Modelling Soil Organic Matter Dynamics – Global Challenges - Sustainable Management of Soil Organic Matter - page 70

Parameterization of APSIM for Systems Simulations
The systems simulations were performed on the residue retention
treatments of the Mackay Late and Mackay Early experiments. The experi-
ments commenced with planting in July 1992 for the Early experiment and
November 1992 for the Late experiment. The sites were managed similarly,
differing only in the harvest date. Crops were grown for 12 months then
harvested and allowed to ratoon (i.e. regrow), four and five times for the
Early and Late experiments, respectively. Information on fertilization and
tillage practices was used to specify crop management in the model. Model
parameters were based, wherever possible, on measured data. The soil
profile was sampled to 1.5 m depth in 1997, following the residue
decomposition measurements, and soil total C and N, mineral N and bulk
density measured to initialize N and C in APSIM-SoilN. Measurements of
microbial biomass (Robertson and Thorburn, Chapter 3.1) were used to
set the biom pool size. All other pools were set equal to values used in
previous APSIM simulations of soil N dynamics in sugarcane residue
retention experiments (Thorburn et al ., 1999). The soil water parameters
(saturation, dul and ll15 ) were estimated from the daily soil moisture data.
Default parameter values (Keating et al ., 1999) for the sugarcane variety
grown in the experiments (Q124) were used.
Results and Discussion
Optimization of decomposition parameters
The optimal APSIM-Residue parameter values differed from the default
values (Table 2.4.1) and gave predictions of residue mass much closer to
measured values (Fig. 2.4.1a) with consequently lower D values (Table
2.4.1). When the optimal values were used to predict residue mass at the
third experiment, Harwood Early, predictions of residue mass were also
much closer to measured values than those resultant from the default
parameters (Fig. 2.4.1b).
The optimal value of E os, max was lower than the default (Table 2.4.1),
indicating a greater moisture limitation to decomposition, whereas
the value of F contact, min indicated no residue-soil contact limitation. This
latter difference was unexpected as initial residue masses were large
(10-20 t ha −1 ) in the field experiments, a condition under which residue-
soil contact should be a limit to decomposition. It may be that a more
appropriate function is required to describe the contact limitation properly.
Another substantial change in the parameter values was the value of
r max , which was reduced from the default value of 0.1 to 0.007 in the
optimization (Table 2.4.1). Since the limitation to decomposition imposed