Articles
THE RATIONALE FOR ADOPTING A MODULAR GENERIC STRUCTURE FOR CROP SIMULATORS
Article number
248_49
Pages
391 – 400
Language
Abstract
Crop simulators (mechanistic models) can be used to capture most of our knowledge of crop behaviour and transfer it to the grower’s holding for use in managing crops.
All the simulators in existence today have been built by individuals or small teams.
The simulators incorporate only a fraction of what the agricultural research community knows collectively; more participation by specialists is needed to improve the simulators.
However, the existing simulators are also monolithic and difficult to understand.
Specialists attempting to work on the models find it difficult to disentangle the part of interest to them so that they can work on that part alone.
By dividing the models into subunits or "modules", each dealing with related groups of processes, and adopting a modular generic structure for all crop models, we can overcome these problems.
Specialists can get involved in translating their knowledge into computer code.
It will be possible to test the individual modules and to interchange modules between crop models.
Using crop simulators created in this manner, the grower can have access to expertise from the whole agricultural research community for making management decisions.
All the simulators in existence today have been built by individuals or small teams.
The simulators incorporate only a fraction of what the agricultural research community knows collectively; more participation by specialists is needed to improve the simulators.
However, the existing simulators are also monolithic and difficult to understand.
Specialists attempting to work on the models find it difficult to disentangle the part of interest to them so that they can work on that part alone.
By dividing the models into subunits or "modules", each dealing with related groups of processes, and adopting a modular generic structure for all crop models, we can overcome these problems.
Specialists can get involved in translating their knowledge into computer code.
It will be possible to test the individual modules and to interchange modules between crop models.
Using crop simulators created in this manner, the grower can have access to expertise from the whole agricultural research community for making management decisions.
Authors
B. Acock, James F. Reynolds
Keywords
Online Articles (59)