Cheng et al. (2026) Atmospheric forcing uncertainty contributes to divergent estimates of China’s terrestrial carbon dynamics

⚠️ Warning: This summary was generated from the abstract only, as the full text was not available.

Identification

• Journal: Environmental Research Letters
• Year: 2026
• Date: 2026-01-01
• Authors: Yue Cheng, Peng Luo, Hao Frank Yang, Mingwang Li, Honglin Li, Yu Huang, Ming Ni, Wenwen Xie, Chuan Luo, Zhigang Hu
• DOI: 10.1088/1748-9326/ae2af8

Research Groups

Not explicitly stated in the abstract, but involves researchers utilizing the Community Land Model (CLM 5.0) for China-focused studies.

Short Summary

This study assesses China's terrestrial carbon budget (1979–2014) using the Community Land Model (CLM 5.0) driven by three different meteorological forcing datasets, revealing that forcing choice significantly alters carbon flux magnitudes and trends, and highlighting the need for improved meteorological inputs for robust carbon budgeting.

Objective

• To assess China’s terrestrial carbon budget (1979–2014) using the Community Land Model (CLM 5.0) driven by different meteorological datasets.
• To systematically assess how uncertainties in atmospheric forcing datasets affect land surface model simulations of terrestrial carbon dynamics.

Study Configuration

• Spatial Scale: China
• Temporal Scale: 1979–2014

Methodology and Data

• Models used: Community Land Model (CLM 5.0), Shapley additive explanations (for machine learning analysis).
• Data sources:
  • Meteorological forcing datasets: CRUNCEP, Global Soil Wetness Project Phase 3 (GSWP3), China Meteorological Forcing Dataset (CMFD).
  • Evaluation data: Over 800 FLUXNET site-months, nine eddy covariance towers.

Main Results

• The choice of meteorological forcing dataset strongly alters the magnitude and trend of carbon fluxes in China.
• China's terrestrial ecosystems were estimated to be either a weak carbon sink or a net source, depending on the forcing dataset used.
• Total ecosystem carbon storage was estimated at 86.30–90.00 petagrams of carbon (PgC), with soil accounting for 84.1% and vegetation for 15.9%.
• GSWP3 performed best overall in simulating gross primary productivity (GPP), while CRUNCEP performed relatively poorly.
• CMFD performed best in high-altitude and cold-dry regions.
• Interannual GPP variability was primarily controlled by precipitation (29.4%), followed by temperature (17.2%), with shortwave radiation having negative effects (11.5%).
• Moisture-related variables (precipitation and humidity) dominate interannual GPP variability.

Contributions

• Systematically quantifies the impact of uncertainties in widely used atmospheric forcing datasets on terrestrial carbon dynamics in China using a process-based model.
• Combines process-based modeling (CLM 5.0) with machine learning (Shapley additive explanations) to analyze GPP variability drivers.
• Underscores the limitations of single-forcing simulations in Earth system modeling and highlights the critical need for improved meteorological inputs for accurate carbon budgeting and climate neutrality goals.

Funding

• Not specified in the abstract.

Citation

@article{Cheng2026Atmospheric,
  author = {Cheng, Yue and Luo, Peng and Yang, Hao Frank and Li, Mingwang and Li, Honglin and Huang, Yu and Ni, Ming and Xie, Wenwen and Luo, Chuan and Hu, Zhigang},
  title = {Atmospheric forcing uncertainty contributes to divergent estimates of China’s terrestrial carbon dynamics},
  journal = {Environmental Research Letters},
  year = {2026},
  doi = {10.1088/1748-9326/ae2af8},
  url = {https://doi.org/10.1088/1748-9326/ae2af8}
}