scispace - formally typeset
Search or ask a question

Showing papers on "Precipitation published in 1968"


Journal ArticleDOI
TL;DR: In this article, the Stanford Watershed Model is used to determine the time delay from when water leaves the snow pack to the streamflow gaging station, and a model was constructed by combining the energy balance equations with relationships describing the other components of the snow accumulation and ablation processes.
Abstract: By making several minor assumptions and using an empirical technique in one situation, the energy balance of a snow pack can be computed on a continuous basis. Net radiation or its components, air temperature, dew-point temperature, atmospheric pressure, amount of precipitation and its temperature, surface snow density, and a wind function that relates vapor pressure gradient to moisture transfer as a function of wind speed, must be either measured or estimated. Comparison of computed versus observed snow pack runoff and snow surface temperature for the Lower Meadow lysimeter studies of 1954 at the Central Sierra Snow Laboratory reveals good agreement. A model was constructed by combining the energy balance equations with relationships describing the other components of the snow accumulation and ablation processes. The Stanford Watershed Model is used to determine the time delay from when water leaves the snow pack to the streamflow gaging station. Tests of the model on 5 years of data from the Central Sierra Snow Laboratory resulted in a reasonable simulation of the observed mean daily flow hydrograph and seasonal areal snow cover plot. It is concluded that the model is at least a good approximation to the actual physical processes and should be a valuable tool in further snow hydrology investigations.

176 citations


Journal ArticleDOI
TL;DR: In this article, the attenuation and scattering of laser beams by rain, fog, and snow have been calculated and measured at 0.63, 3.5, and 10.6μ.
Abstract: The attenuation and scattering of laser beams by rain, fog, and snow have been calculated and measured at 0.63, 3.5, and 10.6μ. Attenuation of the infrared wavelengths by light fog is up to one order of magnitude less than at 0.63μ. But for dense fog, calculation shows that the attenuation at 10.6μ can exceed 40 dB per km. It is found that attenuation by rain can be calculated to good accuracy from average path rain rates provided that forward scattering is taken into account; this scattering reduces the attenuation. Measurements of propagation through precipitation over a 2.6 km path are discussed in detail and are found to be consistent with predictions. The wavelength dependence of attenuation is found to vary from one fog to another because of different drop-size distributions. Attenuation of 0.63μ in rain showers is about 20 per cent less than at 3.5μ and is, of course, much less than the attenuation in fog. But even for extremely heavy rain showers, the 0.63μ attenuation never exceeded 20 dB per km, which is less than the attenuation of millimeter waves under such conditions. Both the attenuation and foreward scattering properties of snow appear to be between those of fog and rain.

113 citations


Journal ArticleDOI
TL;DR: In this article, natural precipitation processes are re-examined on the basis of accumulated knowledge of the microphysical aspects and field observations, with particular attention to the implications for cloud seeding.
Abstract: Natural precipitation processes are re-examined on the basis of accumulated knowledge of the microphysical aspects and field observations, with particular attention to the implications for cloud seeding. It appears that the active lifetime of a convective cell is much the same as the time required to grow precipitation particles and, therefore, that artificial nucleants should be inserted during the inception of the cell. The accretion process appears to be dominant in convective precipitation and there is evidence that the effectiveness of the sweeping action could be enhanced irr many cases by adding more precipitation particles. The dominant precipitation mechanism in the stratiform systems characteristic of extratropical cyclones is the ice crystal process. It is proposed that the uniformly high precipitation efficiencies of such systems, in the face of the great variability of the concentration of natural ice nuclei, result from the exponential increase in active ice nuclei with decreasing t...

93 citations


Journal ArticleDOI
01 Nov 1968-Tellus A
TL;DR: In this article, rain samples from three stations in the Alps (altitude difference 1680 m) show that the evaporation from falling drops causes an isotopic altitude effect of?0.2‰ per 100 m.
Abstract: 1 Simultaneously collected rain samples from 3 stations in the Alps (altitude difference 1680 m) show that the evaporation from falling drops causes an isotopic altitude effect of ?0.2‰ per 100 m. 2 Despite the lack of altitude effect in Alpine snow fall, the summer melting and the wind erosion cause distinct isotopic variations in the net-accumulated material on Kesselwandferner. 3 These variations are reflected in the ice in the ablation zone, and they may be used for studying glacier flow patterns. DOI: 10.1111/j.2153-3490.1968.tb00402.x

91 citations


Journal ArticleDOI
01 Nov 1968-Ecology
TL;DR: To determine the amount of precipitation contributed from cloud moisture, rain gauges were equipped with screen coils to intercept and collect cloud droplets and paired with unscreened gauges set out in forest openings along a mountain slope.
Abstract: Fog and low—lying clouds occur frequently on the higher slopes in the Green Mountains of Vermont. To determine the amount of precipitation contributed from cloud moisture, rain gauges were equipped with screen coils to intercept and collect cloud droplets. Screened gauges were paired with unscreened gauges and set out in forest openings along a mountain slope at 1,800, 2,800, and 3,600 ft. After 8 weeks the screened rain gauge at 1,800 ft collected slightly less water than the unscreened gauge. But at 3,600 ft, where clouds and fog are frequent, the screened gauge collected 66.8% more water than the unscreened gauge. See full-text article at JSTOR

75 citations


01 Jan 1968
TL;DR: A detailed climatological study showed that the average annual precipitation over Lake Michigan is 6 percent less than that of the land portion of the basin this paper, which is contrary to results of most previous studies which indicated that lake and land precipitation were equal.
Abstract: A detailed climatological study shows that the average annual precipitation over Lake Michigan is 6 percent less than that of the land portion of the basin. This is contrary to results of most previous studies which indicated that lake and land precipitation were equal. However, the results of this study are more accurate than the earlier findings because they are based on more detailed information from recent lake precipitation studies, a greater knowledge of the effect of the lake on the atmosphere, and a greater volume of weather data than existed when the previous studies were made. The chief purpose of this study was to obtain the best possible measure of the average precipitation over the lake in order to estimate more accurately the total water resource of the Lake Michigan Basin. The 6-percent lower precipitation over the lake means that on the average the Lake Michigan Basin receives about 700 billion gallons less water from precipitation per year than would be received if the average lake precipitation were equal to that over the land. This report presents the results of four separate investigations. The first three quantify the effect of the lake on thunderstorms, hailstorms, snowfall, and the amount of precipitation from these conditions. In the final investigation these results were combined with land and limited lake precipitation data to derive the average lake precipitation values. The lake both suppresses and enhances precipitation activities, its effects varying seasonally and areally. Lake-effects suppress summer thunderstorm activity by 20 percent over the southern end of the lake, but increase fall thunderstorm activity by 50 percent. During fall, lake-effects cause 400 percent more hail days in lower Michigan than occur in surrounding areas The lake also causes 25 to 100 percent more snowfall and days of heavy snow along the eastern shore than occur along the western shore. The amount of annual thunderstorm precipitation on the east side of lower Lake Michigan averages about 10 percent less than that on the west side. The west-east decrease in thunderstorm precipitation actually begins over the land area immediately west of the lake, and most of the lake-effect decrease occurs over the western half of the lake. Average annual precipitation from snowfall on the east side of the lower lake is 33 percent greater than that on the west side. Most of the west-east increase in snowfall precipitation occurs near the eastern shore with maximization 10 to 25 miles inland in western Michigan. Average lake precipitation in summer is 8.4 inches, which is 14 percent lower than that of the surrounding area. The average winter lake precipitation is 5.4 inches, which is 4 percent higher than that of the surrounding land area. The average spring lake value is 7.4 inches, 7 percent less than the land average, and that in the fall is 8.4 inches, which is equal to the land average. The average annual precipitation over the lake is lowest, less than 27 inches, in the westcentral portion, and highest in the extreme southeastern and northeastern portions. The lake average annual value is 29.6 inches. I N T R O D U C T I O N An exhaustive climatological study of the precipitation regime of the Lake Michigan Basin has been performed to obtain the best possible estimates of the average precipitation over the lake. Increasing national concern with water resources has focused greater attention on the Great Lakes. Scientific evaluation of Lake Michigan as a resource is limited because the water budget of the lake has never been properly measured. The lake surface comprises 30 percent of the basin and only a very few short-term measurements of precipitation have been made on the lake. If accurate calculations of the average annual precipitation could be derived, the lake's water budget could be estimated immeasurably better. The lake comprises 22,400 square miles and is a major influence on the lower atmosphere. At most times of the year the lake water temperature is quite different from the temperature of air masses passing over the lake, so that the lake is constantly acting to either warm or cool the overlying air. Through evaporation the lake also serves as a moisture source for overlying air masses. This

68 citations


Journal ArticleDOI
TL;DR: The results of an analysis of the chemical composition of precipitation and air over the British Isles and Eire are presented in this paper, showing the median concentrations of S, Cl, Na, K, Ca and nitrogen in ammonia and nitrate, and maps are presented to indicate the variations in sources for different chemicals.
Abstract: The results of an analysis of the chemical composition of precipitation and air over the British Isles and Eire are presented Tables are given showing the median concentrations of S, Cl, Na, K, Ca and nitrogen in ammonia and nitrate, and maps are presented to indicate the variations in sources for different chemicals The errors in the sampling technique are discussed

66 citations


Journal ArticleDOI
TL;DR: In this paper, the authors present the application of the square grid system to the estimation of the precipitation, temperature, and runoff distribution in a large area and show how the use of the system enables efficient combination of the meteorologic and hydrologic information available in assessing the precipitation and temperature, as well as runoff distribution.
Abstract: Studies on water resources in the last few years have had the tendency to encompass larger and larger areas The increase in size and sophistication of these studies makes the use of computers almost mandatory To adapt computers to this type of study, the study area must be converted to a schematic representation which is easily manageable by computers A simple and reasonable way of doing this is to cover the study area with a square grid, which may then be considered as consisting of a matrix of squares This system can be used to store, process, and retrieve information and may have numerous applications in the fields of hydrology, hydraulic power, economics, etc The paper presents the application of the square grid system to the estimation of the precipitation, temperature, and runoff distribution in a large area and shows how the use of the system enables efficient combination of the meteorologic and hydrologic information available in assessing the precipitation, temperature, and runoff distribution An example of such a computation for a 43,000-square-mile area is shown and the advantages of the method over the usual techniques emphasized

66 citations


Journal ArticleDOI
TL;DR: Based on airborne observations, estimates of air and moisture flux have been calculated for 18 hailstorms: eight in Colorado, one in Oklahoma, and nine in South Dakota as mentioned in this paper.
Abstract: Based on airborne observations, estimates of air and moisture flux have been calculated for 18 hailstorms: eight in Colorado, one in Oklahoma, and nine in South Dakota. The average air flux for hailstorms in this study was 2.3 × 1011 gm sec−1, while the moisture flux averaged near 2.1 × 109 gm sec−1. Precipitation efficiency for eight thunderstorms was found to be near 55%, though two severe hailstorms did exhibit slightly less efficiency.

49 citations


01 Apr 1968
TL;DR: In this article, a 4-summer hydrologic record from a 1.6-sq km drainage basin at Barrow, Alaska is analyzed, showing that runoff varied greatly from storm to storm, occurring primarily through and over the tundra mat and through an intricate system of polygonal troughs and ponds.
Abstract: : A 4-summer hydrologic record from a 1.6-sq km drainage basin at Barrow, Alaska is analyzed. The watershed, a drained lake basin, is underlain by continuous permafrost within 0.3 m of the tundra surface and is covered by ice-wedge polygons and numerous small shallow ponds. Considerable variations from the 20-yr means of summer climate (thaw period 88 days, precipitation 67 mm) are represented in the data; 1963 - cold, extremely wet; 1964 - cold, extremely dry; 1965 - cool, dry; 1966 - cool, wet. Runoff varied greatly from storm to storm, occurring primarily through and over the tundra mat and through an intricate system of polygonal troughs and ponds. As a result of the subdued coastal topography, varying areas (0.3 sq km to 1.6 sq km) contribute to runoff from different storms. Analyses of hydrographs revealed: (1) lag times generally from 3 to 10 hr; (2) recession constants of about 50 hr, but occasionally as much as 160 hr; and (3) runoff from individual storms between 1 and 70%. About 5% of the thaw season precipitation normally runs off. Comparison of total thaw season precipitation between the U.S. Weather Bureau and a shielded gage located on the watershed indicated no major differences. Precipitation chemistry showed no correlation with storm direction. Assuming all winter precipitation runs off, and the data are spatially and temporally representative, about 50% of the measured annual precipitation in this region runs off into the Arctic Ocean. (Author)

47 citations


Journal ArticleDOI
TL;DR: In this article, the 58-square-mile Walnut-gulch watershed in southeastern Arizona was studied and the average annual precipitation varied from 7.1 to 14.2 inches for the period of record.
Abstract: On the 58-square-mile Walnut Gulch watershed in southeastern Arizona, summer precipitation is characterized by intense thunderstorms of limited areal extent; winter precipitation, usually rain, is characterized by low-intensity storms of wide areal extent. For 11 years of record from 30 recording gages, about 70% of the annual precipitation of 11.22 inches occurred in the summer months June through September. Summer precipitation exceeded winter precipitation in each of the 11 years of record. Average annual precipitation varied from 7.1 to 14.2 inches for the period of record. During the same period, point precipitation varied from slightly under 5 inches to slightly over 20 inches. The lowest annual point precipitation was about 50 to 60% of the highest. Significantly more summer rainfall was recorded on the highest elevations than on the lowest, which indicated an effect of elevation on precipitation. However, the most summer rainfall, for the 11-year period of record, was recorded near the center of the watershed, which indicated possible topographic influences other than watershed elevation. Annual runoff from Walnut Gulch varied from 0.02 inch (1 acre-foot per square mile) to 1 inch (53 acre-feet per square mile). For the same period on a 1-square-mile subwatershed, annual runoff varied from 0.02 inch (1 acre-foot per square mile) to about 6 inches (320 acre-feet per square miles).

Journal ArticleDOI
TL;DR: In this paper, a test for the goodness of fit and an estimate of the order of a Markov chain model may be obtained concurrently for daily precipitation data for two seasons of about 30 years in New Mexico, Colorado, and Oregon.
Abstract: This paper proposes means whereby a test for the goodness of fit and an estimate of the order of a Markov chain model may be obtained concurrently. Application of the method to daily precipitation data for two seasons of about 30 yr. in New Mexico, Colorado, and Oregon suggests two tentative conclusions. First, within a single climatic area, order estimates tend toward zero as precipitation events become rarer. This may occur at the drier stations or at higher thresholds defining a wet day. Second, between climatic areas, the station with the greater diversity of air mass types will tend to have order estimates greater than the station having the same seasonal mean precipitation but a less diverse climate.


01 Jan 1968
TL;DR: In this article, the authors used precipitation data to determine mean benchmark values of contaminants throughout the country and to detect possible trends in these values over long periods of time, and to observe seasonal fluctuations and geographical distributions which may reveal the sources of these atmospheric contaminants.
Abstract: Atmospheric precipitation washes large quantities of particulate and gaseous materials from air, and thus provides a means of studying contamination of the atmosphere. The relationships between the concentration of constituents in air and their concentrations in precipitation are not known with quantitative certainty. However, it is believed that the concentration patterns of contaminants in precipitation reflect their patterns in air. Moreover, though analysis of precipitation water does not reveal absolute values for air contamination, it does provide a means of monitoring changes in contamination with time. The purposes of the project are: (1) to determine mean benchmark values of contaminants throughout the country; (2) to detect possible trends in these values over long periods of time; (3) to observe seasonal fluctuations and geographical distributions which may reveal the sources of these atmospheric contaminants. 48 references, 27 figures, 5 tables.

Journal ArticleDOI
TL;DR: In this paper, the spatial relative variability of storm, monthly, and extended period precipitation in a continental climate typical of midwestern United States was investigated, and the relation of storm variability to areal mean precipitation, storm duration, precipitation type, synoptic weather type, season, and size of sampling area was investigated.
Abstract: Data from four dense raingage networks operated for periods of 7–12yr on areas of 10–550 mi2 in Illinois were used to determine the spatial relative variability of storm, monthly, and extended period precipitation in a continental climate typical of midwestern United States. The relation of storm variability to areal mean precipitation, storm duration, precipitation type, synoptic weather type, season, and size of sampling area was investigated. Storm variability was found to be related exponentially to areal mean precipitation and the relationship was improved only slightly by the addition of storm duration. The storm variability increases with increasing area and is substantially greater with unstable types of precipitation (TRW, RW) than with steady types (R, S). Synoptically, the highest values were obtained with air mass storms and the lowest with low center precipitation. Great interstorm variability was found under apparently similar storm conditions. Monthly precipitation variability is g...

Journal ArticleDOI
S. Nieuwolt1
TL;DR: In low latitudes, where seasonal differences of climate are small and usually limited to factors such as winds and precipitation, the weather is largely controlled by diurnal processes and the diurnal variation of rainfall is the most important result of these processes as mentioned in this paper.
Abstract: In low latitudes, where seasonal differences of climate are small and usually limited to factors such as winds and precipitation, the weather is largely controlled by diurnal processes. The diurnal variation of rainfall is the most important result of these processes and it is a characteristic part of the climate. In Malaya, diurnal rainfall regimes show seasonal variations, which are closely related to the system of the southeast Asian monsoons. The seasonal variations differ according to location, and three regional types can be recognized, which are found, respectively, at the west coast, the east coast, and at inland stations. Singapore, because of its extreme location at the southern tip of the peninsula, does not fit in this classification.

01 Jan 1968
TL;DR: The role of precipitation in N-cycling in three coastal Oregon forest types appears to be minor in comparison with the much greater amounts of N cycled in litterfall.
Abstract: The role of precipitation in N-cycling in three coastal Oregon forest types appears to be minor in comparison with the much greater amounts of N cycled in litterfall. Concentration of NO/sub 2//sup -/ was much greater in stemflow from conifers than in other precipitation-forest type combinations, and that of NH/sub 4//sup +/ was substantially greater in alder stemflow and throughfall. These observations indicate the need for more information on microbial populations on leaf and needle surfaces and on tree stems to elucidate possible differences between tree species in numbers of nitrifying and ammonifying bacteria. Substantial differences between forest types in amount of dissolved solid in precipitation indicate that other nutrient elements might be more important subjects of investigation than N. The role of different tree species in influencing throughfall and stemflow content of exchangeable bases, especially, appears to require further investigation. Other factors that appear to require study include metabolites such as carbohydrates, amino acids, and organic acids in general. Despite the generally greater content of nutrient ions in stemflow than in throughfall, stemflow appears to be of little importance in nutrient cycling by precipitation because of the small volume of stemflow on an area basis. In future studies ofmore » ionic content of precipitation in coastal Oregon forests, sampling effort might better be concentrated on throughfall. Likewise, determination of NO/sub 2//sup -/ -N is not necessary because of its extremely rare occurrence in measurable amounts. Acidity of throughfall precipitation was not different from that of open-collected rainfall. Stemflow was always more acid than the other two forms of precipitation, but the ecological significance of this difference is not believed to be great.« less




Journal ArticleDOI
01 May 1968-Tellus A
TL;DR: The trend of variations in the concentration of chloride ion in rain water with the progress of a rain shower would seem to suggest that the growth of rain drops in convective clouds at Delhi would involve a process which might be analogous to what was envisaged by Woodcock for Hawaiian rains.
Abstract: Rain water samples collected with the progress of rain shower on a number of rain occasions in different seasons at Delhi have been analysed for chloride, sulphate, sodium, potassium and calcium. The trend of variations in the concentration of chloride ion in rain water with the progress of a rain shower would seem to suggest that the growth of rain drops in convective clouds at Delhi would involve a process which might be analogous to what was envisaged by Woodcock for Hawaiian rains. Further, it was noticed that the characteristics of precipitation at Delhi are dependent on the amounts of air-borne soluble particles present in the cloud forming air. Smaller cation concentration ( 1.8) distinguish what may be called “freezing” rain (rain from cold layer clouds) from that of the other type in which both freezing and non-freezing mechanisms operate (rain from cold convective clouds). Smaller ratio values of sulphate to chloride ( 1.8) characterize heavier rainfall. Probable explanations have been suggested for these observed features. DOI: 10.1111/j.2153-3490.1968.tb00369.x




Journal ArticleDOI
TL;DR: In this article, the authors studied the differences between oceanic evaporation and precipitation over the oceans, and how these factors contribute to variations in the circulations of the ocean.
Abstract: THE PERIODIC AND NONPERIODIC REGIONAL VARIATIONS in precipitation over the oceans are fundamental matters. Knowledge of them is mandatory for hydrology, oceanography, and the atmospheric sciences to build upon a firm quantitative basis. The world ocean is the primary (initial) source as well as the ultimate \"reservoir\" for all water that is transient upon the land surfaces of the earth, and a knowledge of the differences between oceanic evaporation and precipitation (E-P) is essential to understanding of the hydrologie cycle: sea-atmosphere-land-sea. The oceanographer, in his analyses of the water budget of the oceans, is deeply interested in the details of the hydrologie cycle. In addition, he is concerned with time and space variations in the balance between evaporation and precipitation themselves. This differential affects the time and space variations in the salinity of ocean waters and, to some extent, the distributions of density ( and mass ) within the oceans. Ultimately, these factors contribute to variations in the circulations of the ocean.

Journal ArticleDOI
TL;DR: In this article, a comparison of the relative effectiveness of the cyclonic circulation pattern at four different tropospheric levels in specifying winter precipitation over the intermountain area of the western United States is made.
Abstract: A comparison is made of the relative effectiveness of the cyclonic circulation pattern at four different tropospheric levels in specifying winter precipitation over the intermountain area of the western United States. This is accomplished by developing the synoptic climatology of precipitation resulting from Lows at the 850-, 700-, 500-, and 300-mb. levels. Twelve-hr. precipitation amounts (expressed as a percent of the 7-day normal) at 280 stations in the Plateau States for 13 yr. are related to the positions of nearby low centers through a computer system of moving coordinates. For each level, the upper Lows are classified into three intensity categories according to the departure from normal of their central heights. Average precipitation amount, distribution, and frequency of occurrence are then calculated for the area of the grid system. The dependence of these precipitation characteristics upon the level, intensity, and location of the upper Low is described and illustrated. The geographica...

Journal ArticleDOI
TL;DR: In this article, a physical production function was developed that relates fence spacing to the most probable annual snow catch, based on a probability analysis of winter precipitation, and the optimum scale of development in terms of fence spacing was determined by standard marginal analysis.
Abstract: Snow fencing promises to be an important means of increasing surface water yield or ground water recharge on windswept watersheds where snow is an important form of precipitation. Assuming an equally spaced series of snow fences, a physical production function can be developed that relates fence spacing to the ‘most probable’ annual snow catch, based on a probability analysis of winter precipitation. The optimum scale of development in terms of fence spacing, determined by standard marginal analysis, indicates that the smaller the marginal value of output with respect to inputs, the greater the probability must be of the fences filling annually if net benefits are to be maximized.


Journal ArticleDOI
TL;DR: In this paper, thermal neutron activation methods have been applied to the determination of silver content in samples of hail, rain and snow, and 80% of the hail and rain samples analyzed contained measurable quantities of silver up to 70 times the minimum detectable amount.
Abstract: Thermal neutron activation methods have been applied to the determination of silver content in samples of hail, rain and snow. Hail and rain samples were collected in South Dakota in regions where AgI cloud seeding was being conducted; the snow was collected in the eastern Sierra Nevada in an area where no such seeding was being conducted. No silver was detected in the snow samples analyzed, indicating concentrations on the average less than 2.5 × 10−11 gm ml−1. Eighty percent of the hail and rain samples analyzed contained measurable quantities of silver up to 70 times the minimum detectable amount. Some of these latter samples were collected in nominally unseeded areas, and although the chemical form in which the silver entered the precipitation is unknown—thereby casting doubt on its consequence—the observations raise important questions which deserve answers, particularly as they may affect statistical evaluations of weather modification experiments.

Journal ArticleDOI
01 Nov 1968-Tellus A
TL;DR: In this article, the authors used the equation of conservation of water vapor in the troposphere to calculate the difference between evaporation at the surface and precipitation in the Northern Hemisphere, which is expressed as a function of total cloudiness (or mean tropospheric relative humidity) and of temperature and horizontal wind at surface and in mid-troposphere.
Abstract: Using the equation of conservation of water vapor in the troposphere, the difference, evaporation at the surface minus precipitation, is expressed as a function of total cloudiness (or mean tropospheric relative humidity) and of temperature and horizontal wind at the surface and in mid-troposphere. Computations of the annual normal values of evaporation minus precipitation using the resulting formulas together with monthly averages of the required parameters for the Northern Hemisphere, agree well in the general patterns with estimates made by Starr, Peixoto and Crisi, as well as with climatological values. However, there are large small-scale discrepancies in the three estimates. The computations have a strong dependence on the divergence of the horizontal wind, and some of the discrepancies must be attributed to errors in evaluating the latter. To determine the importance of anomalies of the difference, evaporation minus precipitation, computations for February 1964 and February 1962 are carried out, showing that the values of the difference between these two months are of the same order of magnitude as the monthly values themselves. If positive values indicate wind from the south, the values of evaporation minus precipitation are inversely correlated to the meridional wind component on a broad scale. Furthermore the computations show that the anomalies in evaporation minus precipitation depend strongly not only on the anomalies of the meridional wind component but also on the anomalies of precipitable water, which in turn depend on the anomalies of surface temperature and cloudiness (or mean tropospheric relative humidity). DOI: 10.1111/j.2153-3490.1968.tb00404.x