TL;DR
This paper presents a global atmospheric model (NCAR CCM2) to simulate the seasonal cycle of atmospheric CO2, comparing its output to observational data from 28 sites. While the model generally captures the seasonal cycle in the Northern Hemisphere, it overpredicts its amplitude in the Southern Hemisphere, highlighting the need for more accurate ocean and biosphere flux representations.
What Problem Does This Paper Address?
Understanding the global carbon cycle, particularly the fate of anthropogenic CO2 emissions, is crucial for predicting future climate change. A key challenge is quantifying how and where roughly half of the released CO2 is absorbed by natural sinks (oceans and terrestrial biosphere). The interhemispheric gradient of atmospheric CO2 is a critical constraint for evaluating global carbon models, as most models tend to overpredict this gradient, suggesting a "missing" sink in the Northern Hemisphere. This paper aims to improve our understanding of the seasonal cycle of CO2, a significant component of its variability, by using a sophisticated atmospheric model and comparing its results to extensive observational data.
Methods
The study utilizes the NCAR Community Climate Model version 2 (CCM2), a global three-dimensional atmospheric model, adapted for CO2 transport. Key features of this model include high spatial resolution (2.8° x 2.8°), 18 vertical levels, and a 15-minute time step. The model incorporates explicit, nonlocal atmospheric boundary layer parameterization. Surface exchange of CO2 with the ocean, terrestrial biosphere, biomass burning, and fossil fuel emissions are included as source/sink terms. The researchers compare results using two different terrestrial biosphere models and two different ocean flux treatments (an annual mean and a monthly varying flux derived from an ocean general circulation model). They also employ a chemical transport model to assess CO2 production from the oxidation of CO.
Datasets
The primary dataset used for evaluation is observational atmospheric CO2 concentration data from 28 sites globally, collected by the National Oceanic and Atmospheric Administration/Climate Monitoring and Diagnostics Laboratory (NOAA/CMDL). This includes flask measurements and aircraft data. The study also uses monthly surface ocean pCO2 maps from the Max-Planck-Institute (MPI-H) ocean general circulation model for one of the ocean flux treatments.
Key Insights and Ideas
- Southern Hemisphere Overprediction: A significant finding is the model's tendency to overpredict the amplitude of the seasonal CO2 cycle in the Southern Hemisphere. This suggests that current representations of Southern Hemisphere sinks, particularly oceanic uptake, may not fully capture the observed seasonal variability.
- Ocean Seasonality Matters: Using a monthly varying ocean flux (from the MPI-H model) significantly improved the agreement with observational data in the Southern Hemisphere, attenuating the seasonal cycle amplitude and bringing it closer to observed values. This highlights the importance of seasonal dynamics in oceanic CO2 uptake.
- Terrestrial Biosphere Dominance: The terrestrial biosphere is identified as the primary driver of the seasonal cycle of atmospheric CO2, especially in the Northern Hemisphere. Differences between two terrestrial biosphere models led to appreciable variations in computed seasonal cycles, particularly at continental sites.
- Vertical Attenuation: The model successfully simulates the vertical attenuation of the seasonal CO2 cycle in the Northern Hemisphere, consistent with boundary layer/free troposphere interactions. However, an increasing amplitude aloft in the Southern Hemisphere suggests interhemispheric transport effects.
- High-Frequency Variability Captured: The model's high temporal resolution allows it to capture infrequent, real extrema in atmospheric CO2 concentrations, such as the "outliers" observed at Bermuda, which are linked to rapid advection of air parcels.
Key Results
- The model generally predicts the timing and magnitude of the seasonal CO2 cycle well for most Northern Hemisphere sites.
- The amplitude of the seasonal cycle in the Southern Hemisphere is consistently overpredicted by the initial model setup.
- Using a monthly varying ocean flux from the MPI-H model halved the seasonal cycle amplitude at three Southern Hemisphere stations, bringing them into better agreement with observations.
- The model's interhemispheric gradient of atmospheric CO2 is about a factor of two too high compared to observational data, with some longitudinal bands showing gradients up to 50% higher than the zonal mean.
- The global tropospheric production of CO2 from the CO + OH reaction is estimated at 0.7 GtC per year.
Limitations and Future Work
The authors acknowledge several limitations and suggest future research directions:
- Interhemispheric Gradient: The persistent overprediction of the interhemispheric gradient suggests a potential "missing" sink in the Northern Hemisphere or an issue with the representation of transport.
- Ocean Flux Uncertainty: While the monthly ocean flux improved results, further refinement is needed, potentially by incorporating more detailed observational data and ocean model predictions.
- Terrestrial Biosphere Models: Differences between terrestrial biosphere models highlight the need for further development and validation of these parameterizations, especially for continental regions.
- "Fertilization Effect": Future work will explore the impact of "fertilization" models, where elevated CO2 enhances terrestrial biosphere growth, on the computed interhemispheric gradient and atmospheric CO2 accumulation rate.
- Isotopic Tracers: A key future direction is the inclusion of stable and unstable isotopes of carbon and oxygen (e.g., 13CO2, 14CO2, C18O16O) to provide more robust validation of CO2 source/sink estimates.
- Synoptic Timescale Analysis: Further investigation into individual events and their impact on model residuals, particularly on synoptic timescales, is planned.
Summary updated