Laserfiche WebLink
<br />" <br /> <br /> <br />00l~55 <br /> <br />-29- <br /> <br />In the case of a long record at a station documented as having had <br />changes in location, and if double-mass analysis shows each to be con- <br />sistent within itself, the record for each location should be treated <br />separately. <br /> <br />Use of data <br /> <br />Determining averages or normals <br /> <br />The World Meteorological Organization has defined the term "normal" <br />as the average computed for a uniform and relatively long period that <br />comprises at least three consecutive 10-year interim periods. The stand- <br />ard periods adopted by the U.S. Weather Bureau for climatological normals <br />are the following consecutive 30 years: 1 January 1921 to 31 December <br />1950, 1 January 1931 to 31 December 1960, etc. "Average" or "long-term" <br />mean" should be used for reference to any other period. The length of <br />record required to establish a suitable climatologic average ~pends to <br />some extent upon the variability of the record. Beginning January 1, <br />1962, the Weather Bureau has used the 19J1-1geOnormal for all stations. <br />If part of a station record is inconsistent, use of all the record may <br />lead to an incorrectriormal or average. <br /> <br />Normal or average precipitation for a period longer than the observed <br />record for a station may be estimated by various comparison techniques. <br />In the Pacific Southwest, five years or more of precipitation records <br />may be sufficient to compute an October-April normal by the double-mass <br />technique; ten years or more to compute a May-September normal by the <br />ratio method. <br /> <br />Average precipitation over an area during any particular storm or <br />period may be determined by computing the average of the records available, <br />by the Thiessen-diagram technique, by the isopercental method, or from an <br />isohyetal map (lines of equal precipitation). In mountainous areas the <br />isohyetal-map approach is recommended if the map is based on a knowledge <br />of topographic influences and characteristics of the particular storm. <br /> <br />Preparing isohyetal maps <br /> <br />All records to be used in preparing isohyetal maps first should be <br />checked for consistency. A specific period of record should be chosen <br />and all averages should be for that period. To arrive at a yearly <br />isohyetal map for any part of the Pacific Southwest, with the possible <br />exception of California, it is desirable first to prepare separate <br />October-April and May-September maps because of the change in source <br />regions and storm paths from winter to summer. <br />