Laserfiche WebLink
<br />- 41 - <br /> <br />and in 90% of cases it does not exceed 74%. The RMS.dis~repancy in all cases was 39%. <br /> <br />2.29 Analysis of the existing data on the errors made by a raingauge network <br />of varying density shows that there is a 27% RMS error in the measurement of the <br />mean precipitation amount for a rain occurrence over an area of 100 km2 covered by <br />one raingauge (Gandin L.S., 1963). <br /> <br />2.30 Since the errors of surface-based and radar precipitation measurements <br />are statistically independent, we assume that the RMS error of a radar measurement of <br />the amount of precipitation during a rain occurrence and over an area of 100 km2 is <br />28%. Thus, when taking special measures to ensure continuous control of the potential <br />of a meteorological radar, radar areal precipitation measurements have an accuracy <br />equivalent to that of a surface raingauge network with a density of one instrument <br />every 100 km2. <br /> <br />3. <br /> <br />RADAR METHODS OF STUDYING CLOUD DYNAMICS <br /> <br />3.1 It is recognized that the formation and evolution of clouds and the <br />weather phenomena connected with them cannot be thoroughly studied without reliable <br />information on the structure of their internal air motions and the nature in their <br />surroundings. A complete model of the microstructure of clouds obviously cannot be <br />constructed without an understanding of how the various microphysical processes are <br />connected with the air-flow distribution and, consequently, weather modification also <br />cannot be successful without such knowledge. This is clear from the fact that the <br />main parameters of precipitation formed in a cloud with a given microstructure, the <br />cloud distribution, intensity and duration are significantly determined by the <br />characteristics of motion both within the cloud and in the meso-scale area, and by <br />the way in which these microphysical and dynamic processes interact. At the same time <br />it should be noted that the structure of turbulent and ordered flows in clouds of <br />various forms has not been described in detail. This is primarily due to difficulties <br />in obtaining experimental data on motions within clouds especially in large convective <br />ones, which are frequently of greatest interest, and it is also important to note <br />that these motions have a very variable structure. <br /> <br />3.2 Measurements obtained directly from aircraft have recently made it <br />possible to obtain a broad picture of the distribution of flows within clouds and <br />around them and to estimate the turbulence and possible speeds of these flows. General- <br />ization of the accumulated data has produced mean statistical models of cumulus clouds. <br />However, the many difficult, still unsolved questions concerning cloud development <br />processes and the problem of their modification, require further study, the results <br />of which should not only be statistical, but it is very important that they should <br />refer to specific clouds in particular stages of development. It is therefore <br />necessary to develop such time-dependent methods of studying cloud dynamics as to <br />enable a cloud to be studied as a whole and at any given time. This can only be done <br />by remote sounding using Doppler radar technology. <br /> <br />3.3 The work begun in the fifties in this field was originally aimed at <br />studying the connexion between radar signal fluctuations of clouds/precipitation <br />which constituted multiple targets, i.e. a set of moving scatterers each with <br />