There are about three to four typhoons striking Taiwan every year. Almost all the typhoons in Taiwan have produced flood and inundation, which subsequently cause damage of property and loss of human life. To effectively reduce the disaster, and support authorities to mitigate the flood, it is crucial to constitute a real-time decision-supporting mechanism to answer for the emergency response. We think the grid technology is a right time to play an important role on this duty.

System Development

During the flood event, four decision-supporting systems are the required tasks for flood mitigation center. They are the flood forecasting system, the real-time monitoring for flood scenario system, the flood warning system, and the emergency response system. Based on these tasks, the flood mitigation grid was formed. The team members participating in this research include two parties; one party (NCHC and central university) is in charge of grid resources and platform; the other one is in charge of (National Taiwan University and III) the application and service. NCHC is responsible for building grid resources and varieties of platforms. Network, storage, and computer are the key resources to form the Cyber-environment and the data grid, computing grid. To enrich the pervasive computing capability in the flood Cyber-environment, the technologies that support real-time communication, remote-taking sensor data, and advanced visualization are incorporated. Access Grid technology has been used for real-time communication for group-to-group interactions across the Grid. The development integrates the modern video codecs, such as MPEG4 and H.264, into the video program. Hence, it is able to do the high-quality video conference. In the sensor network technology, NCHC has designed the embedded system and worked with hardware companies to layout the necessary boards to collect the field river and reservoirs data for flood monitoring. The embedded system is targeted to be remote-controlled via IPv6 and IPv4. All the data, including video, audio and data from data logger, are collected real-time and archived to be analyzed by domain knowledge experts.

      Architecture

      The structure of the Grid application system could be depicted as a three-level architecture based on the functions it provides. They are the fabric service level to connect facilities such as network, computing, and storages; the grid platform level developed by Middleware; and many kinds of application in the grid application level. And the main categories of the grid application level could be divided as the data grid to acquire and transfer the distributed hydraulic data swiftly; the computing grid to apply distributed high-performance computing resources; the sensor network to get the hydrological scenario of the river through remote monitoring images; and the Access Grid to communicate among relevant members for the mitigation policy by holding a multi-users video and data conference easily and smoothly. The job of flood forecasting couples the application of Data Grid and Computing Grid, while the job of dealing with emergency situations resulting from floods might couple the sensor network to assess the flood and hazard, and the Access Grid to have the consultative conference about policy decision.

Grid Infrastructure of KING (Knowledge Innovation National Grid)	Grid Infrastructure for Flood Mitigation
Flood Forecasting and Storage Backup

   Components     

Real-Time River Monitoring  ( http://hmerg.nchc.org.tw/sensornet/)

Information fusion and sharing via Access Grid and distributed GIS and in-house grid middleware

Sensor network development specifically developed for real-time environmental observation.

    Collaborators：