Author(s): Takuya Hamada; Ichiro Fujita; Kojiro Tani; Keisuke Nakayama
Linked Author(s): Ichiro FUJITA
Keywords: Discharge measurement; STIV; MEM; Snowmelt flood
Abstract: The space-time image velocimetry (STIV), which can measure surface velocity distributions from river surface videos, has come to be introduced in Japan as a non-contact discharge measurement technique in recent years. River discharge is usually obtained by integrating local mean velocities estimated by multiplying a surface velocity coefficient. Conventionally, a constant value of 0.85 is used for the surface velocity coefficient, but its validity depends on the estimation of the vertical velocity distribution. Furthermore, its value can be different in transverse direction. Although various types of formula representing a vertical velocity distribution are available, none of them can reproduce a distribution including a velocity dip, except for the formula based on the maximum entropy method (MEM) proposed by Chiu. The advantage of the formula is that it can represent a velocity distribution for a whole depth with a single equation even when the distribution has a velocity dip. In order to examine the performance of STIV with MEM, snowmelt floods of the Shinano and Uono Rivers were measured by ADCP. Firstly, the entropy parameter M, which is a function of the maximum velocity and the cross-sectional mean velocity and assumed to be constant for a channel section, was determined by the ADCP. Secondly, a surface velocity distribution measured by STIV is used to estimate the discharge with the information of M. The proposed method yields discharges comparable to ADCP successfully. In addition, transverse distributions of the surface velocity coefficient and the location of the velocity dip were obtained.
Year: 2020