Global Peatland CO2 and CH4 Flux Data with Water-Table Manipulation Experiments

Yuanyuan Huang's dataset supports research on the tradeoff of CO2 and CH4 emissions from global peatlands under water-table drawdown. It includes flux measurements and site characteristics from manipulation experiments, along with scripts for data pairing, analysis, and global future predictions using random forest models. The dataset provides gridded global predictions under RCP2.6 and RCP8.5 climate scenarios with 95% confidence intervals.

Use Cases

• Training machine learning models to predict future CO2 and CH4 emissions based on water table, climate, and soil data.
• Analyzing the relationship between water table depth and greenhouse gas flux across different climatic zones (Arctic, Boreal, Temperate, Tropical).
• Quantifying uncertainty in emission predictions using ensemble methods and bootstrap resampling described in the code.
• Comparing the global warming potential impact of CO2 versus CH4 emissions from peatlands under different climate scenarios.

Strengths

• Includes ensemble predictions with quantified uncertainty, generating 25,200 members for CO2 and 8,400 for CH4.
• Provides global-scale gridded predictions for two future climate scenarios (RCP2.6 and RCP8.5).
• Data integrates multiple information types: flux measurements, site characteristics, soil, climate, management, and topography.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Last update date is unknown; freshness unverified.

Provenance

Source

Yuanyuan Huang, via paperswithcode.

Collection Method

Collected from water table manipulation experiments listed in a reference document.

Time Range

Temporal coverage of the source experiments is not specified.

Freshness

Last updated date is unknown.

Geography

Global coverage, with site classifications for Arctic, Boreal, Temperate, and Tropical regions.