Geoscience Reference
In-Depth Information
high error under the solution identi
cation is reduced when algorithm of Table
9.8
is used. We see that the risk to have high error is growing with increase of chemical
element concentration. The discrepancy between spectra is decreased with increase
of chemicals concentration. In this case, it is necessary to extend the database of
spectral etalons.
9.3.3 Decision Making and Spectrophotometric Technology
Many tasks related to the water quality assessment can be solved with the use of
simple optical devices that deliver one spectral curve. Now let us consider the case
when spectral images are received with the use of 8-channel spectrophotometer
(Krapivin and Mkrtchyan 2007a, b; Man et al. 2010). According to the rule proposed
by Bickel and Bo (2006) the sorting of the measureme
nt
data {x
i
} is accomplished
by the following procedure. The value R ¼ x
i
x
ð
Þ=r
is calculated, where
q
P
i¼1
ðÞ
P
i¼1
x
i
. Decision about the sample formation
is made, depending on the value R.IfR
2
r
¼
ð
x
i
x
Þ
=
n
;
x ¼ 1
=
measured value xi
i
is not included in
the set {x
i
}. If R
R
o
, value xi
i
is included in the set {xi}
i
} for the following analysis.
Threshold value R
o
is function of number of measurements (Table
2.22
).
Both the Simulation System for Hydrophysical and Hydrochemical Investiga-
tions (SSHHI) (Fig.
9.5
) and Expert System for Ecological Control of Estuary Zone
(ESECEZ) (Fig.
9.7
) were tested in the framework of joint Russian-Vietnamese
hydrophysical experiments (Cao Van Phuong et al. 2009a, b; 2012; Bui Ta Long
et al. 2010; Krapivin et al. 2008b, 2009a; Krapivin and Mkrtchyan 2006a, b, 2007d,
2008; Man et al. 2011b). Measurements were carried out to study different water
reservoirs located in South Vietnam: Dong Nai River, Saigon River, coastal waters
of South-China Sea, lagoons and
R
o
;
[
fishing industry. Spectral
characteristics of these objects are changed with high range. It says that charac-
teristics of water reservoirs were dynamical and their operative control was
implemented by means of SSHHI and ESECEZ. Measurements were obtained by
means of 8-channel spectrophotometer (Figs.
9.14
,
9.15
,
9.16
). Figures
9.17
and
9.18
explain the measuring process.
Spectrophotometer connected with computer with SSHHI and ESECEZ software
according to Klimov (2010) is called Adaptive Identi
fl
flooded reservoirs of the
er. Figures
9.16
and
9.17
explain the measuring procedure in-situ when an educated system is used for
operative evaluation of water reservoir quality.
Figure
9.19
represents normalized spectral characteristics of studied water
objects. It is seen that all spectral curves I/I
0
related to water quality intersect in the
point (4,1). This means that channel 4 (510 nm) is invariant for the studied reservoirs.
It should be noted that the channels 5, 6 and 7 are the most informative. Maximum of
curve S for channel 7 reveals the presence of chlorophyll-a
uorescence on wave-
length 690 nm and accompanying to it the pigments in red spectrum exists.
fl
Search WWH ::
Custom Search