Abstract This paper demonstrates the capability of the model, APSIM-Sugarcane to analyse system performance, by consideration of crop production and related issues for two contrasting sugarcane production systems in northern Australia. Daily output data from short-term model runs are used to demonstrate the capability of the model to represent the diverse management of sugarcane crops encountered in Australia. Output from longer term simulations is used to investigate a number of ‘real life’ sustainability issues, including an assessment of the production risk associated with season to season climate variability at each site, and the long-term implications of crop residue management on system performance. The paper also develops guidelines for methodological issues in long-term simulations of sugarcane production systems, relating to (1) the length of the climate record necessary to capture the ‘true’ season to season variability, (2) the need or otherwise to simulate each crop class (i.e. plant and ratoon crops) over the full climate record, and (3) the impact of initial soil conditions on the time for the system to reach a state of equilibrium. Results from statistical tests comparing pairs of distributions for various simulation run lengths, identified significant differences for run lengths up to 40 years. This suggests that, while adequate for most scenarios, the current practice of basing sugarcane simulations on 40 years’ climate data will not always capture the true season to season variability in response to climate. Results showed that variability associated with season to season climate by management interactions can be adequately captured in a single model run comprised of consecutive crop cycles, without the need to represent every crop class in every year. These findings have the potential to significantly improve modelling efficiency by reducing model run times, simplifying the analysis of model output and, reducing the required computer storage space. Results from long-term simulations, for some model output variables, indicate there is likely to be an initial period of ‘settling in’ of between 5 and 10 years as the model approaches a state of equilibrium, during which the response of certain model variables will be ‘abnormal’ and should be used with caution or discarded.
[1]
James H. Torrie,et al.
Principles and procedures of statistics: a biometrical approach (2nd ed)
,
1980
.
[2]
Graeme L. Hammer,et al.
APSIM: a novel software system for model development, model testing and simulation in agricultural systems research
,
1996
.
[3]
S. B. Milligan.
Intensive Sugarcane Production: Meeting the Challenges Beyond 2000. Proceedings of the Sugar 2000 Symposium, Brisbane, Australia, 20–23 August 199
,
1998
.
[4]
R. Dalal,et al.
APSIM's water and nitrogen modules and simulation of the dynamics of water and nitrogen in fallow systems
,
1998
.
[5]
R. C. Muchow,et al.
Modelling sugarcane production systems I. Development and performance of the sugarcane module
,
1999
.