Transboundary water resource allocation under drought emergencies: A multistage bilevel programming model considering decision-makers' risk perception

Transboundary river basin drought emergencies caused by extreme climate and human activities can increase the risk perception degrees of decision-makers (DMs), causing them to overestimate the consequences of disasters, which could intensify the water competition among riparian countries. To address this issue, this paper treats the basin authorities in upstream and downstream countries as upper- and lower-level DMs, respectively, and a multistage bilevel programming model that considers DMs' risk perceptions is constructed. The model aims to achieve a reasonable water resource allocation scheme by minimizing the risk perception degrees of bilevel DMs that also helps them to pursue the objectives of economic development, social stability, and ecological protection. The value function in prospect theory is modified to describe the risk perception degrees of DMs. To solve the model, a genetic algorithm (GA)-based fuzzy goal programming method is developed, which enables upperlevel DMs to prioritize the minimum satisfactory degree while allowing lower-level DMs to adjust their satisfactory degree accordingly. The proposed model and methods are applied to the Lancang-Mekong River basin, and the comparison results show that the proposed model can effectively reduce the risk perception degrees of upper-level DM (with a reduction of up to 43.15%), while maintaining the risk perception degrees of the majority of lower-level DMs (except for Vietnam) relatively stable (fluctuating within the range of 3.03%). Furthermore, a sensitivity analysis method is used to analyze the impact of DMs' perceived risks on their risk perception degrees. Finally, some management insights are provided to promote water allocation cooperation throughout the whole river basin.