TL;DR
Northern permafrost regions are becoming stronger carbon sinks due to increased early-season plant growth, but this is partially offset by increased late-season respiration, especially in forested areas. Current climate models fail to capture this crucial seasonal compensation, potentially overestimating future carbon uptake in these rapidly warming regions.
What Problem Does This Paper Address?
The Arctic and sub-Arctic regions (Northern High Latitudes, NHL) are warming at an accelerated rate, leading to significant changes in their carbon cycle. While it's known that both plant photosynthesis (carbon uptake) and ecosystem respiration (carbon release) are increasing, there's considerable uncertainty about the net effect on the region's carbon balance. Specifically, scientists are trying to understand:
- How are trends in net CO2 uptake changing across the NHL? Are these changes uniform, or do they vary with factors like vegetation cover and permafrost extent?
- What are the underlying mechanisms driving these trends? Is it primarily changes in plant growth, or are respiration processes playing a more significant role?
- How well do current climate models capture these dynamics? Can they accurately predict future carbon exchange in these sensitive ecosystems?
This uncertainty is critical because the NHL holds vast amounts of carbon in its permafrost, and a shift from a carbon sink to a source could have significant global climate implications.
Methods
The researchers employed a multi-pronged approach to tackle these questions:
- Atmospheric CO2 Inversions (ACIs): They used an ensemble of six ACI products, which provide regional estimates of net ecosystem CO2 exchange by combining atmospheric CO2 concentration measurements with atmospheric transport models. This allowed them to assess trends over large areas from 1980 to 2017.
- Eddy Covariance (EC) Measurements: They utilized data from 48 EC sites across the NHL, offering direct, site-specific measurements of CO2 fluxes. This provided a ground-truth for ACI estimates and allowed for detailed analysis of seasonal dynamics.
- Dynamic Global Vegetation Models (DGVMs): They compared their observational findings with simulations from ten DGVMs (TRENDY intercomparison project). These models simulate carbon cycle processes based on physical and biological principles, offering a "bottom-up" perspective.
- Structural Equation Modeling (SEM): To disentangle the complex interactions between climate, vegetation, and CO2 fluxes, SEM was used. This statistical technique helps to identify direct and indirect causal relationships among variables.
- Productivity Datasets: They used satellite-derived Normalized Difference Vegetation Index (NDVI) and Gross Primary Production (GPP) data to assess trends in vegetation growth.
Datasets
The study utilized several key datasets:
- Atmospheric CO2 Inversions (ACIs): Ensemble mean from six products (Carbon-Tracker 2019B, Carbon-Tracker Europe 2020, CAMS, Jena CarboScope, JAMSTEC) covering 1980-2017.
- Eddy Covariance (EC) Measurements: Data from 48 sites across the NHL with at least three years of continuous measurements (1990-2017) from the FLUXNET2015 database.
- Dynamic Global Vegetation Models (DGVMs): Simulations from ten TRENDY v7 DGVMs (1980-2017).
- Productivity Data:
- Gross Primary Production (GPP) derived from a light use efficiency (LUE) approach (1982-2015).
- Normalized Difference Vegetation Index (NDVI) from GIMMS (1982-2015).
- Climate Data: Monthly gridded air temperatures (CRU TS3.10, 1980-2017).
- Vegetation and Permafrost Data: Percent tree cover (%TC) from AVHRR (1982-2016) and permafrost extent (%P) from ESA CCI Permafrost project (1997-2017).
- Soil Moisture Data: ESA CCI soil moisture product (SM, v4.5, 1980-2017).
- Satellite-driven Carbon Flux Estimates: SMAP L4C product (Nature Run version, 2000-present).
Key Insights and Ideas
- Late-season respiration is the unsung hero (or villain) of Arctic carbon balance: While early-season plant growth is increasing across the NHL, the real determinant of the net carbon sink strength is the respiration that occurs as the growing season ends.
- Forests amplify the seasonal swing: As tree cover increases, so does late-season respiration. This "seasonal compensation" effect means that while forests are strong carbon sinks in the early season, their increased respiration in the late season significantly dampens the overall annual net uptake compared to less vegetated permafrost areas.
- Permafrost regions are gaining carbon faster: Contrary to some concerns, the study found that permafrost regions are actually becoming stronger carbon sinks at a faster rate than non-permafrost, forested regions. This is primarily due to a stronger early-season uptake that isn't as heavily offset by late-season respiration as in forested areas.
- Models are missing a key piece of the puzzle: Current DGVMs struggle to accurately represent the increased respiration during the late growing season, leading them to underestimate the seasonal compensation effect and potentially overestimate future carbon uptake in the NHL.
Key Results
- Net CO2 uptake is increasing faster in permafrost regions: The rate of increase in net CO2 uptake was significantly higher in permafrost regions (0.58 ± 0.086 gC m⁻² yr⁻²) compared to non-permafrost regions (0.13 ± 0.11 gC m⁻² yr⁻²) between 1980 and 2017.
- Tree cover drives seasonal compensation: The trend of respiratory CO2 loss during the late-growing season increased significantly with increasing tree cover. This offset a larger fraction of photosynthetic CO2 uptake, leading to a slower rate of increasing annual net CO2 uptake in areas with higher tree cover.
- Early growing season uptake is similar across gradients: The increasing trends of net CO2 uptake in the early growing season were of similar magnitude across the tree cover gradient in the NHL.
- DGVMs underestimate seasonal compensation: DGVMs failed to simulate the observed increase in late-season CO2 release, leading to an underestimation of seasonal compensation compared to observational data.
Limitations and Future Work
- Model limitations: The study highlights a critical limitation of current DGVMs: their inability to accurately capture the respiration dynamics during the late growing season. This hinders their ability to project future carbon cycle changes in the NHL. The authors suggest that models need to better incorporate direct and lagged effects of temperature, moisture, and productivity on decomposition and soil organic carbon turnover.
- Coarse resolution of ACIs: While ACIs provide valuable regional estimates, their coarse resolution prevents the identification of specific ecosystem types contributing to the observed trends.
- Future research directions: The authors emphasize the need for improved DGVMs that can better simulate seasonal respiration. They also point out that while current evidence doesn't suggest a positive carbon-climate feedback (i.e., warming leading to a net release of carbon), thresholds may exist, and further monitoring is crucial as these ecosystems transition. Understanding the complex interactions between above- and belowground processes in climate-sensitive permafrost regions remains a key area for future research.
Summary updated