An automated system to simulate the River discharge in Kyushu Island using the H08 model

Abstract:

Kyushu Island is located in southwestern part of Japan, and it is often affected by typhoons and a Baiu front. There have been severe water-related disasters recorded in Kyushu Island. On the other hand, because of high population density and for crop growth, water resource is an important issue of Kyushu Island.

The simulation of river discharge is important for water resource management and early warning of water-related disasters. This study attempts to apply H08 model to simulate river discharge in Kyushu Island. Geospatial meteorological and topographical data were obtained from Japanese Ministry of Land, Infrastructure, Transport and Tourism (MLIT) and Automated Meteorological Data Acquisition System (AMeDAS) of Japan Meteorological Agency (JMA). The number of the observation stations of AMeDAS is limited and is not quite satisfactory for the application of water resources models in Kyushu. It is necessary to spatially interpolate the point data to produce grid dataset. Meteorological grid dataset is produced by considering elevation dependence. Solar radiation is estimated from hourly sunshine duration by a conventional formula. We successfully improved the accuracy of interpolated data just by considering elevation dependence and found out that the bias is related to geographical location. The rain/snow classification is done by H08 model and is validated by comparing estimated and observed snow rate. The estimates tend to be larger than the corresponding observed values. A system to automatically produce daily meteorological grid dataset is being constructed.The geospatial river network data were produced by ArcGIS and they were utilized in the H08 model to simulate the river discharge. Firstly, this research is to compare simulated and measured specific discharge, which is the ratio of discharge to watershed area. Significant error between simulated and measured data were seen in some rivers. Secondly, the outputs by the coupled model including crop growth module and reservoir operation module were analyzed. However, there are differences between the simulated value and measured value. We need to improve dam operation, artificial water intake, and parameters such as depth of the soil and flow velocity in the river.