Environmental Engineering Reference
In-Depth Information
another component requires the same variable as an input, the link can be considered
correct if a check of the variable attributes show that these are identical, whereas the
correctness of the variable as an input must be investigated within the component
producing the output. The principle of applying “parsimony” is of course still valid
in model building. For instance, there is no point in coupling two components in which
strong assumptions (and thus the limitations) of one impose an unnecessary burden
on the modelling capabilities of the other. This, however, applies both to monolithic
and component-based system development. As always, the choice of model should be
conditioned by both the intended application of the model and a comprehensive system
analysis, and this is totally independent of the type of implementation.
The SEAMLESS project has developed a framework to integrate analyses of
impacts on a wide range of aspects of sustainability and multi-functionality (Van
Ittersum et al.
2008)
. This requires the evaluation of the agricultural outputs and
system externalities for a wide range of production systems and environments.
Although some indicators of system performance can be provided using static models
derived from existing databases, estimating system behaviour for novel techniques
or existing techniques applied to new environments requires process based simulation.
Also, even for known systems, some of the externalities due to agricultural production
are only available as observational data for a very small number of experimental
sites. The analysis of the biophysical components of agricultural systems thus
requires a simulation framework which can be extended and updated by research
teams, which allows easy incorporation of research results into operational tools,
and which is transparent with respect to its contents and its functionality. The problems
and requirements outlined in the previous paragraph have formed the basis of the
design of the Agricultural Production and Externalities Simulator (APES) which
offers flexibility in being an open modelling environment that allows an extensible set
of modelling choices. The emphasis in APES has been to provide a transparent and
flexible modelling platform that can be adapted to different modelling goals. This
is a quite different rationale from the specific biophysical modelling solutions that
are currently implemented.
APES: The Agricultural Production and Externalities Simulator
APES is a simulation model system for estimating the biophysical behaviour of
agricultural production systems in response to the interaction of weather, soil and
agro-technical management options. The system allows the incorporation, at a later
date, of other modules which might be needed to simulate processes not included
in the existing version, such as the impact of plant pests (see also Fig.
4.1
).
Biophysical processes are simulated in APES using deterministic approaches
which are mostly based on mechanistic representations. The criteria to select
modelling approaches were the need to:
•
Account for specific processes to simulate soil-land use interactions
•
Input data to run simulations, which may be a constraint at the European scale
