Mysteries of the Global Carbon Cycle

Editors’ Vox is a blog from AGU’s Publications Department.

Natural processes cycle carbon between the atmosphere, ocean, and land. This was a finely balanced system until human activities began to increase the atmospheric carbon dioxide concentration to unprecedented levels, disrupting this balance and the ability of natural systems to respond. A recent article published in Reviews of Geophysics explores how well we understand the present-day land-ocean-atmosphere carbon cycles and their interactions. We asked some of the authors to describe how carbon cycle dynamics are observed and measured, our current understanding, and what further work is needed.

In very simple terms, what is the carbon cycle and why does it matter? 

Carbon is exchanged among the Earth’s land, ocean, and atmospheric reservoirs through processes that operate at multiple timescales. These processes form the basis for all life on Earth and are critical to the stability of the Earth’s climate and habitability. Fossil fuel combustion, land use change, and other human activities are now playing an increasingly important role in the carbon cycle; adding about 40 billion tonnes of carbon dioxide (CO₂) to the atmosphere each year. These emissions are increasing the atmospheric CO₂ concentrations by about 0.5% each year, affecting the radiative balance and hence the climate of the Earth. 

How do natural processes regulate anthropogenic CO₂ emissions? 

Natural processes, such as photosynthesis by plants on land and in the ocean, remove CO₂ from the air, while respiration by plants, soils, and ocean microbiota emits CO₂ to the atmosphere. The oceans can also exchange CO₂ with the atmosphere as it dissolves in seawater or comes out of solution. In pre-industrial times when the concentration of CO₂ in the atmosphere was stable, these natural processes were roughly in balance, absorbing as much CO₂ as they emitted to the atmosphere on annual to decadal timescales. Since the beginning of the industrial age in the late eighteenth century, these natural reservoirs have continued exchanging carbon and have become net sinks of CO₂, absorbing as much CO₂ as they emit along with roughly half of the CO₂ emitted by human activities. The other half of anthropogenic CO₂ emissions that remain in the atmosphere are driving the observed climate change.  

What changes have been observed in the fraction of atmospheric CO₂ over the past few decades? 

Over the past 60 years, anthropogenic CO₂ emissions have increased by a factor of four from less than 10 to more than 40 billion tonnes of CO₂ per year. Over this period, the natural land and ocean carbon sinks have increased proportionally, maintaining a near constant airborne fraction, near 45%, when averaged over decadal timescales.

On year-to-year timescales, the airborne fraction can change substantially as the land biosphere and oceans respond to climate fluctuations, such as strong El Niños or La Niñas, or large volcanic eruptions that produce widespread cooling by injecting sulfate aerosols into the stratosphere. In general, strong El Niños, like the ones in 1997/98 and 2015/16 are associated with higher airborne fractions while volcanic aerosol injections are associated with reduced airborne fractions.

What are some recent advances in measurement techniques that enable a better understanding of carbon cycle dynamics? 

Over the past two decades, space-based remote sensing observations have provided dramatic improvements in spatial resolution and coverage of the land and atmospheric components of the carbon cycle. High resolution imaging observations of the land biosphere have provided new insights into land use and land use change and disturbances. Space-based microwave and lidar observations are now used to quantify above-ground biomass carbon stocks. These data are being combined with surface plot-based in situ studies of carbon stocks as well as data from flux towers to understand the processes controlling gross primary production, respiration, and other key drivers of the carbon cycle.

Spatially resolved observations of atmospheric CO₂ from ground-based, airborne, and space-based platforms are being analyzed with atmospheric inverse models. These analyses help us to understand both the natural and anthropogenic processes adding and removing CO₂ from the atmosphere on scales ranging from individual power plants, to basin-scale ocean regions, to the globe. These data are being combined with improved bottom-up estimates of anthropogenic emissions and models of the land biosphere and ocean to diagnose the present-day processes maintaining the airborne fraction, and predict how they might evolve in response to ongoing human activities and climate change.

For the ocean, scientists have built open access databases of surface and interior ocean measurements. With machine learning and other statistical approaches, they are improving quantification of the ocean carbon sink and its temporal variability. The fact that independent observations and numerical models indicate comparable magnitudes of the multi-year mean sink indicates a robust understanding of the ocean sink. Ongoing efforts aim to better quantify inter-annual to decadal timescale variability in the ocean carbon sink, and to understand how physical and biological feedbacks will modify future ocean sink. 

What are bottom-up and top-down methods and what do they tell us? 

Bottom-up methods compile inventories of CO₂ emissions and removals by estimating the contributions from all known sources and sinks, and then summing the results. Bottom-up methods like those recommended by the Intergovernmental Panel on Climate Change (IPCC) Taskforce on Inventories divide human activities into a series of specific sectors, including Energy, Industrial Processes and Products Use (IPPU), Agriculture, Forestry and Other Land User (AFOLU), and Waste. Each sector is then subdivided into categories. For example, the AFOLU sector includes land use categories for Forest, Cropland, Grassland, Wetlands, Settlements, and Other Land.  For each category of each sector, anthropogenic emissions and removals of CO₂ and other greenhouse gases can be measured directly, modeled or estimated by multiplying observed activity data by an assumed emission factor. This approach generally works best for known processes, such as fossil fuel use, where the activity data (i.e., number of liters of oil or tonnes of coal) is well quantified, and the emission factors (i.e., number of kg of CO₂ released per liter of oil) is well known.

Other bottom-up methods are based on “upscaling” of measurements from discrete, surface-based sensors, such as eddy covariance towers, deployed in forest or crop biomes. Estimates of emissions and removals of CO₂ can also be derived from computer simulations of the ocean circulation, biology and chemistry or of the vegetation photosynthesis, growth and mortality and soil decomposition on land. These are also considered to be bottom-up methods because they are based on the scaling up from local processes to regional or global scales.

Another way to track emissions and removals of CO₂ is to directly measure its concentration in the atmosphere at high spatial and temporal resolution and then analyze these data with atmospheric inverse modeling systems to generate top-down estimates of the net carbon fluxes between the surface and atmosphere. These top-down methods are generally not as source specific as bottom-up methods but complement those methods by providing an integrated constraint on the total net amount of CO₂ emitted into or removed from the atmosphere.

Why is it important to understand the processes that control the emissions and removals of CO₂? 

Human activities have increased the atmospheric CO₂ concentration by almost 50% since the beginning of the industrial age, from about 270 parts per million (ppm) in 1750 to 415 ppm today. CO₂ is an efficient greenhouse gas that now accounts for about 70% of the observed 1.1 °C of warming in the global temperatures.

While these changes in atmospheric CO₂ are large, they would have been much larger if natural sinks in the land biosphere and oceans had not absorbed over half the anthropogenic emissions, limiting the airborne fraction to about 0.45. In spite of their importance, the nature and location of these natural sinks in the land biosphere are still not well understood. The ocean sink is better constrained by a variety of data sources and numerical models, but the mechanisms responsible for recent changes from year-to-year remain insufficiently explained. Because of this, we cannot accurately predict how they might evolve in response to continuing human activities and climate change.

Moving forward, what are some of the measurement and modeling gaps that must be addressed to effectively monitor the carbon cycle?

Sustained and expanded measurements of carbon exchanges between the atmosphere, ocean, and land reservoirs are critical for tracking changes in these systems as we embark on efforts to monitor and control the atmospheric carbon dioxide buildup. For the land and atmospheric carbon cycles, expanded surface and airborne in situ measurement networks are needed across the tropics and at high latitudes to monitor the rapidly evolving humid tropical forests and arctic and boreal regions as they respond to climate change. While space-based remote sensing observations have improved the coverage of these areas, the sparsity of accurate in-situ data from these regions currently precludes efforts to validate or interpret their data. For the oceans, ship-based measurements of ocean carbon will continue to provide critical accuracy standards while expanded deployments of autonomous platforms are needed to improve the measurement coverage, resolution, and repeat frequency.

—David Crisp (davidcri@gmail.com; 0000-0002-4573-9998), California Institute of Technology, USA (retired); Han Dolman (0000-0003-0099-0457), Royal NIOZ, Netherlands; Toste Tanhua (0000-0002-0313-2557), GEOMAR Helmholtz Centre for Ocean Research Kiel, Germany; Galen A. McKinley (0000-0002-4072-9221), Columbia University and Lamont-Doherty Earth Observatory, USA;  Judith Hauck (0000-0003-4723-9652), Alfred-Wegener-Institut, Helmholtz-Zentrum für Polar und Meeresforschung, Germany; Ana Bastos (0000-0002-7368-7806) Max Planck Institute for Biogeochemistry, Jena, Germany; and Stephen Sitch (0000-0003-1821-8561) University of Exeter, UK.

Editor’s Note: It is the policy of AGU Publications to invite the authors of articles published in Reviews of Geophysics to write a summary for Eos Editors’ Vox.

Citation: Crisp, D., H. Dolman, T. Tanhua, G. A. McKinley, J. Hauck, A. Bastos, and S. Sitch (2022), Mysteries of the global carbon cycle, Eos, 103, https://doi.org/10.1029/2022EO225018. Published on 22 June 2022.

This article does not represent the opinion of AGU, Eos, or any of its affiliates. It is solely the opinion of the author(s).

Text © 2022. The authors. CC BY-NC-ND 3.0
Except where otherwise noted, images are subject to copyright. Any reuse without express permission from the copyright owner is prohibited.