Data from profiling floats in the Black Sea revealed complex temporal and spatial relationships between physical variables and biogeochemical parameters such as oxygen, chlorophyll, and the backscattering coefficient at 700 nm. To address these interdependencies, as well as limitations in understanding biogeochemical dynamics, a feedforward backpropagation neural network (NN) was developed. Trained on float data, the NN predicted biogeochemical states using physical measurements alone. Its performance was particularly strong for oxygen. Biogeochemical states reconstructed by the NN, using physical data from a coupled physical–biogeochemical model, outperformed the biogeochemical outputs of the model itself. This demonstrates that combining float data, numerical model outputs for physical parameters, and NNs is a powerful method for reconstructing the 4D dynamics of the euphotic zone. Basin-wide patterns and temporal variability in oxygen, backscattering coefficient, and chlorophyll were also examined.