Optimally growing initial error for predicting the sudden shift in the Antarctic Circumpolar Current transport and its application to targeted observation

Time-optimal path for oceanic targeted observation based on multiple mobile platforms using the traveling salesman problem

“…Among the adjoint-based approaches, the Conditional Nonlinear Optimal Perturbation (CNOP) proposed by Mu et al (2003) [13] is designed for the targeted observation study of the oceanic models governing the nonlinear motions. To date, the CNOP-based targeted observation has shown the potential to improve the prediction of multiple oceanic events or processes, including the El Niño-Southern Oscillation (ENSO) [14], Kuroshio related events [15][16] [17], South China Sea western boundary current [18], Antarctic Circumpolar Current (ACC) transport [19] and Yellow Sea vertical thermal structure [20] [21].…”

Application of the Most Sensitive Initial Error in Sampling Scheme Design for a Mobile Platform

“…Targeted observation is primarily concerned with the location of the deployment of the observing assets. In a related work, Zhou et al, 2022 studied the effect of the spatial structure of the CNOP on the prediction of the Antarctic circulation and demonstrated that it can be used to design an observation network for the Drake Passage [8]. Yang et al, 2020 employed intelligent optimization algorithms with a dimensionality reduction module to compute CNOP in the GFDL CM2p1 model, suggesting that the method is effective for calculating CNOP in complex models [9].…”