The effect of freshwater biofilms on turbulent boundary layers and the implications for hydropower canals

The majority of the electricity supplied in Tasmania, Australia, is produced by hydropower. Hydro Tasmania, the power generation utility, operates 29 hydropower stations incorporating 170 km of open channels. These open channels are susceptible to extensive biofilm growth dominated by the freshwater diatoms Gomphonema tarraleahae and Tabellaria flocculosa, which form a gelatinous biofilm several millimetres thick and cause reductions in flow capacity of up to 10%. This thesis presents results of a multidisciplinary study on the effects of freshwater biofilms on hydropower canal capacity and turbulent boundary layer structure. The extent to which the surface roughness affects the structure of the turbulent boundary layer was critically examined, in the context of the wall similarity hypothesis. A recirculating water tunnel, equipped with a floating element force balance and a twodimensional Laser Doppler Velocimetry system, was used to obtain detailed measurements on test plates covered with flow-conditioned freshwater biofilms. Total drag measurements, mean velocity profiles and turbulent Reynolds stresses were compared for smooth, sandgrain and biofouled test plates. An artificial biofilm was developed to study the motion of algae streamers under flow conditions. Each test surface was mapped using digital close-range photogrammetry to provide a three-dimensional surface model. A logarithmic relationship was found between the roughness function and the maximum peak-to-valley height from the photogrammetry measurements for the low-form gelatinous biofilms. Seven methods for determining the wall shear stress were investigated; however, none of the methods examined were entirely satisfactory. A method which can be used for both smooth and rough surfaces with reasonable measurement uncertainty is needed to allow objective comparison of the structure of smooth and rough wall turbulent boundary layers. The presence of biofouling caused a significant increase in the local skin friction coefficient and overall drag coefficient with increases in skin friction of up to 210% measured over smooth plate values. Results for low-form gelatinous biofilms provided support for the wall similarity hypothesis. The biofilms modified the structure of the turbulent boundary layer in the near-wall region. Velocity defect profiles, Reynolds stress profiles and quadrant analyses showed good agreement for different test plate surfaces in the outer region of the boundary layer. The results for an artificial filamentous biofilm were less conclusive, and may not support wall similarity.