ICAROS: AN INTEGRATED COMPUTATIONAL ENVIRONMENT FOR THE ASSIMILATION OF ENVIRONMENTAL DATA AND MODELS FOR URBAN AND REGIONAL AIR QUALITY D. A. SARIGIANNIS 1∗ , N. SOULAKELLIS 2 , K. SCHÄFER 3 , M. TOMBROU 4 , N. I. SIFAKIS 5 , D. ASSIMAKOPOULOS 4 , M. LOINTIER 6 , A. DANTOU 4 and M. SAISANA 1 1 Institute for Health and Consumer Protection, European Commission – DG JRC, B-1050 Brussels, Belgium; 2 Department of Geography, University of the Aegean, Mytilene, Greece; 3 Fraunhofer Institute for Atmospheric Environmental Research, Garmisch-Partenkirchen, Germany; 4 Department of Physics, University of Athens, Greece; 5 Institute for Space Application and Remote Sensing, National Observatory of Athens, Greece; 6 Institut de Recherche pour le Développement, Montpellier, France ( ∗ author for correspondence, e-mail: dimosthenis.sarigiannis@cec.eu.int) Abstract. Integrated environmental management in urban areas is nowadays considered a sine qua non objective of Community and national environmental and development policies. A large amount of scientific information on the state of the environment is now available from a large pool of data sources. This work presents an innovative method for integration of these data sources and effective coupling of environmental information with appropriate models and decision-support tools. State-of- the-art Earth observation techniques, ground-based air quality measurements, atmospheric transport and chemical modelling, and multi-criteria decision-aid systems are used in an integrated information fusion environment in support of environmental and health impact assessment and decision-making at the urban and regional scales. Results of the pilot application of the method in the area of Lombardy in Northern Italy demonstrate the validity and usefulness of this novel approach. Keywords: aerosol optical thickness, air quality, environmental management, information fusion 1. Introduction Environmental systems and, consequently, environmental management problems have several attributes that render their formal representation particular and the proposed solutions multidisciplinary. Some of these distinctive features are high- lighted below: • Dynamics. Environmental systems evolve with time and many of these have memory (for instance, accumulation of toxic pollutants with biocumulative effects). • Spatial coverage. Modelling environmental processes rigorously needs linking spatially referenced variables with time referenced ones. Data are stored in spatial databases often via a geographic information system (GIS) to assist in spatial analysis. Water, Air, and Soil Pollution: Focus 2: 641–654, 2002. © 2002 Kluwer Academic Publishers. Printed in the Netherlands.