By incorporating the sea surface temperature (SST), chlorophyll-a mass concentration (Chl.a), sea surface height anomaly (SSHA), ocean mass and geostrophic current from multi-source remote sensing observations, this paper models and predicts the temporal-spatial distributions of the catch per unit effort (CPUE) of O. bartramii in the northwest Pacific Ocean with the supervised learning algorithm-based back propagation (BP) neural network model. The multi-source remote sensing data were used to build the BP neural network model, and the accuracy of the model-simulated and -predicted O. bartramii CPUE was then evaluated with the historical fishery data during 2004 to 2017 in the northwest Pacific Ocean from the Chinese Squid-Jigging Technology Group of Shanghai Ocean University. Compared with the traditional scheme of predicting the spatial-temporal distributions of O. bartramii CPUE with SST, Chl.a and SSHA only, the accuracy of model-simulated and -predicted CPUE can be further improved after incorporating the ocean mass and geostrophic current into the BP neural network model.Specifically, the standard deviation (STD) and root mean square error (RMSE) of model-simulated O. bartramii CPUE both increased by 22%, and STD of model-predicted O. bartramii CPUE increased by 31% and RMSE decreased by 26%.