QC ? 2 A3 LIBRARY OF THE UNIVERSITY OF CALIFORNIA. GIFT OF Class 8061 'IZ -Nf 'AN - ' saaxvw U. S. DEPARTMENT OF AGRICULTURE. WEATHER BUREAU. SOME CLIMATIC FEATURES OF THE ARID REGIONS. COMMUNICATED TO THE NATIONAL IRRIGATION CONGRESS. AT ITS FIFTH ANNUAL SESSION, PHCENIX, ARIZONA, DECEMBER 15-17, 1896, BY L.. MOORE, CHIEF OP WEATHER BUREAU. WASHINGTON : WEATHER BUREAU. 1896. U. S. DEPARTMENT OF AGRICULTURE, WEATHER BUREAU. SOME CLIMATIC FEATURES OF THE ARID REGIONS. COMMUNICATED TO THE NATIONAL IRRIGATION CONGRESS. AT ITS FIFTH ANNUAL SESSION, PHCENIX, ARIZONA, DECEMBER 15-17, 1896, BY L. MOORE, CHIEF OF WEATHER BUREAU, -odf WASHINGTON : WEATHER BUREAU. 1896. U. S. DEPARTMENT OF AGRICULTURE, WE!THER BUREAU, Washington, D. C., December 1, 1896. To THE NATIONAL IRRIGATION CONGRESS, Phoenix, Ariz. Mr. President and Representatives: Finding it impossible to accept your invitation to deliver an ad- dress before your honorable body on the 17th instant, on the subject of ' : Sensible Temperatures," I herewith transmit for your considera- tion a paper having for its purpose the presentation of a few salient features of climate in their bearing upon the irrigation, sensible tem- perature, and kindred physical features of the arid and subarid re- gions of the great West. In the preparation of the charts, and in collating much of the in- formation contained in the text of this paper, I have pleasure in acknowledging the valuable assistance rendered by Mr. Alfred J. Henry, Chief of the Division of Records and Meteorological Data of the Weather Bureau. Although having served as a member of the Government Depart- mental Board on Irrigation during the past year, I feel that my knowledge of the practical application of irrigation to agriculture is too limited to justify me in intruding my opinions upon the many here assembled who are so much better fitted by experience and edu- cation to consider the problem, still I desire to express my sympathy for, and hearty interest in, your noble efforts to claim for the present generation fruitful dominion over many of the alluvial valleys and plains now unknown to the plow and the. reaper, and to transmit to posterity an empire of fertile fields and thrifty homes. Under the direction of the Honorable Secretary of Agriculture it was my pleasure on September 20, 1895, a few weeks after coming to the head of the Weather Bureau, to issue instructions to the observers of the weather service to begin the telegraphing from observation stations of the readings of the wet-bulb thermometer, more popularly known as the " sensible " temperature. This is about the temperature felt by animal life and may be many degrees below the air tempera- ture, the difference between the two temperatures depending upon the relative humidity of the air the drier the atmosphere the lower the sensible temperature when compared with the air temperature ; 236825 3 the damper the air the higher the sensible temperature. This will be better understood when it is stated that in case the air be satu- rated, the readings of the dry and the wet bulb thermometers will be the same and the sensible temperature and the air temperature will be equal. In the semiarid regions of the West the sensible tempera- ture during the summer months often is 20 to 30 less than the air temperature, which condition is due to the extreme dryness of the atmosphere. In the more humid regions of the eastern part of the country such extreme differences can not occur. Gradually the publication of sensible temperatures has been ex- tended until now the readings of the wet-bulb thermometer appear on all published maps and meteorological tables of the United States Weather Bureau issued from its many stations. Over four million maps are posted annually in conspicuous places in the important cities and towns of the country, and the tables are printed in many of the large dailies of the principal cities. The publication thus made of the thermal conditions of the United States will result in correcting many erroneous impressions in regard to the climate of the West which have found lodgment in the minds of those accustomed to consider only the air temperature. To present anything like a comprehensive discussion of the widely varying thermal conditions of the United States and a comparative view of the effects of solar insolation on the humid sections of the East, the subarid regions of the middle west, and the arid portions of the Rocky Mountain Plateau necessitates a cursory consideration of the climate of the United States in general and of some selected portions of foreign countries. A proper understanding of the climatic conditions of the region west of the one hundredth ^parallel jis of vast importance, not only to the agricultural, industrial, hygienic, and therapeutic interests of the many states represented in this Congress, but of the United States in general. That which is of vital interest to a part should be, and in this case is, of material interest to the whole. The physical characteristics of this vast region and its geograph- ical boundaries have been so fully described elsewhere that it is not necessary to repeat them here. A word, however, regarding the ter- ritory under discussion may not be amiss. The lands west of the one hundredth meridian are generally classed as arid, or semiarid, excepting of course, considerable portions of California, Oregon, and Washington, and mountainous regions of these and other States. The dividing line between arid and agricultural lands (using the lat- ter term as signifying a region of sufficient rainfall for the growth and maturity of the staple crops) is not strongly marked. Indeed no inconsiderable portion of several States west of the Mississippi have, at irregular intervals, sufficient rainfall to yield bountiful crops, NOTE. After the word hundredth," 17th line from the bottom, page 4, read "meridian" instead of "parallel." but in a majority of years the rainfall is below the needs of success- ful agriculture. It would be manifestly improper to class these lands as belonging to the arid regions of the country ; they should rather be referred to as regions of uncertain rainfall in which agri- cultural operations without irrigation are more or less hazardous. Much valuable information respecting the climate and resources of the arid regions has been collected by public-spirited citizens of the Western States and Territories. In the following remarks an attempt will be made to point out some of the more important conclusions that may be drawn from the climatic data collected both by private citizens and Government officials. The available material for this discussion may^be considered in a two-fold sense, viz : first, as defining the climate of the region, par- ticularly with reference to its suitability as a place of residence for those unable to withstand the rigors of the climate elsewhere, and second, in its relation to engineering and other economic projects. It is not difficult to reach a proper understanding of the climates^ of the arid region, but unfortunately the correlation of climatic data and vital statistics has not been effected as fully as might be desired.) The problem, therefore, is not completely solved by an examination of the climatic statistics alone. It is admitted by those eminent in the medical profession that the study of the influence of the various physical elements of climate upon the human organism and the col- lection of morbidity and mortality statistics demand the careful attention of public officials and the active cooperation of private citizens in whatever sphere they may be found. General meteorology, especially that branch pertaining to weather forecasting, is being developed as rapidly as the conservative use of means and appliances at hand will permit. The development of the physiologic aspect, however, is a problem of much wider range, and one requiring the accumulation of a vast fund of statistics outside the realm of either climatology or meteorology. It is a fact to be regretted by every sincere lover of knowledge that the collection of data relating to the classification of climates with reference to cer- tain diseases, the influence and special dangers of various climates upon the health seeker, and other matters of equal importance are not coextensive with the collection of weather reports. The present elaborate system of weather reports is the direct result of private effort and investigation fostered by the Smithsonian Insti- tution. Is it too much to hope that the facts upon which the science of medical climatology must depend for advancement will be con- tributed by the private worker? Within the broad confines of the United States there are many, but not all, shades and varieties of climate. One of the questions most frequently asked the \Veather Bureau is, " Where shall I find a climate possessing both dryness and equability of temperature ?" To this interrogatory reply must be made, that the ideal climate as regards equability of temperature and absence of moisture does not exist in the United States, but that the nearest approach to it will be found in the great Southwest, where all shades of dryness, from a rainfall sufficient for successful agriculture to the aridity of the desert, may be found. The temperature of the Southwest is not equable in the sense of having an extremely small daily range, but, on the other hand, it possesses the quality of uniformity in a greater degree than will gen- erally be found elsewhere except on the seacoast. The most equable temperature on the globe will be found on the high table-lands and plateaus of the Tropics. Santa Fe de Bogota, in the United States of Colombia, has an average temperature of about 59 for all months of the year, and the range for the entire year is less than is often expe- rienced in a single day in these latitudes. But while the ideal temperature may be found on the higher eleva- tions of the Tropics, the rainfall is much greater and more continuous than in this country. The rainfall of the great Southwest varies with location. Less I than 200 miles from the Colorado Desert, where the rainfall is prac- tically nil, places may be found whose annual average rainfall is as great or greater than any point in the Middle States of the East. Generally speaking, however, the greater portion is dry, using that term as indicating a rainfall considerably less than 20 inches per annum on the average. The mountainous portions of Arizona and California have an aver- age annual rainfall ranging between 20 and 50 inches, depending somewhat upon the elevation and geographic position, while the low- land portions and the plateaus, especially east of the Sierras, have a rainfall both small in amount and variable in character. The rain- fall records of the arid region, and other portions of the United States, are published in the monthly bulletins of the various climate and crop centers, and in more convenient form in the annual data volumes of the Weather Bureau. It is not possible to report upon them in detail here. The temperature of a place depends chiefly on three conditions, viz., latitude^elevation, and contiguity toTar^^odje^of^water. At sea level in~the Tropics extreme condition^ of heat and~moisture so combined as to produce very great physical discomfort abound. But even under the equator it is possible to escape the tropical heat of low levels by asceijding from 4,000 to 6,000 feet. In the economy of nature there is a certain limit beyond which the two extremes, dry- ness and equability of temperature, can not coexist; thus we may find a region so deficient in moisture as to satisfy the requirements of the case, but the very lack of moisture is a condition that facilitates radiation and thus contributes to great extremes of temperature. Regions may be found, as on the lower Nile, where there is a lack of rain- fall coupled with a high and moderately uniform temperature. The mean winter temperature of Cairo, Egypt, is 56 ; mean summer tem- perature, 83 ; a range from winter to summer of 27. The mean win- ter temperature of Phoenix, Ariz., is 52 ; mean summer temperature, 87 C ; a range of 35. It is by no means difficult to find a counter- part of the far-famed Egyptian climate in the great Southwest. The dryness of the air and the clearness of the sky are the^ conditions upon which daily ranges of temperature depend; the greater these, the greater the range of temperature from day to , night. While a high summer temperature is characteristic of the Southwest it is a fact long known to residents of that section, and somewhat imperfectly realized in other portions of the coun- try, that the sensation of heat as experienced by animal life is not accurately measured by the ordinary thermometer. The sensa- tion of temperature which we usually refer to the condition of the atmosphere depends not only on the' temperature of the air, but also on its dryness, the velocity of the wind, and other circumstances. The human organism when perspiring freely evaporates the moisture of its surface and thus lowers its temperature. The meteorological instrument that registers the temperature of evaporation and thus in a great measure the actual heat felt by the human body, is the wet bulb thermometer. The latter as indicated by its name is simply an ordinary mercurial thermometer whose bulb is wetted with water at the time of observation. Chart I has been constructed to show the average actual and sensi- ble temperatures of Weather Bureau stations in the United States for the summer season. The broad principle illustrated by this chart is that the greatesf differences between shade and sensible temperatures are found where the air is the driest and the least where the air is most humid. A glance at the chart is sufficient to show the general trend of the lines of equal air and sensible temperatures. The great interior val- leys, and the plains east of the foothills of the Rocky Mountains are uniformly heated under the insolation of summer to an average of from 65 on the northern boundary to about 80 on the Gulf Coast. The northern portion of this vast extent of country is, moreover, in the path of atmospheric disturbances that pass from west to east over our northern boundaries, thus causing an indraught of warm^/ moist air from lower latitudes. Again, the distribution of atmos- pheric pressure over the pnstprnj-,\rn-tVnrdH of the United States is at times such as to cause a more or less complete stagnation of the i * generally eastward drift of the air ; the surface of the ground warms [ 8 I up under intense insolation and loses but little heat by radiation at night ; the winds are light southerly or southeasterly and there is an absence of vertical interchange between the warm surface air and the cooler air aloft. Such conditions sometimes extend over the en- tire Mississippi Valley and eastward to the Atlantic Seaboard. On the other hand, while it is possible for a heated term to prevail over an arid region by day, the relatively great radiation by night lowers the temperature to an endurable degree and there is but little bodily discomfort. The heat of the daytime, moreover, is borne without distress by reason of the great dryness of the air. The red lines of Chart I show the temperature of evaporating surfaces in summer in the United States. It will be seen that the line of 60, which marks the temperature of evaporation of the region of New England and the Great Lakes, passes almost due north and south along the eastern foothills of the Rocky Mountains, and skirts southern New Mexico and Arizona. The line of 55 passes almost due south from eastern Montana to southeastern New Mexico and thence northwesterly. The temperature of evaporation in all of the territory above this line (55), embracing almost two-thirds of the arid region, is below 55; in fact, in almost one-third of the region it is not over 50. The sensible temperature of two-thirds of the United States, or east of the one hundred and fifth meridian, ranges from 55 to 75. West of the one hundred and fifth meridian the range is from 50 to 65. Chart II has been prepared to illustrate the extreme differences that prevail in midsummer, the 8 p. m., seventy-fifth meridian time observation of July having been used. (8 p. m., seventy-fifth meri- dian, corresponds to 7 p. m. central, 6 p. m. mountain, and 5 p. m. Pacific time). There is an objection to the use of synchronous time in depicting climatic elements that have a marked diurnal period. Observations taken at the same moment of local mean time should be used whenever possible, but the exigencies of a service instituted for the purpose of forecasting weather changes demand the use of synchronous time. As regards the data of this chart (II), it may be urged with propriety that a comparison of thermometric readings made at the same moment of time from the Atlantic to the Pacific is misleading, since an accurate estimate can not be made of the amount of increase of temperature for western stations due to diurnal influ- ences alone, and it was mainly with a view of illustrating this fact that the chart was prepared. The thermometer readings on the Atlantic Seaboard are made near the hour of 8 p. m., local mean time; those on the California coast are made near 5 p. m., local mean time. Naturally the Pacific Coast temperatures are considerably higher than those on the other side of the continent, three hours later in the afternoon. The contrast be- tween the two sides of the country is plainly shown by the black lines of equal actual temperature on Chart II, and it will also be observed that the Southwest is the warmest part of the United States. The lines of equal sensible heat, on the other hand, show an en- tirely different condition as regards the location of greatest heat. The arid region is now the coolest part of the United States, judged from the temperature of evaporation only. The line of 60 sensible temperature, starting in New England, skirts the northern boundary as far as the one hundred and tenth meridian ; thence it follows a south-southeasterly course to southeastern New Mexico ; thence west- erly to the neighborhood of Los Angeles, Cal., and thence northerly, with a few unimportant deflections, to the north Pacific Coast. The decrease of temperature from the hour of maximum heat to- nightfall is not regular, nor does it bear any definite relation to an increase in longitude reckoned westward from Greenwich. A com- parison of the normal 8 p. m. seventy-fifth meridian time tempera- tures with the normal maximum temperature of the day shows that on the eastern coast line the temperature at 8 p.m. is, on the average, 8 to 12 lower than at the time of greatest daily heat. In the Lake Region and lower Ohio Valley the difference is from 5 to 8. In the upper Mississippi and Missouri valleys and Texas and the plains region the difference averages from 4 to 7 ; that is to say, the tem- peratures at the 8 p. m. observation (corresponding to about 6.30 p. m., local time) are from 4 to 7 lower than the highest point reached by the thermometer during the day. On the eastern slope of the Rocky Mountains, although the evening observation is made at 6 p. m., local time, two hours nearer the time of greatest heat than at New York and Philadelphia, the difference is as great as at the last-named places. In other words, the temperature falls as much by 6 p. m. at Denver as it does by 8 p. m. in New York and Phila- phia. This would seem to be the result of the greater daily range and more rapid rate of cooling at elevated stations. West of the Rockies the differences range from zero at Red Bluff to less than 4 in the great interior basin and from 5 to 6 in southern Arizona. The local vicissitudes of temperature are well illustrated in the case of Red Bluff, Cal., where the average temperature at about 5 p. m., local time, is but four-tenths of a degree below the maximum of the day. Curiously enough, at Los Angeles, in the lower part of the State, the 5 p. m. temperatures are about 10 lower on the average than the maximum of the day. Chart III has been constructed to show the relative humidity of the United States in summer. The data used in preparing the chart were the synchronous observations at 8 a. m. and 8 p. m., seventy- fifth meridian time, during the eight years 1889-96. The chart itself shows better than mere words the distinctively dry and humid regions. The influence of the ocean is seen on both coasts, as also that of the Gulf of Mexico and the Great Lakes. 10 Broadly speaking, the variation of insolation from day to night, and from season to season, with the changing declination of the sun, is the great controlling agent of climate. The most regular, and at the same time the simplest climate of the world, is that of the Tropics, where the succession of changes from day to day are as monotonous in their regularity as they are enervating on the human system. The great life zone, the seat of business enterprise and activity, is found in temperate climates. Here the simple diurnal changes of the Tropics are largely masked by irregular changes, the result of the passage of cyclonic and anticyclonic systems. The sum total of these changes constitutes the weather of the temperate zone. Between the Tropics and the temperate zone there are in certain longitudes considerable areas where the climate is more or less tran- sitional between the two strongly marked zones. The southwestern part of the United States may be classed as having a climate between the extremes of the Tropics and the temperate zones. Not being within the path of storm frequency, the sequence of weather is more uni- form than in more northern latitudes, or on the same parallel farther east. The rainfall is deficient ; there is an absence of clouds ; insola- tion by day and radiation by night are both strong ; the range of temperature from day to night is large, from 25 to 35, depending upon the elevation and character of the surface of the ground ; the winds are generally light and the evaporation is high. / The climatic data collected by the Weather Bureau, viewed from / an economic standpoint, are mainly valuable as giving definite infor- I mation of the monthly and annual amounts of rainfall, the maxi- \ mum and minimum amounts in varying periods of time, the distri- V bution throughout the year, and the secular variation. It is not proposed to enter into a discussion of the rainfall records, nor of the closely related subject of evaporation from a water surface. The last-named subject is one that requires careful experimental work with a view of determining what form of evaporometer is best adapted to the requirements of the case. The conditions under which obser- vations of evaporation could be made at Weather Bureau stations are largely artificial, and it is doubted if results of scientific or practical value could be obtained therefrom. The use of windmills for pumping and storing water is quite com- mon both in humid and in arid regions ; in the latter, however, the successful application of the forces of the wind to the raising of water for irrigating purposes is a much more important matter than else- where. Irrigable land, so situated that the water of natural streams can not be diverted thereto, must depend upon the flow of artificial reservoirs, and these latter in turn must be supplied from the under- ground flow. When the static pressure on the surface of the latter is not sufficient to force the water to the surface, resort must be had 11 to pumping. The economy of the wind as a motive power can not be questioned ; but, on the other hand, its use is restricted to such classes of work as will admit of temporary cessation during a calm. A knowledge of the strength of the surface winds is essential to a thorough understanding of the limits and possibilities of windmills in connection with irrigation problems. The strength of the surface winds over the arid region varies greatly on account of the broken configuration of the land and its -geographic position with respect to the path of cyclones and anti- cyclones. Surface winds are purely the result of unequal pressure distribution over adjacent areas. The winds blow from a region of high pressure to a region of low pressure, and the steeper the gradient the faster they blow. The velocity of the lowest stratum of air is greatly retarded by friction on the surface of the earth, the retarding effect being most noticeable in the layer of air extending from the surface of the ground to fifteen feet above it. It is exceedingly difficult to combine wind data of meteorological observatories into a homogeneous system. The influence of a border- ing range of hills or of adjoining buildings may seriously vitiate the record. Accurate comparisons of wind velocities can not be made, on account of the lack of uniformity in the elevation of ane- mometers, except by the application of a system of instrumental corrections. Such a system of corrections has not been applied, nor have actual velocities been reduced to true velocities in any case. A comprehensive presentation of the wind velocities of any region should include a statement of the average hourly velocities for the several months of the year ; of the strength of gust velocities for short intervals ; of the diurnal and seasonal changes ; and, as par- ticularly appropriate to the subject of these remarks, a statement of the number of hours in each month of available wind power, say of wind velocities from 6 to 10 miles; 11 to 15 miles, etc. The average hourly velocities for a number of stations west of the one hundredth meridian have been computed and will be found in Table I. These data express the average force of the wind as shown by anemometers exposed on Weather Bureau buildings in the places named for the eight years ended with 1895. The averages have been made from the record for the twenty-four hours and include, as will be more fully shown later, periods when the velocity is above the average as well as below it. It is apparent that the region of greatest velocity of the wind or greatest windiness is in Kansas, Oklahoma, Texas, and Nebraska, some distance east of the foothills of the Rocky Mountains. The slope of this area is to the eastward ; its surface offers compara- tively little resistance to the winds ; the barometric gradients are moderately steep, except during the summer season. In the latter season the air of the plains becomes heated and rarified and a strong 12 vertical circulation is established and the air of surrounding regions is drawn in, thus maintaining active air circulation at a time when the pressure gradients are weakest. The variation in velocity during the year is graphically shown be- low in Fig. 1, the curve " a " representing the plains group. In the diagram just mentioned the months of the year are repre- sented by vertical divisions ; the velocity of the wind in miles per hour by the horizontal divisions, each of which represents a mile of wind. The three curves show the average hourly wind velocity for each month of the year in three separate regions, viz., (1) the great plains- extending from the Dakotas to Texas; (2) the Rocky Mountain re- gion, including the States of Wyoming, Colorado, and New Mexico, and (3) Arizona. The curves are alike in their general characteristics, and it is quite- evident that they are all modifications of one simple type. The plains type is believed to be the fundamental one. Here by reason of the surface configuration there is little friction on the earth's surface ; the sweep of the wind is not broken by forests or other ob- structions, and the high average velocities attained lead us to believe- that the annual march of the wind is much the same as would obtain on a water surface. It should be remembered that the wind veloci- ties observed on the great plains are almost as high as prevail at ex- posed stations on the seacoast. There is an abundance of wind for driving windmills as will be shown by the statistics of Table II. The curves in Fig. 1 show a maximum of wind in April and a mini- mum in August, and this characteristic is common to all sections of the arid region except the interior valleys of California where the- spring maximum is deferred until May and June. The coast winds of California differ in some respects from those of the interior. They are principally from some westerly quarter, quite- steady and strong in summer, especially in the region of San Fran- cisco, giving to the coast a climate much colder than some miles in- land. The effect of these westerly winds is plainly seen on Charts I,. II, and III. The foregoing relates especially to the relative strength of the winds- in the different months of the year. Considering now the relative velocities of different localities as shown by the three separate types, it is observed, first, that the winds of high stations, such as Cheyenne, Denver, and Santa Fe, are not as strong as might naturally be ex- pected. It is true that wind velocities increase both with elevation above ground and with elevation above sea, but the increase does not depend upon elevation alone, nor does it increase in direct proportion to the increase in elevation. The winds of Mount Washington are- much stronger than those of Pikes Peak, although the latter is more- than double the altitude of the former. I I 2 I iH $ OK? 0:OOoOlQO(QO o 10 o 10 pioomo o ^t 'g-coo 13 The diminution of the average velocity of the wind, as shown in curves b and c. Fig. 1, may be ascribed in a great measure to the in- equalities of the land surface, although there are doubtless other causes that contribute to the general retardation. In a region of great contrasts of surface configuration, moreover, the number of faulty anemometer exposures is relatively greater than in a region of comparatively even surface. In the latter case, the only condition required is that the anemometers be placed at a uniform level above the street. And in case they are not of uniform elevation we may apply a correction to the recorded velocities in order to reduce them to some standard height above ground. As illustrating the increase of velocity with elevation above ground, the case of El Paso, Tex., may be cited. The anemometer at this station was exposed for ten years at an average elevation of 30 feet above ground, and the aver- age annual wind movement for the same period was 41,000 miles, or at the rate of 4.7 miles per hour. An increase of elevation to 80 feet gave an average annual wind movement of 71,000, or at the rate of 8 miles per hour, almost double the velocity at the low elevation. A description of some experiments made by Mr. Thomas Steven- son to determine the relative velocity of the wind at different heights above ground, will be found in the Journal of the Scottish Meteoro- logical Society from which the drawing below, Fig. 2, is reproduced. With low velocities the change with altitude up to 50 feet, the high- est elevation at which the experiments were made, was not great. Thus, a velocity of 10 miles per hour at 5 feet above ground became 13 miles at 50 feet. With the higher velocities, however, the increase was much more rapid, a velocity of 20 miles per hour at 5 feet above ground becoming 37 miles at 50 feet. The formula proposed by Stevenson for finding the velocity Fat any point, H feet above ground, from the known velocity v at a height h feet above ground (h being above 15 feet), is v \ h + 72 Using the known velocities at Weather Bureau stations, we may compute from this formula the velocities that should prevail at the elevation of the windmill driving arms above ground. A second -conclusion of no less importance is that in order to obtain the maxi- mum efficiency of the wind in regions of low average velocity, it is necessary to place the driving arms at a much higher elevation than in regions of high velocity. The velocity of the winds in Arizona, the Great Basin, and the interior valleys of California, on the average of the year, does not vary greatly from about 6 miles per hour. This value, however, in- cludes varying periods of greater and less velocities. 14 The average wind velocity is chiefly useful in determining, in a gen- eral way the practicability of using the wind as a motive power. In India, for example, an examination of the recorded wind velocities shows at a glance that windmills can not be employed except at a few- places during the height of the summer monsoon. The average velocities for certain portions of the arid region, given in Table I, indicate clearly that during a part of the time the wind is not sufficiently strong to drive an ordinary windmill. How much wind is available on the average can be determined by a tabulation of the continuous records according to a scale of progressive veloc- ities. Thus it may be assumed that all winds of 5 miles an hour and under are inefficient, and, on the other hand, that winds of 26 miles an hour and upward are too violent for practical use. We have, therefore, as the effective winds those ranging in velocity from 6 to 24 miles per hour. Table II shows for six stations, Dodge City, North Platte, Pueblo, Lander, Salt Lake City, and Yuma, the percentage of time during the five years 1889-1893 (except at Lander, where a period of three years was used), that hourly velocities of 0-5 miles per hour, 6-10,. 11-15, 16-20, 21-25, 26-30, and 31 and upward prevailed. Had time permitted, the data for a greater number of stations would have been tabulated. The greatest amount of available wind, as elsewhere indicated in this paper, is found on the plains from Texas to the Dakotas. The two points selected to represent this region, Dodge City, Kans., and North Platte, Nebr., have an excellent anemometer exposure, both instru- ments being 52 feet above ground. The surface of the surrounding country is generally flat and treeless, and there are no obstructions to the free sweep of the winds. At Dodge City the winds during 72 per cent of the time ranged in velocity from 6 to 25 miles per hour. Assuming that the use of wind- mills is restricted to winds of 6 to 15 miles per hour, then they may be employed 51 per cent of the time. It is manifest, however, that higher velocities may be used with safety. The results at North Platte agree quite closely with those just mentioned. High velocities at the last-named station are not so prevalent as at Dodge City, yet the percentage of effective winds is about the same. There is a popular conception that of all meteorological elements the wind is the most fickle. A study of the figures in the table will doubtless lead to a different conclusion. The general agreement between the percentages of the two stations is reasonably conclusive evidence that such percentages truly repre- sent the actual wind movement of the greater portions of western. Kansas and Nebraska. 15 Pueblo, Colo., in the Arkansas River Valley, elevation 4,653 feet above sea level, was selected as representative of eastern Colorado, where irrigation is extensively practiced. Although the altitude of Pueblo is 2,176 feet greater than Dodge City, the total wind travel is- 36 per cent less and the total effective winds (6-25 miles per hour) but 49 per cent against 72 per cent at Dodge City. It is important to note, however, that the percentage of w r inds of low velocities (6-10) is practically the same at both stations, 30 and 29, respectively. The great difference between the two points being in the higher velocities. The Weather Bureau station at Lander, in the Popo Agie Valley r seven miles east of the Big Wind River chain of the Rocky Moun- tains, has been in operation but a few years. The velocities are unusually low, effective winds prevailing only 24 per cent of the time. The average for April, May, and June, how- ever, is 33 per cent, the highest of the year. The low velocities at this station are doubtless due to the sheltering influence of the moun- tains to the westward and the low exposure of the anemometer, 3 feet above ground. The winds of the Great Basin are not truly presented by the instru- mental records of Salt Lake City, the immediate environment of the latter being such as to vitiate the record in a greater or less degree. A spur of the Wasatch range, but 3 miles distant to the eastward,. shelters the city both on the east and southeast. On the north hills rise to an elevation of 1,000 to 2,000 feet above the city, while Great Salt Lake is but 10 miles distant to the northwest. The combined influence of the mountains and the lake produces a regular diurnal period except when overcome by the stronger gradients of atmospheric disturbances to the northward. The winds of the forenoon are from the southeast (the mountains) and of the afternoon from the north- west (from the lake). Notwithstanding the faulty exposure of the wind instrument at Salt Lake City, there appears to be sufficient wind to drive a windmill 34 per cent of the time. High winds are unusually rare. The percentage of effective winds at Yuma, Ariz., is 49 out of a probable 100. The winds of Arizona differ from those of many other portions of the United States in that there is a greater daily range and less interference with the regular progression from morning to night by cyclonic storms. The winds of the nighttime fall to a very low velocity, but those of the daytime are quite strong, especially on the higher plateaus, where an average of 12 to 14 miles per hour is maintained during the warm season. The anemometer at the Yuma station is about 50 feet above ground and 100 feet distant from the Colorado River. The location is not as advantageous as might be chosen if a high-wind velocity were 16 the only desideratum, but it is similar to a majority of the locations in which windmills would naturally be placed, and for that reason the data are applicable to the problem in hand. The average wind velocities heretofore published for Phoenix, Ariz., were determined from the record of an anemometer exposed 19 feet above ground. When the station was reestablished, in August, 1895, the anemometer was placed 57 feet above ground. The velocities obtained in the new location are considerably higher than those in the old location, the monthly average being about 5 miles per hour. While greater velocities generally prevail during the warmer hours of the day, there will be occasions in the daytime when the wind will fall below the effective limit. The remedy in such cases is a higher exposure of the windmill. Summarizing the foregoing in a few words, it may be said that there is an abundance of effective wind on the plains east of the Rocky Mountains in all months of the year and that no special adaptation of the ordinary windmill is necessary. The amount of effective wind decreases with approach to the Rocky Mountains, but in well-exposed localities there is still suffi- cient for all ordinary needs. On the leeward side of a mountain range and in sheltered valleys it will be necessary to increase the elevation of the windmill support in order to obtain sufficient wind for pumping purposes. In Arizona, New Mexico, the interior valleys of California, and the Great Basin there is generally sufficient wind for ordinary purposes, but in certain locations it will be necessary to select a favorable position for the windmill and, as in the Rocky Mountain Region, to place the driving arms at as great an elevation above ground as may be practicable. Periods of calm or of ineffective velocities occur on an average once or twice every month during the season, April to September. The duration is seldom longer than twenty-four hours ; in extreme cases they may continue forty-eight hours. 17 TABLE I. A i-erage hourly Telocity of the wind at selected stations of the Weather Bureau. Stations. | February. March. f <5 | = - -j >. p Ha August. September. October. Norember. fe Bismarck, X. Dak Huron 8 Dak 9-0 11.2 8.2 9.5 8.5 7.1 9-9 9.0 16.3 10.2 8.0 10.2 12.6 8.3 10.3 9.4 8.1 10.8 10.5 16.7 12.4 9.9 10.5 12.8 12.1 12.9 14.7 13.8 9-6 10-6 10.2 11.6 13.4 12.2 10.8 12.6 11.8 9.6 10.5 9.6 12. s U.t 13.7 11.9 11.4 10.2 19.2 20.6 18.9 13.3 13.2 12. ! 11.4 11.3 10.5 10-6 13.3 10.0 12.2 10.9 8-0 13.8 9.1 18.8 12.0 8.8 9.4 10.3 9.1 10.3 9.0 6.9 11.8 7.8 16.1 9.6 7.5 9.1 11.6 8.4 9.5 8.6 6.0 10.9 6.5 13.4 8.2 7.5 10.8 13.7 9.5 10.9 9.3 7.0 11.9 9.0 16.9 9.5 7.2 10.5 12.7 9.2 10.5 8-8 6.6 10.4 8.8 16.2 10.0 7.2 9.4 11.9 8.8 10.4 8.8 7.3 8.9 9.4 16.4 10.5 8.2 8.8 11.6 8.4 9.5 8.2 7.7 10.4 9.7 16.2 11.2 8 4 Rapid City, S. Dak. . .. Valentine, Xebr North Platte, Nebr . . Concorclia Kans Dodge City, Kans. . . Oklahoma. Okla Amarillo. Tex . . \bilene Tex El Paso Tex Mean 9.7 4.0 12.4 10.8 3.6 12.5 7.4 8.0 7.3 8.9 5.4 7.6 11.4 5.0 4.6 6.8 5.5 6.7 7.6 3.7 11.6 5.0 7.0 7.2 6.2 6.8 6.9 5.7 4.7 2.4 4.3 7.6- 7.7 4.8 5.9 G.5 12.1 5.2 12.8 8-1 8. 7 8.2 8.4 5.9 8.2 10.4 7.0 5-6 7.5 5.7 7.2 9.7 5.0 10.9 6.3 8.0 7-1 6.9 7.0 6.2 6.8 6.6 5.0 2.9 4.8 8.0 7.7 5.8 5.8 6.8 13.2 5.6 11.8 8.6 9.4 8.3 8.6 6.9 8.5 10.8 8.2 6.3 7.8 6.2 7.9 9.0 5.2 11.4 6.5 8.0 9.0 8.5 7.7 7.1 8.1 6.9 5.3 3.6 5.3 ' 7.4 7.8 6.8 6-4 7-1 12.4 5.5 10.8 7.9 8.9 8.4 7.9 7.3 8.1 10.5 8.0 6.5 8.0 5.6 7. 7 8.4 4.3 10.6 6.5 7.4 8.6 8.0 7.4 6.6 7.6 6.5 5.5 3.4 5.1 7.8 8.5 7.5 6.2 7.5 11.6 5.5 10.0 7.2 8.2 7.9 6.3 7.4 7.5 9.6 7.1 6.3 7.6 5.3 7.1 8.1 4.1 10-0 6.4 7.2 8.8 7-6 7.0 6.2 7.4 6.5 5.2 3.6 5.1 ' 7.2 8.6 8.5 6.3 7.6 9.8 4.7 9.0 6.6 7.2 6.7 4.5 6.1 6-4 10.0 6.2 5.4 7.2 5.3 6.8 7.8 3.6 9.6 5.8 6.7 6.9 6.0 6.0 7.0 6.5 6.1 5.1 3.8 5.0 5.8 7.7 7.2 5.8 6.6 9.1 4.0 7.9 6.4 6.5 5.9 4.8 5.7 5.9 8.7 5.0 4.7 6.8 5.4 6.1 7.2 2.9 9.0 5.6 6.2 6.2 5.6 5.8 6.1 5.9 5.3 4.5 3.3 4.4 4.8 6.9' 6.4 5.5 5.9 10.5 5.0 9.1 6.4 6.4 6.1 4.6 5.8 6.2 10.0 5.5 5.2 7.1 5.4 6.6 8-3 3.0 9.4 5.7 6-6 6.2 6.1 7.2 4.8 6.1 5.3 4.9 2.9 4.4 6.3 7.0 5.9 5.5 6.2 10.1 3.7 10.1 6.8 6.5 6.5 5.5 5.2 6.3 10.9 5.7 4.8 7.4 5.5 6.9 6.9 3.3 9.2 5.0 6.1 6.3 6.3 6.9 4.6 6.0 5.0 4.2 2.3 3.8 '6.3 5.1 5.6 10.0 4.1 11.4 7.3 6.6 6.5 5.4 4.6 6.6 12.1 6.1 4.8 7.1 5.5 7.1 7.6 3.2 9.6 4.6 6.2 5.3 5.2 6.6 5.4 5.6 5.3 4.5 2.2 4.0 '6.6 5.8 3-7 4.7 5.2 10.0 4.0 12.0 7.3 7.3 6.6 6.2 4.3 6.8 11.9 5.7 5.3 6.3 5.6 7.0 7.0 3.7 10.0 4.8 6.4 6-0 5.1 6-5 5.8 5.8 6.0 5.2 2.4 4.5 1 6.8 7.3 4.2 4.9 5.8 Cheyenne Wyo Pueblo, Colo Santa Fe. N. Mex Fort Stanton. X. Mex. . . Montrose Colo 7.4 6.7 6.3 4.6 7.0 11.4 5.2 4.2 5.3 5.6 6.3 8.2 3.4 9.8 4.5 6.4 5.5 5.2 6.6 6.3 5.9 5.3 5.2 2.8 4.4 7.2 7.0 4.1 5.0 5.8 Mean Havre Mont Miles Citv Mont. .. Helena, Mont Baker City Oreg Mean Idaho Fall* Idaho Boise City, Idaho Winnemucca, Nev Salt Lake City. Utah ... Mean Prescott Ariz Fort Apache, Ariz Fort Grant Ariz Yuma, Ariz Mean Walla Walla. Wash .... Portland Oreg. Mean Bed Bluff, Cal Sacramento. Cal Fresno Cal San Diego, Cal Mean . 18 TABLE II. Summary of wind movement, in percentages, for the years 1889-93. DODGE CITY, KANS. (Elevation above sea, 2,477 feet; anemometer above ground, 52 feet.i Miles per hour. 0-5. 6-10. 11-15. 16-20. 21-25. 26-30. 2 3 4 6 5 7 3 2 5 3 2 2 4 31+ 1 1 3 6 4 4 1 1 3 2 2 1 2 27 26 17 16 16 17 19 24 23 28 27 22 34 34 28 24 26 26 29 31 25 31 37 36 30 21 19 24 22 23 19 24 21 21 19 18 19 21 10 11 16 16 16 15 16 13 13 10 9 11 13 5 6 8 10 10 12 8 8 10 7 5 6 8 March April May JUly October Year NORTH PLATTE, NEBR. (Elevation above sea, 2,809 feet; anemometer above ground, 52 feet.) 32 42 16 6 3 1 31 37 17 9 4 1 1 March 26 34 21 11 5 2 1 April 20 30 22 14 7 3 4 May 19 31 26 14 6 3 1 23 34 21 12 6 3 1 J U ly 28 35 23 10 3 1 81 38 20 9 2 33 33 19 10 4 1 34 36 17 7 3 2 1 36 36 13 8 4 2 1 37 39 14 2 1 Year 29 35 19 10 4 2 1 PUEBLO, COLO. (Elevation above sea, 4,653 feet; anemometer above ground, 81 feet.) CO 25 8 4 3 1 1 February ... . 48 28 g 6 5 2 2 March 45 29 13 7 3 2 1 April 37 32 14 8 4 3 2 Mav 39 31 15 g 4 2 1 43 29 17 7 3 1 J U ly 43 34 16 6 1 52 31 12 4 1 52 31 13 3 1 October 57 28 8 4 2 1 59 24 8 4 3 1 1 December 54 25 9 5 4 2 1 Year 49 29 12 5 3 1 1 LANDER, WYO. (Elevation above sea, 5,373 feet; anemometer above ground, 34 feet.) so 11 4 g 1 1 1 February 84 10 4 1 1 o o March 75 14 6 3 1 1 o April 65 23 g 3 1 o May 67 22 8 3 o 68 19 8 4 1 July 71 18 rv 3 1 August . 79 15 4 2 71 17 7 3 1 October 84 9 4 2 1 82 10 3 2 1 1 1 80 12 4 2 1 1 15 C .) 1 o o 19 TABLE II. Summary of wind movement, in perce?itages Continued. SALT LAKE CITY. UTAH. {.Elevation above sea, 4,334 feet; anemometer above ground, 90 feet, i Miles per hour. 0-5. 6-10. 11-15. 16-20. 21-25. 26-30. 31+ January .... 83 73 61 56 56 55 58 63 61 72 74 72 65 11 17 25 25 27 29 29 28 27 20 16 18 23 3 6 9 13 11 12 11 7 8 6 6 6 8 1 3 4 4 3 2 2 3 2 3 3 3 1 1 2 2 2 1 1 1 1 1 March \pril May Julv \UTUSt . October Year YUMA, AEIZ. (Elevation above sea, 40 feet; anemometer above ground, 50 feet.) 54 25 13 6 2 o 52 26 11 7 3 1 March 49 ?? 12 8 3 1 \nril 47 29 12 7 3 2 Mav 43 33 14 6 3 1 47 37 13 3 o o J U ly 42 36 16 5 1 45 36 16 3 o o September . 58 31 8 2 1 o October 65 24 8 2 1 o November 61 25 10 3 1 o 58 24 11 6 1 o Year 52 29 12 5 2 o CO 8 I 1 . r UNIVERSITY OF CALIFORNIA LIBRARY BERKELEY Return to desk from which borrowed. This book is DUE on the last date stamped below. 12 1948 JUM 41981 JUI LD 21-100m-9,'47(A5702sl6)476 I