The Determination Of Upper Winds By Electronic Means.
Upper-wind observations are made by following the course of a small hydrogen-filled balloon which rises at a more-or-less constant rate through the atmosphere-and which at the same time is carried along in the wind stream. At certain intervals of time readings are made of at least three of the following four measurments—azimuth angle, elevation angle, slant range and height, and from these measurements it is possible to work out the velocity of the balloon over any interval of time.
There are three main methods of following these balloons, which are:—
With visual methods a special theodolite is employed and measurements of azimuth angle and elevation angle are made while slant range can also be obtained by tachyometry methods if a tail is hung below the balloon. Height can be assumed from the rate of ascent of the balloon. A further refincment in visual methods is the use of two theodolites situated at the ends of a base line, when slant range can be calculated with considerable accuracy. Most of the upper-air observations taken to-day are made with visual single-theodolite methods.
In Radio-direction Finding a radiosonde transmitter is carried aloft by the balloon and the position of this transmitter is “fixed” either by taking the bearings of the signal from two or more ground stations, or by obtaining: azimuth angle and elevation angle from a single station. In both cases use is made of the height which is calculated from the radiosonde measurements of pressure, temperature and humidity.
In Radar methods electromagnetic pulses are transmitted from the ground radar set and a portion of these waves are reflected by a special target that is carried aloft by the balloon—with radar methods, azimuth angle, elevations angle and slant range are all measured.
The accuracy required in upper-air observations, largely depends on the use to which such observations are put. Perliaps the greatest accuracy is-required when these observations are used for gun correction, and artillery experts would like these winds with a vector error not exceeding 1 m.p.h.
Whether such an accuracy has any real meaning when the effect of wind over a distance comparable with the range of modern guns is taken into account is a debatable point and, in any case, such an accuracy is usually beyond the capability of any known method of upper wind determination, with the possible exception of the double-theodolite visual method.
For purely meteorological purposes such as the drawing of upper-air charts and forecasting the maximum permissible vector, error should not exceed 5 knots when the wind is taken over 2,000ft. layers up to 30,000ft. For the simple day-to-day use of these observations in say aviation forecasting this accuracy is probably too stringent, but for research purposes and the more exact forecasting techniques, particularly in tropical regions, this accuracy is not sufficient and the magnitude of the vector error should not exceed 3 knots. It is of interest to state here that the British Air Ministry have specified the maximum permissible vector error as 2.7 knots at a maximum range of 100 miles, these figures being used as a guide in the design of new electronic wind-finding equipment.
Now a balloon ascension rate of 1,000 feet per minute can usually be obtained with a moderate-sized balloon and careful design of the target, and assuming an average wind velocity to 30,000ft. of 60 knots the azimuth-angle accuracy required to give a vector error at 30,000ft. of less than 5 knots over 2,000ft. layers must be at least 0.3°. Therefore, any electronic method employed for wind finding must be capable of measuring bearing accuracy to within 0.3°. The other main requirement in electronic wind-finding equipment is range, and although in our assumption of an average 60-knot wind to 30,000ft. the range at 30,000ft. is only 35 miles this average may be exceeded on several occasions; also observations to over 50,000ft. are most desirable. Hence range requirements have usually been stated as a minimum of 50 miles with a desirable maximum of 100 miles.
The British M.O. Direction Finding Set employs an Adcock H-formed aerial system which is rotated for minimum signal strength which occurs when the horizontal arm of the H is at right angles to the incoming signal. Three of these sets spaced at the corners of an approximately equilateral triangle of side of about 20–30 miles are used to follow the Kew radiosonde, which operates in the frequency band of 27.5 to 28 megacycles per second. The height of the balloon is obtained from the measurements of pressure, temperature and humidity sent out by the radiosonde.
The British M.O. consider these sets give probable vector errors of under 4 knots at 30,000ft. for the better-sited stations and under 6 knots for the poorer-sited stations. The disadvantages of this type of equipment are mainly economic, being the cost of the radiosonde transmitter which must be flown every time a sounding is required, and also the number of operators required to man all three stations.
The American-designed Direction Finding Set S.C.R. 658 was developed to track the Diamond Hinman 375 megacycle radiosonde. Two sets of aerials are provided, one for azimuth and one for elevation and the signals from each set are compared on a separate C.R.T., thus enabling measurements of both azimuth angle and elevation angle to be made. Height is obtained from the radiosonde measurements so it is possible to obtain a complete fix with only one station. The accuracy of this equipment is not known, but it is assumed to give better than 5-knot vector error at 30,000ft.
The advantage of this type of D.F. set over the British M.O. set is that only one station need be operated. The expense of the radiosonde transmitter required for each flight is still a disadvantage, while there is a further disadvantage in that the set cannot give accurate elevation angles below 15° because of ground reflection interference. The U.S. authorities are therefore developing a 1,725 m.c. radiosonde and D.F. set that can be used down to 4°.
The original Radar sets used for following meteorological balloons were not designed specially for that purpose, but anti-aircraft fire-control radar sets were usually used because they gave the necessary accuracy and also measured elevation as well as bearing and range. The longer-wave-length radar sets were originally used, but microwave sets are now employed almost exclusively.
The first of these sets to be described is the American S.C.R. 584. This was originally an A.A. fire-control radar that was used extensively in Britain
in the last few years of the war. It is a 10 c.m. set of 400 K.W. power which can automatically track the balloon reflector. It can measure azimuth and elevation angles to 0.1° and proved quite a suitable set for wind-finding purposes; but the initial cost of these sets was almost prohibitive.
The New Zealand Micro-wave Meteorological Radar, developed by the Dept. S.I.R., also operates on 10 cm., and has a power output of 200 K.W., and a bearing and elevation accuracy of 0.25°. This accuracy is achieved by the use of a narrow beam—hence the largé parabola, and by bisecting an are scribed (or painted) on a long persistence P.P.I. tube.
The chief advantage of radar methods over D.F. methods is that slant range is measured with considerable accuracy and so the vector error in wind determination at extreme ranges is usually less than with D.F. methods. The economic aspect of lower operating costs is of considerable importance. Another advantage is the use of wind-finding radar sets for storm detection and though this has more immediate use in tropical latitudes it can still prove a very valuable tool to the meteorologists in temperate latitudes if suitable techniques are worked out. Generally warnings of the onset of extensive rain areas, especially those associated with cold fronts where there are usually strong convectional currents, can be obtained from upwards of 100 miles away.
The chief disadvantage of radar methods, apart from their large initial cost, is their rather limited range especially when using reflectors of moderate size. This defect can to some extent be overcome by the use of pulse-repeater equipment. This pulse-repeater equipment was developed for a rather special purpose, namely, to enable warships to make upper-wind observations readily and conveniently by using their existing radar installations. Three models were developed at 200, 700 and 900 megacycles respectively. The 200 megacycle model was specifically designed for use aboard weather and guard ships which carried only search radars incapable of measuring angles of elevation. The 200 megacycle pulse repeater was always used with a standard radiosonde so that height was obtained from the radiosonde measurements. The 700 and 900 megacycle models were designed for tracking by fire-control radars, and thus elevation angle as well as range and azimuth angle was obtained; hence these pulse repeater equipments could be used by themselves. They could be followed for up to 100 miles, and, as slant range was always measured, the accuracy was considerably better than obtained with standard D.F. methods using the same frequency.
In conclusion I would like to stress that the design of electronic equipment for wind determination has by no means reached a static state, and new developments and designs are being tried out in many of the larger countries, particularly Britain and U.S.A. The latest British development in wind finding uses a microwave radiosonde which is triggered by a U.H.F. ground transmitter. In America, as mentioned previously, the development is towards straight direction finding on a centimetric radiosonde transmitter.