link to homepage

Institute for Advanced Simulation (IAS)

Navigation and service

ERC Advanced Grant for Martin Schultz

Deep learning is rapidly gaining importance for the analysis of large data volumes and has achieved spectacular successes recently. However, there are only a few researchers, who have started to use the methods of artificial intelligence in the area of environmental research. Dr. Martin Schultz, who heads the research group on Earth System Data Exploration at JSC, has now received a prestigious Advanced Grant from the European Research Council to apply deep neural networks to improve the understanding of the global distribution and trends of air pollution.

In this project entitled IntelliAQ, Schultz wants to link measurements of air pollutants with high-resolution geographic data and data from numerical weather prediction models to obtain detailed maps of regional and global air quality. Furthermore, he expects that deep learning will allow for improved air quality forecasts and automated control of data quality.

The foundations for IntelliAQ were laid in the development of the TOAR database, which contains the world's largest collection of surface ozone measurements and is hosted at JSC. Together, the JSC division Federated Systems and Data, and the Simulation Laboratories Climate Science and Terrestrial Systems, provide an ideal environment to realize the project objectives. IntelliAQ is very demanding in terms of efficient storage and processing of huge data volumes and poses interesting challenges for developers of neural network architectures due to the heterogeneity and complexity of the data involved. The project will receive funding of € 2.5 million over 5 years.
(Contact: Dr. Martin Schultz,

from JSC News No. 258, 30 May 2018