Geoscience Reference
In-Depth Information
Tab l e 5 . 1 .
Wavelength intervals for accounting for the molecular absorption of the atmo-
spheric gases
Gas
Wavelength interval (nm)
H
2
O
444-446, 468-470, 502-510, 538-552, 566-602, 626-666, 684-746, 784-978
O
3
330-356, 426-848
O
2
626-632, 686-694, 758-774
NO
2
330-616, 638-656
NO
3
598-672
out the concrete observations with the concrete K-3 instrument. In principle,
application of these data to other experiments and instruments is admissible,
but the analogous estimations and calculations are desirable to implement.
Inspiteofusingthefastalgorithmforthecalculationofthevolumeco-
efficient of the molecular absorption as per (5.6), its application takes time
unacceptable for the computer codes of the operative data processing. There-
fore, the passage, which is based on using the look-up tables of the preliminary
computed absorption cross-sections
C
a
(
λ
,
P
,
T
), could essentially simplify the
scheme of calculations in the applied algorithm.
The spectroscopic data of O
3
,NO
2
and NO
3
gases have been initially pre-
sented as look-up tables (Sect. 1.2). The temperature dependence of the ab-
sorption cross-section of O
3
in 330-356 nm is described according to Bass and
Paur (1984) as follows:
λ
j
,
T
)
=
λ
j
)+
C
1
(
λ
j
)
T
+
C
2
(
λ
j
)
T
2
.
C
a
(
C
0
(
(5.7)
This very interval should be used for the interpolation over wavelengths and
temperature dependence of the absorption cross-section (1.28). The absorp-
tion cross-sections
C
a
(
λ
,
P
,
T
)ofH
2
OandO
3
are also computed preliminarily
as look-up tables. The grid over wavelength is selected as an irregular one to
describe all specific features of the spectrum on the one hand and to mini-
mize the wavelength quantity in the grid on the other hand. Some ideas of
the algorithm presented in the study by Pokrovsky (1967) constitute the basis
of this grid selection. The centers of the spectral lines are selected as basic
points
λ
j
because they correspond to maximal values
C
a
(
λ
,
P
,
T
). Other points
are defined by the consequent dividing of the basic grid proceeding from the
comparison of exact and interpolated values
C
a
(
λ
,
P
,
T
). Then the following
λ
parameterization of dependence
C
a
(
,
P
,
T
)isused:
λ
j
)
P
C
1
(
λ
j
)
T
∗
T
C
2
(
λ
j
)
λ
j
,
P
,
T
)
=
λ
i
)
C
0
(
C
a
(
R
(
,
(5.8)
P
∗
λ
j
) is the correcting factor. To minimize the systematic uncertainty
of (5.8) this factor is selected after the preliminary computing of every indi-
cation of the instrument corresponding to wavelength
where
R
(
λ
j
. Equation (5.8) has
beenwrittenonthebasisoftheknownapproximationofthecoefficientofgas
Search WWH ::
Custom Search