In this study, we conducted a series of numerical Lagrangian experiments in the Porcupine Abyssal Plain region of the North Atlantic and developed a machine learning approach to predict the surface origin of particles trapped in a deep sediment trap. The data contain :

- I. Probability density function of the particles position from the Lagrangian experiments.

-II. The dynamic variables (temperature, vorticity, u, v, sea surface height) associated with each Lagrangian experiments and used for the training/ testing.

-III. The saved parameters and logs of the machine learning models.

-IV. Some processed data such as kinetic energy and okubo-weiss parameter used for analysis.