Environmental Decision-Making - New Ecoinformatics Tools in Environmental Science: Applications and Decision-making - page 255

Entering indicator I brings a scheme to monitor and forecast natural catastrophes
much closer. We now present the possible structure of a monitoring system that has
functions to search for, forecasts, and plots the course of a natural catastrophe.
Three levels are selected in the system: holder, resolver and searcher, the blocks of
which have the following function:

(1) periodic examination of the Earth ' s surface elements;

(2) memorizing suspicious elements;
(3) forming of traces of the moving anomaly from suspect elements;
(4) accumulation in time of data on fixed elements of the Earth ' s surface to carry

out a statistical analysis in order to decide whether the fixed suspicious ele-

ments are of noise or signal in character; and

(5) stage-by-stage localization of the procedure of the search for an anomaly.

ciency of such a monitoring system depends on the parameters used by
technical facilities to take measurements and on the algorithms for data processing.
The environmental model, used in parallel with formation and statistical test of the
row {I ʩ (t)} and adapted to mode of the monitoring in accordance with GIMS-based
adaptation scheme, plays an important role.
Employing these criteria when studying natural catastrophes reveals that both
form and behavior I ʩ (t) have a typical type for each process in the environment.
One of the dif

The ef

cult problems consists in determination of these forms and corre-
sponding to their categorizations. For instance, a such often appearing of dangerous
natural phenomena, as landslides and mud ows, has typical signs of the preliminary

change in the relief and landscape, which are successfully registered with satellite in
optical range. Moreover, aerial images and in situ measurements of relief slopes,
hillsides, and hydrologic networks permit forecasting such events several days
before. However, the limited possibility of the optical range under the cloudiness or
vegetable cover has to be overcome by introducing the systems of the remote
sensing in microwave range of the electromagnetic spectrum. Then, in addition to
the above-mentioned indicators of landslides and mud

fl

ows, it is possible to add
such informative parameters as soil moisture and biomass, since increase in soil
moistureleads to the appearance of landslides, while increase in the biomass is
indicative of strengthening the restraining role played by vegetation cover in pre-
venting soil rocks moving. Particularly, this is important when supervision is
snowed-stone or simply snow avalanches. Creating a catalog of such indicators for
all possible natural catastrophes and adding them to the knowledge base of a
monitoring system is a necessary stage in increasing its ef

fl

ciency.
The knowledge of the informative indicators set x i characterizing the natural
catastrophe of j-th type and a priori determination of its cluster X j in space of these
indicators allows to calculate the velocity v j of the approach of the point x i to the
center of X j during the satellite observation and, thereby, to evaluate the time of the
coming of the catastrophe. Other algorithms can be used to forecast natural
catastrophes. For instance, forecasting of the forest

fire is possible by means of
dependencies of forest radiothermal radiation on different length of the waves from