Page images
[graphic][merged small]

A STREET BARBER OF BISKRA, THE BEAUTIFUL OASIS VILLAGE OF ALGERIA The author of the accompanying article is now visiting Algeria in search of a suitable place for the establishment of the National Geographic Society's solar-radiation observatory (see text, page 113).


A MIXED QUARTET AT BASSOUR, ALGERIA (SEE, also, pages 125 AND 126) This farmer seems to be carrying out the Mosaic injunction set forth by St. Paul in his first letter to Timothy, that "thou shalt not muzzle the ox that treadeth out the corn." However, the yoke prevents the hungry-looking horse from taking any part in the feast.

One is spelled with a capital C. That is the one used by food experts. We use the kind spelled with a small c. It is 1/1000 as large a unit as the other. In units of heat, the intensity of solar rays at mean solar distance outside our atmosphere is called "The Solar Constant of Radiation."



From several thousands of measurements of it, some from sea-level, others from mountains of from one to three miles' altitude, and from one set of automatic measurements from a free balloon at an altitude of more than 15 miles, we state the mean value of the solar constant as 1.938 calories per square centimeter per minute.

This means that if we should take a cube of water I centimeter (about threeeighths inch) on edge, blacken it so that it would completely absorb solar rays, and expose one of its surfaces at right angles to the sunbeam from a station upon the

moon, in March or September, when the solar distance is at its mean value, the rise of temperature would be 1.938 degrees centigrade.

"Solar constant" is a misnomer, for we find solar-radiation values variable between the limits 1.85 and 2.03 caloriesa range of nearly 10 per cent. Such extreme values are very rare, but fluctuations of 2 or 3 per cent from the mean are not uncommon. What evidence can we bring forward to support this claim?

Not only do the two independent stations of the Smithsonian Institution, in California and Chile, which are 4,000 miles apart, in opposite hemispheres, support each other, but visible changes in the sun go hand in hand with our measurements.


Increased radiation is to be expected when sun spots are numerous.

The sun rotates on its axis in about 27 days. Hence, sun spots and other visible features cross the solar disk in about half

[graphic][merged small]

AN OBSERVATION SHELTER ON MOUNT WHITNEY, CALIFORNIA On the summit of this, the highest peak in the United States outside of Alaska (14,501 feet), representatives of the Smithsonian Institution have at various times made solar-radiation observations.

that time and remain the other two weeks on the sun's invisible hemisphere. When in this rotation a sun spot crosses the central part of the visible disk, we almost invariably observe a depression of our solar-radiation values.

A notable case attended the great sunspot group of March, 1920, which caused magnetic storms on the earth and great displays of Northern Lights. So we see that while the presence of many sun spots betokens high solar-radiation values, each individual spot, as it crosses the sun's center, pulls the value down.

We think there is a sort of cloudiness over each sun-spot group which intercepts a part of the sun rays when it passes between the earth and the sun. Sometimes the depressing effect greatly exceeds the positive one due to increased solar activity. This is doubtless why, when spots are very numerous, the average solar-radiation values are sometimes made lower instead of higher.

Besides the sun spots, we see on the solar surface bright patches called facu

læ. When these are numerous, solar radiation values run high. Mr. Clayton has even made use of this relation to predict for five days in advance what values of solar radiation we are going to observe. He examines the sun with a telescope and sends a letter to the Smithsonian Institution, giving his prediction. The results of seven months' consecutive daily predictions of this character run strongly in his favor.

Hence, we see that many visible phenomena upon the solar surface undergo changes closely associated with our measurements. This fact cannot but strongly support the view that we are on the right track, and that our indications of changes in the sun's output of energy are real. The sun, in other words, is a variable star. There is evidence that some of the other stars vary in a manner similar to the solar variation.

But what of the practical applications of all this? Can we make use of solar variation to predict the changes of weather upon the earth?

A good many people

during the past cen-
tury, since sun spots
came prominently into
the field of study,
have endeavored to
base weather forecasts
upon their appearance.
Among these sun-spot
prophets in our coun-
try have been the late
Henry C. Maine and
Father Ricard. As we
have shown, there is a
real relation, though
not a very close one,
between the appear-
ance of sun spots and
the heating effect of
the solar rays. It is
not, therefore, surpris-
ing that some measure
of success may have
attended these sun-
spot forecasters.


About ten years ago, however, Mr. Clayton, who was then chief forecaster for Argentina, undertook to study the exact relations which exist between the weather and our results on the intensity of solar radiation. Realizing that our observations of that time were faulty, he did not attempt to use them individually,



While many of the original observations of the solar-radiation experts were made at the now abandoned Harqua Hala station, in Arizona, conditions in California were found to be better adapted to the work. The apparatus at the right, on top of the tower, is a cœlostat, which by a double-mirror arrangement reflects the solar beams down into the laboratory, at the base of the tower (see p. 114).

but took means of groups of high, medium, and low solar values for his researches.

Thus, with the years 1913, 1914, 1915, and 1918 grouped in this manner and compared to the temperature of Buenos Aires, he obtained certain results. The departure of Buenos Aires' temperature from normal having been tabulated, corresponding to each of the days of high solar radiation, and not only for the day itself, but for every succeeding day to the

twentieth, he then took the mean values for all these tabulated days of high solar radiation and for their successors.

In this way he determined the average march of temperature accompanying and following high values of solar radiation. He made similar determinations for the medium and the low solar conditions. The comparison is very striking. Starting from a low value on zero day, the temperature soon reaches normal, and not till the tenth day do the highest effects

[blocks in formation]
[merged small][ocr errors]


6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 1 2 3 4 5 6 7 8 9 1112



[ocr errors]

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 July


Predicted temperature and rain

Observed temperature and rain


The dotted lines show predicted temperature and rain in the Weekly Weather Forecast from May 29 to August 20, 1924, issued from Buenos Aires. The solid lines show the observed temperature and rain for the same period (see text below).

prevail. Then follows a decline and a new rise.

Almost exactly opposite consequences follow states of low solar radiation. Mean solar values yield continuously. mean temperatures for Buenos Aires. The astonishing thing is that the maximum difference in temperature at Buenos Aires between high and low conditions of the sun occurs, not on zero day, but on the tenth day after.

It is probable that the average solar values. stated by Mr. Clayton show excessive range, because our Mount Wilson observations of that time were very faulty, and many days must have been high or low from error, not from true solar changes. The range of 5 per cent indicated in the solar values should very probably be regarded as no more than 22 per cent, and corresponding to it we find 6° Fahrenheit ten days after the


Here, indeed, was a bright forecasting prospect. Mr. Clayton continued his studies, and after the Smithsonian Institution had established its Chile station the Argentine Weather Bureau arranged to receive our daily solar observations regularly by telegraph.


By December, 1918, Mr. Clayton was ready for a public trial of a new method of long-range forecasting. From that time to the present, the Argentine weather service has issued every Wednesday an official bulletin giving an exact forecast of the temperatures to be expected each morning and evening at Buenos Aires for the week beginning on Thursday. The

bulletin also includes the dates and intensities of expected rainfall. These bulletins are not given away, which, of course, would be easy, but are sold to clients.

The chart on this page shows the forecasts for twelve consecutive weeks in the year 1924 and the actual temperatures and rainfall which occurred.

Retiring to private life in 1922, Mr. Clayton returned to the United States and published a notable book, "World Weather," in which a summary of his studies of the relations of solar changes to weather conditions is included.

Later, by the aid of a grant from Mr. John A. Roebling, Mr. Clayton began the study of solar variation as related to the weather of North America. As it seemed the best way to prove the usefulness of these researches, he undertook to forecast experimentally for the city of New York.

In order that the experiment might lack no element of trustworthiness, it was arranged that the Smithsonian Institution should forward to Mr. Clayton each morning its telegraphic advices of solar conditions of the day preceding, received from the field stations of California and Chile. He thereupon prepared a definite forecast of the exact maximum temperature to be expected at New York for the third, the fourth, and fifth days from the solar observation, and mailed the forecasts about noon to the Institution.

Thus, Mr. Clayton had no further possible control of his forecasts for two, three, and four days, respectively, before they were to mature.

After a full year of this experiment, the forecasts were compared with the official records by a computer at the Smithsonian Institution. A very decided

« PreviousContinue »