There are many large-scale problems which require new approaches to computing, such as earth observation, environmental management, biomedicine, industrial and scientific modelling. The CrossGrid project addresses realistic problems in medicine, environmental protection, flood prediction, and physics analysis. These applications are oriented towards specific end-users:

· medical doctors, who could obtain new tools to help them to obtain correct diagnoses and to guide them during operations,

· industries, which could be advised on the best timing for some critical operations involving risk of pollution,

· flood crisis teams, which could predict the risk of a flood on the basis of historical records and actual hydrological and meteorological data,

· physicists, who could optimise the analysis of massive volumes of data distributed across countries and continents.

Each of these efforts will contribute to improving the quality of life and environment.

These new developments and applications could be complex and difficult to use, even by experienced users. This problem is recognised, and the CrossGrid project plans to develop several tools which will make the Grid more friendly for average users. Portals for specific applications will be designed, which should allow for easy connection to the Grid, create a customised work environment, and provide users with all necessary information to run their jobs.

As the first application, we will develop a Grid-based prototype system for pre-treatment planning in vascular interventional and surgical procedures through real-time interactive simulation of vascular structure and flow. The system consists of a distributed real-time simulation environment, in which a user interacts in Virtual Reality (VR). A 3D model of the arteries, derived using medical imaging techniques, will serve as input to a real-time simulation environment for blood flow calculations. The user will be allowed to change the structure of the arteries, thus mimicking a surgical procedure and the effects of this adaptation will be analysed in real time while the results are presented to the user in a virtual environment.

The second application will be a Grid-based Support System for flood prevention and protection. The kernel of this system is numerical flood modelling that requires an appropriate physical model and robust numerical schemes for a good representation of reality, as well as grid distributed supercomputing aspects for realistic simulations when dealing with large problem sizes.

As the third application we will develop final user applications for physics analysis running in a distributed mode in a Grid-aware environment using large distributed databases for High-Energy Physics (HEP). The challenging points are: seamless access to large distributed databases in the Grid environment, development of distributed data-mining techniques suited to the HEP field, and integration in a user-friendly interactive way, including specific portal tools. As indicated in the CERN LHC Computing Review Report, interactive data analysis tools is one of the areas where joint efforts amongst one or more experiments, resulting in common projects and products, might lead to cost savings, or decreased risk, or both. The proposed applications explore and develop the use of advanced simulation and interactive data mining techniques.

The fourth application is weather forecast and air pollution modelling. An important component will be a data mining system. It will be used for the analysis of the archive of operational data from a mesoscale model. It will
be also used for meteorological reanalysis of data bases which include homogeneous meteorological information from a single numerical weather prediction model integrated over decades. The system will also include an atmospheric pollution chemistry module.