Ocean surface drifter and drifter simulation data

Many practical problems in fluid dynamics demand an empirical approach, where statistics estimated from data inform understanding and modelling. In this context data-driven probabilistic modelling offers an elegant alternative to ad hoc estimation procedures. Probabilistic models are useful as emulators, but also offer an attractive means of estimating particular statistics of interest. In this paradigm one can rely on probabilistic scoring rules for model comparison and validation. Stochastic neural networks provide a particularly rich class of probabilistic models, which, when paired with modern optimisation algorithms and GPUs, can be remarkably efficient. We demonstrate this approach by learning the single particle transition density of ocean surface drifters from observations using a mixture density network. This provides a comprehensive description of drifter dynamics, from which we derive maps of various single-particle statistics. Our model also offers a means of simulating drifter trajectories as a discrete-time Markov process. A drifter release simulation using our model shows the emergence of concentrated clusters in the subtropical gyres, in agreement with previous studies on the formation of garbage patches.
The dataset is intended to accompany the code repository archived at doi.org/10.5281/zenodo.7737161 . They are both related to the upcoming paper Brolly, M.T. (in submission), "Inferring ocean transport statistics with probabilistic neural networks".