9511 - Laserfiche WebLink

Home Browse Search

236 Student t-tests to determine if there was a signifi- cant difference between the control and treatment casts. In addition, we used casts of two species lacking humps, but closely related to each of the treatment fishes, to compare drag and force direction of nuchal humped and non-humped casts. The more streamlined catostomid was the flannehnouth sucker, C. latipinnis, and the more streamlined cyprinid was the roundtail chub, Gila robusta, 439 and 403 nun TL, respectively. We also determined drag coefficients of two well-known objects, a 324 mm TL rainbow trout, Oncorhyn- chus mykiss, and a 57.17 mm diameter sphere for comparison with other studies. We compared humped vs. non-humped mor- photypes by analysis of drag coefficients, which are widely used for comparing the hydrodynamic drag among fishes. The drag coefficient (CD) is defined as CD = 2DIPSU2 where D is the drag force, p is the fluid density, S is the reference area, and U is the velocity (Denny 1988, Alexander 1990, Vogel 1994). The drag coefficient is a dimensionless function of the Rey- nolds number that can be used to compare the effects of hydrodynamic drag on objects of differ- ent morphologies, including fishes (Webb 1975, Denny 1988, Alexander 1990). We compared drag coefficients of our fish casts with nuchal humps to those of more streamlined fishes and the casts with the nuchal humps removed to help evaluate the hydrodynamic role of nuchal humps of the Colo- rado River fishes. Such hydrodynamic compari- sons have often been made between species, which may make it difficult to separate effects from other interspecific differences (Pettersson & Bronmark 1999). Hence, we tested both models of conge- nerics and models with humps removed. We used two different methods for determining the surface area in this study: frontal area (the maximum projection of the body onto a plane normal to the direction of flow) and wetted area (the total surface exposed). We used the frontal area to calculate the drag coefficient of the sphere, and the wetted area to determine the drag coeffi- cients of the fish casts. We calculated the reference area measurement in this manner to keep to con- vention and produce quantities that could be compared to values previously reported in the lit- erature. Frontal surface area is considered easier to measure accurately, while wetted surface area is difficult to measure in complex shapes. We tested drag and lift components of fish casts in an 18.3 in x 0.9 in x 0.6 in flow tank at the U.S. Bureau of Reclamation Hydraulics Laboratory in Lakewood, Colorado (Figure 2). Two 100 hp centrifugal pumps delivered water into a 950 000 1 reservoir and then into the tank baffle through a 30 cm diameter pipe. A stone baffle within a screen helped to control the flow of water at the filling point, and 4.6 in below the baffle water also passed through an array of flow-straighteners consisting of 200 galvanized pipes (5.1 cm diameter and 51 cm long) stacked upon one another to fill the channel. We used a Sontek Adv-4 Doppler veloc- ity probe to measure the flow rates, which reached 1.0 in s-1 in the flume. We tested effects of lower velocities (i.e., 0.3 in s-1) on the fish casts in a smaller flow tank at the University of Colorado at Boulder. Water flow was generated by a 0.5 hp electric motor which operated a propeller in a 15 cm PVC recir- culating pipe. The water entered the flow tank through a pipe baffle to disperse flow into a larger upper end (0.3 in x 0.3 in x 0.64 m), which nar- rowed to a 0.3 in x 0.23 in x 0.64 in working sec- tion. The water entered the working section through a flow-straightener constructed of approximately 2000 drinking straws of 0.48 cm ',,.,Drag Figure 2. Experimental flume at the U.S. Bureau of Reclama- tion Hydraulics Laboratory in Lakewood, Colorado illustrating experimental design, including mounting and orientation of the fish casts along with various flow components.