An Event-Based Approach to Process Environmental Data

Environmental sciences studies deal with a large amount of heterogeneous data sources. In each application or scientific study there is a need for a set of tools to manage the data, from the capture, processing to data storage. In particular, in environmental applications there is a need to combine distinct kinds of sensor data from those aboard satellites to those in ground networks. While the first sources are static and usually processed via image management, the second kind is dynamic and processed in streams. Solutions for combining them are usually tailored to a specific problem and geographical region. This work is concerned with solving some of these problems by enabling the integration of heterogeneous data sources whether they are static or streams. The proposed solution uses a pre-processing phase in order to filter the data sources needed for the study, Enterprise Service Bus (ESB) to enable the integration of different data sources and Complex Event Processing (CEP) to process the events that emerge from the integrated environmental data. Events are detected through patterns that are specified by experts and inserted into the CEP engine. Upon detection of an event pattern, events are disseminated, stored in a database or sent to other systems. The main contributions of this work are: (i) a framework to help environmental scientists cope with heterogeneity problems that allows integration of static and stream data sources in a generic way (as opposed to specific solutions in the literature), (ii) treatment of environmental data events, and processing patterns, (iii) application of the aforementioned findings in ecological studies showing how scientists can use our proposal to acquire data of interest from the available data sources.