A Data Warehouse for earthquakes signal precursors analysis

Earthquake analysis and monitoring requires complex processing of raw data acquired from several kind of sensors or extracted from data repositories. Management and processing of huge amount of data produced by different sources, i.e. remote sensing and measurement stations of ground and atmospheric physical parameters, the complexity of the environmental models adopted to understand observed phenomena, the necessary distributed collaboration among scientists, results sharing and validation, require specific middleware and applications. A number of researches propose to transform acquired data into assessments of current ecosystem conditions and its evolution trends in time and space; many algorithms, protocols, and applications have been presented to collect, process and analyze ground precursors representing changes in the physical and chemical state of the Earth. These projects deal with data obtained i.e. from satellite, VLF signal and INGV database, integrated with meteorological data and stored in ad hoc repositories. No work deals with a general and flexible approach to integrate and manage different kind of data. In this paper we present a data warehouse to analyze earthquake signal precursors, whose goal is to fuse and exploit efficiently different kind of data and share/display the processing results. The proposed platform allows users/researches: to access data collected by different sources, to record data for future use, to normalize and visualize data for exploring and analyzing complex structure and relationships, and to process data using distributed computing resources and web applications. A case study related to data fusion of INGV data, magnetic field variations acquired by a network of VLF radio stations, MODIS data to evaluate modification in the ground temperature by satellite and geo-located sensor data of weather and climatic parameters is presented.