Geoscience Reference
In-Depth Information
In the process of measurements or calculations, for each pollutant of the type
ʾ
the function of concentration distribution P
ðu; k;
h
;
t
Þ
is formed. Depending on
P
2
P
n
time, at each point
ðu; k;
is formed. Because
of measurements and calculations errors due to incorrect data, the choice of situ-
ations P
n
\
h
Þ
a series of data P
n
1
;
n
; ...;
n
; ...
MCL
n
or P
n
MCL
n
is made at each time moment t on the basis of the
respective statistical assessment in distinguishing between two versions:
(1)
the decision is made on the basis of a totality of indicators, and
(2)
the decision is made on the basis of each indicator separately.
In the latter case, the inequality P
n
MCL
n
is tested for stability during a
prescribed time interval
ʔ
t. In the case of meeting the condition P
n
MCL
n
during this time interval the display gives a signal with the respective information.
The time
is exploitation
to minimize the amount of false alarms or, from condition of maximum probability
not to miss a situation with a violated MCL. The actual choice of
ʔ
t is determined at expert level in the process of the system
'
ʔ
t depends on the
multitude of local anthropogenic and other factors and refers to the user
s com-
petence. To make it easier, as to obtain an expert estimate or to make a decision in
the real time regime, especially in the case of the use of an integral criterion, the
user receives estimates of all indicators on a homogeneous normalized scale.
Let ql
'
ð Þ
be the indicator of the atmospheric quality in the compartment
ʩ
i,j,k
by the l-th parameter. To reduce all indicators to a single scale, we normalize
them:
;
i
;
j
;
Q
l
¼ Ql
ð
;
i
;
j
;
k
Þ
ql
ð
;
i
;
j
;
k
Þ=
max
i
;
j
;
k
ql
ð
;
i
;
j
;
k
Þ
Then the results of normalization are smoothed by height (k), latitude (i), longitude
(j) and the whole territory
ʩ
. The smoothed results are used to assess the air quality of
territories by the user
final decision is made on the basis of comparison
of Q
l
estimates with a set of threshold values included in data base. The complex
assessment of the state of the atmosphere over a given part of the territory
'
is choice. The
ʩ
is made
by the vector parameter Q ¼ Q
1
; ...;
ð Þ
. This n-dimensional vector belongs to
some indicator space X, each point in which corresponds to a concrete combination of
MCL
ʾ
(prescribed by the user). The whole space X is divided into two parts: X
0
—
pollution does not exceed acceptable levels; X
1
—
Q
n
pollution reaches a dangerous level.
A concrete realization of Q is estimated by belonging to one of these sub-spaces.
Thus, the main stages of ESPAP functioning are described by the scheme pre-
sented in Fig.
5.22
. Of course, the quality of the decisions made depends on the
adequacy of information in the data base. Table
5.20
shows a fragment of the data
base. These database fragments and various modi
cations of parametric descrip-
tions of partial processes of aerosol transformation are controlled by a totality of
identi
ers which the user disposes of through such schemes of interaction with the
system as scrolling windows and hierarchical menu. Therefore, a multitude of
identi
ers is used at the level of the system
'
is manager who sees to the compatibility
of its functions.
Search WWH ::
Custom Search