Global ocean phosphate dataset based on the World Ocean Database 2023

Global ocean phosphate dataset is generated by following the protocol published by Ito et al., (2024) Journal of Geophysical Resarch, Machine Learning and Computation. The dataset covers the global domain at 1 degree longitude-latitude resolution from the surface to 1000m depth at yearly frequency for the period of 1990 to 2020. The quality controlled bottle phosphate data is taken from the World Ocean Database 2023, and only "good" data points are used for our calculation. There two ensemble members based on two different implementation of the neural network algorithm. Cited Reference Ito, T., Cervania, A., Cross, K., Ainchwar, S., & Delawalla, S. (2024). Mapping dissolved oxygen concentrations by combining shipboard and Argo observations using machine learning algorithms. Journal of Geophysical Research: Machine Learning and Computation, 1, e2024JH000272. https://doi.org/10.1029/2024JH000272