Laserfiche WebLink
<br />3-02 <br /> <br />.'1, <br /> <br />the record, of the frequency arrays in the order of ' 'magnitude, and <br />of corresponding plotting positions. In arranging the .!lata in order <br />of mago.1 tude, much time can be saved by taking the event's in the <br />order in which they occurred, and placing them rapidly on a..blank <br />sheet of paper in the order of magnitude, thus _lr-h'g an irregular <br />tabulation. This tabulation is then recopied onto the form ,shown <br />on exhibit 3 lI.Ild cross-checked with chronologic values. <br /> <br />3-03. <br /> <br />SELECTION AND ARElANGEMmT OF DAn <br /> <br />a. The primary consideration in selection of an array of <br />data for a frequency study is the use .to which the frequency esti. <br />mates will be put. If the frequency curve is to be used ,-for esti- <br />mating damages that are related to instantaneous peak flpws in a <br />stream, peak flows should be selected frOll1 the record. /H the <br />damages are related to maxiJIlum mean-daily flows or to ~. 3-da.y <br />flows, these items should be selected. If' the behavior'. of a reser- <br />voir under investigation is related to the 3-da.y or 10~ re.inflood <br />volume, or to the seasonaJ. snOWlllelt volume, that pertinent i tam <br />should be selected. Occasionally, it is necessary to !lelect a <br />related variable in lieu of the one desired. For exam;ple, where , <br />mean-de.ily flow records are more complete than the records of peak <br />flows, it ~ be more desirable to derive a frequency curve of <br />.l <br />" mean-daily flows and then, from the computed curve, derive a peak- <br />\ flow curve by means of an empirical relation betweenmean-daily <br />flows and peak flows. AD. reasonably independent values should be <br />selected, but the annual maximum events should ord1l1arily be segre- <br />gated when the application of analytical procedures discussed in <br />section 4 is contemplated. <br /> <br />\. <br /> <br />. <br />aspect of each event (such as peak flow, mean-daily flow, or flood <br />volume for a specified duration), and each event must be controlled <br />by a uniform set of hydrologic and operational factors. For example, <br />it would be improper to combine items frOll1 old records that are <br />reported as peak flows but are in fact only daily readings, with <br />newer records where the peak was actusJJ.y measured. Similarly, <br />care should be exercised when there has been .significS!lt change in <br />upstream storage regulation during. the period of record. so as. not <br />inadvertently to canbine unlike .events into a single series. . In <br />such a case, the entire record should be adjusted to a standard con- <br />dition. <br /> <br />c . Hydrologic factors lI.Ild relationships operating during a . <br />generil winter rainfi.ood are usually quite different from those <br />operating during a spring snowmelt flood or during a local sUlllller <br /> <br />- 10 - <br />