, to be expected on the average with a given cloudiness, cloud density and solar air mass. These relations are of the form T=(a/m)e−bm where m= air mass (i.e. secant of zenith distance of the sun), e = base of natural logarithms, a and b are constants which depend on the cloudiness and cloud density. The effect of the cloud density on the insolation is of the same importance as cloudiness if the sky is largely covered. The following table shows the insolation for different values of cloudiness and density and for air masses 1 and 3 in cal/cm2/hr. It will be seen that, especially with larger values of the cloudiness, the effect of the density is very important.

LUNAR AND SOLAR BAROMETRIC TIDES IN AUSTRALIA

Abstract The lunar semidiurnal barometric tide L2 and the solar 24-, 12-, and 8-hr. oscillations of the surface pressure have been determined for 10 stations in Australia and on adjacent islands. At Rabaul and Moresby L2 is considerably smaller than elsewhere in these latitudes. The characteristic annual variation of the phase—late high tide during the D season—is found at most Australian stations. But the annual amplitude minimum occurs only at half the Australian stations during this season, contrary to the behavior of L2 over most of the globe.

LUNAR AIR TIDE IN THE CARIBBEAN AND ITS MONTHLY VARIATION

Abstract The lunar air tide and the solar 24-, 12-, 8-, and 6-hourly oscillations have been determined for Willemstad, N.W.I. and Trinidad, B.W.I. Monthly means of these oscillations have been computed for Puerto Rico.

Wave formations in noctilucent clouds

Abstract The implications of the appearance of wave and billow structure in noctilucent clouds are discussed. While in some cases the billows can be explained as effects of the angle under which the clouds are viewed, the wave motion itself with its varying horizontal convergence and divergence must also give rise to billows. In any case the presence of billow structure presents no difficulty to the hypothesis that the noctilucent clouds consist of dust particles. Observations of the wind and temperature distribution at the heights of the noctilucent clouds and during the time when waves are visible could give information about the dynamics of the atmospheric layers, in conjunction with observations of the lengths and velocities of these waves.

INTERNAL WAVES IN THE ATMOSPHERE AND CONVECTION PATTERNS

Wave motion a t a surface of discontinuity in the atmosphere may become visible in the form of cloud banks, if the vertical displacement of the air and its water vapor content are large enough to permit condensation in the regions where the wave motion is ascending. Similarly, atmospheric convection patterns in cloud layers can be observed directly when the sky clears in the areas where the air descends, while clouds are observed in the regions of ascending motion. Since convection patterns are discussed elsewhere in this publication, attention will mainly be given here to the internal waves and their cloud patterns, with the ultimate purpose of showing that the two phenomena are closely related. Laboratory experiments with unstable fluids have shown that, with suitable vertical shear of the motion, polygonal cells, transverse vortices, crossed vortices, or longitudinal vortices may develop. In view of these results, the opinion has been expressed that long cloud rolls which are often observed and commonly referred to as billow clouds or mackerel sky, and which have largely been attributed to internal wave motion, may rather be such transverse or longitudinal vortices as observed in convection experiments. The direction of the cloud rolls in these experiments differs according to the experimental conditions. Similarly, the orientation of internal waves does not necessarily have to be normal to the wind shear, as will be discussed under the next heading, and i t will be seen that a decision about the physical nature of the observed cloud rolls may not always be possible, even if wind observations are available. Hence, it is pertinent to point out that the theory of the internal waves gives wave lengths for the distances between the cloud rolls which are in excellent agreement with

The Perturbation Equations in Meteorology

Before entering into a discussion of the systems of hydrodynamic equations suitable for the investigation of atmospheric dynamics, it is appropriate to make some general remarks on the typical difficulties of investigations in theoretical meteorology and on the general principles on which the formulation of the perturbation equations is based. Such a discussion naturally includes an enumeration of the types of problems where the application of perturbation methods is particularly suitable.

THE LUNAR AND SOLAR AIR TIDES AT SIX STATIONS IN NORTH AND CENTRAL AMERICA

Abstract The lunar semidiurnal tide and the solar 24-, 12-, 8-, and 6-hour oscillations have been determined for the six stations Balboa, Panama; San Juan, P.R.; Aguadilla, P.R.; Burbank, Calif.; Oklahoma City, Okla.; and Greensboro, N.C.

ON THE RELATION BETWEEN THE WIND FIELD AND PRESSURE CHANGES

Abstract It is shown on the evidence of observational material that the simplifications necessary in order to derive the equation for the isallobaric wind are not justified, because the neglected terms in the equations of motion, viz., the convective terms and the local derivatives of the geostrophic deviation, are of the same order of magnitude as the terms retained in the equations. Hence the concept of the isallobaric wind has to be abandoned. Consequently the convergence ahead of moving cyclones and the divergence behind, in the lower troposphere, cannot be explained by means of the isallobaric-wind relation. It is shown that the distribution of the acceleration of motion in moving pressure systems offers an explanation of the observed distribution of convergence and divergence.