This paper describes the building of an integrated simulation tool based on a systems approach, and its contribution to local political discussion of the mitigation of microbiological contamination of the water in a coastal area. Local management schemes view water quality as a high-priority environmental objective. In practice, how far this objective is achieved depends on trade-offs between the costs of improved water treatment facilities and the acceptable impacts of water contamination. An in-the-field experiment in collaboration with local managers was carried out in the Thau lagoon on the French Mediterranean coast during the SPICOSA (Science and Policy Integration for Coastal System Assessment) project, from 2007 through 2011. It consisted of building a modeling platform and an integrated assessment framework for simulating exploratory scenarios. The modeling platform combines a dynamic contamination model, which represents the sources of microbiological contamination, wastewater treatment facilities, and physical mechanisms of lagoon contamination, with a prospective economic model, which estimates the patterns of development of economic activities in the area through a holistic approach. Exploratory scenarios are used to assess the risk of water contamination and the efficiency of management measures, under various assumptions about the evolution of the system. The contamination simulations suggest that the work currently planned by local authorities will be inadequate for preventing increased water pollution, and that additional but fairly inexpensive management measures for maintaining the current level of water quality should be considered. The integrated assessment framework estimates the ecological and socio-economic impacts of the various pollution mitigation policies in the broader context of possible local development patterns. The results illustrate how the systems approach may aid in the design of an applicable water policy based on operational objectives and feasible technical options. (C) 2013 Elsevier Ltd. All rights reserved.