Environmental Engineering Reference
In-Depth Information
simulation, namely to estimate production and system externalities in response to
detailed agricultural management applied in specific soil-weather combinations.
Modelling approaches selected and implemented in APES were mostly developed
at field scale. Simulation outputs at this scale have been used in the literature to
provide outputs at the regional scale by linking to Geographical Information
Systems holding information on the spatial distribution of soils and weather. In such
cases the most frequent recommendation is to use simulation outputs to make
relative comparisons between different agro-management options. Other options
are to use simulation outputs at the field scale as “cell” data to be integrated in
spatially explicit models, as in some catchment models. In this case, the increased
number of inputs needed generally limits the use of these models to case studies.
All uses at scales other than the field scale involve additional assumptions that may
be difficult to justify. Moving across scales is being addressed in SEAMLESS with
specific actions, but it is outside the modelling domain of APES.
The optimum temporal scale is still a matter of debate, as opinions differ about the
significance of possible drift in multi-year simulations without the re-initialization
of state variables. However, this use is both a given and implicit in the simulation
of multi-year crop rotations and is accepted in peer-reviewed publications on the
use of tools like APES. In any case, the issue is not about APES itself, but about all
model tools built with modelling approaches similar to APES.
As the simulation tool has been developed with a focus on modularity, APES
versions including different modelling engines and components (modelling solutions)
can be made available as “closed” modelling solutions to be used for situations
where the assumptions made by their developers (modellers) apply. A set of options
may be made accessible (e.g. to simulate reference evapotranspiration using
different approaches), but in order to protect system integrity, APES users will not
be able to access model composition (in their role of model users). However, APES
is an open system so that the same individual, with a different role, may access
model building, in this case taking the responsibility for the choices made. This is
the expected use beyond the end of the SEAMLESS project. Simulations can be run
using long series of either generated or observational weather data, to account for
the stochastic variability of weather. Outputs can be evaluated as means and deriving
measures of variability.
Inputs
Whenever alternative options are available to simulate a given process and such
options perform almost equally well, the less demanding model in terms of
parameters and inputs should be selected. However, data availability and quality
cannot be allowed to limit the implementation of model capabilities when it
prevents the achievement of the goals of SEAMLESS. APES releases minimize
data requirements and use options, such as pedo-transfer functions and weather
generators, to estimate missing variables and parameters from the available data.
Search WWH ::




Custom Search