Laserfiche WebLink
<br />defined by the current criteria in the algorithm are used in forming the gauge-radar pair. In <br /> <br />agreement with previous conc:lusions, Figs. 12a and 12b show that the radar has overestimated <br /> <br />the rainfall relative to the gauges regardless of whether the center radar bin or 9-bin average is <br /> <br />used as evidenced by the majority of points to the right and below the I: I diagonal line. This <br /> <br />overestimation is more pronounced as the rain rate threshold increases from 49 to 55 dBl which <br /> <br />is expected if hail is contaminating the radar rainfall estimates, Referring back to the sample <br /> <br />biases shown in columns four and five of Table I (corresponding to Fig, 12a and 12b, r~'sp.), the <br /> <br />storm-total sample biases decrease from about 0,7 to 0,5 as the rain rate threshold increases from <br /> <br />49 to 5:5 dBl, indicating increasing radar overestimation. Since a sample bias near 1.0 is <br /> <br />desirable, even with the lowest setting of 49 dBl the radar estimates for this case are still too <br /> <br />high relative to the gauges by at least 40% (\.0 / 0.7) on the average. <br /> <br />These results depend on the chosen l-R relation. Had another l-R relation been used, <br /> <br />the points in Fig, 12 would bt: shifted, and the optimum choice ofrain rate threshold would be <br /> <br />different. Clearly the optimum Z-R parameters and rain rate threshold are mutually dependent v' <br /> <br />on each other as well as other meteorological factors that are not fully understood. <br /> <br />The relative insensitivity of the choice of rain rate thresholds on the gauge-radar pairs that <br /> <br />would be used in the existing Adjustment algorithm is illustrated in Fig. 12c. For many of the <br /> <br />W <br />.-0 r'J). <br />,~ <br /> <br />pairs, changes in the rain rate threshold made no difference at all on the radar-estimated rainfall <br /> <br />as many of the plotted symbols for each gauge overlay each other. Thus the current Adjustment <br /> <br />algorithm is relatively insensitive to the choice of rain rate threshold, and this less-than-desirable <br /> <br />result is caused by the way the gauge-radar pairs are formed. This unfortunate result suggests <br /> <br />that the Adjustment algorithrn would not likely be able to capture the true gauge-radar bias <br /> <br />18 <br />