Reproduce “Advancing global crop simulations with differentiable modelling” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Yunan Lin, Sebastian Bathiany, Maximilian Gelbrecht, Jens Heinke, Christoph Müller, Niklas Boers
Agrocosm makes the crop-and-soil processes of LPJmL differentiable, allowing gradients to tune selected physical parameters against daily observations of gross primary production, ecosystem respiration, and evapotranspiration. Calibration reduced normalized error by 24.8–36.2% across these fluxes at held-out sites. Without using yield data for fitting, it improved historical yield-anomaly correlations across 61.2–64.3% of evaluated crop area, not every location.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.