Australasian Science: Australia's authority on science since 1938

Myths about Carbon Storage in Soil

By Robert White and Brian Davidson

Goals of sequestering carbon in agricultural soil ignore the law of diminishing returns.

The idea that an increase in the carbon content of the world’s soils could substantially offset greenhouse gas emissions has been enthusiastically promoted by politicians and environmental groups. It arises because the amount of organic carbon in the world’s soils is impressive – some 1500 billion tonnes to a depth of 1 metre, which is about twice the amount of carbon in atmospheric carbon dioxide.

This carbon is stored in soil primarily as organic matter, comprising the residues of plant material, animal excreta and dead organisms. Although it’s also present in insoluble carbonates, these are much more stable than soil organic matter and not very amenable to human manipulation.

A prominent example of this idea is the “4 per 1000 Initiative” launched by the French Ministry of Agriculture at the COP21 meeting in Paris last December ( The thinking is that even a small annual increase of 0.4% in soil carbon, averaged over all soils, would not only improve soil fertility and agricultural production but would also help to limit the global temperature increase to 1.5–2°C, as advocated by the Intergovernmental Panel on Climate Change.

While this may be a laudable aspiration, the concept is flawed. It implies that soil carbon will increase by a slightly bigger increment each year as the amount of carbon in the soil increases, in an analogous way to the effect of compound interest on money in a bank.

As shown by the blue line in the figure, the effect can be approximated by an exponential model. However, we know from many field studies of soil carbon changes that, following the addition of extra organic residues, the soil carbon content rises asymptotically to a new steady-state maximum according to the “law of diminishing returns” (as illustrated by the orange line in the figure). Although the result of this behaviour is little different to the “4 per 1000” model over the first few decades, the end result after 100 years (when the change is deemed “permanent”) is very different.

A similar flaw occurs in the Australian government’s methodology for estimating carbon sequestration in soil as part of the Carbon Farming Initiative ( In this methodology, a linear increase in soil carbon is calculated according to prescribed land management actions and site location for a period as long as 100 years. Again, although the assumption of a linear increase in soil carbon is reasonable for 25 years or so, it does not hold for longer periods up to 100 years.

Although “sustainable intensification” (e.g. nutrient management, new irrigation, managing soil acidity and pasture renovation) and stubble retention (where plant residue is left on top of the soil after harvest) are desirable from an agronomic viewpoint, we have reported that this will not achieve much net carbon abatement that can be set against Australia’s greenhouse gas emissions ( The main reason for this is that at a carbon price of $12–14 per tonne of CO2-equivalent there is no financial incentive for a farmer to make the change, except in the case of liming acid soils and stubble retention.

With regard to stubble retention, national uptake is also limited because many cereal farmers in southern Australia are already retaining stubble and hence would be ineligible under the Carbon Farming Initiative. Furthermore, when stubble is retained, a new crop needs to be seeded by direct drilling (commonly called “no-till”). A recent review in Nature Climate Change of no-till globally has shown that the benefits of no-till for carbon accumulation, and hence carbon abatement, have been overstated (

We conclude that although actions to build up soil carbon as organic matter are desirable to improve soil fertility and agricultural productivity, sequestering carbon in agricultural soils will not provide a major offset for greenhouse gas emissions. A focus on these actions should not distract policy-makers from more direct methods of reducing emissions in agriculture and other industries.

Robert White is Emeritus Professor and Brian Davidson is Associate Professor in The University of Melbourne’s Department of Agriculture and Food Systems.