Identifying the Transitions in the Stable Socio-Environmental System Due to Extreme Events

The critical challenge in a hydrological system is to predict whether the system approaches a critical threshold. The urban centres are grappled by the extreme events especially floods with the shifts from one stable state to another in an urban socio-environmental system. Here, we identified the critical transitions of hydrological processes, including precipitation and runoff, by analyzing their shifting nature. Structural break-regression models, incorporating shifts in both mean and trend, are applied to the series.  The point of change indicates the transition within the system.  These models are then evaluated using two widely employed penalized likelihood criteria for multiple changepoints. These criteria strike a balance between the quality of model fit (measured by likelihood) and the consideration of parsimony. Two models are tested i.e., bisegmentation and penalised maximum likelihood with the white noise detection. The latter was found to be better fit to both precipitation and runoff for the three cities (Guwahati, Mumbai, and Dehradun) in India.