Theme Leader: Luk Peeters
A major challenge in imaging the deeper earth is the requirement to assimilate multiple data. These mostly sparse data are not only of different type, but also of differing resolution in space and time. We are developing and applying robust tools to assimilate these data and prior knowledge in our probabilistic earth models.
The probabilistic assessments of resources allows us not only to systematically quantify model uncertainty but also to determine the value of information of various data sources. Value of Information analysis compares the cost of data acquisition with the potential payoff. These results can then be used to compare different gathering different types of data or acquiring data at different locations.
An important topic within this research theme is the visualisation and communication of data and models. We are improving techniques for researchers to intuitively interact with large high-dimensional, spatio-temporal data sets using techniques such as dimensionality reduction, pattern recognition and immersive visualisation and analytics.