Modelling Soil Organic Matter Dynamics – Global Challenges - Sustainable Management of Soil Organic Matter - page 71

Fig. 2.4.1. Comparison of measured (symbols; bars represent ± 1 SD about the measured
values) and predicted (lines) decomposition of sugarcane residue following harvest of a crop.
The solid lines are the predictions using the parameter values optimized on the data from the two
experiments shown in (a) ( z - Mackay Late, r - Mackay Early), while independent predictions for
the third experiment (Harwood Early) are shown in (b). The dashed lines are the predictions using
the default parameter values (Probert et al., 1998b).
by the high C : N of the sugarcane residue was implicit in the optimal value
of r max , the optimal value is not directly comparable with the default. How-
ever, the value of F C:N (from Equation 3a) was ~0.5 at both the Mackay
experiments, implying an r max value of ~0.014 without the C : N limitation,
a value that is still substantially lower than the default. Snow et al .
(1998) also had to reduce the default value of r max (to 0.025) to simulate
the decomposition of litter in a Eucalyptus plantation accurately. These
experiences suggest that the value of r max and other parameters in Equation
2 are not yet generic across the wide range of environments (e.g. semiarid
temperate areas to the humid tropics) and agricultural systems (cereals,
legumes, sugarcane and forests; both rain-fed and irrigated) where APSIM
is being applied.
Impact of different decomposition rates on cropping system
simulations
In simulation of the Mackay Early experiment, the different residue
decomposition rates produced by the different parameterization schemes
substantially affected simulated sugarcane yield (Fig. 2.4.2a). In all but
the first crop, the simulation with the optimized parameter scheme gave
higher yields, closer to those measured in the experiment. The effect of