Data-driven modeling on the global annual soil nitrous oxide emissions: Spatial pattern and attributes.

Area of Science:

• Environmental Science
• Soil Science
• Atmospheric Chemistry

Background:

• Previous global terrestrial soil nitrous oxide (N2O) emission estimates were divergent and poorly matched observed data.
• Accurate N2O emission mapping is crucial for understanding its role in climate change and developing mitigation strategies.

Purpose of the Study:

• To generate a data-driven global map of terrestrial soil N2O emissions using extensive field observations.
• To identify key environmental variables influencing global N2O emission variations.
• To compare data-driven estimates with previous model-based assessments.

Main Methods:

• Utilized a dataset of 5549 global field observations of soil N2O emissions.
• Employed the random forest approach to create spatially explicit maps of annual soil N2O emission rates.
• Quantified the predictive importance of various factors including soil nitrogen substrates (nitrate, ammonium, fertilizer) and climatic variables (temperature, precipitation).

Main Results:

• The global mean soil N2O emission rate was estimated at 0.059 ± 0.006 g N m⁻² year⁻¹, lower than previous model ensemble estimates.
• Soil N2O emissions were found to be higher in the Northern Hemisphere compared to the Southern Hemisphere.
• Cropland soils exhibited higher average annual N2O emission rates (0.094 ± 0.009 g N m⁻² year⁻¹) than forest (0.039 ± 0.004 g N m⁻² year⁻¹) and grassland (0.045 ± 0.007 g N m⁻² year⁻¹).
• Soil nitrogen substrates (nitrate, ammonium, fertilizer) were more influential in predicting N2O emissions than mean annual temperature and precipitation.

Conclusions:

• Data-driven models based on extensive field observations provide more reliable estimates of global soil N2O emissions.
• Previous process-based models may overestimate N2O emissions due to limited validation data and simplified assumptions.
• Global field observations are essential for constraining and improving process-based models for accurate N2O emission predictions.