Laserfiche WebLink
<br />31 <br /> <br />OllD145 <br /> <br />With the IFG4 Method, 72% of the time the magnitude of error <br />between average measured and predicted velocities was less when <br />measured at high discharge calibration flow than at a low dis- <br />charge calibration flow. This indicates that the results of the <br />IFG4 Method are more reliable at high flows on larger streams <br />than they are at lower flows on small streams. <br /> <br />In a final test the magnitude of error in predicted average <br />velocities was compared to the known' field values for average <br />velocities with results of the R-2 Cross and the IFG4 methods <br />being stratified into groups measured at high Q (>10 cfs) and <br />low Q (<10 cfs) calibration flows (Table 16). Two significant <br />departures from this classification were on th~ Fryingpan and <br />the Lake Fork of the Gunnison rivers where low Q calibration <br />flows were 75 and 68 cfs, respectively. In those two streams <br />high Q flows were 250 and 170 cfs, respectively. The magnitude <br />of error for the IFG4 Method was less than that of the R-2 Cross <br />Method (Table 16). The differences in magnitude of error were <br />smaller between the two methodologies when velocities determined <br />at low Q calibration flows were used. <br /> <br />Table 16. Comparison of the R-2 Cross with the IFG4 methods <br />for the magnitude of error in predicted velocities <br />at high and low discharge (Q) calibration flows. <br /> <br /> High Q Low Q Total <br />Error magnitude N percent N percent N percent <br />R-2 Cross >IFG4 26 60.5 17 53.1 43 57.3 <br />R-2 Cross < IFG4 17 39.5 15 46.9 32 42.7 <br /> TOTAL 43 100.0 32 100.0 75 100.0 <br /> <br />I conclude that the IFG4 Method gives more reliable r_esu1ts <br />at higher discharge levels than does the R-2 Cross Method and still <br />-~intains a slight edge at low Q calibration flows. <br />