SPATIAL MEASUREMENTS OF SNOWMELT FLOOD BY IMAGE ANALYSIS WITH MULTIPLE-ANGLE IMAGES AND RADIO-CONTROLLED ADCP

Measurements of flood flow have been conducted intensively at major control sections in Japan for storing reliable hydrological data for use in a long-term river planning. However, such a measurement pays attention only to flow passing through one cross-section; thus, spatial flow features are not available, while river channel changes its feature at every flood event. In this research, we performed concurrent measurements of a snowmelt flood of the Uono River using an image analysis and radio-controlled acoustic Doppler current profiler (ADCP). In the image analysis with STIV (Space-time Image Velocimetry), three video cameras were used to cover a river reach of about 300m by changing their view angles at every location. On the other hand, the boat-mounted ADCP was remotely controlled to form a zigzag trajectory to cover the same reach. The accuracy and limitation of STIV was made clear through a comparison with ADCP data and a spatial distribution of correction factor from surface to depth-averaged velocity was found to have a weak correlation with a large-scale bed slope.