What was done
Soil carbon contents were measured at 2179 locations scattered across England and Wales between 1994 and 2003 adjacent to points where similar measurements had been made between 1978 and 1983, after which rates of change of soil carbon content were calculated for the quarter-century period 1978-2003.

What was learned
The authors found that "carbon was lost from soils across England and Wales over the survey period at a mean rate of 0.6% yr^-1," which phenomenon they attributed to "climate change," noting that over the period of study "the mean temperature across England and Wales increased by about 0.5°C."

What it means
Bellamy et al. conclude that "losses of soil carbon in the UK, and by inference in other temperate regions, are likely to have been offsetting absorption by terrestrial sinks," while Associated Press writer Michael McDonough leads off his review of the study (7 Sep 2005) by stating that "rising temperatures resulting from climate change are likely causing soil in England and Wales to lose large amounts of carbon, possibly further contributing to the greenhouse gas effect."  These conclusions, however, are not as well supported as they are portrayed to be.

For starters, Bellamy et al. only resampled 38.5% of the original sites that were sampled between 1978 and 1983, so that the vast majority of England and Wales was not assessed for changes in soil organic carbon (SOC) content, much less "other temperate regions" of the globe.  In addition, they claim that "the relationship between rate of carbon loss and carbon content is irrespective of land use," which was key to their being able to claim "a link to climate change."  However, in an accompanying article that raises several other concerns, Schulze and Freibauer (2005) say that in "re-inspecting the results, we think that the land-use factor has played a role - for example, only alteration in land use and gradual changes in land management can explain why croplands lost more carbon than other areas."  Also in this regard, they say that studies conducted in China, Finland and Flanders "attribute most of the SOC loss to changes in land use and management."  In fact, even Bellamy et al. admit that "various changes in land use will have contributed to carbon losses from soils across England and Wales over the survey period, both under agricultural uses (drainage schemes, post-war grassland conversion, increased stocking rates) and non-agricultural uses (afforestation on wet soils, increased erosion, increased burning of upland vegetation)."  However, they say they "do not have sufficient data at the scale of the National Soil Inventory to explore these effects," so they really do not know the role played by land use, which means they really do not know the role played by climate change.

Another complaint of Schulze and Friebauer is that the SOC losses observed by Bellamy et al. "occurred independently of soil properties, challenging our knowledge about SOC stability," as this observation is at odds with what we have learned about the subject over the years.  They also note that the carbon losses were proportional to SOC concentration, which implies "a first-order decay of a homogeneous pool" that "contradicts the view that SOC in carbon-rich soils contains a higher fraction of stable carbon than does that in carbon-poor soils."

Last of all, Schulze and Friebauer note that SOC contents may have changed in deeper soil layers than the top 15-cm layer measured by Bellamy et al., possibly in compensating ways; and they are firm in their opinion that "increased temperature alone seems to be too weak a driver" to have caused the observed changes in SOC.  We agree, feeling that Bellamy et al. have merely "scratched the surface" of the topic in a way that has yet to reveal its true nature.

Reference
Schulze, E.D. and Freibauer, A.  2005.  Carbon unlocked from soils.  Nature 437: 205-206.