The BERLIN project: Delving into behavioral modelling for evidence-based adaptation policies under complex human-water interactions

Water storage systems are crucial for achieving several Sustainable Development Goals amidst evolving climatic and societal conditions. Today, over 58,000 large dams regulate about 46% of the world’s main rivers, which are shaped by both their inherent hydrologic patterns and the choices made by human sectors, who determine how much water to hold in reservoirs and how much to release for different stakeholders’ needs. While mathematical models for water resources processes benefited from centuries of study and development across spatial and temporal scales, the research on human behavioural models and their integration in socio-hydrology frameworks are less advanced. Thus, there is a pressing need to highlight the critical role of human behaviours concerning water systems with coexisting and conflicting water purposes, asking for more accurate and valid water balance simulations when exploring alternative, robust water management strategies able to satisfy competing and multisector societal needs.  

In this context, the BERLIN project aims to construct innovative behavioural models of the human intentions and preferences by leveraging the recent advances in Machine Learning, which allow exploiting the full potential of the unprecedented availability of big observational data, with insights from Social Learning, incorporating stakeholders’ experiences and preferences from a triple-loop approach (risk awareness, risk perception, and risk adaptation) to reinforce the model-based exploration of adaptation policies. Their combination in different climate change hotspots representing semiarid regions, river deltas, and snow-dependent river basins will support the development of behaviourally explicit hydrologic models, providing rigorous retrospective assessments of observed decision-making processes and the generation of reliable and credible projections of the future co-evolution of complex water systems. At the same time, BERLIN will promote knowledge exchange by involving key stakeholders through co-creation processes, collaborative frameworks, and participatory indicators, ensuring that place-based knowledge and end-users’ priorities are embedded in global modelling efforts.

Particular emphasis will be dedicated to elucidating how water systems and societies co-evolve through feedbacks between hydrological dynamics, infrastructural operations and institutional/behavioral drivers. For this purpose, socio-hydrological modelling and hydro-social analysis become increasingly more important to unpack how policies, risk perceptions, inequality, and power-interest (im)balance shape water availability, hazards, and resilience over time. From a sociohydrology perspective, these approaches improve water resource allocation, sustainability, and conflict resolution by integrating human decision-making with physical processes. From a hydrosocial research angle, they support context-specific, equitable, and robust strategies that anticipate behavioral responses, unintended consequences, and long-term dynamics under uncertainty and change across scales and decision-making settings, thus supporting better water resource governance.

Against this background and aligned with the HELPING vision of the Science for Water Solutions Decade, BERLIN promotes anchoring hydrological science in real-world decision-making processes and integrating global datasets with national and local information sources, including in-situ observations. The integration of sociohydrology and hydrosocial research concepts and methods contributes to the understanding of the key interactions and potential loops between global drivers and locally specific water management and governance challenges, explicitly accounting for human responses, non-linear dynamics, feedbacks, and evolving system trajectories.