MONTHLY WEATHER REVIEW.
Editor : Prof. CLEVELAND ABBE. Assistant Editor : CLETEZAND ABBE, jr.
VOL. .xxxvI. SEPTEMBER, 1908. No. 9.
The MONTHLY WEATHER REVIEW summarizes the current man-
uscript data received from about 3,600 land stations in the
United States and about 1,260 ocean vessels; it also gives the
general results of the study of daily weather maps based on
telegrams or cablegrams from about 200 North American and
40 European, Asiatic, and oceanic stations.
The hearty interest shown by all observers and correspon-
dents is gratefully recognized.
Acknowledgment is also made of the specific cooperation of
the following chiefs of independent, local, or governmental
services: R. F. Stupart, Esq., Director of the Meteorological
Service of the Dominion of Canada; Sefior Manuel E. Pastrana,
Director of the Central Meteorological and Magnetic Observa-
tory of Mexico; Camilo A. Gonzales, Director-General of Mexi-
can Telegraphs; Capt. I. S. Kimball, General Superintendent of
the United States Life-Saving Service; Commandant Francisco
5. Chaves, Director of the Meteorological Service of the Azores,
Ponta Delgada, St. Xchaels, Azores; W. N. Shaw, Esq., Director
Meteorological Office, London; Maxwell Hall, Esq., Govern-
ment Meteorologist, Kingston, Jamaica; Rev. L. Gangoiti,
Director of the Meteorological Observatory of Belen College1
Havana, Cuba; Luis G. y Carbonell, Director, Meteorologioa,
Service of Cuba, Havana, Cuba; Rev. Jose Algu6,S. J., Direc-
tor of the Weather Bureau, Manila Central Observatory, Phil- ippines; Maj. Gen. BL A. Rykachef, Director of the Physical
Central Observatory, St. Petersburg, Russia; Carl Ryder,
Director, Danish Meteorological Institute, Copenhagen, Den-
mark.
As far as practicable the time of the seventy-fifth meridian
is used in the test of the MONTHLY WEATHEB REVIEW.
Barometric pressures, both at land stations and on ocean
vesbels, whether station pressures or sea-level pressures, are
reduced, or assumed to be reduced, to standard gravity, as well
as corrected for all instrumental peculiarities, so that they
espress pressure in the standard international system of meas-
ures, namely, by the height of an equivalent column of mer-
cury at 32O Fahrenheit, under the standard force, i. e., apparent
gravity at sea level and latitude 45'.
FORECASTS AND WARNINGS.
By Prof. E. B. QAEEIOTT, in charge of Forecast Division.
September opened with low barometric pressure and severe
gales over the British Isles and the northwestern coasts of
continental Europe. Over the Atlantic and Pacific oceans
pressure was near the normal. An extensive area of high
barometer covered western interior portions of the North
American Continent, and a storm area was central near the
middle Atlantic coast of the United States. The barometer
had risen to 30.18 inches at Vladivostoli and pressure was
decreasing over the interior of Siberia.
During the first decade of the month there was a gradual
tho well-marked change from summer to fall types of at-
mospheric pressure over the great Asiatic continental area.
I n the United States barometrio movements were as a whole
abnormally slow, and resulted in a period of stagnated dry
weather over middle and northern districts east of the Rocky
Mountains during which forest fires caused considerable dam-
age in localities in the North-Central States, and a serious
shortage of water was experienced in many sections.
From the 1st to 3d the Atlantic coast storm moved rapidly
northeastward to the Canadian Maritime Provinces, and during
the succeeding five days apparently crost the Atlantic in the
middle latitudes. From the 2d to 4th the interior American
high area moved south of east to the middle Atlantic coast
attended by a marked fall in temperature and light frost in
the upper Lake region, in mountain districts and in lowlands
of the Middle Atlantic and New England States. During the
5th and 6th a shallow barometric depression attended by heavy
rains moved rapidly from the Gulf States along the Atlantic
coast.
On the morning of the 9th there was evidence of a storm
formation near the Leeward Islands of the Lesser Antilles,
and during the afternoon and night of that date the center of
the disturbance past on a northwesterly course near and to
the eastward of Porto Rico. By the morning of the 10th the
storm-center had advanced to a position north of Porto Rico,
and by the morning of the 11th had past to the westward
of Turk's Island, where wind velocities estimated at 80 miles,
39-1
or more, an hour caused destruction of life and .property,
Continuing a north-of-west course during the 12th and 18th.
the center of the storm recurved northward during the 14th
and past to the eastward of Nassau, Bithamas. From this
region the disturbance moved northeastward between Bermuda
and the American coast during the 15th and 16th, past
south of the Canadian Maritime Provinces during the 17th,
and disappeared over the Atlantic east of Newfoundland after
the 18th, after which it apparently merged into an extensive
@rea of low barometer that extended southward from Iceland.
Beginning the morning of the 10th advices regarding this
storm were telegraphed dailv until the 16th to Atlantic and
Gulf ports. The advices of the 10th stated it would be dan-
gerous for vessels during the next few days in the subtropical
region of the Atlantic off the south Atlantic coast of the
United States, north of the West Indies, and thence to the
longitude of Bermuda. I n view of a possible recurve of the
storm somewhat farther to the westward than the longitude
in which the turn to the northward was actually made, advices
urged precautionary measures along the coasts of the Florida
Peninsula. The exceptional severity of the storm during its
westward passage over the Bahamas and attending its subse-
quent northeasterly course over the Atlantic is shown by
reports of vessels that were caught within its vortex.
A remarkable period of dry weather over the northern half
of the United States east of the Rocky Mountains set in dur-
ing the latter portion of August and continued well into the
third deoade of September. I n two or three instances during
this period indications that as a rule presage rain partially or
wholly failed. The rather remote causes of the dry spell are
now recognized. It was not possible to detect and interpret
them with previous imperfect knowledge of the operative in-
fluence of the greater barometric areas. On September 22d
the following forecast based upon radical changes in pressure
was issued:
A barometrio dlftturbanoe will cross the country from about the 24th
to 28th, attended by rains that will set in over the oentral valleys about
276
276 MON!CHLY WEBTHER REVIEW. ~EPTEYBER, 1908
t h e close of this week and extend over the Atlantic States by the begin-
nlng of next week. Following the rains there will be a sharp fnll in tem-
perature, with frost in the central valleys and Eastern States north of the fortieth parallel.
The rains that attended this disturbance occurred as fore-
cast and relieved the drought in northern and northeastern
districts. The frosts that followed its passage extended over
the Middle Western States and the States of the Ohio Valley
and middle Atlantic coast.
The following comment on this storm and cool wave is
made by the Market Growers Journal, Louisville; Ky., of Sep-
tember 30, 1908:
A general area of rain set in over practically the entire Rocky Moun- tain region the latter half of the week and moved gradually eastward,
bringing rain to the Mississippi Valley States by Saturday night. The raln reached the Ohia Valley Sunday night of this week and the indica- tions at thls wrlting are that before this issue reaches our readers the
drought in all sections of the country which have been suffering will
have been brought to an end. An interesting fact of this raln period is that i t was predicted early last week in a bulletin sent out by the Weather Bureau at Washlngton. * The general rains, which
marked the end of the drought, are being followed by a period of cold weather which will mark the end of the unusually warm weather of
September.
An editorial in the Albany, N. Y., Journal of September 29,
reads as follows:
Just now there is in evidence the fulfllment of a forecast made a weekago. Early last week it was announced from Washington that condltlons were favorable for the development of B general rain area in
this part of the country by about the 28th Instant. Because of the long absence of raln that prediction was of unusual interest, and the arrival
of the time appointed for its fulfllment was awaited with mingled hope and apprehension. There was a widespread feeling of relief when the sky became over- cast and pwclpitation began, gradually as it nearly always does after a
long period of dry weather. The Weather Bureau is to be congratulated upon the accuracy of a “long di*@nce” forecast, made at a critical time when all ordinary
signs. even to that old standby, the sun’s crossing the line,” seemed to fail.
The Kansas City, Mo., Star of September 27; remarks as
follows :
An interesting fact ahout the storm area that is now moving across the country Is that i t wits accurately predicted by the Weath:: Bureau last Tuesday evenlng (September ’22). when a long distance fawecast was put o u t saying that rains would fall in the centrd vallrys about the
closn o f this week, and in the Atlantic States at the beginning of next week, folbiwed by frnbt north of the fortirth parnllel.
The Weather Bureau’s forecrbts for a week ahead are &Ill in the ex- perimental etage, but they promise to be nf great value. They are based
on reports of bammt-trio pressures in various parts of the world, lndics-
ting the progress o f storm areas. Last Tuesday’s forecast was hased on reports of low brrnmetrio pressure at the time i n home and Sitka,
Alaska, and in Honolulu. But the present storm area firat nppeared over the Rocky Mountain regions with a bank of high barometer all week along the Paciflc coast, sn thnt it seernsqunctionable as to whether
it cnme from I he ccmditimson which the Weather Bureau based its long distance forecast of last Tuesday.
The forecast was based on reports for several days preceding
the date of its issue. On September 17 Pacific pressure was
high over Honolulu and low over Nome. Three days later
pressure was high over the Bering Sea region and low over
the Hawaiian Islands. It has been observed that pressure
changes over the Pacific Ocean forerun by several days certain
changes on the Pacific coast and the American Continent as a
whole. On September 23 a decided fall in the barometer
occurred over the middle and south Pacific coast districts, and
on the following day the barometric disturbance appeared, as
stated, over the Rocky Mountain districts. It is true that
pressure continued high over the north Pacific coast. That
was expected. The predictions are based, not necessarily upon
the progress of individual storm and high barometer areas,
but upon a study of atmospheric conditions over the whole
Northern Hemisphere, and more directly, at this season of the
year, on the. general circulation of the atmosphere over the
Pacific and Atlantic oceans. In winter the great continental
areas of high barometer, and more especially the Asiatio high
mea, appear to dominate the general atmospheric changes of
the Northern Hemisphere. By a study of the association and
interrelation of the greater areas of high and low barometric
pressure is the forecasting of weather changes for a week, or
more, in advance made possible.
On September 24 West Indiau stations were advised of the
presence of a cyclonic disturbance east of the Lesser Antilles
in latitude about 1 5 O north. On the following morning West
Indian ports and Atlantic aud Gulf shipping interests were
informed that a disturbance of marked iutensity near the Lee-
ward Islands of the Lesser Antilles was moving in a westerly
direction. During the succeeding two days the hurricane cen-
ter moved on a west-northwest course, and at G a. m. of the
28th past near Port au Prince, Haiti, with a reported minimum
barometer reading at that place of 29.24 inches. Continuing
a west-northwest course the vortex of the storm advanced over
or near the Great Bahama Bank by the close of the month
and recurved thence northward over the western Bahamas by
Cctober 1, with reported minimum barometric pressure 28.68
inches at 10 a. m., and wind exceeding 80 miles an hour from
the south at Nassau. Assuming a northeasterly course the
storm then advanced over the Atlantic in the direction of Ber-
muda. Further advices that may become available regarding
this storm will be given in the October, 1908, MONTHLY WEATHEB
REVIEW.
It is interesting to note that during the present season three
West Indian hurricanes have occurred simultaneously with
typhoons in Asiatic waters. I n the third decade of July a
destructive typhoon struck Hongkong, and a severe storm that
had its origin in the Caribbean Sea moved northward along
the Atlantic coast .of the United States. I n the second decade
of September a typhoon advanced from the Philippine Islands
northward along the eastern Asiatic coast, and a hurricane
devastated the eastern islands of the Bahama group, moving
thence northeastward. In the third decade of September a
hurricane swept west-northwest from the Lesser Antilles to
the western Bahamas and recurved thence northeastward, and
a typhoon past from the Philippine Islands westward over
the China Sea.
BOSTON FORECAST DISTRICT.*
[New England.]
The drought that
prevailed was one of the most severe in many years, particu-
larly in Vermont, and the average rainfall for New England
was the smallest for September since the establishment of the
New England Climatological Service in 1888. The first k i l h g
frost of the season occurred in Maine, New Hampshire, Ver-
mont, and parts of Massachusetts on the 16th. Frost warn-
ings were sent to cranberry growers on the 16th. Stirm
warnings were issued on the let, 17th, and 28th, and there
were no storms without warnings.4. W. Smith, Diatrict Fore-
caster.
The month was unusually warm and dry.
HEW ORLEANS FORECAST DISTRI(IT.*
[Louisiana, Texas, Oklahoma, and Arkansas.]
A disturbance for which storm warnings were ordered ap-
peared in the west Gulf the morning of the 17th. This dis-
turbance was attended by a wind velocity of 64 miles an hour
at Galveston, Tex., and by general showers over Louisiana,
eastern Texas, Oklahoma, and Arkansas. The first frost of the
season was reported in Oklahoma and northeastern Texas on
the 28th, for most of which warnings had been iseued-I. M.
Cline, Dith-kt Forecader.
LOUISVILLE FORECAST DISTRIOT.*
[Kentucky and Tennessee.]
The month was remarkable on account of the severe drought
and unueual warmth that prevailed, except in southeastern
Tennessee where rainfall was about normal. Over the balance
of Kentucky and Tennessee practically no rain fell from the
S E P T ~E R , 1908. MON!CEliY WEATHER BEVIEW. 277
6th to 27th. A decided change to colder came with the rain-
storm of the 27th, and the last three days of the month were
unusually cold, with frost generally in Kentucky and parts of
Tennessee. Special warnings were isclued for the frost.-
F. J. Walz, District Forecaster.
[ Indlana, Illinois. Mlchlgan, Wlsconsln, Minnesota, Iowa, Ml6SOUrf, North Dakota, South Dakota, Nebraska, Kansas, and Montana.]
Except during the last few days of the month September
was unusually warm. The change to cooler weather set in
over the Western States on the 25th and advanced slowly
eastward, bringing unseasonably cold weather and general
frosts for which warnings were issued well in advance. Frost
warnings were again issued on the 30th in advance of another
cool area. Special frost warnings were issued to the cranberry
marshes of Wisconsin on the 1st. 2d, and (ith, and in each
case frost and freezing temperatures were reported in the
bogs. The drought conditions continued from the previous
summer months, and they were not effectually broken until
the passage of the storm of the last week. The only disturb-
ance that justified storm warnings crost the upper Lakes on
the 30th, and warnings were issued for this storm on the
morning of that day.
The following action was taken by the South Dakota State
Board of Agriculture in connection with a special forecast
telegraphed from Chicago to the Local Office of the Weather
Bureau at Huron, 8. Dak., on September 8:
By resolution of the State Board of Agriculture, it is my duty and pleasure to express to you, and through you to the U. S. Weather Bureau, our sin ere thanks for the long forecast given us for the week d our
Fair. It was of great value for us to know this as i t saved muoh expense in preparing for rain as we felt we should. Besides this it was a great
relief of mind to the management to know that we could expect such fine weather.
(IHICAQO FORECAST DISTRI(IT.*
J. W. CAMPBEEL, Preaidmt,
H. J. Car, Proj>sanr and District Forecastrr.
DENVER FORECAST DISTBICT.*
[Wyoming, Colorado, Utah, New Mexico, and Arizona.]
Except during the closing week the month was warm and
dry. From the 23d to 26th a heavy storm of snow and rain
moved from Wyoming to southern New Mexico. The storm
was followed by one of the most severe cold spells on record
during September in eastern portions of Wyoming, Colorado,
and New Mexico. Timely warnings of frost and freezing tem-
perature were issued in connection with the cold spel1.-
P. IllcDonough, Local Forecaster, teniporarily in charge.
SAN FRANCISCO FORECAST DISTRI0T.t
[California and Nevada.]
The most striking feature of the month was the storm that
prevailed over southern California on the 23d to 25th. Gen-
erally speaking the rainfall was the heaviest during September
since records have been kept. It varied in amount from half
an inch to several inches. The raisin-making section had
ample warnings of the rains and the benefit of the service bas
been acknowledged. No frost nor storm waruings were issued
during the month.-A. a. Mctldie, Professor and DWrict Fore-
caster.
PORTLAND, OREG., FORE(IAST DISTBICT.~
[Oregon, Washington, and Idaho.]
The month was unusually dry in western and northern sec-
tious, and temperature was slightly above normal east of the
Cascade Mountains. A moderate disturbance crost the northern
portion of the district the last day of the month. Light frosts
occurred on the 23d and 24th and heavy frosts on the 25th
and 26th. Warnings of the storm and frosts were issued in
time to be of service to those interested in them-E. A. Real#,
Daltlrict Forecaster.
RIVERS AND FLOODS.
The feature of the month was the general drought that pre-
vailed over the middle and northern district3 east of the Rocky
Mountains. Little or no rain fell over this extensive area
until the end of the month, and all streams, escept the Missis-
sippi and Missouri, were at very low stages. The two larger
rivers were not lower than usual for the season of the year.
The drought conditions were most severe in the Ohio Val-
ley and the Middle Atlantic States, and in many places rivers
were lower than ever before. Navigation was practically sus-
pended on the Ohio, and many manufacturing plants in the
upper Ohio Valley were compelled to suspend operations on
account of lack of water.
Delayed reports of the flood of August and early September
in the rivers of eastern South Carolina show that the damage
caused thereby amounted to over $900,000, divided as follows:
Property loss, excluding crops, $200,000; losses of crops,
$700,000. The losses due to erosion of land and suspension
of businew were reported great, but detailed reports were not
available.
The highest and lowest water, mean stage, and monthly
range at 211 river stations are given in Table IV. Hydro-
graphs for typical points on seven principal rivers are shown
on Chart I. The stations selected for charting are Keokuk,
St. Louis, Memphis, Vicksburg, and New Orleans, on the Mis-
sissippi; Cincinnati and Cairo, on the Ohio; Nashville, on the
Cumberland; Johnsonville, on the Tennessee; Kansas City, on
the Missouri; Little Rock, on the Arkansas; and Shreveport,
on the Ked.-H. G. Frankenfild, Profwor of Meteorology.
* Morning forecasts made at district center; night forecasts made at
+ Morning and night forecasts made at dietrict center. Washington, D. C.
-
~
SPECIAL ARTICLES, NOTES, AND EXTRACTS.
R &S m OF EXPERIMENTS IN AEBODYNAMICS
By Dr. A. F. ZAHM. Dated Waahlogtoo, D. C., August 8.1, 190S.
INTRODUCTION.
Aerodynamics may be defined broadly as the science of
motion of air, or an aeriform fluid. Commonly air alone is
implied in the word. This is especially true when the name
ie used by engineers. With them it is the analog of hy-
draulics, which is the science of motion of water. Both
sciences treat not only of the movement of their peculiar
media, but also of its effects on objects, or machinery, con-
nected with the fluids.
Bn important function of aerodynamics is to determine the
velocity and stress of air at every point of this medium, when
it flows past an obstacle, the physical conditions of the fluid
being given or observed by means of suitable instrumeuts.
From the point-velocity the stream-lines may be mapped; from
the point-stress about an object the resultant pressure and
friction may be found by integrating over its surface.
Equivalent results may be obtained if the object move
against the fluid, since only the relative motion is of conse-
quence. Devices are in use, also, for revealing these inte-
grated effects directly, without first finding the point-velocity
and point-stress. Some of these will be described presently.
Experimental aerodynamics may be studied in its elements,
as diatinguished from its applications, by considering it under
these heads: (1) velocity and stream-lines; (2) normal stress
and resultant pressure; (3) shearing stress and resultant
friction; (4) combined pressure and friction. To trace the
development of even this much of the science would require
a large volume.
The following pages present a brief sketch of such of the
writer’s experiments as may be classed under the above heads,