Subsurface system behaviour is driven and controlled by the interplay of physical, chemical, and biological processes which occur at multiple temporal and spatial scales. Timely predictive understanding of this behaviour is needed for operational purposes such as subsurface remediation, ecosystem services optimization, or water and soil management as well as for scientific purposes. Current approaches to develop system understanding generally rely on manual approaches for data reduction, numerical modeling and data visualization. Such approaches have fundamental limitations in how timely, effectively and reproducibly they can provide the required understanding. In addition, manual approaches are ill suited for dealing with the large volumes of heterogeneous (and often streaming) data which are increasingly available for subsurface systems. In response to this challenge Subsurface Insights has developed cloud based software known as predictive assimilation framework (PAF). PAF ingests and store heterogeneous subsurface data and can visualize and process this data to provide information on the current state and evolution of the subsurface system. PAF is modular and customizable: capabilities can be activated as users get more and different data, and users get to control access to data and results.
PAF has five components ((1) data and meta data location and ingestion (2) data qa/qc and management (3) data processing, analysis, mining and assimilations (4) a powerful user interface which provides for interactive data and information access and (5) orchestration software which controls the other elements and their interplay) which seamlessly act together. PAF is organized around the project model where data and users are associated with one or more projects. Appropriate access levels provides for fine grained control of data and information access.
For a demonstration of PAF capabilities please contact Roelof Versteeg at Subsurface Insights for further information on PAF.
Development of PAF was supported by the US Department of Energy SBIR program through award DE-SC0009732.