Comparison of thermal, salt and dye tracing to estimate shallow flow velocities: Novel triple-tracer approach

Abstract

The accurate measurement of shallow flow velocities is crucial to understand and model the dynamics of sediment and pollutant transport by overland flow. In this study, a novel triple-tracer approach was used to re-evaluate and compare the traditional and well established dye and salt tracer techniques with the more recent thermal tracer technique in estimating shallow flow velocities. For this purpose a triple tracer (i.e. dyed-salted-heated water) was used. Optical and infrared video cameras and an electrical conductivity sensor were used to detect the tracers in the flow. Leading edge and centroid velocities of the tracers were measured and the correction factors used to determine the actual mean flow velocities from tracer measured velocities were compared and investigated. Experiments were carried out for different flow discharges (32–1813 ml s−1) on smooth acrylic, sand, stones and synthetic grass bed surfaces with 0.8, 4.4 and 13.2% slopes. The results showed that thermal tracers can be used to estimate shallow flow velocities, since the three techniques yielded very similar results without significant differences between them. The main advantages of the thermal tracer were that the movement of the tracer along the measuring section was more easily visible than it was in the real image videos and that it was possible to measure space-averaged flow velocities instead of only one velocity value, with the salt tracer. The correction factors used to determine the actual mean velocity of overland flow varied directly with Reynolds and Froude numbers, flow velocity and slope and inversely with flow depth and bed roughness. In shallow flows, velocity estimation using tracers entails considerable uncertainty and caution must be taken with these measurements, especially in field studies where these variables vary appreciably in space and time.