REPORT OF THE CHIEF OF THE WEATHER BUREAU.

UNITED STATES DEPARTMENT OF AGRICULTURE,

CENTRAL OFFICE OF THE WEATHER BUREAU, Washington, D. C., October 1, 1911. SIR: I have the honor to submit a report of the operations of the Weather Bureau during the fiscal year ended June 30, 1911.

Hon. JAMES WILSON,

WILLIS L. MOORE,

Chief of Weather Bureau.

Secretary of Agriculture.

MOUNT WEATHER RESEARCH OBSERVATORY.

The work of the observatory has been carried on along practically the same lines as for the previous year; at this time it is chiefly concerned with the investigation of the upper air over the United States. The work under this head may be divided into three separate branches, as follows:

1. Soundings of the upper air over Mount Weather, Va., by means of kites and captive balloons.

2. Soundings of the air at great altitudes by means of free balloons carrying meteorological instruments. While all preliminary testing and the later computing in this branch of the work are carried on at Mount Weather, the actual ascensions are made in the West, since the proximity of Mount Weather to the Atlantic Ocean on the east makes it inadvisable to send up free balloons from that point. The immediate supervision of all of the work enumerated under headings 1 and 2 is assigned to Research Director William R. Blair.

3. Study of the temperature and pressure changes in the lower layers of the air by means of summit and base stations in the mountains of Colorado.

PROGRESS IN AERIAL INVESTIGATIONS.

Considering first the aerial work at Mount Weather, it may be remarked that four years of nearly continuous kite or balloon records have been secured and published. During the fiscal year just ended 299 soundings were made by kites and 69 by captive balloons. The average altitude attained by the kites was 2,929 meters (9,609 feet); by captive balloons, 2,150 meters (7,054 feet) above sea level. Mount Weather itself is 525 meters (1,725 feet) above sea level. The number of kite flights in which an altitude of a mile above the mountain top (2,134 meters above sea level) was reached during the year was 250; 2 miles, 85. In but 10 flights was an altitude of 3

miles reached or exceeded. The mechanical equipment used in kite flying has been brought to a high state of perfection, as evidenced by the few "breakaways" of the kites during the year, but a kite meteorograph satisfactory in all respects remains to be constructed. The classification and compilation of the Mount Weather kite and balloon data necessary in order to study the information secured under different weather conditions have been a part of the office work during the year, and a summary of the mean results for three years has been prepared and is now in the hands of the printer. It will appear as part 2, Volume IV, of the Mount Weather Bulletin.

A most important piece of work was accomplished by the Mount Weather Observatory during the year in the completion of two sounding-balloon campaigns, the first at Huron, S. Dak., and the second at Fort Omaha, Nebr. Some account of the earlier work in this direction was given in my last annual report. In order to present the subject intelligently, I shall repeat some of the facts given in previous reports.

The use of small free balloons to carry meteorological instruments into the upper regions of the atmosphere dates from 1893, nearly 20 years ago, when Messrs. Hermite and Besançon in France sent up varnished paper balloons carrying registering instruments which brought back a record of the meteorological conditions encountered in the ascension. To Assman, of Germany, however, is due the substitution of small rubber balloons in these ascents. The expansion of the confined gas at great altitudes bursts the balloon, the landing of the instrument being effected by means of a light parachute with which the balloon is covered.

The first series of sounding-balloon ascents in the United States was made at St. Louis, Mo., in the years 1904 to 1907, inclusive, under the direction of Prof. A. Lawrence Rotch, of Blue Hill Observatory. Prof. Rotch conducted 77 ascensions, the instruments being recovered in all but 5 cases. Thirty-seven of the ascensions reached an altitude of 10,000 meters (6 miles) or greater. In the series by Prof. Rotch heights exceeding 10 miles (16 kilometers) were attained 5 times. The Mount Weather Observatory has sent up 91 sounding balloons, of which number 81 were recovered. Heights exceeding 10 miles were attained in 37 cases, the greatest height attained being 18.9 miles at Huron, S. Dak., on September 1, 1910.

The exploration of the atmosphere by means of sounding balloons has become an international work and is carried on through an international commission of which Prof. H. Hergesell, of Strassburg, is president. Through this commission are collected and published the results of aerial observations made quite generally at appointed times by all meteorological services in the Northern Hemisphere. The most important single result that has come from the observations is the discovery of a region in the atmosphere, about 7 miles above the earth's surface, where the fall in temperature with increasing altitude ceases. On the contrary, there may be a slight rise in temperature on entrance to this region. Although various names have been assigned to this region, none fully describes its characteristics. In this report it will be referred to as the "upper inversion." Inversions of temperature are frequently found in the atmosphere next to the earth, but they are generally small in amount and fleeting in character. The upper inversion, however, appears to be a world-wide

phenomenon. It was discovered in northern Europe, and its presence has since been established in the United States, within the Arctic Circle, north of Europe, over tropical Africa and the Indian Ocean, and in Java.

The matter which follows is largely taken from a preliminary report on the data secured by sounding-balloon ascensions in the United States. The full report will be published in the Mount Weather Bulletin, Volume IV, probably in part 3, which will be sent to press in August, 1911. The statements refer to the most obvious facts ascertained by the balloon ascensions.

The original plan of sounding-balloon ascensions in this country contemplated simultaneous ascents from two points on an east and west line. Notwithstanding the great area of the United States, regions adapted to the work are remarkably few, outside of the Western Plains and the Central Mississippi Valley. Fort Omaha, Nebr., was selected as the western station largely because of the presence there of a detachment of the United States Signal Corps, which maintains a hydrogen-gas plant. The thanks of the bureau are due to Gen. James Allen, Chief Signal Officer, and the local officials at Fort Omaha, for many courtesies shown while operations were conducted at that point. Indianapolis, Ind., was selected as the eastern station. The board of public parks of that city kindly placed at our disposal ground in the park system on which to conduct the ascensions. For various reasons the number of ascensions at this place was less than at Fort Omaha, but all of the balloon meteorographs sent up were eventually found and returned to Mount Weather.

The second and fourth series of ascensions were made at Fort Omaha, Nebr. In the second Omaha series the balloons used had been on hand about six months and the rubber had so deteriorated during that time that heights much above 6,000 meters (3.7 miles) were not attained.

The third series was made at Huron, S. Dak. The advantages of the latter station lie in its geographic position, being farther north, and thus more directly in the path of cyclonic and anticyclonic areas. A successful series of ascensions was made from the State fair grounds at Huron, 24 of the 26 instruments sent up being recovered.

During the fourth and last series of ascensions, at Fort Omaha, from February 8 to March 4, 1911, inclusive, 25 meteorographs were sent up, of which 22 were found and returned. This was also an excellent series, but unfortunately no well-marked cyclonic areas passed over Fort Omaha during the three weeks the party was there. Naturally the first thought in connection with the upper inversion is its relation to terrestrial weather conditions. Thus far ideas on the subject are quite hazy, but certain facts have been established, as follows: The lowest temperatures of the upper inversion are found in equatorial regions and the highest in the middle latitudes. In other words, temperature increases with increase of latitude, contrary to the rule which prevails on the earth's surface. In tropical Africa, Berson, of the German expedition, found a temperature of -83.9° C. (-119° F.) at an altitude of 19 kilometers (11.8 miles).1 At the same elevation in the United States the temperatures range

Later ascensions made at Java, Batavia, confirm the existence of low temperatures over equatorial regions. See van Bemmelin in Met. Zeit., May, 1911.

between -55° and -60° C. (-67° to -76° F.). The discovery of the low temperatures aloft over the equator serves to increase, rather than diminish, the complexities involved in the accepted theories of the general circulation of the atmosphere. Another fact of great interest in connection with the upper inversion is that its temperature, while practically constant from season to season, varies greatly from place to place and from day to day. In the United States the mean of 50 ascensions made under the direction of the Mount Weather Observatory, all of which entered the region, gives for the lower limit of the upper inversion a temperature of -52.1° C. (-61.8° F.) regardless of season. The mean temperature of the lower limit of the upper inversion, as deduced from the ascensions made under the direction of Prof. Rotch at St. Louis, Mo., for all seasons is 56.0° C. (-68.8° F.). The lower temperatures registered over St. Louis may be due in part to the latitude effect. The mean temperature of the upper inversion in Europe is not far from -55° C. (-67° F.).

It is said that in Europe the beginning of the upper inversion is found at a less altitude over cyclonic than over anticyclonic areas; also that it is higher in summer than in winter. In this country the lower limit of the upper inversion does not appear to be at a less altitude in cyclonic than in anticyclonic areas, although the evidence is not absolutely conclusive either way. In the Huron series the upper inversion was reached at an altitude as low as 9,328 meters (5.8 miles) on the front of an anticyclone. It was also reached at the low elevation of 9,712 meters (6 miles) in the transition region between a cyclone and an anticyclone, and at an altitude of 9,372 meters (5.8 miles) in a cyclone, while on other occasions in cyclones it has been reached at altitudes ranging from 10,000 to 14,000 meters (6 to 8.7 miles). The greatest altitude at which it was encountered, 14,983 meters (9.3 miles), on September 28, 1909, was not in an anticyclone, but in the transition region between a northern cyclone and a southern anticyclone.

In the United States the seasonal distribution of the ascensions has not been so good as might be wished. If the year be divided into two portions, the warmer half, or from April to October, inclusive, and the colder half, from November to March, inclusive, the following results are obtained for the average height of the lower limit of the upper inversion:

Thus it is seen that the lower limit of the upper inversion in the United States is found at a slightly less altitude in winter than in summer, agreeing in the main with European observations. The winter series of 21 ascensions was made from February 8 to March 4, inclusive. The summer series was made mostly in September and October.

By reason of the clear skies and relatively dry air of South Dakota and other western States it was possible to make observations on the motions of the balloons after they had gotten well into the region of the upper inversion, and thus to obtain some interesting facts con