Data management and especially the transformation and preparation of heterogeneous data for scientific analyses is still a demanding task. The preparations for hydrological models need typically the transformation of different data of varying sources, extensive work with geoinformation systems (GIS), the generation of computational grids and the preparation of initial and boundary conditions. A uniform, standardized and scalable data management system will reduce the expenditure of time significantly and allows a faster and more extensive learning.
The Upper Rhine Rift is due to its long history of environmental monitoring, and the diversity of landforms and land use best suited for the setup of a long term observatory with a shared virtual research environment. The essential idea is composed on the target-oriented and balanced development of existing research areas of the universities and the integration of the LUBS for the long-term and extensive basic monitoring. Vital is the administration of the research data in a shared and lasting data infrastructure and research environment, that facilitate quick access to the data.
The main challenge is to improve the fast scientific usability of these heterogeneous and partial incoherent data
- Within the scope of synoptic analysis e.g. using methods of geo- or multivariate statistics or of machine learning
- For the parametrization, verification and conduction of process models for predictions of the terrestrial water and elements balance.
This needs a fast scalability of data grids in space and time to generate coherent data matrices, a connection to geographic information systems to manage georeferenced data and interfaces to analytics software lie R or Matlab, to use their functionality directly for data interpolations. On the other hand is therefore a need to develop pre- and postprocessors to parameterize models and verify them.