Widespread carbon accumulation in mature forests remains underestimated by satellite observations: source data and reproducible code

作者：Lei Ma, G. C. Hurtt, Hao Tang, Dalei Hao, Philippe Ciais, Yude Pan, Stephen Sitch, Pierre Friedlingstein, Ralph Dubayah, Matheus Henrique Nunes, Xinyuan Wei · 发表于：Zenodo (CERN European Organization for Nuclear Research) · 年份：2026 · DOI：10.5281/zenodo.21192282 · 被引用次数：1 · 研究领域：Computer science、Database、Remote sensing、Data mining

This repository contains the source data and reproducible Python code used to generate all figures in the manuscript Widespread carbon accumulation in mature forests remains underestimated by satellite observations. The repository is organized by figure. Each figure has a dedicated folder (e.g., Figure_1, Figure_2, Figure_S1) containing the processed source data used to generate the figure, together with publication-quality exported figures. A single Jupyter notebook (Figures.ipynb) located in the repository root reproduces all figures in the manuscript from the corresponding source data. Each section of the notebook is clearly labeled by figure number (e.g., Figure 1, Figure 2, Figure S1) and generates the associated figure directly from the processed data provided in this repository. The repository contains only the processed data required to reproduce the published figures. Original datasets (e.g., GEDI, FIA, NEON airborne lidar, ESA CCI Biomass, TRENDY, and other publicly available products) should be obtained from the original data providers as described in the manuscript Data Availability section. All analyses and figure generation were performed in Python 3.9.23 using the following packages: NumPy 1.26.4, xarray 2024.7.0, rioxarray 0.15.0, Rasterio 1.4.3, GeoPandas 1.0.1, Shapely 2.0.7, Matplotlib 3.9.4, Cartopy 0.23.0, cmocean 3.0.3. This repository is intended to facilitate full reproducibility of all figures presented in the manuscript.