A participatory system dynamics modeling approach to facilitate collaborative flood risk management: A case study in the Bradano River (Italy)

Participatory methods to support successful policy decisions regarding the complex and dynamic interactions of social, ecological, and physical processes involved in flooding must be simple, easy-to-use, and cost-efficient. Accordingly, a stepwise methodological framework based on causal loop diagrams (CLDs) was developed to address the challenge of context-sensitive initialization of key stakeholders in the collaborative flood risk management process. The methodology consists of five main stages: (i) problem definition, (ii) stakeholder analysis and identification of key groups, (iii) interviews with key stakeholders to construct individual CLDs, (iv) merging of individual CLDs to form a holistic qualitative model representing the entire system, and (v) implementation of an order-oriented reduction process to simplify the final merged CLD, thereby increasing understanding of the most important processes and feedbacks. The proposed approach for flood risk management was tested in a coastal area of southern Italy, a region historically affected by flood events. Given its simplicity, the proposed method was seen as a valuable tool to elicit and map mental models, especially when working with stakeholders who did not have prior modeling experience, i.e., farmers, agricultural companies, tourist complexes, and infrastructure authorities. Indeed, all stakeholders in the current study were able to understand the process and proposed different flood risk management policies, such as land-use changes, management of damage insurance payouts, floodplain activities, and improvement of public awareness. The proposed methodology overcame multiple barriers in initializing stakeholder engagement, including the technical focus of most flood management agencies, the additional cost and time requirements for stakeholder involvement, as well as institutional structures that impede collaborative management. Moreover, the results point to socio-economic aspects of flood risk management that have not been considered in previous modeling studies.