Selection algorithm for observation points in environmental data assimilation based on the quantum squeezing effect

Abstract

In this paper, a quantum-enhanced framework is proposed to optimize observation point selection in environmental data assimilation. The method transforms the task into a QUBO problem, balancing uncertainty reduction and spatial diversity. By leveraging a quantum-inspired optical Ising machine, it avoids the exponential complexity of classical optimization. Tests on the Lorenz-1996 model demonstrate its superiority over traditional methods, enhancing computational efficiency without loss of accuracy. The findings underscore the potential of quantum-inspired optimization for scalable, real-time assimilation in high-resolution weather prediction, reducing dimensionality and computational cost.