2017 Volume 23 Pages 305-310
We indicated two issues such as the error including in the precipitation and in the predicted water surface level in this study. Radar precipitation after improving the observed radar precipitation of MLIT revised by the rain gauge and the amount of precipitation analyzed by radar-AMeDAS, it is shown that the calculation results from the distributed cell runoff model almost accords with observation data. It shows that it is important to filter the several object water gauge stations adding on the Hirakata station to improve the accuracy of the prediction discharge at the upstream end. Furthermore, it is shown that the accuracy of the predicted water surface level was improved by adding the manning coefficient in the state values such as the discharge at the upstream end of the branches.
