Participants: Edzer, Jan, Lubia and Karine

Title: Analyzing spatio-temporal phenomena from data: necessary representations and operations.

Different phenomena:

1. continuous physical phenomena (temperature, precipitation)
2. continuous vectors with a magnitude and direction (wind)
3. evolving objects (clouds)
4. moving objects: trajectories (animals), and their aggregates (flows; through a network? through free space?)
5. space-time point patterns (volcanic eruptions, scientific conferences), possibly marked

Topics to address:

1. (Karine: implementing/designing an algebra for a spatio-temporal database)
2. implementing a full algebra for spatio-temporal data analysis
3. web services to access spatio-temporal databases and web services for spatio-temporal data processing
4. dealing with uncertainty
5. usability of visualisation tools for uncertain spatio-temporal data
6. go beyond the time snapshot = animation analogy –> Q: What that means?, ST = time added to space.

What we want:

1. find/characterize/identify spatio-temporal patterns (change)
2. spatio-temporal patterns in streams/flows, understand spatio-temporal interactions

Spatio-temporal data is used to represent dynamic spatial phenomena. Some examples are physical processes like temperature and volcano eruption, moving objects like animal tracking and iceberg movement as well as evolving objects like oil spill evolution on the ocean. This spatio-temporal data is generated by different technologies, such as fixed and mobile geosensors, GPS and Earth observation satellites or radio-frequency identification (RFID) and can be distributed in different data sources. In order to analyse this kind of data specific operations are necessary, for instance, extraction of trajectories, calculation of the variation of distance between two moving objects over time and extraction of time series of a specific location or area. In this context, service oriented architecture is a good option to create interoperable frameworks in order to handle distributed spatio-temporal data sources. Therefore, this work aims to evaluate if and how existing web service specifications, such as Sensor Observation Service (SOS), Web Feature Service (WFS), Web Coverage Service (WCS), Web Processing Service (WPS), Open Location Service (OpenLS) and Sensor Planning Service (SPS), deal with spatio-temporal information. The main goal is to assess them by considering the following aspects: supported spatio-temporal data and operations, usability and compliance of existing implementations.

• TerraLib (database, elementary topological operations)
• TerraHS/Haskell (prototype implementation)
• PostGIS
• Secondo (database for moving objects), Aguila (visualisation)
• TerraME, PCRaster (environmental modelling)
• TerraView (wants to do animation)
• R, aRT
• OGC web services – 52N suite, …

1. web service (access and processing) specifications (Master thesis)
2. usability of visualisation of spatio-temporal data (Q: bachelor or master?)

- Q:Is goodness of fit a possible mobility measure proposal?