Konferenzbeitrag
BRISEIDE a spatio-temporal framework to support environmental analysis and emergency management
Vorschaubild nicht verfügbar
Volltext URI
Dokumententyp
Text/Conference Paper
Zusatzinformation
Datum
2011
Zeitschriftentitel
ISSN der Zeitschrift
Bandtitel
Verlag
Shaker Verlag
Zusammenfassung
Large scale natural disasters such as landslides, earthquakes, forest fires or fresh floods, require adequate IT tools to help operators better plan resources and improve reaction to unexpected events by deploying the most appropriate response model and contingency plan. To do so decision makers, civil protection operators, public administrators typically need to be capable of accessing, distributing and processing a wide range of information including, but not limited to, Earth Observation data, geographical information within GIS repositories, as well as live sensor data. Being able to access these resources from a unique 3D interactive environment, which acts as a one-stop-shop, it ensures essential improvement in terms of efficiency, thus reducing possibility of errors. This work presents the results of the ongoing EU-funded project BRISEIDE - BRIdging SErvices, Information and Data for Europe (www.briseide.eu), which is being financed by ICT-PSP Programme. BRISEIDE specifically addresses civil protection scenarios by developing an interoperable Spatial Data Infrastructure (SDI), which provides not only interoperable access to data but also to processing functionalities. To do so BRISEIDE has developed a number of interoperable web-services that are not only able to deal with spatial features but also with temporal dimension. This way it becomes possible for the civil protection operators to perform spatio-temporal analysis, essential to run comprehensive spatio-temporal queries of historical data and to perform simulations of scenarios of crisis. The possibility to retrieve or query datasets based on time properties is a fundamental requirement of the BRISEIDE framework. For this the fundamental step forward has been the definition of a data model which allows operating on spatial datasets which consider time as true dimension. With regard to this issue, it must be noted that Open Geospatial Consortium (OCG) standards such as WMS (Web Map Service) and WCS (Web Coverage Service), in their latest versions, formally support time as dimension. The requirement, set by the project, was to define a data/metadata model that considered time as a variable in a way that could be supported through OGC-compliant requests. The aforementioned protocols support in fact the TemporalDomain element, which may or may not be referring to a spatialDomain element. In this specific case the TemporalDomain element describes the valid time constraints that can be associated to requests sent by the BRISEIDE 3D client to create a virtual scene (e.g. through GetMap or GetCoverage requests). The data model developed by BRISEIDE, which is at the basis of the interactive 3D client developed, includes the TemporalDomain item. This may be structured either as a sequence of time instants (using Geography Markup Language – GML’s TimePosition) and/or time intervals (using timePeriod with beginPosition and endPosition, both of the GML TimePosition type). These time periods may refer to regular samples (indicated by the optional timeResolution element) or continuous reading (when timeResolution is absent). The 3D client can then represent raster spatio-temporal geographic information, through a complex structure based on a number of gridded datasets having the same domain and the same band layout. This is typically the case of remote sensing observations, imagery data or model runs/executions referring to time intervals. The 3D client can handle, in an interoperable manner, through WCS requests, time-dependent multidimensional raster data, such as Earth Observation data, whose models are represented by several 3D hypercubes for a predefined number of specific times, separated by a certain time period. Given a specific run time, a hypercube of data is in general multi-dimensional by itself. Its extent can span for instance over Longitude, Latitude, Time (various forecasts time) and Elevation. The basic idea is that such a multidimensional object can be built by wrapping instances of the