2015-01-012024-05-17https://scholars.lib.ntu.edu.tw/handle/123456789/680027Abstract: Frequent disasters caused by extreme weather events have resulted in considerable economic and social losses in Taiwan and all over the world in the recent decades. The major objective of this project is to develop an integrated smart platform for disaster reduction which seamlessly integrates information of hydro-metrology, fluvial-pluvial flood, soil erosion and landslide disasters in decision-making process. After analyzing the information, lesson learned center is built for disaster education and communication by employing various display tools. In the first and second year of this project, we have collected data such as: typhoons in the past, heavy rain events from the rainfall stations, meteorological satellite data of different bands, radar quantitative precipitation, historical positioning of the typhoon paths. Then, we have developed some simulation modeling systems for calibration and verification, and provided the system to collect information from multiple sources and to enhance the understanding of the flood potential. During the third year of this project, we have completed some test cases, and planed presenting and visualizing disaster information, automatic hydrology observation vehicle. This project comprises five research foci: (1) Hydro-meteorology under climate change research: development of real-time spatiotemporal rainfall estimation by using artificial intelligent technologies with QPESUMS radar-rainfall data. The impacts of urban heat island effects on the hydrological pattern, including typhoon, frontal and convective rainfall; (2) Fluvial-pluvial flood research: development of flooding simulation technology by integrating numerical models of river routing, overbank flow, overland flow, and storm water management system to make risk map of dike failure; (3) Soil erosion and landslide research: development of landslide and debris flow model for the prediction and assessment of sediment-related disasters; and (4) Presentation and visualization for disaster information: development of an integrated disaster reduction smart platform as a hub of all natural disaster information including the above mentioned researches and building the lesson learned center and the Interactive Data Visualization System (IDVS) for flood warning. (5) The automated hydrology observation system has advantage of not being affected to the above problems, and the data quality is better than manual observation. Therefore, establishing automatic sensoring system is a future trend. A stationary automated hydrology observation system was installed at the Tai-tung Bridge to measure flow depth, turbidity, surface velocity, bed erosion during high flow discharge, and the system can deliver live video back to the control center with a CCTV. Another target is to use the concept of stationary automated hydrology observation system to integrate instruments on a car; it is highly automated and mobile and achieves a goal of "one person can complete the measurement of river hydrology". To forecast weather patterns more accurately, this project will apply different data sources to improve simulation models (such as the rainfall estimation), and integrated multiple systems (such as meteorological and hydrological disaster prevention information) to improving predictions. In the following years, we will research in: (1) a novel combination of different traditional artificial intelligence theories in the climate prediction models to improve accuracy in early warning of extreme typhoon rainfall; (2) applying different sources rainfall in overland flow simulation to integrate the different modules; (3) combining critical rainfall models and soil erosion predictions to simulate the landslide area and volume of sediment yield from the watershed by inputting rainfall variables; (4) analyzing data to enrich the completeness of the disaster information and support timely disaster responses.