Forecasting nonlinear adaptation in human-water systems

Environmental and socioeconomic transformation is triggering unprecedented and abrupt changes in the behavior of individuals, or nonlinear adaptive responses. The decision to incur water theft under growing scarcity, or the dam construction paradox where new infrastructures drive demand growth that offsets the supplies made available, are both examples of nonlinear responses with severe consequences for societies that are challenging to anticipate. This study critically assesses gaps in our understanding of human behavior, human interactions, institutions, and uncertainty that limit our ability to model and forecast nonlinear adaptation responses. Next, I propose a research agenda to address these gaps and illustrate its feasibility leveraging relevant pioneering research. I argue that a key aspect to achieving forecasting skills should be generalizability, building on unified theories to design models that can be widely adopted rather than resorting to ad-hoc approaches.