Sign in Subscribe
Machine learning

Artificial Intelligence Study Reveals Uncertainty in Soil Carbon Pools Across Ecosystems in China

Strategic Insights

— 2 min read

A new study using machine learning models to project spatiotemporal changes in soil organic carbon density has revealed uncertainty in China's soil carbon pools across various ecosystems. The study, conducted by a team of researchers from the University of Chinese Academy of Sciences, found that wetlands exhibited the highest soil organic carbon density but contributed minimally to total carbon stock due to their limited area. In contrast, forests and grasslands served as the dominant carbon reservoirs, particularly in deeper soils. The research also showed that climate was the most critical determinant of soil organic carbon density in both soil depth, although its explanatory power was relatively weaker for deep soil organic carbon density.

Key Takeaways:

  • The study focused on projecting spatiotemporal changes in soil organic carbon density by 2050 and 2100 under three Coupled Model Intercomparison Project Phase 6 (CMIP6) climate scenarios.
  • The national average soil organic carbon density in the 0-20 cm soil layer was 4.08 kg C m2, while in the 0-100 cm layer it was 9.42 kg C m2.
  • Wetlands exhibited the highest soil organic carbon density (27.47 kg C m2 in the 0-100 cm layer) but contributed minimally to total carbon stock due to their limited area.
  • Forests and grasslands served as the dominant carbon reservoirs, particularly in deeper soils, with a total carbon stock of 39.18 Pg C and 90.46 Pg C, respectively.
  • Climate was the most critical determinant of soil organic carbon density in both soil depth, although its explanatory power was relatively weaker for deep soil organic carbon density.
  • Human activities significantly reduced surface soil organic carbon density in forests and grasslands.
  • Future climate change would exacerbate the decline of surface soil organic carbon density, particularly in wetlands, posing substantial challenges to achieving the 4 per mil initiative goals.

Statistics:

  • National average soil organic carbon density in the 0-20 cm soil layer: 4.08 kg C m2
  • National average soil organic carbon density in the 0-100 cm layer: 9.42 kg C m2
  • Total carbon stock in forests and grasslands: 39.18 Pg C and 90.46 Pg C, respectively
  • Maximum soil organic carbon density in wetlands (in the 0-100 cm layer): 27.47 kg C m2
  • Explanatory power of climate on soil organic carbon density in deep soils: weaker ( approximately 30%)

Sources:

  • Zhang, S., Song, B., Wang, M., Zhang, L., Lu, Y., Guo, H., Guo, X., Zhang, Y., Zhou, X. (2025). Spatial distribution, drivers, and future variation of soil organic carbon in China's ecosystems: A meta-analysis and machine-learning assessment. Ecological Indicators, 179, 114255. doi: 10.1016/j.ecolind.2025.114255
  • NewsRx. (2025, October 21). University of Chinese Academy of Sciences Researchers Detail Research in Machine Learning. China Weekly News, 580.

Read more