Studies show that the warming of the climate system is altering the movement and storage of carbon in the far north of the Earth. And these changes carry global implications. Among the many questions that scientists such as myself are investigating is whether the Arctic will continue being a net absorber of carbon, or shift to become a net emitter.
The Earth’s carbon cycle – the movement and storage of carbon between the land, atmosphere and oceans – is a fundamental element of the climate system. Oceans are currently the Earth’s greatest carbon sink, meaning they absorb more carbon than they emit.
However, there is a seasonal rise and fall of emission rates which is largely attributed to the summer “green up” of northern ecosystems from vegetation growth during the short warm season, and CO2 emitted by plant respiration for growth and soil carbon decomposition. Anthropogenic emissions from, for example, people burning fossil fuels add to the land-to-atmosphere transfer.
Measurements of several key variables are the principal source of information on carbon cycle processes. Scientists monitor land emissions of carbon dioxide and methane using flux chambers, unmanned devices placed on the ground which make continuous measurements of gases released from soils at each location. However, field sites are sparse, particularly across remote parts of Eurasia. Satellites provide observations over broad regions, but must be calibrated with ground data in order to translate the remotely sensed data into meaningful physical quantities.
Measurement data are then incorporated or “assimilated” into computer models, which help to understand how the climate system operates and allow us to predict how the climate may change in the future.
An analysis of measurements and climate models from earlier this year showed that the models are more accurate in their simulations of current air temperatures than previously thought. This raises confidence in the model projections of future climate.
Yet, while they do well at estimating current temperatures, the computer models’ simulations of ecosystem processes that affect carbon emissions, such as growth of woody stems and leaves and breakdown of carbon in soils, are relatively simple. Scientists are looking to fine-tune those to improve our ability to forecast future changes.
Instrumenting the Arctic
In a synthesis study published in Biogeosciences earlier this year, other researchers and I from the Permafrost Carbon Network used estimates from nine land surface models, ground-based measurements and remote sensing data to investigate the land-to-atmosphere fluxes, or transfers, of CO2 across northern Eurasia.
We found that while the models tended to disagree on the amount of carbon accumulated in the Arctic each year, when averaged together they show that while the region’s carbon sink strengthened in the 1960s to 1990s period it has begun to weaken since the late 1990s. In other words, ecosystems throughout northern Eurasia are absorbing less carbon. We tried to find out why.
When we compared the computer model estimates with available field observations, we found that the models tend to overestimate carbon emissions from land, such as plant respiration during growth and soil carbon decomposition. This result supports earlier studies that suggested that models slightly overestimate respiration rates, or how quickly plants produce carbon dioxide. It also implies that the present-day sink is somewhat stronger than the models predict.
The Arctic has been a sink for atmospheric CO2 since the end of the last Ice Age and presently accounts for up to 25% of the Earth’s carbon sink. Our study shows, however, that the northern region’s status as a carbon sink is weakening due to growing emissions, in the form of both CO2 and methane, as frozen soils in the Arctic thaw.
Soils in areas of permafrost contain twice as much carbon as there is currently in the atmosphere. As the climate and permafrost soils have warmed, microbes have started to break down this organic carbon, which has been frozen and fixed in the permafrost. That has led to a rise in land emissions of CO2 and methane.
Another recent expert assessment by scientists from the Permafrost Carbon Network, published in Nature, concluded that as much as 5%-15% of the terrestrial permafrost carbon pool is vulnerable to release in the form of greenhouse gases during this century.
This is equivalent to between 130-160 petagrams (or billions of tons) of carbon, which is similar in magnitude to carbon losses from historical land use change, such as deforestation, but far less than fossil fuel emission rates.
Emissions of greenhouse gases also come from disturbances such as wildfire, forest dieback due to drought and logging, and other land use changes. It has been estimated that with continued warming, releases of carbon from microbial decomposition and other sources will overwhelm the capacity for plant carbon uptake in the Arctic, leading to net carbon emissions from permafrost ecosystems to the atmosphere, possibly by the middle of the 21st century.
If rising air temperatures were thought of as a car rolling down hill, a strengthening northern carbon sink would be analogous to the breaks being applied. A weakening sink is like easing up on those breaks. The Arctic switching from a sink to a source will be equivalent to switching from the break pedal to the accelerator. That is, warming increases will become more rapid, which will further increase emissions, accelerating climate change in a self-reinforcing warming cycle.
By one estimate, the Arctic switching from a carbon sink to a source would be strong enough to cancel 42%-88% of the total global land sink – that is, the absorption of CO2 into forests, soil and other sources on land.
Role of forests
In the future, forests and ecosystems will continue to play a major role in the carbon cycle of the Arctic. Atmospheric measurements made at NOAA’s Mauna Loa observatory show that amid the long-term increase of CO2 levels in the atmosphere, the concentrations of CO2 rise and fall each year by about two parts per million (ppm) as forests and ecosystems in the northern hemisphere draw in and release CO2.
Atmospheric CO2 concentrations, particularly over the northern hemisphere, tend to rise in autumn and fall in spring each year as a result of vegetation decay and growth. In a study published in Science, researchers used ground-based and aircraft measurements to document an increase in the amplitude of the seasonal cycle in atmospheric CO2 concentrations – that is, the seasonal ups and downs of CO2 flows – since 1958. The researchers suggested that large ecological changes in the northern forests could be causing the observed alterations in historical atmospheric CO2 concentrations.
This trend is akin to the Earth taking deeper breaths now as compared to decades past, which could be driven by changes in boreal and temperate forests.
For example, observations show that evergreen shrubs and trees are migrating northward in response to warming. Those species absorb and release more CO2 than the tundra vegetation they replace. Disturbances are also changing the Arctic landscape. A shift in age to younger forests that experience more vigorous seasonal carbon uptake was also identified in the Science study as a potential cause of the increased seasonal cycling of CO2.
All these measurements and models are revealing alterations to the Arctic carbon cycle, which threaten to accelerate warming. Going forward, research in Arctic regions will need to focus on understanding the vulnerability of the permafrost carbon pool, in particular. As society considers ways to reduce carbon emissions and avoid the most harmful effects of climate change, we must consider these future carbon emissions.