A risk assessment of river levee breach in gravel-bed rivers under climate change: A case study of the Otofuke River, Japan

Climate change is a critical social concern in many research and engineering fields, and one of the vital issues in this regard is its significant impact on water and sediment-related disasters. Further increases in precipitation intensity and associated river discharge increases will change the flood characteristics and associated morphological changes in rivers, causing critical damage to river training structures and residential areas along rivers. Here, we assess the risk of the river levee breach caused by significant bank erosion due to in-channel morphodynamic processes using a large ensemble hydrological dataset and a physics-based morphodynamic model. The target river reach is the upper Otofuke River, Japan, a typical steep, gravel-bed river, and some river training structures along this river were damaged due to the significant morphological change of bed and bank during several huge flood events. Our observed dataset is highly limited, making it essentially difficult to assess risks related to river disasters, especially under future climate conditions. The use of a large ensemble climate calculation dataset, including precipitation, discharge, etc., provided in the context of climate change research will overcome the aforementioned difficulties and contribute to a comprehensive understanding of possible significant flood events in current and future climate conditions. We use a large dataset of river discharge hydrographs provided from the large rainfall calculation based on the dynamically downscaling climate calculation of the d4PDF. The thousands of hydrographs are categorized by the k-shape clustering method to evaluate the several important hydrograph shapes since the morphological change of rivers is highly affected by both peak discharge and overall shape characteristics of hydrographs. By using a physics-based morphodynanic model, iRIC-Nays2DH, and characterized hydrographs, we then simulate the possible river morphodynamic processes in the current and future climate conditions. Based on the river morphodynamic calculations, the risk of river embankment caused by morphological change of the river under current and future climate conditions is analyzed. The results show that the hydrograph shape has an important role in the morphological evolution of the river, such as sandbar and meandering development, so the damage intensity of river embankment caused by such morphological change is highly dependent on the hydrographs. More specifically, under the same peak discharge, a hydrograph with a single sharp peak causes less bank erosion; on the other hand, a relatively longer hydrograph drastically increases the risk of bank erosion and river embankment breach. Such a longer hydrograph that greatly impacts the river disaster has a small possibility of occurrence, but it does take place in the future climate conditions. It indicates that climate change does increase the risk of river levee breach caused by in-channel morphodynamic processes in steep, gravel-bed rivers.