News & Views | Published:

Carbon cycle

A warm response by soils

Nature volume 464, pages 499500 (25 March 2010) | Download Citation

The flux of carbon from soils to the atmosphere has apparently increased with climate warming. But does this reflect a net loss of carbon to the atmosphere that could exacerbate climate change?

The world's soils contain twice as much carbon as the atmosphere1. As a result, small increases in organic carbon loss from soils could greatly enhance carbon dioxide concentrations in the atmosphere2, potentially creating a positive feedback on climate3. On page 579 of this issue4, Bond-Lamberty and Thomson add to our understanding of how soils might respond to climate change.

The climate is warming5, and it has long been assumed that, as soils warm, the rate of carbon flux to the atmosphere (through soil respiration; RS) will increase. However, this has been very difficult to confirm from measurements because RS is highly spatially variable, it cannot be measured by large-scale remote sensing, and the soil medium is not easily accessible4. Bond-Lamberty and Thomson present an analysis of a five-decade record of global RS chamber measurements, matched with high-resolution historical climate data and corrected for other factors, to show that, over the past 20 years, RS has increased with climate warming. From the results of their meta-analysis, the authors also estimate that RS is much larger than previously reported.

But these findings do not necessarily mean that soils are losing a greater proportion of their large carbon stores to the atmosphere. Increases in RS can also arise from the loss of fresh carbon inputs to the soil, driven by increased plant growth owing to climate change (at least in cooler regions; Fig. 1, overleaf). As Bond-Lamberty and Thomson point out4, their results may well represent an increase in the rate of carbon cycling, rather than a big shift between the global carbon pools in soils and the atmosphere.

Figure 1: Possible mechanisms driving increased soil respiration (RS) in a warming climate.
Figure 1

a, An increase in RS could occur through a rise in the decomposition rate of old soil organic carbon, leading to a net loss from the global pool of soil carbon to the atmosphere. b, If carbon inputs to the soil increase, higher RS could derive from enhanced release of fresh carbon. In this second case, there would be no net loss of carbon from the global soil pool. The findings of Bond-Lamberty and Thomson4 leave open which of these processes — if not a combination of them — is likely to dominate.

Assessing the balance between increased soil carbon inputs through greater plant growth due to climate warming, and increased carbon losses through higher decomposition rates, should be a research priority. There are several methodological challenges to doing so.

One is accurately determining the carbon input to the soil, especially below-ground inputs from fine-root turnover and exudation. Another is to detect a small change in the content of soil organic carbon (its state) within a reasonable time period. A third is measuring the components of carbon output, particularly in distinguishing between output due to increased respiration from plant roots and the immediate root environment, and output due to respiration from free-living microbes in the bulk soil. These have often been referred to as autotrophic and heterotrophic respiration (RA and RH, respectively), but the distinction between them is blurred because of the close association of plants and soil organisms. Bond-Lamberty and Thomson's findings show that total RS has increased, and suggest that both RA and RH have also risen. A barrier to reliably predicting the future of global fluxes of soil organic carbon to the atmosphere is the uncertainty associated with measuring all three components of soil carbon cycling (input–state–output).

Bond-Lamberty and Thomson's analysis4 also throws up a curiosity in northern soils. The authors fit a single global model to all of the RS data from 1989 to 2008, which gives a positive global relationship between RS and temperature. An unexpected result, however, is that the RS data from the boreal and Arctic regions, when examined in isolation, show a significant negative relationship with temperature, which is unexpected from previous studies6,7,8,9,10.

This finding adds an intriguing aspect to the debate about how northern soils will respond to climate change, in which the following factors have to be taken into account. The boreal and Arctic regions are projected to experience greater than average climate warming1; they contain some of the largest soil carbon stocks on Earth11; the permafrost experiences seasonal temperatures either side of 0 °C such that small levels of warming could have profound effects on biological activity and carbon fluxes9; and permafrost processes and highly organic soils such as peats are relatively poorly described in current coupled carbon-cycle–climate (C4) models12,13. For all these reasons, the response of northern soils to climate change is a fertile area for future research.

At the global level, however, Bond-Lamberty and Thomson's elegant analysis lends strong support to the hypothesis that soil carbon fluxes will increase in a warming climate. The outcomes of their meta-analysis should further drive efforts to develop methods to disaggregate the underlying processes contributing to RS. Without understanding these processes, it will not be possible to accurately predict the net response of soil carbon stores to climate change — and that is a central question for determining the biospheric feedbacks between the carbon cycle and climate.


  1. 1.

    Eur. J. Soil. Sci. 47, 151–163 (1996).

  2. 2.

    , , & Adv. Agron. 97, 1–43 (2008).

  3. 3.

    , , , & Nature 408, 184–187 (2000).

  4. 4.

    & Nature 464, 579–582 (2010).

  5. 5.

    PCC Working Group I Climate Change 2007: The Physical Science Basis (Cambridge Univ. Press, 2007).

  6. 6.

    et al. J. Ecol. 94, 740–753 (2006).

  7. 7.

    Glob. Planet. Change 53, 222–232 (2006).

  8. 8.

    & Soil Biol. Biochem. 39, 1689–1698 (2007).

  9. 9.

    , & Science 312, 1612–1613 (2006).

  10. 10.

    et al. Nature 460, 616–619 (2009).

  11. 11.

    Ecol. Appl. 1, 182–195 (1991).

  12. 12.

    et al. J. Clim. 19, 3337–3353 (2006).

  13. 13.

    et al. J. Ecol. 97, 851–863 (2009).

Download references

Author information


  1. Pete Smith is at the Institute of Biological and Environmental Sciences, School of Biological Sciences, University of Aberdeen, Aberdeen AB24 3UU, UK.

    • Pete Smith
  2. Changming Fang is at the Coastal Ecosystems Research Station of Yangtze River Estuary, Institute of Biodiversity Science, Fudan University, Shanghai 20043, China.

    • Changming Fang


  1. Search for Pete Smith in:

  2. Search for Changming Fang in:

About this article

Publication history



Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Newsletter Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing