83 FEBRUARY, 1923. MONTHLY WEATHER REVIEW.
age and other painted instruments of the Weather Bureau equipment had the paint sand-blasted from them, as did some houses about town. The rain ga e on. the
Figure 1 gives an idea of the velocity of the pebbles that were carried by the wind. The window is riddled
Weather Bureau grounds registered 0.15 inch o f sand.
FEB. 12, 1323.
FIQ. 3.-Barograph trace at time of windstorm at Independence, Calif., Feb. 12, 1933.
as if a machine gun had been turned on to it. The pebbles passed through without cracking the panes to the edge. Figure 2 shows a ranch house at Fort Inde- pendence after the storm. The loss of the roof of the courthouse is explained as the result of internal pressure. No doubt if the windows
had not blown in the roof would have held intact. All windows in this buildin were heavy plate glass. It was
and not the force of the wind. The writer experienced a 110-mile hurricane, with air free of dhbris, a t Tatoosh Island, Wash., during which no windows in the Weather
the foreign objects in t a e wind that broke the windows
the storm was felt to the westward and northward only. warm months the prevailing direction of the valley, and in the other months it is During the winter months moderately the northwest fre uently occur. These
Winds from the northeast or east seldom occur. The northwest winds are cold and dry tlnd bring clear weather, and often continue a t a rate from 25 to 50 miles an hour for several cla Y at a time, but rarely
from the west or southwest down from the mountains. Since 1914 wind velocities of 60 miles an hour, or more, have occurred at this station six times. Previous to this storm Owens River Valley has suffered but little from windstorms.
winds, whether up or down the val ? ey, are regular winds.
reach destructive velocities. A e destructive winds come
PREDICTING MINIhlUM TEMPERATURES.
By WALTER J. BENNETT, MeOeorologist.
[Weather Buresu Office, Tampa, Fls., February 19,1923.1
The relation between relative humidity and dew point a t the p. m. observation and the minimum temperature the next morning has been shown by Prof. J. Warren Smith and others to be a very real one. (Sup lement 16,
that for some localities the mean solar noon observation may be used with great accuracy, and thus give a deter- mination of the mnimum temperature early enough in the day to be of ractical benefit. (E. M. Keyser,
MONTHLY WEATHER & EVIEW, October, 1922.) But it would be a mistake to put too much faith in the relative humidity and dew point observations, especially in cases when a dro in tem erature is to be espected
of a large mass of cold air. Such cases are of great im ortance in Florida. tolerably exact forecast of the minimum temperature is a real necessity for this section of the country. A study of past records shows that when the tempera- tures mentioned in the next following paragraph are reached the results as indicated will follow. The tempera- tures are those recorded in the Weather Bureau ther- mometer shelter exposed on the roof of the postroffice
MONTHLY WEATHER REVIEW.) Also it has % een sliown
not from radiation af one, but ! rom the bodily movement
building in Tampa. Ground and grove temperatures will run from 2' to 10' lower within a few miles, depend- ing upon local conditions. A temperature of 40' to 42' indicates light frost with 1itt.le damage; 35' to 40' means considerable damage to tender ve etables; 36' to 38" much greater damage;
32' to 34', almost complete destruction of vegetables, except when rotected or in s ecially favored localities;
on citrus trees; 28' to 30' will kill youn trees un ruiked:
seriously damago young growth; 24' to 2G' will freaze many oranges and damage trees; 22' to 24' means practically all oranges frozen and many trees frozen to
Below 33O means extremely great damage t e a 5 i m d oves not protected by heaters. To etermine how valuable the indications of the noon relative humidity and dew point would be in determining the minimum temperatures, dot charta were made for the several months of December, January, February, and March covering five years past, and curves were drawn free hand. Separate charts were h t made for clear,
34' to 36" a, f amage of 50 per cent or more to vegetables;
30' to 32' kil P s Vegetables and R amages the youn growth
26' to 28' will freeze some oranges, kil f young trees and B
.
84 MONTHLY WEATHER REVIEW. FEBRUARY? 1928
artly cloudy, and cloudy weather, but there was so .
%ttle difference in shape and location of the several curves, that all observations were used for each month, r ardless of the state of the sky. %he records of the office were then searched for all predict,ed miiiimum temperature during the past five Minima .we predicted only when frost or near- E?temperntures are expecbed, and such predictions are seldom made on Sundays, because few or no re orts
57 cases in which a definite prediction of the minimum temperature was made. dicted temperature and the actual minimum was t en taken for each case. point a t the noon observation were then consulted and the de arturc of the temperature indicated by the curve
The actually predicted minimum was within 4’ of the real minimum 36 of 57 times, or 63 per cent. The mini- mum indicated by the dew-point humidity curve showed exactly the same percentage of accuracy. This was for the 57 special cases only, when the degree of cold was of real importance. If minimum temperatures were pre- dicted by the curve eve da , the average percentage of
The average error, taken regardless of sign, of the tem- erature forecast from the a. m. weather map was 4.1.
$he average error of the temperature indicated by the dew-point humidity curves was 4.3, nearly the same. Considering plus and minus departures, it was found that the minimum temperature was below the curve-indicated
are received Sunday mornings. In all, there were P ound
Bf“-
The error between this
Tho relative humidity and the dew .
from t P le actual minimum was taken for each case.
accuracy would be cons1 -7 erab s y greater.
temperature 27 times, with an average minus departure of 4.4. It was above 24 times, and exactly equal 6 times. Incidentally, this almost equal distribution of plus and minus de artures furnishes a fair check on the accuracy
Taking the temperatures actually forecast from the weather map, it was found that the real minimum went lower than the predicted minimum only 14 times out of the 57, and the average minus departure was only 1.9’.
This was undoubtedly due to a conscious intention on the art of the forecaster to predict a temperature just a little rower than he actually ex ected.
this locality, we find that the temperature orecast for Tampa on the morning of February 2,1917, for the morn- in of February 3 was 25’, and the actual minimum was
26 . This was before the mean solar noon observations were begun, but computations from corrected hygrograph readings applied to the curve, show that the minimum indicated by the curve would have been 33’. Estimatin the minimum temperature in this case 7’ too high woul have been a calamit .
the minimum temperature is of real importance and not merely of scientific interest, it may be estimated from the S a. m. weather map more accurately than it can be indi- cated by dew-point humidit curves from observations at noon. But these noon rea&ngs may be useful in check- ing or confirmin the minimum forecast, and i t may be a correction curve based u on the garometric gradient that would be of practical v ue.
with whic ?h the curves were drawn.
F
Going back now to the r ast serious freeze ex erienced in
B
i
It would seem t E a t for this particular locality, when
ossible to deveop K 89
NOTES, ABSTRACTS, AND REVIEWS.
Paul Frederick Maxlcell (1892-1923).
Paul Frederick Maxwell, in the strength and vigor of earl manhood was overwhelmed b a snowslide on one
Wheel Ga Colo. Ex erinient Station on March 5, 1923.
Mr. Maswell left camp about 9 a. m. to make the regu- larly scheduled snow depth and density measurements on the denuded watershed. Not returnin at the ex-
and Torrence immediate1 set out in search of him. On
occurred on what is known as “Snowstake area B-11;”
and when they found snowshoe tracks leading to the slide their worst fears were realized. An hours search failed to reveal the body, but finally the heel of a s:iow- shoe was discovered pro’ectin from a snow bank at the
Slowly and with difficulty the body was borne back to the camp, arriving there a t 7:30 p. m.! a little less than
12 hours from the time of his departure in the morning. Mr. Maxwell entered the service of the Weather Bu- reau in 1916, and had seen service at North Head, Wash.; Boston, Mass.; and New Haven, Conn. From the last- named station he was transferred to Wagon Wheel Gap and was in his second year of dut at that station. Snowslides have occurred in loth watersheds a t the Wagon Wheel Gap roject while yet in timber due to the
removal of the timber from the B-area the hazard of slides greatly increased and this fact was full realized by the observers, who, nevertheless, with a fi B elity that is extremely gratifying, carried on under these circum- stances.
of t T e steep slopes of the denude1 area of t,he Wagon
When his % ody was P ound 6 liouis later life was estinct.
petted time his companions at the camp, & essrs. Weld
axriving at the B-area t i ey saw that a snowslide had
bottom of the slide, an 4 % the ody was soon uncovered.
very great angle o P the slopes, 35’ in places. With the
It was the writer’s privilege to have known Mr. Maxwell personally and to have discussed with him the observational material of the pro’ect. His mind was keenly alive to the roblems invo ll ved and he gave his
His memory will be treasured by his associates in the Weather Bureau as one who made the supreme sacrifice, just as truly as did those who ga-re up their lives on the scarred battlefields of France and Belgium. Mr. Max- well is survived b his arents, hfr. and &. W. D. Maxwell, of Baker &ty, &eg., and by his wife and three small children. To all of these, his associates in the Weather Bureau extend their deep and lasting sym athy in the loss sustained by this t r w c event.-A. J. E!
best efforts toward t 1 eir solution.
WARMER AIR IN REAR OF CYCLONE OF FEBRUARY 8,
1923.
The morning weather map for the United States, February 8, shows a depression of the barometer accom- panied by the usual cyclonic wind circulation centered over the Great Basin. Immediately in the rear of the cyclone center is the legend ’‘ Warmer 20 de ees.” The orthodox temperature distribution in cyc Y ones which visit the United States is warm in front and cold in the rear; it is well known, however, that this distribution does not hold for the Pacific coast and the northern Plateau regions. The present case is of sufficient interest to warrant a few words of explanation. On February 7 an anticyclone, sea-level pressure,
30.40 inches, occu ied the regon in question, but, by the
‘on was occupiefi by a cyclone as above mentioned.%e next mornin , it g ad vanished and the identical