Soil Carbon and Agricultural Productivity: Perspectives from Sub-Saharan Africa - Soil Carbon: Science, Management and Policy for Multiple Benefits - page 136

(Palm et al ., 2001). According to Palm et al .
(2001), N, polyphenol and lignin contents
are the major residue quality factors deter-
mining the decomposition rate of organic
materials and the subsequent nutrient re-
lease. These factors influence the manage-
ment options of organic materials. Organic
resources can thus be classified as of high,
low or intermediate quality based on the
proportions of the above parameters. High-
quality organic resources contain high pro-
portions of N and low proportions of lignin
and polyphenols. Materials with such char-
acteristics decompose fast and release nutri-
ents immediately for plant use, but contribute
less to the build-up of SOC. At the other ex-
treme, low-quality organic resources con-
tain low amounts of N but high proportions
of lignin and polyphenols. Such materials
are slow in decomposition and nutrient re-
lease, and hence can contribute more to the
build-up of the SOM pool. The contribu-
tions of these legumes to nutrient supply
and SOC depends on the quality of the
organics used. Legume residues have a low
C:N ratio and a low lignin and polyphenol
content, and are regarded as high quality
(Swift et al ., 1994). These therefore decom-
pose relatively fast, releasing nutrients to
the current crop. Their contribution to SOC
build-up may be low relative to other or-
ganic resources such as maize stover and
manure, which have a relatively high C:N
ratio. As a result, manipulation of resource
quality, particularly N and polyphenol con-
tent, or organic resources used as nutrient
sources is a potentially important way of
managing SOM (Mafongoya et al ., 1998;
Palm et al ., 2001; Kimetu et al ., 2004).

et al ., 2002). With respect to soil carbon se-
questration, it is most desirable to fix atmos-
pheric C in those pools having long turnover
times. Eswaran et al . (1995) defined four
pools based on carbon dynamics: first, an
'active or labile pool' of readily oxidizable
compounds. The formation of this pool is
dictated largely by plant residue inputs
(and hence management), and climate. Se-
cond, a 'slowly oxidized pool' associated
with soil macroaggregates. The dynamics
and size of this pool are affected by soil
physical properties such as mineralogy and
aggregation, as well as agronomic practices.
Third, a 'very slowly oxidized pool' associ-
ated with microaggregates, where the main
controlling factor is the water stability of ag-
gregates and agronomic practices have only
little effect. Fourth, a 'passive or recalcitrant
pool', where clay mineralogy is the main
controlling factor and there are probably no
effects due to agronomic practices. Indica-
tive residence times of the above pools are
termed as 'labile', 'moderate', 'slow' and
'passive', respectively ( Table 10.3 ) .
Changes in total SOC proportions in the
soil take a long time to manifest. Partitioning
SOC into the different pools using fraction-
ation procedures offers a better understand-
ing of the dynamics of soil carbon otherwise
masked by the native pool. Kapkiyai et al .
(1999), in a study in Kenya, observed that
organic matter and microbial biomass among
treatments were proportionately larger than
changes in total SOC. Similar observations
have been reported by other studies employ-
ing SOM fractionation procedures (Cambar-
della and Elliot, 1992; Christensen, 1992;
Woomer et al ., 1994).
The recent development of models has
improved understanding of the dynamics of
SOM and the nutrients associated with it.
Models can be used to provide a better under-
standing of decomposition and accumulation
processes and to predict future conditions
from previous experience of SOM. Soil car-
bon models such as the Rothamsted carbon
model (Jenkinson, 1990) and the CENTURY
model (Parton et al ., 1988) have been devel-
oped for temperate regions and validated
against data from long-term experiments
across the tropics. These models have been

Soil Organic Matter Pools

Soil organic matter consists of diverse frac-
tions ranging from young, biologically ac-
tive pools to more recalcitrant, 'passive'
pools (Schimel et al ., 1985). These fractions
and their dynamics are particularly useful
in understanding the link between SOM dy-
namics and nutrient availability (Motavalli
et al ., 1994). SOC pools have different turn-
over rates (Buyanovsky et al ., 1994; Six