WSP04021 - Laserfiche WebLink

Home Browse Search

28 Average velocity figures for the entire study reach were high for the R-2 Cross Method in 13 of 14 instances when compared against the output for the same parameter from the IFG4 Method. Many investigators have cited the importance of water velocities in streams in relation to the fish population, fish spawning, and aquatic invertebrate production (Baldes, 1968; Baldes and Vincent, 1969; Dodds and Hisaw, 1924; Hooper, 1973; Hoppe and Finnell, 1970; Lewis, 1969; McNeil, 1962; Orcutt, Pulliam, and Arp, 1968). It is of utmost importance to determine the relative reliability of the various computer models in predicting average velocities across a stream transect. If grave errors are inherently a part of the presently used computer models, then derived minimum-Ilow re~Qm~_ mendations could also be erroneous. not only in Colorado. but iQ other states as well. To evaluate the differences in average velocity outputs between the R-2 Cross and the IFG4 methods, several tests and comparisons between the two methodologies were used. The first comparison was done using known average water velocities actually determined in the field. These average water velocities, determined on over 90 stream transects across a wide range and large number of cali- bration flows, were used as a standard against which predicted average water velocities from the R-2 Cross and IFG4 methods were compared. The results (Table 12) do not give a clear cut indica- tion of any real differences in the magnitude of error between the IFG4 and the R-2 Cross methods. The findings of Elser (1976), Bovee and Milhous (1978), and Bovee, Gore, and Silverman (1977) do not agree with these results. They found the IFG4 stage dis- charge approach to be superior to the Manning equation one point approach used in the Single Transect R-2 Cross Method. The next comparison involved separating out the direction (positive or negative) of the percentage of error between the IFG4 and R-2 Cross methods and comparing the results against known field measurements. The only clear cut distinction result- ing from this comparison (Table 13) was that in more than 77% of the instances the R-2 Cross Method overestimated the known aver- age velocity while the IFG4 Method underestimated the known aver- age velocity 67% of the time. A third test was made by comparing the absolute magnitude of error for a given average velocity at a given transect for both the R-2 Cross and the IFG4 methods and comparing each to the known average measured velocity at that transect to see which methodology produced a lower percentage error. In this comparison the IFG4 did slightly better than the R-2 Cross giving a lower percentage error when compared to measured field velocities 52.6% of the time (Table 14). Once again, there was no clear cut superiority demonstrated by the IFG4.