Abstract
Surface failures induced by rainfall cause most sediment disasters. Many lead to extensive damage. Assessment of the risk of such disasters is crucially important for the safety and security of society. Such risk assessment is becoming increasingly important in this era of global warming, when torrential rains and localized heavy rains are proliferating. Instead of an index such as accumulated rainfall amount, we attempted to conduct risk assessment using the soil moisture deficit, i.e., the remaining capacity of the soil column to accept rainwater. After a numerical model was developed to calculate the soil moisture deficit in any or all MESH3 (about 1 km resolution) grid cells with existing routine data, the model was applied to 446 MESH3 grid cells in Niigata prefecture, where failures occurred during 2010-2011. The model showed that the soil moisture deficit near the failure occurrence became smaller, i.e., the soil column was closer to complete saturation. This finding suggests that the soil moisture deficit can be a good indicator of risk. Furthermore, a probability distribution model was constructed from minimum soil moisture deficit data near the failure occurrence. Using this probability model, the threshold of the soil moisture deficit to put an area on alert, corresponding to the disaster coverage ratio, can be calculated easily. The possibility of early warning is also demonstrated by combining the soil moisture deficit at a certain time with a short-time rainfall forecast for several hours in the future.
