Laserfiche WebLink
<br />It can be seen from Table 1 that, for the case shown, both Models 1 and 2 are <br />reasonable approximations of the observed rainfall pattern, with Model 2 <br />being slightly better. In other rainfall patterns examined, Modell fit the <br />data best. Therefore, we shall proceed with the analysis of sampling <br />variance using Modell but it should be recognized that any model that <br />reasonably fits the data could be used for this analysis and the results <br />(discussed in section 7) with respect to the relative magnitudes of sampling <br />and natural storm variability would probably be similar. <br /> <br />Equation (1) was used to computer-simulate 100 different isohyet,al patterns, <br />generally spanning the real-world range of variation of the input parameters <br />(see section 5), that is E = 1, 2, 3, 4, and 5; G = 0.15, 0.6, 1.35, 2.25, <br />and 4.95 (A50 = 40, 25, 15, 10, and 5); and (X offset, Y offset) = (.01, <br />.01), (.01, .35), (.01, .75), (.35, .01), and (.35, .35) with only (X offset, <br />Y offset) = (.35, .01) used for E = 4 and 5, and all values of G. The value <br />of PM was arbitrarily set equal to 3 mm, but it should be recognized <br />that the results of the study do not depend on the value assigned to PM <br />since each precipitation valUl~ in the pattern is the same percentage of PM <br />whatever value PM may have. For each isohyetal pattern simulated, the <br />absolute total amount of preC"ipitation (RTOTL on Fig. 2) was calculated by <br />numerical integration. This is the true value (ground truth) of total <br />raincell precipitation against which the gage-estimated total raincell <br />precipitation values are compared. <br /> <br />8 <br />