Machine-learning based acceleration of land surface models

SPINacc

A spinup acceleration tool for land surface model (LSM) family of ORCHIDEE.

Concept: The machine-learning (ML)-enabled spin-up acceleration procedure (MLA) predicts the steady-state of any land pixel of the full model domain after training on a representative subset of pixels. As the computational efficiency of the current generation of LSMs scales linearly with the number of pixels and years simulated, MLA reduces the computation time quasi-linearly with the number of pixels predicted by ML.

Peer-reviewed scientific Documentation in Global Change Biology [open-source]

Code is freely available on github