RESEARCH RECORD / Carbon-ECS
METHOD / FINDINGS
A 17-year, approximately 1 km benchmark and a physics-inspired model that pair large-scale transport modeling with iterative refinement of local carbon-flux detail.
The dataset spans 2003–2019 (204 months) and repairs gaps through anomaly removal, wavelet denoising, and K-nearest-neighbor interpolation. The Frequency Global Module forecasts amplitude and phase; the Diffusion Local Refiner iteratively restores local detail.
17 years of monthly data at approximately 1 km. Benchmark specification, not a comparative performance claim. (Table 1, p. 2.)
A hierarchical mixture-of-experts model that separates environmental regimes from physical mechanisms for interpretable global ocean carbon-flux forecasting.
A 17-year, approximately 1 km benchmark and a physics-inspired model that pair large-scale transport modeling with iterative refinement of local carbon-flux detail.
