Carbon cycle confidence and uncertainty: Exploring variation among soil biogeochemical models

Area of Science:

• Soil Science
• Biogeochemistry
• Computational Modeling

Background:

• Soil organic matter (SOM) is crucial for soil health and global carbon cycling.
• Current soil biogeochemical models face uncertainties in predicting SOM stabilization and decomposition.
• Advancements are needed to evaluate and improve these models at larger scales.

Purpose of the Study:

• To isolate the impact of model structural uncertainty on global soil carbon stocks and turnover.
• To develop and utilize a soil biogeochemical testbed for comparative model evaluation.
• To identify key factors influencing soil organic matter dynamics.

Main Methods:

• Developed a soil biogeochemical testbed forcing three distinct models (CASA-CNP, MIMICS, CORPSE) with uniform climate and plant productivity data.
• Compared model outputs for initial soil carbon stocks, temperature-turnover relationships, and 20th-century carbon stock changes.
• Conducted single-forcing experiments to assess the impact of temperature, moisture, freeze-thaw, and soil texture.

Main Results:

• Models agreed on initial global soil carbon stocks (~1,400 Pg C) but diverged on temperature-turnover relationships and 20th-century carbon stock changes (±20 Pg C).
• Uncertainties from freeze-thaw processes and soil texture had a greater impact than direct temperature effects.
• Models projected distinct seasonal heterotrophic respiration rates, reflecting structural differences in environmental sensitivities and stabilization assumptions.

Conclusions:

• Model structural uncertainty significantly impacts global soil carbon stock and turnover predictions.
• Freeze-thaw and soil texture are critical factors influencing SOM dynamics, requiring careful model parameterization.
• The developed testbed provides a framework for understanding model uncertainties and advancing soil organic matter research.