Ecoinformatics and Soil-Plant Formations - New Ecoinformatics Tools in Environmental Science: Applications and Decision-making - page 707

capacity, fraction of photosynthetically active radiation absorbed by the green
portion of the vegetation canopy, canopy re ectance, transpiration, and other

important environmental characteristics. These data and the results of

fl

field exper-

iments allow to generate model experiments.
The GIMS-based method uses a priori information to start the model experiment.
Under this procedure, a set of algorithms operate over the whole data set and
determine the input parameters of the model used. The model structure and its
coef

fluctuation of the difference between the
experimental and model estimations of the EMW attenuation effect. Usually, the
central part of the model describes the fundamental processes within the SPF, such
as the radiation balance, water circulation (evapotranspiration, water uptake by
roots), photosynthesis, and mortality.
In a common case of environmental investigations, combined use of algorithms,
computer models and remote sensing technologies is possible within the hybrid
information-modeling system with the structure described by Soldatov (2011,
2012). This structure links experimental optical and microwave techniques with
computer modeling supporting the correlations between different stages of simu-
lation experiment.
The model-based analysis and optimization of computational experiments have
speci

cients are changed depending on the fl

c features that certainly demand the determination of new design points.
Unfortunately, in many scenarios of simulation experiments, the objective function,
e.g., the response of survivability indicator to the impacts on the environmental
system cannot be assessed exactly. Thus, the true quality of plan can be certainly
obtained only by repeating the corresponding experiment and performing an esti-
mation of the distribution of the responses. The GIMS-technology based on the
evolutionary modeling and sequential analysis as follows from Fig. 1.5 optimizes
the investigation process.
Complex engineering problems clearly cannot be resolved without using sim-
ulation experiments, which are usually brought about by numerical solution of a
complex system of equations. In this case, the researcher runs into a con ict

between his desire to enhance the accuracy of a simulation experiment and the
limited capabilities of the used algorithms. The biggest dif

fl

culties arise when the
investigator cannot use mathematical equations to parameterize the studied pro-
cesses. These dif culties are overcome by the evolutionary computer technology

that helps to design a new type of the model (Bukatova et al. 1991; Ren é and

Hassane 2005).
In common case, the GIMS-based approach provides the reproduction of a
numerical simulation experiment, derived from a given set of quantitative models
and algorithms independently from the complexity of problem for the study. Of
course, the investigator has to compose the information about different parts of the
simulation experiment including information about input and output data as well as
about the models to use. All models used in the experiment must be identi ed,

accessible, and fully described. The models required for the simulations must be
provided with all governing equations, parameter values, and necessary conditions