carteronsoilcredits

Australian FArm Institute
Conference Papers

John Carter
Principal Scientist
Queensland Department of Natural Resources and Water

John Carter completed a Masters of Science at the University of
Queensland after spending several years working on mining rehabilitation
in central Queensland. In 1986, on the completion of his degree, he
joined the Queensland Department of Primary Industries in Brisbane.
During his research career John has worked on issues of woody weed
invasion, dynamics of tropical woodlands, green house gas inventory,
climate data and pasture production. During the last seven years John
was a program leader in the CRC for Greenhouse Accounting. He led
programs on soil carbon and bio-geochemistry of the carbon cycle.
John currently runs the AussieGRASS pasture model that provides
pasture production forecasts and drought analysis for the continent and
maintains research activities on soil carbon in rangelands.
Current and future systems developed for
trading carbon at an international and
national level are likely to impose a degree of
uniformity in the rules applied to accounting
at state and project scales. Accounting under
tightly-defi ned rules makes it more diffi cult for
individuals to participate in trading, but also
better defi nes the commodity, makes carbon
more marketable and tends to increase the value
of that carbon.
Rules for trading carbon in Australia are likely
to be based on International Panel on Climate
Change (IPCC), Kyoto and International
Standard (ISO 14064) accounting constructs
(Henry et al. 2005). This may place some
important constraints on the ability to profi tably
trade soil carbon at the project scale or in
state-based schemes. In practice, the following
attributes of trading schemes make soil carbon
accounting a diffi cult task:
• inclusion of all gases, all pools
• gross-net or net-net accounting
• time periods for locking up carbon in
commercial contracts (up to 125 years)
• avoidance of leakage
• accounting for future risk and measurement
uncertainty
• need for auditing, certifi cation and
verifi cation.
In any trading scheme, the importance of all
gases and all pools, and a net-net accounting
construct, make trading in soil carbon
complex, as trading necessarily includes all
carbon stocks and greenhouse gas fl uxes on
an area of land. Under the net-net accounting
rules of the Kyoto Protocol for Article 3.4
activities, there is a requirement to increase
the rate of sequestration each accounting
period and not just the amount of carbon
stored, for each commitment period.
1 Beverley Henry, Queensland Department of Natural Resources and Water,
co-authored this paper.

24
Land Use Change, Land
Management and Soil
Carbon Stocks
Increasing soil carbon relies on increasing
inputs of carbon to the soil or reducing losses
of carbon from the soil pools. Reducing
losses is diffi cult as this is largely controlled
by prevailing temperature and moisture
conditions, as soil carbon is mainly lost by the
action of soil micro-organisms. It is diffi cult
to manipulate these independently of factors
controlling plant growth.
Increasing the input of organic matter to soils
(through better plant growth) is possible in some
systems, however, economics and increased
production of other greenhouse gases often
mitigate against the trends to net sequestration
of carbon and reduction of greenhouse gas
emissions. It is useful to categorise areas for
carbon accounting by land use change and land
management activities within a land use category
as outlined in Tables 1, 2 and 3.
Practical Issues
Soil carbon is more diffi cult and expensive
to measure and verify than carbon in tree
plantations. To assess carbon stocks, sampling
has to both measure carbon concentration
in the soil and the soil density. Many land
management options change soil bulk density
either mechanically or by changing the amount
of organic matter in the soil. Sampling to a
constant depth is not adequate and adjustments
need to be made to soil carbon estimates to
account for compaction or ‘fl uffi ng up’ of
surface soils (Gifford & Roderick 2003).
Accounting needs to be to a minimum depth of
30 centimetres and preferably to 1 metre.
These types of measurements require
expensive mechanical sampling and laboratory
measurements. Soil carbon is spatially variable,
even at quite fi ne scales. Different soils within
a project boundary would probably need to
be individually sampled and erosion and
deposition accounted for separately.
In both pasture and cropping systems variability
is at the scale of a few centimetres (grass
tussocks vs. compacted wheel tracks). High
spatial variability increases the amount of
sampling (and, hence, analysis costs) required
to precisely estimate soil carbon stocks. If soil
carbon change is small or slow, it may not be
possible to statistically differentiate between
real change and sample variability. The carbon
stored as a result of changed agricultural and
grazing activities is predominantly labile and
readily degradable, and so can be rapidly lost by
microbial decomposition.
In grasslands and croplands where fi ne roots
and recent particulate organic matter are a large
part of new carbon stocks, climate variability
effects may be of a similar magnitude to changes
in soil carbon due to management practices.
In the longer term, climate change may affect
soil carbon stores with the resulting balance
being determined by plant growth response
to increasing CO2 and impacts of changes in
rainfall and temperature.
26
There is a possibility of net loss of carbon.
The implication of this variability in time and
space is that a signifi cant amount of the carbon
increase would need to be retained as a buffer
against measurement uncertainty and future risk.
Summary
Accounting constraints around the net-net
construct result in a diffi cult barrier for trading
under long-term contracts for sequestered
carbon. Similarly, climate change and costs, and
uncertainty of measurement, are disincentives for
trading in soil carbon. Australia’s decision not
to ratify the Kyoto Protocol limits prospects for
international trading and the size and dynamics of
trading at the state and national levels.
Specifi c areas have greater certainty of
large gains and less risk and might be likely
candidates for trading – these are soil carbon
build up where: old crop lands are converted
to forestry; severely degraded areas are
revegetated; and there is continued application of
bio-solids that would have otherwise been burnt
without use for bioenergy.
Many efforts to build carbon in agricultural
systems may not generate a return from carbon
trading. However, such efforts are likely to
generate other environmental and productivity
benefi ts such as improved infi ltration rates,
reduced erosion, increased nutrient availability
and more sustainable production. There may also
be non soil carbon greenhouse benefi ts.
References
Eckard, RJ, Chen, D, White, RE & Chapman,
DF 2003, ‘Gaseous nitrogen loss from temperate
perennial grass and clover dairy pastures in
south-eastern Australia’, Australian Journal of
Agricultural Research, vol. 54, no. 6, pp. 561–70.
Gifford, RM & Roderick, ML 2003, ‘Soil
carbon stocks and bulk density: Spatial or
cumulative mass coordinates as a basis of
expression’, Global Change Biology, vol. 9,
pp. 1507–14.
Guo L-B, Halliday MJ, Siakimotu SJM &
Gifford, RM 2005, ‘Fine root production and
litter input: Its effects on soil carbon’, Plant and
Soil, vol. 272, pp. 1–10.
Henry, B, Mitchell, C, Cowie, A, Woldring, O &
Carter, J 2005, ‘A regional interpretation of rules
and good practice for greenhouse accounting:
northern Australian savanna systems’, Australian
Journal of Botany, vol. 53, no. 7, pp. 589–605.
Wang, WJ, Dalal, RC & Mitchell, C 2005,
‘Importance of N-related greenhouse gas
emissions in a cropping system under contrasting
farming practices’, Third International Nitrogen
Conference, Science Press USA Inc, Monmouth
Junction, US, pp. 753–59.
Acknowledgments
Much of the information in this paper was drawn
from work by the Cooperative Research Centre
(CRC) for Greenhouse Accounting.
Various tools are available to landholders
to investigate the greenhouse profi les
of their enterprises. Specifi c calculators
exist for grains, dairy, sheep and cotton:
http://www.greenhouse.crc.org.au/tools/
calculators/ and http://www.isr.qut.edu.au/
tools/index.jsp The Australian Greenhouse
Offi ce has developed a carbon accounting
toolbox for use as a guide for carbon
accounting at the project scale.

Table 1: Land use change options.

Crop to pasture
The transition of cropping lands to pasture can increase soil carbon stocks. There is also likely to be an
increase in methane emissions from grazing, which may be offset to some degree by a reduction in nitrous
oxide emissions if crops were fertilised (1 kg of methane is equivalent to 21 kg of CO2).

Crop to forest

This transition is likely to provide signifi cant capacity for carbon storage especially on areas cultivated
for more than 30 years. However, the transition may not be economically viable compared with continued
cropping. Change from annual crops to fruit or nut production could be more economically viable than
timber production.

Pasture to forest

Some small gains in soil carbon may be possible if grazing land is converted to Eucalypt forest. Conversion
of improved pastures to pine forest has been shown to result in a loss of soil carbon (Guo et al. 2005). In both
cases, signifi cant carbon would be stored in timber and other environment advantages might accrue.

Degraded land

Normal restoration processes are likely to result in increased amounts of soil carbon in degraded sites.
However, reclamation areas tend to be small and measurement would be expensive relative to the value
of sequestered carbon.

Table 2: Land management actions: cropping.

Crop (general)

In all circumstances, any net-net accounting construct would make economic gains unreliable. All
biological systems eventually ‘fi ll up’ with carbon and the rate of sequestration slows as systems
approach equilibrium. The main option with net-net accounting would be to increase the area over
which new management activity is started each year.

Reduced or zero till

The data suggest that there is little gain in soil carbon under Australian conditions, since extra plant
production is compensated for by increased microbial degradation of soil carbon. However, zero till
might help reduce the ongoing carbon losses associated with conventional tillage. When combined
with adequate fertiliser application and stubble retention, small gains in soil carbon are possible – 2.3
tonnes of carbon per hectare over 33 years measured on the Darling Downs.

While changes in soil carbon are minimal, there are good ‘whole enterprise’ greenhouse gains through
reduced use of fossil fuels (diesel and in manufacture of machinery), and advantages in increased yields
and reduced erosion.

Irrigation

Irrigation is a method of increasing plant production, however, in Australia the future for expanding the
area of crop and pasture under irrigation is small. Costs of water are likely to increasingly decrease the
economic viability of irrigation.

Fertiliser

Nitrogen and, to a lesser extent, phosphorus fertiliser can improve pasture and crop growth, profi tability
and soil carbon. There has been a signifi cant trend to increased fertiliser use in Australia over the last 30
years. The increasing rates of nitrogen fertiliser application will increase the production of nitrous oxide
powerful greenhouse gas (Wang et al. 2005) – 1 kilogram (kg) of N2O produces the same greenhouse
impact as 310 kg of CO2, so 85 kg of soil carbon would need to be stored permanently to compensate
for each kilogram of N2O emitted. With fertiliser response, it is likely that carbon gains will eventually
approach zero, but emissions related to N2O will be ongoing, creating a debit at some point in the future.
Another downside of increasing nitrogen fertiliser use is a signifi cant energy cost in the production of
fertiliser, which further offsets carbon storage.

Organic matter addition

Applying organic matter (manure etc) to soils can build soil carbon and sometimes supply other nutrients.
Improvements in soil chemistry, such as cation exchange capacity, are also possible. The test for these
activities is that the change in soil carbon must be more than just a transfer of carbon from one location to
another. For example, if the residue would otherwise be burnt, there could be signifi cant opportunities to
build soil carbon without risk of leakage.
Reducing stubble burning

Burning of crop stubble was once a common procedure, but is now relatively rare. Burning of stubble
signifi cantly reduces carbon input to the soil and also results in emissions of non-CO2 greenhouse gases.
Reduction of burning should increase soil carbon, especially when combined with zero till and fertiliser
application, however, a project would probably have to demonstrate a change from a practice of regular
stubble burning to no or reduced burning after 1990.


Table 3: Land management actions: pasture.

Conversion from native to improved pasture, and use of nitrogen fertiliser

Improved pastures are those with a regular input of fertiliser, addition of a legume or regular tillage.
Nitrogen fertiliser additions will increase soil carbon relative to non-fertilised pastures. However, as
with cropping systems, the greenhouse costs associated with extra nitrogen delivered through fertiliser
or legumes must be accounted for (Eckard et al. 2003). Some additional downsides may also be of
concern, eg soil acidifi cation; emissions of methane from additional livestock introduced to take
advantage of extra pasture production.

Modifi cation of stocking rates

Reducing stocking rates may, in some instances, increase soil carbon. In very degraded systems, it may
be possible to increase soil carbon by controlling grazing pressure of domestic stock, feral animals and
macropods. Total de-stocking in moderately stocked systems may not increase soil carbon as grasses that
do not have a degree of disturbance (fi re or grazing) may be less productive. Increased pasture biomass
also increases the risk of higher fi re frequency and affects woody biomass stocks. Reducing the number
of domestic livestock may reduce methane emissions. Verifi cation and certifi cation of net reductions in
stocking rates could potentially be diffi cult as there is a need to demonstrate that the reduction of stock on
the project area has not led to increases in stock numbers in other areas (leakages). The National Livestock
Identifi cation System (NLIS) might be adapted to such applications, but has yet to be proven.