[1] The continuous measurement of river discharge for long periods of time is crucial in water resource studies. However, the accurate estimation of river discharge is a difficult and labor-intensive procedure ; thus, a robust and efficient method of measurement is required. Continuous measurements of flowrate have been carried out in a wide, shallow gravel bed river (water depth % 0.6 m under low-flow conditions, width % 115 m) using Fluvial Acoustic Tomography System (FATS) that has 25 kHz broadband transducers with horizontally omnidirectional and vertically hemispherical beam patterns. Reciprocal sound transmissions were performed between the two acoustic stations located diagonally on both sides of the river. The horizontal distance between the transducers was 301.96 m. FATS enabled the measurement of the depth-and range-averaged sound speed and flow velocity along the ray path. In contrast to traditional point/transect measurements of discharge, in a fraction of a second, FATS covers the entire cross section of river in a single measurement. The flow rates measured by FATS were compared to those estimated by moving boat Acoustic Doppler Current Profiler (ADCP) and rating curve (RC) methods. FATS estimates were in good agreement with ADCP estimates over a range of 20 to 65 m 3 s À1 . The RMS of residual between the two measurements was 2.41 m 3 s À1 . On the other hand the flowrate by RC method fairly agreed with FATS estimates for greater discharges than around 40 m 3 s À1 . This inconsistency arises from biased RC estimates in low flows. Thus, the flow rates derived from FATS could be considered reliable.Citation: Kawanisi, K., M. Razaz, K. Ishikawa, J. Yano, and M. Soltaniasl (2012), Continuous measurements of flow rate in a shallow gravel-bed river by a new acoustic system, Water Resour. Res., 48, W05547,

Continuous measurements of streamflow and water temperature were carried out in a shallow gravel-bed river during dam flushing operations using a new shallow acoustic system with two crossing transmission lines. The fluvial acoustic system (FAS) was equipped with four 25 kHz broadband omnidirectional transducers. The reciprocal sound transmissions were performed between the two pairs of acoustic stations, located on both sides of the river. The four-station FAS enabled the measurement of the stream direction in addition to the depth- and range-averaged sound speed, and therefore water velocity. Discharge values computed from FAS reports were compared to those estimated by the moving-boat acoustic Doppler current profiler (ADCP) and rating curve (RC) methods. FAS estimates were in good agreement with ADCP and RC estimates over a range of 50 to 180 m3 s−1. The temperature gradients induced strong sound scattering for the 25 kHz FAS. The fine suspended sediment particles also induced a gradual decrease in the signal-to-noise ratio of the FAS. Nevertheless, the FAS has been proved to be a reliable and accurate technique for the continuous measurement of cross-sectional average velocity/water temperature.