Content uploaded by E. A. Miller
Author content
All content in this area was uploaded by E. A. Miller on May 31, 2022
Content may be subject to copyright.
ISSN 10248560, Atmospheric and Oceanic Optics, 2011, Vol. 24, No. 3, pp. 280–287. © Pleiades Publishing, Ltd., 2011.
Original Russian Text © V.A. Gladkikh, A.E. Makienko, E.A. Miller, S.L. Odintsov, 2011, published in Optica Atmosfery i Okeana.
280
INTRODUCTION
In the first part of the paper, we outlined the pur
poses of the studies and considered the interlevel cor
relations of the wind velocity of the atmospheric
boundary layer (ABL) over Moscow. The continuation
of the work concerns the analysis of the ABL air tem
perature profiles and certain turbulence characteris
tics near the underlying surface. The paper discusses
the measurements of the ABL characteristics at two
observation sites (OSs) on the territory of Moscow:
(1) Tushino Meteorological Station (Tushino OS) in
the northeast part of the city, and (2) an observation
site at the center of the city at ul. Shabolovka (Shab
olovka OS). The characteristics of these sites were pre
sented in [1].
At Shabolovka OS, the measurements were per
formed from May 30 to July 6, 2006. We employed
data from the Meteo2 ultrasonic meteorological sta
tion (UMS) (which is located at a height of 5 m and
measures air temperature, humidity, and pressure, as
well as wind velocity components, approximately ten
times per second) and the Volna4 sodar (which esti
mates the height of the layer of intense turbulent heat
exchange and altitudetemporal profiles of wind
velocity components, covers the altitude range of 50–
500 m, and obtains a single instantaneous profile
approximately every 9 s with a 10m step in altitude).
At Tushino OS, the measurements have been per
formed routinely since June 2006. In addition to the
Meteo2 UMS and Volna4 sodar, the MTP5 meteo
rological temperature profiler is used (which measures
the temperature profiles with steps of 5 min in time
and 50 m in height, covering the altitude range up to
600 m).
AIR TEMPERATURE
The air temperature at different heights could be
estimated only at the Tushino OS, which operated the
temperature profiler. Analysis of the processes of tem
perature variations is an independent class of problem;
so it will not be pursued here. We will only consider the
results concerning comparison of specific features of
the ABL structure at two spatially separated observa
tion sites on the territory of Moscow.
Figure 1a presents the variations in temperature at
heights of 5, 100, and 600 m for the period from Sep
tember 24 to 30, 2007. In this episode, the diurnal
behavior of temperature at a height of 600 m is more
poorly defined than at heights of 5 and 100 m. This
behavior of the temperature profile is quite typical for
the Tushino OS in the warm (and dry) season.
In the episode mentioned above, there were regular
formation and subsequent breakup of radiation
(nighttime) temperature inversions, whose lower
boundaries were adjacent to the Earth’s surface
(before onset of the process of their breakup due to
convective heat fluxes), and whose upper boundaries
reached 250–350 m (Fig. 1b). Approximately this
same altitude range was also observed on other days in
September 2007. The height of the inversion upper
boundary reached 500–600 m in only three cases dur
ing the month.
Study of the Atmospheric Boundary Layer Parameters under Urban
Conditions with Local and Remote Diagnostics Facilities.
Part. 2. Air Temperature and Heat Flux
V. A. Gladkikh
a
, A. E. Makienko
a
, E. A. Miller
b
, and S. L. Odintsov
a
a
V.E. Zuev Institute of Atmospheric Optics, Siberian Branch, Russian Academy of Sciences,
pl. Akademika Zueva 1, Tomsk, 634021 Russia
b
Central Aerological Observatory, ul. Pervomayskaya 3, Dolgoprudnyi, Moscow oblast, 141700 Russia
Received July 7, 2010
Abstract
—We considered the results of the complex measurements of meteorological parameters at two
observation sites in Moscow with the help of Volna4 sodar, the Meteo2 ultrasonic meteorological station,
and the MTP5 meteorological temperature profiler. The frequency of occurrence and the heights of temper
ature inversion boundaries under the conditions of woodland–park zone are analyzed. The obtained results
are compared with analogous estimates of some other authors for the central part of the city. It is found that
a stable stratification of the urban roughness sublayer takes place considerably more often in the woodland–
park zone of the city than in the central part of the urbanized territory.
DOI:
10.1134/S1024856011030092
OPTICAL MODELS
AND DATABASES
ATMOSPHERIC AND OCEANIC OPTICS Vol. 24 No. 3 2011
STUDY OF THE ATMOSPHERIC BOUNDARY LAYER PARAMETERS 281
A characteristic feature of the episode mentioned
above (from September 24 to 30, 2007) is a longterm
persistence of the temperature inversions (including
nearground and elevated inversions) up to 73% of all
observation time (163 h). For the entire sample in Sep
tember 2007, the frequency of occurrence of the inver
sions was 48% (331 h out of 687 h of observations, of
which 303 h were accounted for by a nearground
inversion). The maximum temperature difference at
the inversion boundaries reached 7.5
°
C. It is notewor
thy that the inversions broke quite rapidly. For
instance, in the episode presented in Fig. 1b, it took
0.2–1.5 h after the lower boundary was detached from
the level of 5 m (an exception was September 26, 2007,
when this process lasted for more than 2 h). We note
that the temperature inversion was identified accord
ing to the signs of the gradient of an instantaneous dis
crete temperature profile. Possibly, after an additional
processing of the initial profiles (such as via their
smoothing and interpolation), the statistics of the
inversion characteristics may be somewhat altered.
We stress that at the Tushino OS, the temperature
inversion had started to form in the immediate vicinity
of the Earth’s surface. This fact indicates that at night,
the urban cover (model of layering of the atmosphere
as described in [1]) at Tushino OS, unlike that at Shab
olovka OS, is typically characterized by negative verti
cal turbulent heat flux. It is just this flux that had led to
formation of the nearground temperature inversions.
400
350
300
250
200
150
100
50
0
28 292726 01
2524
(a)
(b)
00:00 Local time
Temperature,
°
C
Tushino OS, September 2007
Height of upper boundary of
Date
temperature inversion, m
450
500
550
30
18
16
14
12
10
8
628 292726 01
2524
20
22
30
Fig. 1.
Air temperature: (a) temperature variations at different heights for the Tushino OS on September 24–30, 2007: 5 m (solid
line), 100 m (circles), and 600 m (dashed line); and (b) height of the upper boundary of the temperature inversion at the Tushino
OS (circles) and at the Dolgoprudnyi OS (stars).
282
ATMOSPHERIC AND OCEANIC OPTICS Vol. 24 No. 3 2011
GLADKIKH et al.
The turbulent heat fluxes at the Tushino and Shab
olovka observation sites will be discussed in more
detail on pages 990–993 of this paper.
The specific features of formation and characteris
tics of the temperature inversions
over the central part
of Moscow are presented in work [2], which also used
the MTP5 temperature profiler. The results of work
[2] (Fig. 2, solid line) suggest that the frequency of
occurrence of inversions at the center of Moscow is
less than that predicted by the existing methods (the
prediction results are shown in Fig. 2 by the dashed
line).
The reader may wish to refer to work [3] for a con
firmation that this situation is typical in big cities; that
work, on the basis of experimental data (Zürich, Swit
zerland), found that in the urban roughness sublayer
stable stratification virtually did not form (except in a
few winter episodes).
The frequency of occurrence of temperature inver
sions over Tushino OS in 2007 is shown in Fig. 2 by the
line with symbols; estimates of persistence of temper
ature inversion within a 1month period without seg
regation according to inversion types (squares) are
shown, as well as the persistence of nearground inver
sions only (dots). We note quite a rapid breakup of
inversions in the warm season (most so in April). The
frequency of occurrence of inversions over Tushino OS
in 2007 agrees well with results in [2], which were
obtained in 2000–2001 (see Fig. 2). As expected, the
frequency of occurrence of inversions at the center of
Moscow is less than over Tushino OS, though this is
not the case in separate months (note that results for
different years are compared!).
We should point to the following specific feature of
variation in the height of the upper boundary of tem
perature inversion
H
i
at Tushino OS (see Fig. 1b): the
height
H
i
is markedly lower in the period of inversion
breakup. This
H
i
behavior is typical in those cases
when classical diurnal variations in the temperature
within ABL are observed. This effect is partly respon
sible for a faster inversion breakup. It is noteworthy
that
H
i
subsidence begins quite synchronously with the
process of detachment of the lower inversion boundary
from the surface. It should be noted that this
H
i
change
is typical
not only
for Tushino OS. For instance, Fig. 1b,
presents synchronous measurements of
H
i
in Dolgo
prudnyi (north of Moscow). Despite considerable dis
tance between these observation sites, one and the
same type of change in the upper boundary of the tem
perature inversion during its breakup is well evident.
As far as we know, the specialized scientific literature
so far paid no attention to this
H
i
behavior. Possibly,
this highlighted feature is associated with turbulent
inversion breakup not only on the side of the lower
inversion boundary, but also aloft.
We should also note that the inversion may some
times break without detachment of its lower boundary
from the underlying surface; i.e., something like an
inversion melt takes place (in this case, the upper
boundary gradually subsides toward the underlying
surface).
The main conclusion of this section may be that
diverse vertical distributions of air temperature may
exist within Moscow city limits. In particular, at the
Tushino OS the presence of the
nearground
tempera
ture inversions is rather frequent, even in the warm
season. This distinguishes it from the observation sites
in the central part of the urbanized territory, where a
stable stratification in the urban roughness sublayer is
rare [2–4].
This spotty pattern of the territory of Moscow in
terms of the temperature gradients of the lower atmo
spheric layer should be taken into consideration when
heat and momentum fluxes and emissions of aero
sol/gases and other monitored substances are mod
eled.
It should be noted that, for the period of 2006–
2009, at the Tushino OS we performed over 25000 h of
observations of the temperature profile with steps of
5 min in time and 50 m in height in the altitude range
up to 600 m. In addition, in this time period we mea
sured the average and turbulent characteristics of tem
perature and wind fields at the levels of 5 and 7 m, and
obtained episodic sodarbased data on the structure of
the layer of intense turbulent heat exchange (within
50–500 m) and on the profiles of the wind velocity
components in this layer.
70
60
80
50
40
30
20
10
0XIIXIIXVIIIV
II
Frequency of occurrence of
Month
temperature inversions, %
XVIIIVIVIIII
641
607
459
557
683
662 709
671
687
665
702
640
Fig. 2.
Frequency of occurrence of the temperature inver
sions (persistence of the inversions in a 1month observa
tion period) at the Tushino OS during 2007 and at the cen
ter of Moscow in 2000–2001. The abscissa axis indicates
only 2007. For the Tushino OS, nearground inversions are
indicated by circles, and inversions of all types by squares
(numbers near data points indicate the lengths of the
observation periods during a month, in hours). For the
center of Moscow (reproduced from [2]), the solid line
shows observations and the dashed line shows calculations.
ATMOSPHERIC AND OCEANIC OPTICS Vol. 24 No. 3 2011
STUDY OF THE ATMOSPHERIC BOUNDARY LAYER PARAMETERS 283
SECONDORDER MOMENTS
OF TEMPERATURE AND VERTICAL WIND
The wind velocity components, measured by sodar
and UMS, carry information not only on the average
values, but also on the turbulent characteristics. The
Meteo2 UMS may provide information at a fre
quency of about 10 readings per second, while the
Volna4 sodar retrieves one instantaneous profile of
the wind velocity at a frequency of about every 9 s (for
a sensing height up to 500 m). By virtue of this, UMS
data can be used to analyze quite a wide range of pul
sations of wind velocity components and temperature,
while sodar data may provide information only on the
lowfrequency turbulence component, which never
theless contains a major part of the flux energy. It
should be noted that UMSs of different types are
actively used to estimate the statistical characteristics
of the meteorological parameters in the lower layer of
the atmosphere and, as such, are currently the chief
tool for studying the thermodynamic processes under
urban conditions (see, e.g., [5–7]).
We will consider certain results of analysis of the
vertical turbulent heat flux
Q
S
and the standard devia
tions of the temperature
σ
T
,
S
and vertical wind velocity
component
σ
w
,
S
at the level of 5 m, as well as their
relations with the wind velocities at the overlying lev
els. We determined the aforementioned parameters
within a 10min time interval. They are interesting, in
particular, because the heat flux
Q
S
may serve as an
indicator of directionality of the processes of admix
ture redistribution in the nearground atmospheric
layer (increase or decrease in their concentration),
while
σ
T
,
S
and
σ
w
,
S
may characterize the intensity of
these processes.
The turbulent heat flux
Q
S
is calculated according
to the UMS data with the use of the formula
Q
S
=
(W/m
2
), where is the air density;
w
' and
T
'
are turbulent components of the vertical wind and
temperature, respectively;
c
p
is the specific heat capac
ity of air at constant pressure; and the overbar repre
sents time averaging. The reader is referred to [8] for
some details and results of
Q
S
determination with the
help of Meteo2 UMS.
The variations in
Q
S
and air temperature in
June 2006 at Shabolovka OS are exemplified in Fig. 3.
It is seen that the
Q
S
variations, on the whole, coin
cide with the diurnal behavior of temperature. At the
same time, the total (synopticscale) change in the
temperature has little effect on the value of the former.
The fact that
Q
S
depends weakly on the average air
temperature on an urbanized territory was already
noted elsewhere (see, e.g., [9]). More detailed diurnal
variations of the temperature and
Q
S
can be seen in
Fig. 4, which presents a fragment of Fig. 3 from
June 20 to 24, 2006.
We should note that there is a time lag between
maxima of temperature and flux
Q
S
: the maximum in
the temperature is delayed in time relative to the max
imum in
Q
S
. This is especially apparent in the cases
when the
Q
S
variations are quite smooth. This situa
tion was also observed at the Tushino OS, and it is also
characteristic for some other regions [10].
ρcpw'T'
ρ
80
40
120
0
–40
160
–80 17151396
21 105
28
12
8
Vertical turbulent heat flux, W/m
2
Temperature,
°
C
Date
24
20
16
3 7 11 14 18 21 22 23 25 2619 27 29 30
Fig. 3.
The Shabolovka observation site in June 2006. The heat flux (line) and air temperature (stars) at a height of 5 m.
284
ATMOSPHERIC AND OCEANIC OPTICS Vol. 24 No. 3 2011
GLADKIKH et al.
At Tushino OS, the negative heat fluxes at the level
of 5 m are more frequent than at Shabolovka OS, with
a correspondingly more frequent occurrence of near
ground temperature inversion compared to the center
of the city. This conclusion is supported by Fig. 5,
which presents the histograms of the
Q
S
distribution
for the period of synchronous observations at Tushino
and Shabolovka sites from June 13 to 30, 2006 (total
observation period was 411 h at each site).
According to results presented in the figure, nega
tive
Q
S
values were observed in approximately 25% of
the cases at the Shabolovka OS and in approximately
50% of the cases at the Tushino site.
The redistribution of heat, moisture, aerosol, and
gases in urban cover is associated, in particular, with
dynamic factors; therefore, it is interesting to estimate
how the vertical heat flux
Q
S
and standard deviations
of the vertical wind velocity component
σ
w
,
S
and tem
perature
σ
T
,
S
are correlated with the wind velocity at
different levels.
Quantitatively, the relation of the secondorder
moments and wind velocity at different heights was
estimated with the help of the linear correlation coef
ficient (LCC) defined as the value of the normal
ized crosscorrelation function of the timedis
u
Ky
x,
By
xτ()
crete samples
X
(
t
i
) and
Y
(
t
i
) for zerovalued argument
τ
(for zero time lag):
Ky
xBy
xτ = 0() 1
NB
x0()By0()
==
×Xt
i
()Yt
iτ+().
i1=
N
∑
Vertical turbulent heat flux, W/m
2
Temperature,
°
C
Local time
Date
80
40
120
0
–40
22:57
18:10
08:35
18:10
08:35
18:10
13:23
22:57
03:46
28
18
24
20
22:58
13:23
03:46
13:23
03:46
13:22
18:10
22:58
03:46
08:35
08:35
13:22
18:10
22:57
100
60
20
–20
30
22
26
20 20 20 21 2121 21 21 22 23 2422 22 22 22 23 23 23 23 24 24 24 24
Fig. 4.
The Shabolovka observation site June 20–24, 2006. The heat flux (stars) and air temperature (line) at a height of 5 m.
Frequency of falling within bin, %
Tushino, June 13–30, 2006
Shabolovka, June 13–30, 2006
Vertical turbulent heat flux bins, W/m
2
45
40
35
30
25
20
15
10
5
–80 –40 0 40 80 120 160
0
Fig. 5.
Distribution histograms of the vertical turbulent
heat flux
Q
S
. The frequency of occurrence was calculated
for 20 W/m
2
flux bins.
ATMOSPHERIC AND OCEANIC OPTICS Vol. 24 No. 3 2011
STUDY OF THE ATMOSPHERIC BOUNDARY LAYER PARAMETERS 285
Here,
B
x
and
B
y
are autocorrelation functions of
the samples
X
(
t
i
) and
Y
(
t
i
) respectively (samples were
not centered); –1
≤
≤
1. It should be stressed that,
for
τ
= 0, the LCC value remains the same for any
order of summation (for arbitrarily rearranged terms
of the time series); therefore, LCC can be calculated,
in particular, using observation time series which are
preliminarily segregated according to any of the char
acteristics (such as according to time). It is only
important that the segregation leave the products
X
(
t
i
)
Y
(
t
i
) nondecoupled.
Henceforth, by the function
X
(
t
i
) will be meant 10min
average values of the longitudinal wind velocity
(
H
,
t
i
) at different heights, and by the function
Y
(
t
i
)
will be meant either value of
Q
S
,
σ
T
,
S
, or
σ
w
,
S
for the
corresponding instants of time. As an example, for the
wind velocity (
H
) at certain height
H
and heat flux
Q
S
at the height of 5 m, the linear correlation coefficient
will be written as
Analysis of for different heights shows that the
level of correlation between and
Q
S
depends essen
tially on sign changes of
Q
S
in the sample: the larger
the
Q
S
values of one and the same sign are in the sam
ple, the closer the correlation between them and wind
velocities at different levels are. As an example, we
present the estimates for those same episodes con
sidered in LCC analysis of winds at different levels in
the first part of our paper [1]. In particular, positive
Q
S
fluxes somewhat prevail (62% of all observation time)
in measurements at Shabolovka OS in the period from
May 30 to June 2, 2006 (about 93 h of observations).
At the same time, the negative heat fluxes were small
in value (
Q
S
≤
–20 W/m
2
only in one quarter of cases).
Figure 6a presents a plot of (line plus squares).
Evidently, the correlations are high, especially when
Q
S
are compared with wind at heights higher than
140 m. However, if are calculated only for positive
(Fig. 6a, line plus triangles) or only for negative (line
plus circles)
Q
S
values, its values substantially increase
in both these case and for all heights.
The LCC regularities, obtained for episodes from
September 24–30 and November 11–14, 2007 (a total
of about 182 h of observations), were also characteris
tic for the Tushino OS (Fig. 6b), where no significant
correlation was obtained when calculated for a sample
with all
Q
S
fluxes included. However, the situation
changes when the LCC analysis is performed only for
positive (44% of the observation time) or only for neg
ative (56% of observation time)
Q
S
values. In these
cases, the correlation increases to quite large values. It
should be noted that, at Tushino OS, the linear corre
lation coefficient at the level of 5 m has nearly the
Ky
x
u
u
KQ
u.
KQ
u
u
KQ
u
KQ
u
KQ
u
KQ
u
same value as it is at other heights, independent of
whether we consider just onesigned
Q
S
samples or the
entire observation time series as a whole.
We will consider how standard deviations of tem
perature (
σ
T
,
S
) and vertical wind velocity (
σ
w
,
S
) are
related to the longitudinal wind velocity at different
levels, by calculating the respective linear correlation
coefficients and Remembering that the sign
of the turbulent heat flux at the 5m level was impor
tant for the estimates, we, likewise, had analyzed
and using a segregation with respect to stable
(negative
Q
S
) and unstable (positive
Q
S
) stratifications.
Calculation results showed that at the Tushino OS,
the dependence of on the height is nearly the
same for any sign of
Q
S
(Fig. 7a).
u
KσT
u
Kσw
u.
KQ
u
KσT
u
Kσw
u
KσT
u
1.0
0.9
0.7
0.6
0.5
0.3
0.2
0.1
022020016012080
40
(a)
(b)
Linear correlation coefficient
Shabolovka OS
Tushino OS
Height, m
0.8
0.4
20 60 100 140 180
1.0
0.9
0.7
0.6
0.5
0.3
0.2
0.1
022020016012080
40
0.8
0.4
20 60 100 140 180
Fig. 6.
Linear correlation coefficients of the vertical turbu
lent heat flux
Q
S
at a height of 5 m and average longitudinal
wind velocity at different heights: (a) The Shabolovka
OS (May 30–June 2, 2006); (b) The Tushino OS (Septem
ber 24–30 and November 11–14, 2007). The calculation
without segregation according to stratification type (
䊏
), for
only negative
Q
S
(
䊉
), and for only positive
Q
S
(
䉱
).
u
KQ
u
286
ATMOSPHERIC AND OCEANIC OPTICS Vol. 24 No. 3 2011
GLADKIKH et al.
We note a minimum of at the height of 60 m,
which is also characteristic for the coefficient of the
interlevel correlation of the wind velocity at this obser
vation site (see work [1, Figs. 5 and 6]). Unlike, at
Shabolovka OS there is a substantial dependence of
on the type of stratification; the correlation is
markedly smaller for negative than positive
Q
S
values
(Fig. 7b). At the same time, the character of the
dependence on altitude differs between the observa
tion sites considered here, for as yet unclear reasons.
The altitude dependence of the linear correlation
coefficient (which relates the standard deviation
of the vertical wind to the average longitudinal wind
velocity), without segregation according to the condi
tions of stratification, on the whole, is similar for both
observation sites, though with some difference in the
correlation strength. This is demonstrated in Fig. 8
(lines plus squares). After segregating the data accord
ing to the conditions of stratification at the 5m level,
and after the subsequent calculation, we can con
clude that, at the Shabolovka OS, the type of the near
ground stratification has little effect on the strength of
correlation between
σ
w
,
S
and longitudinal wind at the
KσT
u
KσT
u
KσT
u
Kσw
u
Kσw
u
overlying levels
(Fig. 8b). However, applying this
same segregation procedure to data for the Tushino
OS, we reveal a somewhat different regularity:
virtually does not depend on the stratification type up
to the height of 90–100 m, above which the correla
tion markedly decreases (somewhat increases) for pos
itive (negative)
Q
S
values (Fig. 8a). We also note the
minima at heights of 50–60 m at both observation
sites and strong correlation between longitudinal wind
velocity and standard deviation of the vertical wind at
the 5–m level at the Shabolovka OS.
CONCLUSIONS
A comparison of results obtained at two spatially
separated observation sites shows that different vertical
air temperature distributions may exist within Moscow
city limits. In particular, at the Tushino OS the
near
ground
temperature inversions may quite frequently be
present even in the warm season, which distinguishes
it from observation sites in the central part of the
urbanized territory, where a stable stratification in the
urban roughness sublayer is a quite rare occurrence.
Kσw
u
Kσw
u
(a)
(b)
Linear correlation coefficient
Tushino OS
Shabolovka OS
Height, m
0.8
0.6
0.5
0.4022020016012080
40
0.7
20 60 100 140 180
0.8
0.6
0.5022020016012080
40
0.7
20 60 100 140 180
Fig. 7.
Linear correlation coefficients of the standard devi
ations of temperature
σ
T
,
S
at height of 5 m and average
wind velocity at different heights (a) The
Tushino OS (September 24–30, 2007 and November 11–
14, 2007), and (b) the Shabolovka OS (May 30–June 2,
2006). The calculation without segregation accord
ing to the stratification type (
䊏
), for only negative
Q
S
(
䊉
),
and for only positive
Q
S
(
䉱
).
u
KσT
u
():
KσT
u
(a)
(b)
Linear correlation coefficient
Shabolovka OS
Tushino OS
Height, m
1.0
0.8
0.7
0.6022020016012080
40
0.9
20 60 100 140 180
0.7
0.6
0.5022020016012080
40
0.8
20 60 100 140 180
Fig. 8.
Linear correlation coefficients of the standard devi
ations of vertical wind velocity component
σ
w
,
S
at a height
of 5 m and average wind velocity at different heights
(a) The Tushino OS (September 24–30 and
November 11–14, 2007), and (b) the Shabolovka OS
(May 30–June 2, 2006). The calculation without
segregation according to the stratification type (
䊏
), for
only negative
Q
S
(
䊉
), and for only positive
Q
S
(
䉱
).
u
Kσw
u
():
Kσw
u
ATMOSPHERIC AND OCEANIC OPTICS Vol. 24 No. 3 2011
STUDY OF THE ATMOSPHERIC BOUNDARY LAYER PARAMETERS 287
The results obtained here show that, for the consid
ered time period of observations at the Tushino OS,
the effect of the stratification type on the correlations
of the wind velocity and nearground heat flux comes
primarily through change in the statistics of the verti
cal wind velocity component, in view of the fact that
the statistics of the temperature at this observation site
remains unchanged for any stratification. The Shab
olovka OS shows an opposite regularity; the stratifica
tion type influences the correlations of the wind veloc
ity and nearground heat flux only through change in
the statistics of the temperature.
The found regularities require extra studies with the
use of a largervolume statistical experimental mate
rial. These studies are promising in that they poten
tially can relate the wind characteristics over the city
(at the levels above the atmospheric nearground layer,
where wind can be predicted to a good justifiability)
with the characteristics of the processes in urban cover,
thereby providing important predictors in support of
the existing methods for predicting the distribution of
gas–aerosol emissions in the atmosphere over urban
ized territory.
ACKNOWLEDGMENTS
The authors are grateful to the officials of the State
Nature Conservation Organization Mosekomonitor
ing for providing access to experimental data from
diagnostic complexes used in this study.
This work was supported by the Ministry of Science
and Education (State contracts nos. 02.740.11.0674
and 14.740.11.0204) and by the Presidium of the Rus
sian Academy of Sciences (project no. 4.1).
REFERENCES
1. V. A. Gladkikh, A. E. Makienko, E. A. Miller, and
S. L. Odintsov, “Study of Parameters of Atmosphere
Boundary Layer in Urban Conditions Using Means of
Local and Remote Diagnostics. Pt. 1: Interlevel Corre
lations of Wind Velocity,” Opt. Atmos. Okeana
23
,
978–986 (2010).
2. E. N. Kadygrov, I. N. Kuznetsova, and G. S. Golitsyn,
“Heat Island Above Megalopolis: New Results on the
Base of Remote Sensing Data,” Dokl. Akad. Nauk
385
,
541–548 (2002).
3. M. W. Rotach, “Profiles of Turbulence Statistics in and
above an Urban Street Canyon,” Atmos. Environ.
29
,
1773–1486 (1995).
4. S.E. Gryning, E. Batchvarova, B. Brummer, H. Jor
gensen, and S. Larsen, “On the Extension of the Wind
Profile over Homogeneous Terrain Beyond the Surface
Boundary Layer,” BoundaryLayer Meteorol.
124
,
251–268 (2007).
5. P. G. Mestayer, P. Durand, P. Augustin, S. Bastin,
J.M. Bonnefond, B. Bénech, B. Campistron, A. Cop
palle, H. Delbarre, B. Dousset, P. Drobinski, A. Druil
het, E. Fréjafon, C. S. B. Grimmond, D. Groleau,
M. Irvine, C. Kergomard, S. Kermadi, J.P. Lagouarde,
A. Lemonsu, F. Lohou, N. Long, V. Masson, C. Mop
pert, J. Noilhan, B. Offerle, T. R. Oke, G. Pigeon,
V. Puygrenier, S. Roberts, J.M. Rosant, F. San
ї
d,
J. Salmond, M. Talbaut, and J. Voogt, “The Urban
BoundaryLayer Field Campaign in Marseille
(UBL/CLUESCOMPTE): SetUp and First
Results,” BoundaryLayer Meteorol.
114
, 315–365
(2005).
6. A. Christen, M. W. Rotach, and R. Vogt, “The Budget
of Turbulent Kinetic Energy in the Urban Roughness
Sublayer,” BoundaryLayer Meteorol.
131
, 193–222
(2009).
7. M. A. Nelson, E. R. Pardyjak, J. C. Klewicki, S. U. Pol,
and M. J. Brown, “Properties of the Wind Field Within
the Oklahoma City Park Avenue Street Canyon, Part I:
Mean Flow and Turbulence Statistics,” J. Appl. Mete
orol. Climatol.
46
, 2038–2045 (2007).
8. G. I. Gorchakov, V. A. Gladkikh, P. S. Glyadkov,
A. E. Makienko, S. L. Odintsov, and R. S. Ozeretskii,
“Application of Acoustic Diagnostics Means of Low
Atmospheric Layer for Solution of Gas Exchange Prob
lems in Urban Conditions,” in
Proceedings of the
19th Session of Russian Acoustical Society
(GEOS,
Moscow, 2007), Vol. 2, Pp. 168–172.
9. M. Piringer, S. Joffre, A. Baklanov, A. Cristen,
M. Deserti, K. De Ridder, S. Emeis, P. Mestayer,
M. Tombrou, D. Middleton, K. BaumannStanzer,
A. Dandou, A. Karppinen, and J. Burzynski, “The Sur
face Energy Balance and the Mixing Height in Urban
Areas—Activities and Recommendations of COST
Action 715,” BoundaryLayer Meteorol.
124
, 3–24
(2007).
10. V. A. Gladkikh, A. E. Makienko, and S. L. Odintsov,
“Experimental Studies of Turbulent Heat Fluxes in
Ground Atmosphere Layer,” in
Proceedings of the
XIth Working Group on Siberia Aerosols
(IOA SO RAN,
Tomsk, 2004), p. 30.