60 INSTRUCTIONS FOR MAKING PILOT BALLOON OBSERVATIONS 



189. The observations. — After a little practice, it is found that the 

 three readings — azimuth, elevation, and graticule — can all be taken 

 at each minute. It is best to read the graticule first. Alternatively, 

 the graticule can be read at the half minute between each minute 

 reading of the azimuth and elevation, afterward deducing the minute 

 readings of the graticule by interpolation. The upper pendant should 

 only be used while the lower one is outside the field of view. When 

 the reading of the graticule for the full tail has fallen to 5.0 no further 

 minute readings should be taken, but the time should be noted when 

 the remaining division of the graticule is passed as a check on the 

 assumption of a uniform rate of ascent. After .the limit of 5.0 has 

 been passed, the balloon should be assumed to rise at 180 meters per 

 minute; or, at the discretion of the observer, the vertical velocity 

 may be assumed at some other rate, based on his experience during 

 the initial stages of the ascent. 



190. Owing to the swinging of the tail, a single observation, how- 

 ever accurate it may be, will not afford a reliable value of the angle 

 subtended by the tail at the observer, especially if the angle of eleva- 

 tion is large. Oscillations in the vertical plane through the line of 

 sight affect the accuracy of the results more than oscillations in the 

 plane perpendicular to the line of sight. If the angle of elevation is 

 small, the errors will, in general, be small; but the error produced in 

 the micrometer reading by the swinging of the tail is a systematic 

 error, and since this is, in general, too low, the computed heights and 

 velocities are correspondingly exaggerated. When the balloon is at 

 a fair height the values deduced from consecutive readings of the 

 micrometer become rather irregular, and it then becomes necessary 

 to smooth them out. The exact point at which the smoothing process 

 should be adopted must be left to the discretion of the observer. 

 Smoothing should not be employed indiscriminately, especially in the 

 lower layers of the atmosphere, because by so doing the irregularities 

 which represent actual irregularities or discontinuities in the atmos- 

 phere will thus be lost. In the higher levels, although consecutive 

 tail readings may be irregular, they can still be used (by smoothing) 

 to check the rate of ascent of the balloon, although they can no longer 

 be used to define actual irregularities or discontinuities existing in the 

 atmosphere. Smoothing is done by plotting the tail readings on 

 squared paper with time as abscissa and drawing a smooth curve 

 among the points so obtained. 



