| Abstract: | The field of ecohydraulics tries to link biological‐ and physical‐based processes in order to describe better the distribution of plants and animals in rivers. We tested the hypothesis that the influence of cover in pools and riffles would not be detectable using average velocity, turbulent kinetic energy and turbulent intensity, and compared these measurements to locations distant from cover. We measured water velocity fluctuations using an acoustic Doppler velocimeter. We found that turbulent intensity in the downstream direction (u′) and the transverse or cross‐stream direction (v′) were the most useful in detecting the presence of cover in pools and riffles. Differences were apparent between locations near cover in pools and riffles. Turbulent kinetic energy (k) and vertical turbulent intensity (w′) detected cover in pools but not in riffles. Average downstream velocity ( ) detected cover in riffles but not in pools. Average cross‐stream ( ) and vertical ( ) velocities did not detect any differences at all. We rejected the null hypothesis and concluded that turbulence caused by habitat features such as large rocks, wood or other channel complexities results in a statistically meaningful difference in flow characteristics in locations near cover. This finding was tempered by the fact that knowledge about how fish respond to turbulence is limited in comparison to our understanding of average velocity values. Despite the potential benefits to habitat modelling of incorporating turbulence‐based metrics, application of these findings will be challenging because turbulence modelling is difficult and current models may not be appropriate for application to rivers. Copyright © 2007 John Wiley & Sons, Ltd. |
