Measuring Uncertainty in Lichen Biomonitoring of Atmospheric Pollution: The Case of $\hbox {SO}_{2}$

Large-scale biomonitoring surveys on atmospheric air pollution face the following challenges: (1) how to select which biomonitor and how this organism reflects the atmospheric pollutant of interest and (2) what methods to use for interpreting the vast data that will be gathered. This paper addresses these issues through an integrative fuzzy knowledge-based system for environmental biomonitoring applications with lichens. The system gathers and combines geographical, ecological, and physicochemical data of lichen responses to pollution within a computer program that (1) recognizes groups of indigenous species suitable for long-term pollution monitoring and (2) estimates pollution levels from species distribution data. Thereby, the proposed scheme provides a means to convert species field data to measurements under conditions of uncertainty. The proposed biosurveillance program has been successfully tested at a small scale, proving its functionality even under conditions of increased uncertainty. Within a suitable management framework, it could further be utilized in environmental impact studies and risk assessment (positive or analytic approach), short-term decision making (normative or tactical approach), and long-term policy making (normative or strategic approach).