TL;DR
This study presents a comprehensive "bottom-up" global carbon budget, integrating inventory and satellite data across nine major land regions. The findings reveal a global land CO2 uptake of -2.2 ± 0.6 PgC yr⁻¹, consistent with "top-down" atmospheric estimates, and crucially, suggest that global soil heterotrophic respiration (SHR) is a smaller CO2 source (39 PgC yr⁻¹) than previously thought.
What Problem Does This Paper Address?
Accurately quantifying regional and global carbon budgets is fundamental for understanding the Earth's carbon cycle and informing climate change mitigation policies. However, significant discrepancies exist between different modeling approaches and observational data. Specifically, "bottom-up" estimates, which rely on inventories of carbon stock changes, often struggle to account for all carbon fluxes, particularly lateral transport (e.g., trade, riverine export). This makes it difficult to reconcile these estimates with "top-down" atmospheric measurements. Furthermore, the contribution of soil heterotrophic respiration (SHR) to the global carbon cycle, a major flux, has been a significant source of uncertainty, with previous estimates suggesting it accounts for a larger proportion of net primary productivity (NPP) than this study finds.
Methods
The authors developed a "bottom-up" approach to estimate Net Ecosystem Exchange (NEE) for nine large land regions. NEE was calculated as the sum of carbon stock changes (AC) and lateral carbon fluxes (Frivers and Ftrade).
- Carbon Stock Changes (AC): This involved compiling inventory estimates of changes in living and dead biomass, soil carbon, and wood/crop products. For regions lacking inventory data, satellite estimates of biomass changes were utilized.
- Lateral Carbon Fluxes:
- Ftrade: This accounted for the net flux of carbon from crop and wood products due to international trade.
- Frivers: This represented the flux of dissolved organic and inorganic carbon exported by rivers to the ocean, excluding carbon from weathered carbonate minerals.
- Soil Heterotrophic Respiration (SHR): SHR was inferred as a residual flux by subtracting Net Primary Productivity (NPP) and all other non-SHR carbon exchanges from the calculated NEE. This was a key novel aspect, allowing for a more direct estimation of SHR.
- Uncertainty Analysis: A Monte Carlo approach was employed to propagate uncertainties from individual flux estimates to the regional and global NEE and SHR values.
Datasets
The study utilized a variety of datasets:
- Inventory Estimates: Data from the REgional Carbon Cycle Assessment and Processes (RECCAP) project and other sources for carbon stock changes.
- Satellite Data: Remote-sensing data (Lidar and optical measurements) were used to estimate biomass changes in regions lacking inventory data.
- Trade Statistics: FAO data and GTAP-MRIO data were used to quantify carbon fluxes from crop and wood trade.
- Riverine Export Data: Data from the GLOBALNEWS model and other sources were used to estimate riverine carbon export.
- Net Primary Productivity (NPP) Data: Three satellite-based products (MODIS, GIMMS, BETHY/DLR) and one data-assimilation product (CARDAMOM) were used.
- Site-Level SHR Data: Data from a global database of site-level measurements were used for independent verification of SHR estimates through upscaling methods (random forest and climate-based semi-empirical models).
Key Insights and Ideas
- Reconciliation of Top-Down and Bottom-Up Budgets: The study successfully bridges the gap between top-down atmospheric measurements and bottom-up inventory-based estimates, providing a more robust global carbon budget.
- Importance of Lateral Fluxes: The research highlights that lateral fluxes (trade and riverine export) are crucial for accurately calculating regional carbon budgets and reconciling them with atmospheric observations. Ignoring these fluxes can lead to significant discrepancies.
- Reduced Global Soil Respiration: A major finding is that global soil heterotrophic respiration is a smaller CO2 source than previously estimated, implying that land carbon models may have overestimated the soil carbon sink's vulnerability to warming and CO2.
- Decoupling of NPP and Soil Carbon: The study emphasizes that soil carbon dynamics are not solely driven by NPP but are significantly influenced by lateral carbon fluxes, leading to a temporal and spatial decoupling.
Key Results
- Global Bottom-Up NEE: A global "bottom-up" NEE of -2.2 ± 0.6 PgC yr⁻¹ was calculated for the period 2000-2009, which is consistent with independent "top-down" atmospheric estimates of -2.4 ± 0.7 PgC yr⁻¹.
- Global Soil Heterotrophic Respiration (SHR): The estimated global median SHR is 39 PgC yr⁻¹ with an interquartile range of 33-46 PgC yr⁻¹. This is significantly lower than previous estimates of ~50 PgC yr⁻¹.
- SHR as a Fraction of NPP: The median ratio of SHR to NPP is 0.78, which is smaller than the IPCC assessment report's value of 0.9 and DGVM model estimates of 0.89.
- Regional Differences: Northern-hemisphere regions (Russia, North America, Europe, East Asia) tend to be larger CO2 sinks, contributing significantly to the global land sink. Tropical regions showed smaller sink strengths.
Limitations and Future Work
- Uncertainty in Tropical Soil Carbon: The study acknowledges that soil carbon changes in tropical regions were ignored due to data limitations, introducing a potential underestimate of 26% in tropical AC.
- Model Representation of Lateral Fluxes: The authors suggest that current land-carbon models often lack adequate representation of lateral fluxes and their responses to climate change and human pressure.
- Future Model Development: They recommend incorporating these lateral fluxes into future Earth System Models (ESMs) and using spatially explicit data for model evaluation.
- Improved Carbon Budgets: National, regional, and global carbon budgets for mitigation targets should account for lateral carbon flows and their impact on vertical carbon exchange.
- Further Investigation of Reservoirs: The development of reservoirs and their impact on carbon burial and outgassing were not fully considered, suggesting an area for future research.
Summary updated