THE

AMERICAN JOURNAL OF SCIENCE

[THIRD SERIES.]

ART. XXXV.-Daily March of the Wind Velocities in the United States; by FRANK WALDO, Princeton, New Jersey.

[The following matter is extracted from a paper prepared by the writer for the Weather Bureau of the Agricultural Department, and is published with the kind permission of the proper authorities.]

IN the Appendix No. 14 to the Chief Signal Officer's Annual Report for 1890, the average wind movement is given, for a large number of stations, in miles per hour for each hour of the day (1 to 24), for each month of the year, and also the averages for all of the months of the seven years 1883-89.

This presents most valuable data and is certainly the most unique of the tabular compilations published by the Weather Bureau. Hitherto we have relied mainly on the papers published by Hann and Köppen for collected data concerning the daily period of wind velocities, and even in these there are comparatively few places of observation taken into account. The publication of such data as those which we are considering, from a large number of stations having a variety of immediate exposures, and distributed over so large a portion of an entire continent, furnishes material for a very complete treatment of the subject of hourly winds. The present paper is mainly devoted to a view of the conditions of the geographical distribution of some important phases of the daily march of the wind velocities. The material is sufficiently rich to serve as a basis for a number of similar and more complete investigations. The hourly wind velocities as originally published are arranged according to synchronous hours of the 75th meridian. time. It is inconvenient in that form for many kinds of inves

AM. JOUR. SCI.-THIRD SERIES, VOL. XLIX, No. 294.-JUNE, 1895.

tigation and so I have arranged the data for the four midseasonal months and for the year, according to the local times; and, moreover, have converted the published anemometer miles (with constant 3.00) into true miles by means of Marvin's Table published by the Signal Service. I have also grouped the stations geographically and not alphabetically as given in the original table of hourly winds. This table cannot be reproduced here on account of its length, but I may remark that the relations of the maxima and minima have been investigated; and in the unpublished table are given the amplitudes or ranges in miles per hour, the excesses and deficiencies in terms of the average, and the amplitudes in terms of percentage of the averages. An account of this will shortly appear in the American Meteorological Journal.

The curves showing the daily march of the hourly wind velocities for January, July and the Year, I have also drawn. for individual stations, but these cannot be reproduced here on account of the expense of drawings. The characteristics of these curves show marked variations with changes of geographical position, as we should expect: and while the number of years of observations which have been employed, which for the most cases is 7, is not sufficient to remove all irregularities from some of the curves, yet in most cases a sufficiently good idea of the daily march is given. I have however given the curves for January and July for 20 stations more or less representative of the various sections of the U. S. See the curves and explanation of the diagram at the end of this paper. This material is of such importance that the curves deserve to be taken up for discussion for individual stations, but it is only possible, in the present instance, to treat them in groups. Without further preliminary remarks I will give the main characteristics of these curves in what seems to me to be a proper order. In counting the hours, midnight is given as 0. The wind velocities are given in miles per hour: written m. p. h. In mentioning the characteristics of these curves, it will be remembered that a sharp ascent or descent indicates a rapid change in the wind velocity from hour to hour; a flat curve indicates no change; a sharp crest or trough shows an extreme maximum or minimum of short duration, while when well rounded they indicate a period of several hours during which the conditions at these phases continue to prevail before the swing to the opposite phase sets in.

Atlantic Coast.-On the exposed Atlantic Coast (Curve 1) there is in January but a slight variation in the wind velocity from hour to hour during the whole day, the average wind being 15 or 16 m. p. h.; but for July there is a strongly defined maximum (12 or 13 m. p. h.) at about 16a, and a mini

mum (9.5 m. p. h.) at 4h on the northern and 6 on the Central Coast (7.5) and preceded in this last case by a slight secondary maximum at 3h; while for the Year the maximum is well defined at 16 on the northern (14.5 m. p. h.) and at 151 on the southern (12.5 m. p. h.) coasts, and the minima occur at Oh and 5 for the former (nearly 13 m. p. h.) and about 22h and 4h. for the latter (10-5 m. p. h.), and in both cases there is a slight secondary maximum at 3.

For the ordinary or partially sheltered Atlantic Coast stations (Curve 2) at which the land influence is strongly felt, there are in January well defined primary phases of max. (about 12 m. p. h.) and min. (about 9 or 10 m. p. h.), but the secondary phases are very weak and the curves become quite flat at these times. The max. phase becomes more marked with the southward progress from the north, not only on account of actual increase in the absolute height of the curves (which increase perhaps does not extend to places south of the Carolina Coast) but also on account of the fact that the crest of the curve becomes sharper and the rise and fall more abrupt. The daily curve during the hours of deficiency of wind (below the average) becomes more flattened out and of greater extent in the south. For July the upward swell of the curve is very much broader than for January, throughout the whole coast, but it is especially so at the south. The extreme maximum is more pronounced in all cases, but is most so at the north. At this season there is a remarkable similarity in the length of the swell of the curves (about 16 hours duration) throughout the whole coast.

For the depressions of the curves (i. e. periods of least wind) a greater irregularity is noticeable at the north, where slight secondary phases are visible.

For the Year, as might be expected, the curves possess characteristics between those for January and July: but they resemble the latter very much more closely than they do the former. The crest of the curve is, however, more rounded for the Year than for July; but the amplitude is not quite so great, nor the period of excess above the nearly level portion of the curve at the minimum quite so long, the latter for the Year being about half of the twenty-four hours.

Gulf of Mexico Coast.-The eastern, western, and northwestern coasts are represented (Curves 3 and 4). For January there is but slight absolute change in the irregular curves for these three coast sections. The curves for the extreme eastern and western coasts are somewhat similar as to the times of the phases, but the early afternoon maximum (at about 15") is much more strongly marked in the west than in the east. In the northwestern part (Galveston, Texas) the amplitude is not

so great, and the ill defined crest is several hours earlier than in the other sections just mentioned.

For July, in the eastern part, a well defined and gradual but not excessive maximum rises above a fairly level period of deficiency, there being no strongly marked individual minimum; but in the northwestern part the maximum and minimum are very sharply marked, in each case the curve comes to a point, the amplitude is rather large for a coast exposure and the time of maximum (at 17") is considerably retarded as compared with that in the eastern part (at 12). In the extreme western part (at Corpus Christi) a curve of very marked peculiarities presents itself; it has a single rather sharply defined minimum of 5 m. p. h. at (51) about the same time as in the northwestern part, and a very high but round-crested maximum of nearly 15.5 m. p. h. (at 14h to 18h) with very steeply inclined sides which extend to the sharp angle of the minimum with as rapid a slope throughout the whole length as is ordinarily observed for inland stations at about (or a little after) the noon hour. It is seen that the whole amplitude of fluctuation thus becomes enormously great (over 10 m. p. h.).

For the Year, in the eastern and northwestern parts the maximum is well marked but the crest of the curve is very much flattened; and there is in both cases a nearly level minimum for about half of the twenty-four hours, with a rather steep but slight increase to a nearly level maximum which lasts for six hours. At the extreme west there is an open and very well rounded maximum and a well rounded minimum with a steep ascent, but with a more gradual descent connecting the two: the amplitude is about the same as that for an ordinary inland station.

The Great Lakes (Curve 5).-In January the curves are nearly all irregular, but the amplitudes of the irregularities are not great. The early afternoon maximum, although slight, is plainly marked in all cases: it is sometimes rounded and sometimes sharp-crested; that for the lesser absolute wind velocity being usually the more rounded, and that for the greater wind, which indicates a better water exposure, becoming sharper. The nearly flat minimum portion of the curve is usually somewhat lower for the hours succeeding midnight, than for the hours just preceding it. Secondary phases are quite plainly shown in some of the curves.

In July the max. is well rounded and strongly marked on all of the curves, and this period covers more than half of the twenty-four hours in most cases. When the minimum portion follows midnight it is in many cases a little higher than when preceding it, and is consequently somewhat of a reversal of the conditions for January. Secondary phases are not present.

For the Year, as is usually the case, the curves form a sort of mean between those for January and July. There is a nearly level min. period and a very well marked max. ; this last with characteristics very similar to those for July. The period of the mid-day rise above the level minimum is about half of the twenty-four hours. Secondary phases are not noticeable.

Some peculiarities which distinguish the Upper from the Lower Lake Regions are mentioned farther along in their proper place among inland stations, and the above remarks are offered for comparison with those pertaining to the Gulf coast. Pacific Coast (Curves 6, 7 and 8).-For January, at the north there is a nearly mid-day principal minimum, with a slight min. shortly after midnight, and with two nearly equal maxima from four to six hours before and after midnight; the changes are gradual and relatively small. On the Central coast the rise of the single max. above a somewhat level min. period is gradually accomplished in about twelve hours, the slope of the curve being gentle but the actual crest sharp. the south a single early afternoon max. rises rapidly from a nearly level minimum; the slopes of the sides of the max. which separate to eight hours apart at the base are steep, and the actual crest slightly rounded. (On the high bluff at Cape Mendocino, near the center, the sharp crested max. occurs shortly after noon, and the sharp pointed trough of the min. at about midnight; and while the descent from the max. is steep and regular, the ascent is at first steep and then from 4 to 11h there is little change, then another steep ascent; the whole range being excessive and greater than that for mid-summer.)

At

For July, at the north there are both primary and secondary phases of max. and min. ; but the secondary ones are slight, and the primary occur at nearly a reversal of the times for January, although the amplitude is slightly greater in July. On the central coast there is a single max. and min., but with an enormous amplitude. The crest is slightly rounded, but with a very steep slope, while the trough is more rounded, and the slope of descent becomes more gradual as the trough is approached. At the south there is a high max. with rather rounded crest, and steep sides, especially for the ascent; the curve for the period of minimum wind is quite flat during seven or eight hours, when the rise for the max. begins abruptly and finally ends nearly as abruptly. (The curve for the high bluff at Cape Mendocino, near the center of the coast, is quite similar in shape to that for the south, but the maximum is not quite so pronounced.)

For the Year, at the north the reversion of the phases noticed for January and July is such as to cause practically almost an erasure of the phases of max. and min., and the curve nearly becomes merely an irregular line with little variation of