Information-Modeling Technology for the Environmental Monitoring - New Ecoinformatics Tools in Environmental Science: Applications and Decision-making - page 41

forecast is produced with 75 95 % reliability covering several temporal steps. The

extent of a forecast is determined by the depth of the history against the background
of the saturation of the effect. Given the need for a forecast under the conditions of
change in the trends of human economic activities, an evolutionary processor is
adjusted to the assigned scenario, thus automatically providing for simulation of the
corresponding response of the biosphere to this change.
The suggested structure of the global model thus ensures a

-

flexible combination
of models of the traditional and evolutionary types. The proposed approach helps
escape the need to model unstable processes (climatic, socio-economic, demo-
graphic, etc.) and provides for overcoming uncertainty. A model of this new type
makes it possible to go from learning experiments to the assessment of the viability
of the biosphere with regard to actual trends of anthropogenic stresses in all regions
of the globe.
A departure from the established global modeling techniques based on the new
information technology makes it possible to proceed to creating a global system of
monitoring with the global model as a portion of the support for the system. The
structure of such a system is represented in Fig. 1.19 . Application of the evolu-
tionary computer technology provides for categorization of the whole system by a
class of subsystems with variable structure and for making it adaptable to changes
in the natural process or entity under observation. Furthermore, it becomes possible
to detail heterogeneously the natural systems under study in the space of phase
variables, and to select nonuniform geographical grids in a sampling analysis of the
planetary surface, i.e. arbitrary insertion of significant regularities at the regional
level becomes possible.
The automatic system for processing of global information is aimed at the
acquisition of combined models re

fl

ected in the real-time scale of the climatic and
anthropogenic changes in the biosphere and based on the known history (or, rather,
its simulation). The system relies on the set of models of biosphere processes and,
using software of other units with the help of the scenario of anthropogenic
behavior formulated at input, provides for prompt assessment of the environmental
state and for forecast assessments within the framework of this scenario. The
automatic system for processing of global information also consists of the advan-
tage in that it formulates the entry and bound conditions for particular built-in
models in the study of regional systems and virtually substitutes for

fl

field mea-

surements of those conditions.
The

first version of the global model (Krapivin et al. 1982) was oriented towards
rigid spatial-temporal detailing and therefore required a large quantity of infor-
mation. The subsequent development of an automatic system for processing of
global information has made it possible, owing to the evolutionary technology, to
reject the generally accepted regular geophysical grid in archive development and to
solve this problem using algorithms for the recovery of spatial-temporal

information.
Global model aims to parameterize the main processes that take place in the
biosphere/climate/society system. The existing global knowledge and databases