Laserfiche WebLink
<br />. <br /> <br />. <br /> <br />J <br /> <br />. <br /> <br />FREQUENCY CURVES <br /> <br />9 <br /> <br /> , <br /> / <br /> / <br /> . . 1/ ../' <br /> / -' <br /> Graphical cur... ~/ <br /> e v~earson Type III curve <br /> Ie" ..,Ij{ i <br /> . <br /> - , <br /> ~ V <br /> ~ <br /> I <br /> I <br /> ~ I <br /> I <br /> .~ I <br /> I <br /> I <br /> . I <br /> I <br /> I <br />. I <br /> <br />the particular n, identified by order number. <br />Thus, it is unnecessary to compute plotting <br />positions by this procedure, and plotting can <br />be done rapidly. Most attractive of all is a <br />machine program developed by the U.S. Bureau <br />of Public Roads; this program produces the <br />probability graph with the points plotted on it. <br />The plotting positions given in table 3 are <br />estimated recurrence intervals; probabilities <br />would be their reciprocals; n is the number <br />of items (1950-1915+1=36), and m is the <br />order number. Plotting positions may be com- <br />puted by slide rule or read from a table pre- <br />pared for the purpose. <br />Discharge is plotted against recurrence in- <br />terval on Geological Survey Form 9-179a or <br />9-179b. The former has an arithmetic ordinate <br />scale, and the latter a logarithmic ordinate <br />scale. The abscissa scales are alike and are <br /> <br />700 <br /> <br />600 <br /> <br />" <br />z <br />o <br />" <br />w <br />"' <br />" <br />~ 500 <br />>- <br />w <br />w <br />~ <br /> <br />" <br />iii <br />::> <br />" <br />z <br />LLi400 <br />'" <br />" <br />'" <br />:I: <br />" <br />"' <br />Ci <br /> <br />300 <br /> <br />200 <br />1.01 <br /> <br />based on the Gumbel Type I distribution. <br />Figures 6 and 7 show plots of the data from <br />table 3. Flood data are usually plotted on Form <br />9-179a, and minimum-flow data on Form <br />9-179b. The points are averaged by eye in draw- <br />ing the lines, with the extreme points given <br />less weight than the others because their recur- <br />rence intervals are not so well defined. <br />Interpretation of the curves would be as <br />follows: <br />On figure 6 the discharge defined by the curve <br />at lO-year recurrence interval would be ex- <br />ceeded as an annual maximum at intervals <br />averaging 10 years in length. Conversely, figure <br />7 indicates that the discharge at intervals <br />averaging 10 years in length will be less than <br />the curve value corresponding to the 10-year <br />recurrence interval. Further interpretation of <br />a frequency curve is given on page 12. <br /> <br />2 <br /> <br />3 5 10 20 <br />RECURRENCE INTERVAL, IN YEARS <br /> <br />30 <br /> <br />50 <br /> <br />200 <br /> <br />1.1 1.2 <br /> <br />100 <br /> <br />Figure 6.-Frequency curve based on dala from table 3, assuming that data are annual maximums. <br />