Laserfiche WebLink
<br />001353 <br /> <br />30 <br /> <br />These findings indicate that in particular basins the physically- <br />based models would, on the average, predict the runoff response more <br />accurately than the conceptual models. The prediction of peak discharge <br />would be higher than actual and the prediction of the hydrograph shape <br />would range from poor to good. <br />2) Overall test - This test compares the overall predictive <br />capability of the two model categorie5 for all basins. The writers <br />use all the published data listed in Appendix B. The results of this <br />test are listed in Table 1I-5. Examination of the table reveals the <br />following: <br />a) Both physically-based and conceptual models generally <br />overestimate the actual peak runoff rate. The mean value of peak <br />discharge ratios for the physically-based models is 1.10, and the <br />mean value of the peak discharge ratios for the conceptual models is <br />1.04. <br /> <br />b) The standard deviations or variances from the peak <br />discharge ratio mean are generally higher for physically-based models <br />than for conceptual models. The average standard deviation of the <br />physically-based models is 0.40, and the average standard deviation <br />of the conceptual models is 0.30. <br />c) The modified correlation coefficient (RMOD) for both <br />the physically-based and conceptual models are poor (see Figure 1I-3). <br />d) The special correlation coefficient (RS) for both of <br />the physically-based models is poor. For the conceptual models, the <br />coefficient ranges from poor to good (see Figure 1I-3). <br />