Resilience Aspects in the Sensor Web Infrastructure for Natural Disaster Monitoring and Risk Assessment Based on Earth Observation Data

This paper focuses on enabling resilience of the Sensor Web system for disaster monitoring and risk assessment that is based on the use of Earth Observation (EO) data. Resilience is the ability of the system to maintain trustworthy service delivery in spite of changes. Resilience of the EO infrastructure becomes an extremely important issue, especially in the disaster management domain. We propose to enable resilience through: 1) the use of grid resources and services to meet high performance computing requirements; and 2) assessment of resource reputation to select reliable resources. Grid services enable redundancy through integration of heterogeneous resources administrated by geographically distributed organizations. The same services could be provided by different organizations within the infrastructure, and could be of different levels of quality. Therefore, we propose to incorporate a utility-based reputation model to assess reliability of the resources. The proposed approach is implemented within the Namibia Sensor Web Pilot Project that was created as a testbed for evaluating and prototyping key technologies for rapid acquisition and distribution of data products for decision support systems to monitor floods.