AMERICAN INSTITUTE OF CROP ECOLOGY
A RESEARCH ORGANIZATION DEVOTED TO PROBLEMS OF
PLANT ADAPTATION AND INTRODUCTION
WASHINGTON, D. C.
AICE* SURVEY OF USSR AIR POLLUTION LITERATURE
Volume V I
AIR POLLUTION IN RELATION TO CERTAIN ATMOSPHERIC AND
METEOROLOGICAL CONDITIONS AND SOME OF THE METHODS EMPLOYED
IN THE SURVEY AND ANALYSIS OF AIR POLLUTANTS
Edited By
M. Y. Nuttonson
The material presented here is part of a survey of
USSR literature on air pollution
conducted by the Air Pollution Section
AMERICAN INSTITUTE OF CROP ECOLOGY
This survey is being conducted under GRANT 1 RO1 AP00786 - APC
AIR POLUTION CONTROL OFFICE
of the
ENVIRONMENTAL PROTECTION AGENCY
'AMERICAN INSTITUTE OF CROP ECOLOGY
809 DALE DRIVE
SILVER SPRING, MARYLAND 20910
-------
-------
PUBLICATIONS of the AMERICAN INSTITUTE OF CROP ECOLOGY
Ref
No
1
10
11
12
13
14
15
16
17
18
19
20
21
UKRAINE-Ecologicol Crop Geography of the Ukraine and the
Ukrainian Agro-Climatic Analogues in North America
POLAND-Agricultural Climatology of Poland and Its Agro-
Climatic Analogues in North America
CZECHOSLOVAKIA-Agricultural Climatology of Czechoslo-
vakia and Its Agro-Climatic Analogues in North America
YUGOSLAVIA-Agnculrural Cllmatologyof Yugoslavia and Its
Agio-Climatic Analogues in North America
GREECE-Ecological Crop Geography of Greece and Its Agro-
Climatic Analogues in North America
ALBANIA-Ecological Plant Geography of Albania, Its Agri-
cultural Crops and Some North American Climatic Analogues
CHINA-Ecological Crop Geography of China and Its Agro-
Climatic Analogues in North America
GERMANY-Ecological Crop Geography of Germany and Its
Agro-Climatic Analogues m North America
JAPAN (1 (-Agricultural Climatology of Japan and Its Agro-
Climatic Analogues in North America
FINLAND-Ecological Crop Geography of Finland and Its Agro-
Climatic Analogues in North America
SWEDEN-Agricultural Climatology of Sweden and Its Agro-
Climatic Analogues in North America
NORWAY-Ecological Crop Geography of Norway and Its Agro-
Climatic Analogues in North America
SIBERIA-Agncul rural Climatology of Siberia, Its Natural Belts,
and Agro-Climatic Analogues in North America
JAPAN (2)-Ecological Crop Geography and Field Practices of
Japan, Japan's Natural Vegetation, and Agro-Climatic
Analogues in North America
RYUKYU ISLANDS-Ecologicol Crop Geography and Field
Practices of the Ryukyu Islands, Natural Vegetation of the
Ryukyus, and Agro-Climatic Analogues in the Northern
Hemisphere
PHENOLOGY AND THERMAL ENVIRONMENT AS A MEANS
OF A PHYSIOLOGICAL CLASSIFICATION OF WHEAT
VARIETIES AND FOR PREDICTING MATURITY DATES OF
WHEAT
(Based on Data of Czechoslovakia and of Some Thermally
Analogous Areas of Czechoslovakia in the United States
Pacific Northwest)
WHEAT-CLIMATE RELATIONSHIPS AND THE USE OF PHE-
NOLOGY IN ASCERTAINING THE THERMAL AND PHO-
TOTHERMAL REQUIREMENTS OF WHEAT
(Based on Data of North America and Some Thermally Anal-
ogous Areas of North America in the Soviet Union and in
Finland)
A COMPARATIVE STUDY OF LOWER AND UPPER LIMITS OF
TEMPERATURE IN MEASURING THE VARIABILITY OF DAY-
DEGREE SUMMATIONS OF WHEAT, BARLEY, AND RYE
BARLEY-CLIMATE RELATIONSHIPS AND THE USE OF PHE-
NOLOGY IN ASCERTAINING THE THERMAL*AND PHO-
TOTHERMAL REQUIREMENTS OF BARLEY
RYE-CLIMATE RELATIONSHIPS AND THE USE OF PHENOL-
OGY IN ASCERTAINING THE THERMAL AND PHOTO-
THERMAL REQUIREMENTS OF RYE
AGRICULTURAL ECOLOGY IN SUBTROPICAL REGIONS
22 MOROCCO, ALGERIA, TUNISIA-Physical Environment and
Agriculture
23 LIBYA and EGYPT-Physical Environment and Agriculture
24 UNION OF SOUTH AFRICA-Physical Environment and Agri-
culture, With Special Reference to Winter-Rainfall Regions
25 AUSTRALIA-Physicol Environment and Agriculture, With Spe-
cial Reference to Winter-Rainfall Regions
26 S. E. CALIFORNIA and S. W. ARIZONA-Phyiical Environment
and Agriculture of the Desert Regions .
27 THAILAND-Physicol Environment and Agriculture
28 BURMA-PhysicaJ Environment and Agriculture
28A BURMA-Diseases and Pests of Economic Plants
28B BURMA-Chmate, Soils and Rice Culture (Supplementary In-
formation and a Bibliography to Report 28)
29A
298
29C
30A
30 B
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
VIETNAM, CAMBODIA, LAOS-Fhysical Environment and
Agriculture ,
VIETNAM, CAMBODIA, LAOS-Diseases and Pestsof Economic
Plants . ..
VIETNAM, CAMBODIA, LAOS-CCmatologlcal Data (Supple-
ment to Report 29A)
CENTRAL and SOUTH CHINA, HONG KONG, TAIWAN-
Physical Environment and Agriculture . . $2000*
CENTRAL and SOUTH CHINA, HONG KONG, TAIWAN-
Ma|or Plant Pests and Diseases ....
SOUTH CHINA-lts Agro-Climatic Analogues in Southeast Asia
SACRAMENTO-SAN JOAQUIN DELTA OF CALIFORNIA-
Physlcal Environment and Agriculture .
GLOBAL AGROCLIMATIC ANALOGUES FOR THE RICE RE-
GIONS OF THE CONTINENTAL UNITED STATE
AGRO-CLIMATOLOGY AND GLOBAL AGROCLIMATIC
ANALOGUES OF THE CITRUS REGIONS OF THE CON-
TINENTAL UNITED STATES
GLOBAL AGROCLIMATIC ANALOGUES FOR THE SOUTH-
EASTERN ATLANTIC REGION OF THE CONTINENTAL
UNITED STATES
GLOBAL AGROCLIMATIC ANALOGUES FOR THE INTER-
MOUNTAIN REGION OF THE CONTINENTAL UNITED
STATES
GLOBAL AGROCLIMATICANALOGUES FOR THE NORTHERN
GREAT PLAINS REGION OF THE CONTINENTAL UNITED
STATES
GLOBAL AGROCLIMATIC ANALOGUES FOR THE MAYA-
GUEZ DISTRICT OF PUERTO RICO
RICE CULTURE and RICE-CLIMATE RELATIONSHIPS With Spe-
cial Reference to the United States Rice Areas and Their
Latitudinal and Thermal Analogues in Other Countries
E. WASHINGTON, IDAHO, and UTAH-Physical Environment
and Agriculture
WASHINGTON, IDAHO, and UTAH-The Use of Phenology
in Ascertaining the Temperature Requirements of Wheat
Grown in Washington, Idaho, and Utah and in Some of
Their Agro-Climatically Analogous Areas in the Eastern
Hemisphere
NORTHERN GREAT PLAINS REGION-Prelimmary Study of
Phonological Temperature Requirements of c Few Varieties
of Wheat Grown in the Northern Great Plains Region and in
Some Agro-Chmaticolly Analogous Areas in the Eastern
Hemisphere
SOUTHEASTERN ATLANTIC REGION-Phenological Temper-
ature Requirements of Some Winter Wheat Varieties Grown
in the Southeastern Atlantic Region of the United States and
in Several of Its Latifudmally Analogous Areas of the Eastern
and Southern Hemispheres of Seasonally Similar Thermal
Conditions
ATMOSPHERIC AND METEOROLOGICAL ASPECTS OF AIR
POLLUflON-A Survey of USSR A,r Polluhon Literature
EFFECTS AND SrMPTOMS OF AIR POLLUTES ON VEGETA-
TION, RESISTANCE AND SUSCEPTIBILITY OF DIFFERENT
PLANT SPECIES IN VARIOUS HABITATS, IN RELATION TO
PLANT UTILIZATION FOR SHELTER BELTS AND AS BIO-
LOG 1C Ai I'sDICATORS-A Survey of USSR A,r Pollut.on
Literature
-------
AICE* SURVEY OF USSR AIR POLLUTION LITERATURE
Volume VI
AIR POLLUTION IN RELATION TO CERTAIN ATMOSPHERIC AND
METEOROLOGICAL CONDITIONS AND SOME OF THE METHODS EMPLOYED
IN THE SURVEY AND ANALYSIS OF AIR POLLUTANTS
Edited By
M.Y Nuttonson
The material presented here is part of a survey of
USSR literature on air pollution
conducted by the Air Pollution Section
AMERICAN INSTITUTE OF CROP ECOLOGY
This survey is being conducted under GRANT 1 R01 AP00786 - APC
AIR POLUTION CONTROL OFFICE
of the
ENVIRONMENTAL PROTECTION AGENCY
*AMERICAN INSTITUTE OF CROP ECOLOGY
809 DALE DRIVE
SILVER SPRING, MARYLAND 20910
-------
TABLE OF CONTENTS
Page
PREFACE * v
Maps of the USSR
Orientation vii
Climatic, Soil and Vegetation Zones viii
Major Economic Areas ix
Major Industrial Centers x
Principal Centers of Ferrous Metallurgy and Main
Iron Ore Deposits xi
Principal Centers of Non-Ferrous Metallurgy and
Distribution of Most Important Deposits of
Non-Ferrous Metal Ores xii
Principal Centers of the Chemical Industry and of
•tine Textile Industry xiii
'Principal Centers of Wood-Workingj Paper3 and Food
Industries xiv
Main Mining Centers xv
Principal Electric Power Stations and Power Systems xvi
PROPAGATION OF ATMOSPHERIC IMPURITIES UNDER URBAN CONDITIONS
M. Ye. Berlyand 1
DANGEROUS CONDITIONS OF POLLUTION OF THE ATMOSPHERE BY
INDUSTRIAL DISCHARGES
M. Ye. Berlyand 15
THEORY OF THE DEPENDENCE BETWEEN THE CONCENTRATION OF AEROSOLS
IN THE ATMOSPHERE AND THEIR FLOW ONTO A HORIZONTAL BOARD
M. Ye. Berlyand, Ye. L. Genikhovich, and
G. Ye. Maslova 28
METEOROLOGICAL OBSERVATIONS IN THE STUDY OF INDUSTRIAL
POLLUTION OF THE GROUND LAYER OF AIR
B. B. Goroshko, V. P. Gracheva, G. P. Rastorguyeva,
B. V. Rikhter, and G. A. Fedorova 42
CHARACTERISTICS OF THERMAL STABILITY IN THE GROUND LAYER
OF AIR
V. P - Gracheva 56
-------
BASIC PRINCIPLES OF ORGANIZATION OF THE SURVEY OF
ATMOSPHERIC POLLUTION IN CITIES
B. B. Goroshko and T. A. Ogneva 84
ORGANIZATION AND METHOD OF OPERATION OF ATMOSPHERIC
POLLUTION OBSERVATION POSTS
I. A. Yankovskiy, A. A. Gorchiyev, and
D. R. Monaselidze 98
USE OF STATISTICAL METHODS FOR THE TREATMENT OF
OBSERVATIONAL DATA ON AIR POLLUTION
E. Yu. Bezuglaya 105
STATISTICAL ANALYSIS OF DATA ON AIR POLLUTION IN
CITIES BY MEANS OF NATURAL FUNCTIONS
N. G. Vavilova, Ye. L. Genlkhovich, and
L. R. Son'kin 112
-------
PREFACE
Much of the background material presented in the prefaces to the
preceding volumes of this series is also relevant to the present volume.
The papers and data presented in this volume deal with a number of
aspects of air pollution developed under the wide range of environmental
conditions prevailing throughout the vast land area of the USSR with its
numerous extractive and manufacturing industrial enterprises. Some back-
ground information on the distribution of the Soviet industry's production
machine may be of interest in connection with that country's present and
potential pollution problems and investigations. The planned distribution
of production in the Soviet Union favors effective exploitation of the
natural resources of the USSR, especially in its eastern areas where enor-
mous natural resources are concentrated, and has led to the creation of
large industrial centers and complexes of heavy industry in many of the
country's economic areas (see page ix). The many diverse climatic conditions
of the country and its major economic areas as well as the geographical
distribution of the Soviet Union's principal industrial and mining centers
and of its principal electric power stations and power systems can be seen
from the various maps presented as background material in this volume.
Contamination of the natural environment constitutes a major problem
in all industrial regions of the USSR. The country's industry and trans-
port are continually bringing about massive qualitative changes in the
habitat of man and vegetation through an ever-increasing pollution of air,
soil, and streams. Pollution and the need to control it have become a
matter of great concern among Soviet conservationists and scientists and
they, like their colleagues in the West, have been warning their government
of the colossal and sometimes irreparable damage that is being done to the
environment and urging that serious and effective steps be taken to avert it.
Public awareness of the environmental crisis and the pollution problem
has been greatly stimulated in the USSR by the description, in the local
press, of such phenomena As dirty urban air, polluted rivers, ravaged for-
ests and public parks, and poisoned wildlife as well as by the revealing of
the causes of these conditions. "Pravda", the official Communist party news-
paper, stated in a recent article that "... we are turning the atmosphere of
our major industrial regions and large cities into a dump for poisonous in-
dustrial wastes". In the Soviet Union, like in the West, pollution now
poses for the leaders of the country some fundamental choices between the
economics of production, on one hand, and the progressively worsening living
conditions, on the other. There appears to be, at present, a greater appre-
ciation and a better understanding of the immense problems of air and water
pollution on the part of the urban and rural administrative agencies. As a
result of a mounting demand for the maintenance of a high quality physical
-------
taking shape in the USSR and much relevant air pollution research data
are being developed in the various industrial regions of that country.
Studies of atmospheric diffusion and air pollution constitute a
rapidly developing area of meteorological sciences in the USSR. Determ-
ination and analysis of the complex set of meteorological factors causing
the processes of atmospheric diffusion are being extensively developed
there in conjunction with theoretical and experimental studies of the pat-
tern of progagation and distribution of contaminants in the atmosphere.
Most of the material brought together in this volume deals with some
atmospheric and weather conditions as factors in the dispersal of air
pollutants in a number of the industrial regions of the USSR, regions that
are geographically far apart from each other and subject to different
natural and man-made environmental conditions.
A number of papers presented here deal with the basic principles
involved in the organization of air pollution surveys in cities. Other
papers consist of reports relating to the operation of air pollution obser-
vation posts and to the statistical methods employed in the analysis of the
observational data.
It is hoped that the papers selected for presentation in this volume
will be conducive to a better appreciation of some of the air pollution
investigations conducted in the USSR. As the editor of this volume I
wish to thank my co-workers in the Air Pollution Section of the Institute
for their valuable assistance.
M. Y. Nuttonson
March 1971
-------
ADMINISTRATIVE DIVISIONS
SSR
1 RS.FS.R
2 Karate-Finnish S.S.R
3 Estonian S.S.R.
4 Latvian S.S.H
5 Lithuanian S.S R
6. White RIASIBR SSR.
7 Ukrainian S.SR
a MMoavianSS-R
9 Georgian S.S R.
10 Armenian SSR
11 Aierbaydzhan SSR.
12 Kazakh SSR.
13 Uzbek S S.R
14 Turkmen S.S.R
15 T.dlhikSSR
16 Kirgu S.S.R
A.S.S.R
A. KomlASSR
B Udmurbkaya ASSR
C. Manyskaya ASSR
D Chuvaihskaya ASSR
E. Mordmkaya ASSR
F Tatarsfcaya ASSR
G Bashkirskaya ASSR
H Degestanskaya ASSR
1 Savero-Osetinskaya ASSR
K. Kabardmskaya ASSR
L Abkhazskaya ASSR
M AtMiarskaya ASSR
N Nakhichevenskaya ASSR
0 Kara Kalpakskaya ASSR
f Buryat Mongol skaya ASSR
-------
CLIMATIC ZONES AND REGIONS* OF THE USSR
^ ,X^>SK n-~- ^>^>:
-N/^*f /{*—*££^5
",-KK)5^fe
XV /•sy-^ --.—I
(A jv^^i
J >
-------
MAJOR ECONOMIC AEEAS OF THE U.S.S.R.
I North-Weslern
II Central
III Central Chernozem
IV Volga-Vyatka
V North Caucasian
XI Bailie
XII Soulh-Weslern
XIII Donets-Dnieper
XIV Southern
XV Transcaucasian
VI Volga
VII Urals
VIII West Siberian
IX East Siberian
X Far Eastern
XVI Kazakhstan
XVII Central Asiar
XVIII Byelorussian
PLANNED DISTRIBUTION OF INDUSTRIAL PRODUCTION IN ORDER
TO BRING IT CLOSER TO RAW MATERIAL AND FUEL SOURCES
An example of the planned distribution of industrial production in the USSR is the creation of large
industrial centers and complexes of heavy industry in many of the country's economic areas: the North-West
(Kirovsk, Kandalaksha, Vorkuta), the Urals (Magnitogorsk, Chelyabinsk, Nizhny Tagil), Western and Eastern
Siberia (Novosibirsk, Novokuznetsk, Kemerovo, Krasnoyarsk, Irkutsk, Bratsk), Kazakhstan (Karaganda, Rudny,
Balkhash, Dzhezkazgan).
Large industrial systems are being created - Kustanai, Pavlodar-Ekibastuz, Achinsk-Krasnoyarsk,
Bratsk-Taishet and a number of others. Ferrous and non-ferrous metallurgy, pulp and paper, hydrolysis and
saw-milling industries are being established in the Bratsk-Taishet industrial system. The Achinsk-Kras-
noyarsk industrial system is becoming one of the largest centers of aluminum and chemical industries, and
production of ferrous metals, cellulose, paper, and oil products.
Construction of the third metallurgical base has been launched in Siberia, and a new base of ferrous
metallurgy, using the enormous local iron and coal resources, has been created in Kazakhstan. A high-
capacity power system is being organized in the same areas. Non-ferrous metallurgy is being further
developed in Kazakhstan, Central Asia and in Transbaikal areas. The pulp and paper, as well as thp timber,
industries are being developed at a fast rate in the forest areas of Siberia and the Far East.
Ferrous metallurgy is also developing in the European part of the country by utilizing the enormous
iron ore resources of the Kursk Magnetic Anomaly and the Ukrainian deposits. Large new production systems
are under construction in the North-West, along the Volga, in the Northern Caucasus and the Ukraine.
(After A. Lavrishchev, "Economic Geography
of the U.S.S.R.", Moscow 1969)
-------
THE MAJOR INDUSTRIAL CENTERS OF THE USSR
^^^^^'•'•'•fM
^^i-.^Z-j&sjR
• Mam centres of ferrous metallurgy
non-ferrous metallurgy
O Centres of chemical industry
-------
KMA» Blipelsk
Karaganda .•*"•*
gograd Aktyubinsk
.
^-^=^5 A
rganels^^^ /^
^~^^r
Complete cycle metallurgy
QSIeel smelling and me la I
rolling
Smelling of ferroalloys
Mining of:
iron ores
coking coa!
manganese ore's
I1AIN IRON ORE DEPOSITS IN THE U.S.S.3
40 60 80 TOO 120 140 160
(After A. Lavrishchev, "Economic Geography of
the U.S.S.R.", Moscow 1969)
-------
PRINCIPAL .^CENTERS OF NON-FERROUS METALLURGY IN THE U.S.S.R.
40 60 80 100 120 NO 160
DISTRIBUTION OF MOST IMPORTANT DEPOSITS OF NON-FERROUS METAL ORES
40 60 80 100 130 140 160
O Cold
Ft Plalinum
• Copper ores
Tin ores
Complex ores
Ni Nickel ores
B Bauxites
H Nephelines
A Aluniles
M Mercury ores
xii
(After A. Lavrishchev, "Economic Geography
-------
PRINCIPAL CENTERS OF THE CHEMICAL INDUSTRY IN THE U.S.S.R.
Chemical ind
Oil-relining i
© Production o
O 'Production o
PRINCIPAL CENTERS OF THE TEXTILE INDUSTRY
IN THE U.S.S.R.
40 60 80 100 1?0 H.O
Figures on Ihe map sho
12 Vichuga
13 Tbilisi
14 Kirovabad
15 Nukha
16 Margelan
11 Bryansk 17 Noginsk
O Silt induslry
O Olhcr branches ol Ih
Xlll
(After A. Lavrishchev, "Economic Geography
-------
PRINCIPAL CENTERS OF WOOD-WORKING AND PAPER INDUSTRIES IN THE U.S.S.R.
ej^»Ri9r«!?
^opluovTk f ' ^HM!;^^
i|^SarL©;JE ©-
Volgograd KuibyshevdDu(
Industry:
Timber-sawing and wood-working
© Paper
[^ 1 Principal lumbering areas
Foresls
PPINCIFAL CENTERS OF THE FOOD IWDUSTFY IN THE U.S.S.P.
40 60 80 tOO 120 140 160
nduijry
O Fish »nd lilh packing induilry
(After A. Lavrishohev, "Economic Geography
of the U.S.S.R.", Moscow 1969)
-------
THE MAIN MINING CENTERS OF THE USSR
Oil refining
Oil pipes
Gas pipes
Power stations
-------
PRINCIPAL ELECTRIC POWER STATIONS AOTJ POWER SYSTEMS IN THE U.S.S.R.
Principal Electric Power Stations
Hydro-power
in operation
under construction
and planned
Groups of electric
power stations
^ Operating atomic electric power stations
Areas of operation of single power grids
European part of the USSR
Central Siberia
Areas of operation of integrated power grids
and We"' KS5S Northern Kazakhstan
Caucasus Illlllll Central Asia
nlaii
Lower Tunguska
S^Uya\--lfku*>
_ ^s^-^it, a^
&w r 1 fl >
S^'i V,J1> •*
igures indicate following power stations.
9 Dnieprodzerzhmsk 17 Shatura
tO Dnieproges 18 Elektrogorsk
11 Kakhovka 19 Ivankovo
•••G~*T
1 Baltic
2 Narva
3 Kegum 11 naxnovica i» ivanxovo «p~
4Plavmas 12 Slarabeshevsk 20 The 22nd C P.S U. Congress S I)*
S Novaya Byeforusslcaya 13 Zuyevskaya HEPS on the Volga 07 rktoa I
iSDubossary 14 Shlerovka 21 The Lenin HEPS on the Volga M 1luOSi
7Kanev IS Krasnodar 22 Chardarmskaya * 3M 0
B Kremenchug It Kashira 23 Chirchik-Bozsu **?".. i
340-km
(After A. Lavrishehev. "Economic Geography
-------
PROPAGATION OF ATMOSPHERIC IMPURITIES UNDER URBAN CONDITIONS*
Professor M. Ye. Berlyand
From Meteorologiya i Gidrologiya. No. 3, p. 45-56, (Mart 1970).
The article discusses some aspects of the variation of meteorological conditions in cities
(formation of a "heat island" in the center of the city, increase in the frequency of fogs,
elevated inversions, surface calms, etc.). It is noted that the majority of completed theoret-
ical studies of atmospheric diffusion pertain to the conditions of an open, featureless land-
scape. The article explains to what extent the results obtained are applicable to the evaluation
of impurity dispersal under urban conditions. It analyzes the characteristics of atmospheric
diffusion and the role of factors determining the dispersal of impurities from high and low urban
sources.
The analysis of the influence of meteorological factors on the study
of impurities in the atmosphere is being given an increasing amount of
attention at the present time. Knowledge of the relationships governing
the spreading of an impurity is important for the formulation of recommend-
ations aimed at protecting the air reservoir from pollution for the purpose
of creating conditions where the concentration of pollutants would not ex-
ceed the permissible values. These values form the basis of an efficient
organization of control of atmospheric purity, and in particular, the basis
of the principles used for selecting a representative location and time of
observation and analyzing the data obtained.
The solution of these problems, which pertain to urban conditions, in-
volves the consideration of at least two key features. One is that under
certain conditions, urban factors alone can have an appreciable effect on
the meteorological conditions; the other has to do with the necessity of
evaluating the overall effect of a large number of different polluting
sources.
Studies of atmospheric diffusion of impurities and the methods developed
for calculating the dispersal of discharges into the atmosphere pertain for
the most part to individual sources and to conditions of an open flat country.
The problems involved in the consideration of the total effect of a group of
sources and of the change of the meteorological regime in the city have re-
ceived relatively little study. Nevertheless, the results obtained for an
open region are necessary, first of all, for evaluating the air pollution of
populated areas. For this reason, they are frequently used formally without
special basis also for the conditions of residential and, in particular, urban
areas.
*Based on data of a paper given at a symposium on urban climate in Brussels (October, 1968).
-------
To explain this possibility, it is necessary to consider the results
of studies of urban meteorological conditions. An extensive literature on
this problem is now available [13, 20, 34, 41, etc.], so in the present
paper we shall attempt to discuss only some basic conclusions having a
direct bearing on the problem under consideration.
It should be noted first of all that air pollution in many cities has
reached such proportions that it is of itself one of the basic causes of
the variations of the meteorological conditions mentioned here.
Dust frequently accumulates over a city, thus decreasing the trans-
parency of air and reducing the solar radiation by 10-20%, and sometimes
even more. Such an attenuation of the solar radiation, including its ultra-
violet portion, is an important indicator of the adverse effect of air pol-
lution, and should be considered by hygienists. The radiation effect is
significantly related to changes in the heat balance and temperature con-
ditions of the ground layer of air.
Many climatological studies have established that air in cities is on
the average 0.5-1°C. warmer than in the surrounding area, this being fre-
quently referred to as the "heat islands." In [13, 20, 30, 34, 41], a number
of characteristics of these "islands" and their relationship to weather con-
ditions have been discussed. It is shown that a temperature rise in cities
is observed mainly at night in the presence of a slight wind and an almost
cloudless sky-, most frequently in winter, and almost never in daytime hours.
A decrease of the temperature differences with increasing cloudiness and also
on Sundays, when the air pollution is less than on working weekdays, indicates
a direct relationship between the "heat island" and radiation factors. This
relationship, noted by many authors, has received little study from the stand-
point of the physics of the phenomenon and its quantitative evaluation. Leav-
ing aside the problem of slight variation of the diurnal temperature maximum
as the solar radiation is reduced, the increase of the nocturnal minimum temp-
erature is frequently attributed to an attenuation of the long-wavelength
radiation by urban aerosols. However, there are as yet no reliable data on
the change of the long-wavelength balance in cities. Some estimates of the
radiation effect have been made on the basis of theoretical investigations
of the daily temperature variation [1, 27] for an open region in the presence
of vegetation.
It follows from the results of calculation that a decrease in solar
radiation, all the other conditions remaining the same in the case of clear
weather, may lead to a 2-3°C. increase of the nocturnal temperature minimum
in the ground layer of air. At the same time, the diurnal temperature maxi-
mum also decreases, but, because of the advective transfer and reinforcement
of turbulent exchange in the daytime hours, this decrease is somewhat less.
The existing experimental data on the temperature conditions in a city
pertain chiefly to the ground level. In the last few years, observations
from television towers and special aerological observations have led to some
-------
conclusions regarding the vertical temperature profile above a city. It
turns out that in daytime hours, this profile is close to that over an
open region. In the presence of a "heat island", the temperature strati-
fication in the layer of air up to a height of several tens of meters is
close to the equilibrium stratification or slightly unstable, whereas out-
side the city an inversion is observed during that time [34, 37, 41, etc.].
Consequently, the formation of elevated inversion layers is more probable
above a city.
In the streets and between buildings, the velocity and direction of
the wind change considerably. In the general case, it is difficult to pin-
point any definite patterns to these changes, since they depend considerably
on the specific elements of the city's structure. However, above the city's
buildings, the wind profile and its variations with time acquire the features
prevailing over an open region relatively rapidly. As follows from data now
being accumulated from observations made from television towers, the influ-
ence of the city manifests itself chiefly as an effect of increase in the
roughness of the underlying surface. Because of the diversity of the forms
of urban construction, considerable possibilities lie in the simulation of
the flow around them and the use of wind tunnels.
In this connection, interesting results have been obtained in joint
studies made by the Main Geophysical Observatory and Moscow University [16],
dealing with the distribution of the wind velocity and turbulence for various
types of urban construction. They show that the wind velocity changes with
the height, and more so in the presence of unlike buildings with different
numbers of stories than in the case of construction with the same number of
stories. On the average, the wind profiles are similar in both cases to the
results of observations under natural urban conditions. Such conclusions
pertain primarily to cases with well-defined horizontal flows of air over
the city. In the presence of slight winds, the nature of the air currents
may be substantially determined by the presence of a "heat island". The
theory of this phenomenon has not yet been worked out. It follows from
general considerations that a convective circulation arises in this case, and
in the ground layer the wind velocity is directed toward the center of the
city. The presence of these currents and their velocity, which amounts to
2-3 m/sec, have been confirmed. These results pertaining to variations of
the wind temperature and velocity make it possible also to evaluate the
nature of changes in the turbulent exchange over the city.
Summarizing the above, one can conclude that as a result of horizontal
transfer and an intense vertical exchange of air over the city, there are
frequently produced temperature and wind velocity distributions and hence
turbulence coefficient distributions that are similar to those over an open
region. Only in isolated cases are special conditions created which are
unfavorable from the standpoint of atmospheric diffusion of impurities and
require a special analysis.
-------
This makes it possible to carry out an approximate calculation of
the dispersal of impurities in a city (mainly, from fairly high sources)
without a detailed consideration of the urban construction. In many cases,
this approach can be validated by using the characteristics of diffusion
of impurities from high sources. Analysis shows that as the distance
from the source increases, the vertical impurity concentration profile is
transformed in such fashion that in the zone where the ground concentration
maximum is reached, the impurity becomes almost uniformly distributed in
height. Consequently, even if the construction types do distort the con-
ditions of mixing, they still cannot cause a considerable redistribution
of the impurity (in this zone). As an example, we can refer to experimental
studies in the region of the Shchekino SEEPP (State Regional Electric Power
Plant) [22]. Here, for the same wind directions, a change in the concen-
tration of the impurities discharged from the SREPP's stacks was carried
out in the city, and for other directions, over an open region. The results
obtained showed that the influence of the city was slight.
Special experimental studies [40] dealing with the spreading of a
tracer under urban conditions were made in St. Louis, USA. They showed
that during the course of a day, the dispersions of impurities in the hori-
zontal and vertical directions at a distance of 10 miles from the tracer
source over the city and outside it differed little.
Thus far, (a) many studies have been made with the aim of developing
various schemes and formulas for calculating the concentration of impurities
around the sources [21, 41, etc.]. The effectiveness of the development of
these studies is closely connected with the integration of the atmospheric
diffusion equation:
for suitable boundary and initial conditions.
Here u is the wind velocity,
w is the vertical velocity of propagation of the impurity,
ky, kz are the horizontal and vertical components of the volume
coefficient, respectively,
axis x is oriented along the direction of the average wind, axis y along the
perpendicular to axis x in the horizontal plane, and axis z along the vertical,
In a city, because of the relatively large area it covers and also be-
cause of the presence of high sources, the layer of air in which the main
transport of the impurity takes place extends to a height of several meters
and higher. In such a case, and especially in the presence of elevated temp-
erature inversions, the coefficients of (1) may be very complex functions of
the coordinates. For this reason, considerable importance is assumed by
-------
In calculating the dispersal of pollutants in the atmosphere, it
is necessary to consider the initial ascent of the impurity above the
stack, caused by the entrainment velocity and the overheating of the
stack gases. Because this ascent is a function of the wind velocity and
of the meteorological factors, the dependence of the ground concentration
on the weather conditions assumes an even more complex character. In par-
ticular, if we deal with the influence of the wind velocity u, then, on the
one hand, at a fixed height of the discharge, the ground concentration q
decreases with increasing u. On the other hand, a strengthening of the
wind leads to a decrease of the initial ascent, as a result of which q in-
creases. Consequently, there exists some unsafe velocity i^ at which the
highest value of the maximum ground concentration C is reached.
If the discharge sources are located near the ground, the maximum
concentrations are reached in the presence of ground inversions character-
ized by a weak turbulent exchange. When the impurities are discharged
from stacks, the highest concentrations Cm near the ground are reached under
conditions of convection with a developed turbulent exchange causing an
intense transport of the impurity from the stacks downward, into the life-
sustaining layer of air. In accordance with the theoretical studies cited
[5, 11], the value of (^ for a uniform discharge of impurities into N stacks
located close to each other may be determined from the formula
(2)
v '
Here M and V are the amount of impurity and volume of stack gases discharged
per unit time,
H is the stack height,
F and m are dimensionless coefficients. F depends on the settling rate of
the impurity. F=l for a light impurity and F > 1 for a heavy impurity. The
value of m depends on the characteristics of the ejection of the gases from
the stack. Coefficient A defines the influence of the vertical and hori-
zontal distribution of the air temperature at the unsafe wind velocity, when
the maximum \ralue of the ground concentration is reached. Large values of
A correspond to regions with a pronounced continental climate, which are
characterized by an intense turbulent exchange as a result of large supera-
diabatic gradients in the summertime.
The value of the unsafe velocity is round from the formula
(3)
To evaluate the total effect of pollution from a group of sources with dif-
ferent unsafe velocities, it is desirable to determine the weighted mean
value [11]:
1-1
-------
where u.., and V are, respectively, the values of the unsafe velocity and
highest concentration for the i-th source.
In the general case, when the sources are not grouped around a point
or a straight line, as is frequently the case in large cities, it is neces-
sary to sum up the concentration fields of the individual sources. It is
necessary to consider the different wind directions since they determine
the change of the relative positions of the sources. Such computations,
particularly for cities with large numbers of sources, are extremely cum-
bersome. It is possible to simplify them to a certain extent because
the concentrations in the direction perpendicular to the wind decrease
much faster than those parallel to it. A computer program has been written
for the computations in the case of a large number of sources according
to the above [11].
The theoretical results cited pertaining to the patterns of the spread-
ing of impurities from their sources have been confirmed by an extensive
experimental material [17, 18, 21]. They were used as the basis for the
development of a procedure and recommendations for calculating the dispersal
of impurities in the atmosphere in connection with the design and operation
of industrial enterprises [14, 25].
The results pertained to comparatively frequent unfavorable meteorolog-
ical conditions. The calculations made use of the logarithmic law of
variation of the wind velocity with height, and for the temperature, of its
decrease with height, so that the exchange coefficient increased linearly with
height in the ground layer and remained constant above it. At the same time,
the turbulence is vigorous and an intense mixing takes place between the
layers of air located above the stacks and those adjacent to the underlying
surface.
The theoretical studies performed indicate, in accord with earlier
studies by Hewson and others, that ground concentrations may reach even
higher values in the presence of elevated inversion layers with an attenuated
turbulence. This point is particularly essential in connection with the
above-mentioned tendency toward an increased frequency of elevated inversions
over a city.
It follows from the calculations of [3, 6] that under conditions where
a layer with an attenuated turbulence is located immediately above the
sources, the maximum concentration of light impurities sometimes increases
by a factor of more than 2. However, in cases where such a layer is located
at a height of 100-200 m above the sources, the concentration increase is
much less. It should be noted that in the presence of a layer with an atten-
uated turbulence, there is not only an increase in the concentration maximum,
but also a considerable increase in the area where it is observed. Moreover,
the decrease of the concentration past its maximum takes place very slowly.
Hence, as was noted in [9], the effect of mutual superposition of concentra-
tion fields from individual sources is enhanced. In addition, even if the
concentrations from single sources are comparatively small, the city's total
-------
pollution caused by a large number of these sources may be very considerable.
Elevated inversions may have a much greater effect in the case of cold
discharges [7]. Under such conditions, the initial ascent does not exceed
a certain limit, independently of the decrease of the wind velocity and,
hence, the value of the unsafe velocity decreases, and the ground concentra-
tions increase abruptly.
Under urban conditions, it is essential to take into account the devia-
tions in the vertical wind profile from the logarithmic profile, in particu-
lar, a possible attenuation of the wind velocity to a calm in the lowest
layer of air. Such deviations in cities are because the buildings cause
the air current to slow down. Up to the level of the height of buildings,
the average wind velocity may frequently be close to zero, whereas turbulent
exchange is fairly developed here. On the other hand, above the buildings,
the wind velocity increases rapidly in approximately logarithmic fashion.
The calculations performed [6, 28] show that the presence of still layers
near the underlying surface leads to a substantial increase of the concen-
tration.
Of major importance in evaluating the meteorological factors of urban
air pollution is the analysis of the influence of fogs. They and their
modification, the smog, are held responsible for cases of mass morbidity and
increased death rate in cities, since the frequency of fogs under urban con-
ditions is substantially higher than in rural areas. Their chief cause is
heavy air pollution. This is not only a question of an increased quantity
of condensation nuclei (analysis shows that there are enough of them for
the formation of fogs, provided that moisture saturates the air outside
the cities as well), but rather that the urban impurities contain a consid-
erable amount of hygroscopic particles. The moisture condensation on such
particles may begin at a relative humidity below 100%, and hence, the proba-
bility of formation of a fog increases.
The condxtions of air pollution in the presence of fogs have received
little study. The available experimental material has been inadequately
analyzed. In [9], on the basis of the indicated numerical analyses, some
theoretical aspects of the diffusion of gaseous impurities in the presence
of fogs have been discussed. Fogs forming on the banks of rivers and
water reservoirs have been investigated. The height of the fogs and the
vertical and horizontal distributions of the moisture content and exchange
coefficient were determined theoretically. Also studied were cases of radi-
ation fogs, bearing in mind that an elevated temperature inversion might
be located above them. The calculations revealed some interesting effects,
in particular, the fact that in addition to a redistribution of the pollu-
tants because of their absorption by water droplets, there occurs a substan-
tial increase of the ground concentrations as a result of the transport of
-------
Urban fogs are formed more frequently in the mornings, thus increas-
ing the probability that they will be associated with elevated inversions.
This in turn reinforces the effect of air pollution, particularly from many
sources.
Topographic inequalities may have a substantial influence on the spread-
ing of an impurity in a city. Under hilly topographical conditions, the
character of the motion of air changes considerably. The use of modern methods
of theoretical analysis has provided an approach to the solution of this diffi-
cult problem.
Thus far, calculations have been made for individual examples of a hilly
topography. According to [6, 12], it has been found that under such condi-
tions, the maximum ground concentration is mostly higher than on level ground.
For a height of irregularities in excess of 50-100 m with slope angles of
about 5-6° to the horizon, the difference in the concentration maximum is as
high as 50% or more, depending on the location of the source in the different
forms of the relief. An increase of the concentration is sometimes observed
even when the pollution sources are in high locations, but the latter are in
the vicinity of leeward slopes, where the wind velocity decreases markedly
and descending currents are generated.
Analysis has shown that in the case of smooth relief forms, the latter
are almost completely surrounded by air currents, and an increase of the
concentration is manifested in areas where the wind velocity changes substan-
tially at a fixed height. The combination of the boundary layer method and
the method of construction of potential flows has provided an explanation
for certain characteristics of the velocity field necessary for calculating
the turbulent diffusion [12]. The wind tunnel experiments briefly discussed
above were set up under close to self-similar conditions, permitting the
extension of the results of the simulation to atmospheric processes. As a
result, data were obtained on the vertical wind profile in various parts of
the relief depending on the slope angle of the underlying surface, the height
drop, etc. [16, 19, 28].
In modelling studies conducted in wind tunnels, there is a certain
common approach to the investigation of the flow around the irregularities
of the relief and to the detetmination of the structure of the air current
around residential and industrial structures. In both cases, perturbations
in the field of vertical and horizontal velocities are studied. The experi-
ments cited, performed on models of individual industrial enterprises and
buildings, revealed the zones in which descending currents and stagnations
of the impurity are possible.
In order to avoid a substantial increase in the concentration of the
noxious substances discharged from the stacks in such zones, Khokinsom and
Naneblom and others recommend that the stacks be 2.5 times as high as the
nearest building.
-------
The results mentioned above, pertaining to the wind profile and turbu-
lence, obtained with models may be used directly in accordance with the
above scheme of numerical analysis of atmospheric diffusion. For the study
of the transport of impurities from low sources, it is convenient to deter-
mine the velocity and direction of the air currents in the streets and be-
tween buildings. To this end, the studies made by the Main Geophysical
Observatory and Moscow University dealt with the ground flow field on models
of urban constructions. It was found that the character of the motion in
different parts of the city blocks depends not only on the direction of the
wind over the city but also on the degree of turbulence of the ai;r flow.
The above formulas made it possible to calculate the values of the
impurity concentration pertaining to a twenty-minute withdrawal of air
samples. In the derivation of the original formulas, particularly for
Cjn (2) , account was taken of the probability ) of deviation of the wind
direction with time in the horizontal plane at angle tp, usually described by
the Gauss law [5, 15]:
If the duration of the sampling, is increased, the dispersion of the
oscillations of the wind direction
-------
case.
t£», the impurity may fall at a point for any wind direction. The con-
centration at this point over a long period of time will be determined as
the result of the combined action of all the surrounding sources. This
principle underlies a series of simple schemes for calculating air pollu-
tion in a city, for example, in the work of Clarke [31], and others.
A somewhat different scheme, but essentially similar to the above, was
developed by Turner [43], Pooler [39], and others. The authors divided the
city into a grid of squares and carried out an approximate evaluation of the
discharge of pollutants in each square (inventory of the sources). Then,
at a point of the city under consideration, they determined the total sur-
face concentration produced as a result of the action of the sources sur-
rounding that point, assuming their height and the initial ascent to be
the same all over the city. Lucas [36], Miller and Holtzworth [37] and
others treated the city as a set of fine sources distributed in comparatively
uniform fashion over the urban area. The total concentration near the ground
was determined by integrating the expressions for the concentration from a
point source at a certain average height over the area. In an evaluation
of the concentration from a flow of moving automobiles, Neiburger [42] treats
this as a stationary plane source.
The results of the theoretical studies discussed provide certain grounds
for predicting the degree of air pollution. On the basis of forecasts of
weather conditions, in particular, elevated inversions, calms, fogs, and
also intense turbulence, in the case of comparatively high sources, one can
identify the periods when large values of the ground concentration of pollu-
tants should be expected.
It should be noted at the same time that, despite the considerable
possibilities of application of the theory of atmospheric diffusion to the
calculation and forecasting of the spreading of impurities in a city, sub-
stantial difficulties still exist. They are due to the scantiness and lack
of certain data, including data on the nature of discharges and on a number
of meteorological parameters necessary for such calculations. It is also
necessary to take into consideration the partially random nature of the
operating factors.
For this reason, the organization of the most complete possible set of
observations in a city and a physical-statistical analysis of the data ob-
tained assume an essential importance. In many cities throughout the world,
areas have now been set aside for systematic observations of atmospheric
pollution. Special pavilions are being set up for this purpose in the USSR;
they are placed at intersections of streets or in city squares in areas of
highest air pollution. Moreover, observations are made several times a day
on the concentration of the most common ingredients and a number of meteoro-
logical elements. Automobiles are also used to measure the concentrations
along certain routes and under the plumes of high-capacity industrial plants.
A number of cities have already collected relatively large amounts of data
which permit the analysis of certain patterns of variation of the maximum
concentration, frequency of heavy air pollution, etc. [10, 26].
-------
In the USSR, as in other countries, the heaviest air pollution is, of
course, observed in large industrial centers. Moreover, observational data
indicate that urban air is most heavily polluted in areas where unfavorable
weather conditions prevail [8, 23]. According to measurements of the con-
centration of dust, sulfur dioxide, etc., such regions show maximum concen-
tration values and high frequencies of days when the concentration exceeds
the maximum permissible values. At the same time, these areas are character-
ized by the development of intense turbulence, which causes the ground con-
centrations in regions of large and high sources to increase. In the calcu-
lations of the maximum concentrations, the values of coefficient A in formula
(2) are assumed to be the highest for these areas.
It is common knowledge that a considerable portion of ash and sulfur
dioxide enters the atmosphere as a result of the combustion of fuel, the fuel
consumption being much lower in summer than in winter. Nevertheless, in many
cities the concentrations of these impurities were observed during the warm
half of the year in the presence of an intense turbulent exchange [23].
According to observational data, particularly for regions of low and cold
impurity sources, there are, as was indicated above, unsafe pollutiop levels
under temperature inversion and air stagnation conditions as well.
Analysis of the experimental data is also used to study the influence
of the synoptic situation, washing out of the impurities by precipitation,
etc. It is of interest to study cases in which either an increase or a
decrease of the concentration of one of several ingredients is simultaneously
observed at a large number of points in the city. It turns out that cases
of heavy air pollution, particularly during the cold half of the year, are
mostly observed during stationary anticyclones in the zone of low atmospheric
pressure gradients. A comparatively clean air is observed in cyclonic
weather. This is illustrated by Table 1, compiled on the basis of observa-
tional data for Moscow, Leningrad and Magnitogorsk [24].
Table 1
Deviation From Average Frequency of the Impurity Concentration Greater than
the Maximum Permissible Value During Stationary Anticyclones, %.
City
Moscow ....
Leningrad
MagnitogWrsfc
DUE
Cold-
Period"
-1-7
+21
+34
5t
Warm
PefftSf
+ 1
+13
+21
Sulfur Dii
Cold
Period'
+18
+13
+19
MtidS-
Warm
Period
-2
—1
+4
The studies performed, essentially on a statistical plane, pertain
basically to the investigation of the correlation dependence between the
concentration of the impurity at one or several points and individual fac-
tors, provided that the influence of other factors can be kept constant.
Estimates of the correlation factors, analyses of certain structural func-
-------
the indicated studies it was found that for a number of cities, the corre-
lation factor between the frequency of the concentration above the per-
missible values in the course of a month and the monthly pressure anomalies
was about 0.5. In [33] it was found that under certain urban conditions.
the correlation factor between the dust concentration and certain meteoro-
logical elements was higher than between this concentration and the amount
of fuel burned.
At the present time, several approaches to a more complete statistical
analysis of observational data are being developed at the Main Geophysical
Observatory. One approach involves the study of a multiple correlation,
i.e., taking into account the influence of a set of factors. Here the
separation of the principal factors is not always successful, and the effec-
tiveness of this approach proves limited. Another approach consists in
using the method of expansion in a statistically orthogonal system, that is
to say, in real functions of the concentration field of the impurity. To
this end, the covariant matrices are calculated from data on unnormalized
correlations between the concentrations at different points of the city,
and the eigenvalues are obtained. The first terms of the expansion reveal
the influence of the set of principal factors. The study of the behavior
of these terms with time may serve as the basis for a statistical prediction
of urban air pollution. In order to exclude random effects, first a certain
averaging of the coefficients over time and a normalization of the measured
concentrations to their average values over a sufficiently long period of
time are carried out. Analysis of variance and factor analysis may also be
useful in the study of the concentration field of impurities in a city.
In addition, it should be kept in mind that urban air pollution is
caused by the action of a very large number of factors, and the observa-
tional data are limited. For this reason, the use of even the most perfect
modern methods of statistical analysis, particularly for purposes of fore-
casting air pollution, is inadequate, and in order to increase their effec-
tiveness, it is important to separate the influence of a number of principal
factors on the basis of physical considerations.
In the study of the weather conditions associated with urban air pol-
lution, it is also very important to examine the problem of the spreading
of the impurity beyond the city limits, both vertically and horizontally.
Of late, in addition to ground observations, the impurity concentrations
have also been measured by means of helicopters and airplanes [18, 35].
Thus it was found that several concentration maxima may be observed in the
vertical distribution of an impurity. In particular, high concentrations
of carbon monoxide were found over Leningrad and Budapest at heights of
100-300 m. This is due to discharges of carbon monoxide from high stacks,
and also to the characteristics of the temperature stratification, which
determines the variations of the exchange coefficient in height. The study
of the vertical distribution of an impurity is now assuming a special
interest in connection with the construction of air intakes that supply
clean air to industrial plants [10].
-------
The spreading of an impurity from the city as a whole is now being
followed over distances of several tens of kilometers. Given the modern
tendency toward urban sprawl and the merging of cities in many countries,
there is danger of polluting vast territories extending over hundreds of
kilometers.
The results discussed above demonstrate certain advances in the study
of the relationships governing the spreading of impurities in the atmos-
phere. Nevertheless, the investigation of many important aspects of urban
air pollution is only beginning, and their solution will require a further
development of the theory of atmospheric diffusion, physical-statistical
analysis, and experimental studies under natural and laboratory conditions.
LITERATURE CITED
I. Bep-mna M E. FIpeACxaaamie H pery;mpOBaHne Ten.ioBoro peaonia
c.ios aTMOHpepbi. FjupoMeieoHsaaT, JI , '1956
2 Bep.iHHj. M. E K TeopHH aTMoccpepHofi .iHtJxpysHH — Tpyjbi JTO, sun 138, 1963
3 B e p ;i H H A M E.FeHHXOBHqE. JI.JIojKKHHa B iFI , O H H x y .1 P H. MKC-
.icHHoe jiccjiejoBamie aTMOccpepnofi jHtptpyaiiii inpii Hop\iajibHbi\ M aHoMa.ibHbix
>C.IOBIIH\ CTpaimpHxauHH — Tpvaw ITO, BHII. 158, 1964.
4 B e p .1 H H A M E.iFeHHXOBHqE JI.JIoxxiiHaB 'FI , O H H K v A P. H. Oco-
OCHHOCTH jncp(py3mi Tsme.iOH inpiiMecH a aiMoccpepe — Tpyau FFO, sun 156,
1964
5 B e p .1 a H j M E.reHHxoaiiqE .H.OHHhy^P H O pacMere 3arp«3HeHjm
aT\iuc4>epbi Bbi6poca\iH 113 JUMOBUX ipy6 3^ckTpocTaHunfi — Tpyabi FFO,
sun 158, 1964
6. E e p .1 a H a M E.FeHiixoBHqE JI,/[eMi.flnoBHqB iK HeKoropue aKTva^b-
Hbie Bonpocu Hcc.ieAOBaHHH ai\ioc(J)epHoi"i AH(J)*V3iiH — TPVAH FFO, 3bm. 172,
J965.
7 Bep.iana M. E O6 onacHwx VC.IOBHH\ sarpnsnemia arvocejiephi npOMbiiue4>a
iia pacnpocTpancHiie .npiiMeor OT HCTO'iiiiiha — Tp\aw FFO, nun 234, 1968
13 Beep B TeMiiiqecxaa Mcrcopoionia FiupOMCTCoiisaar, JI , 1966
'4 BpcMeniiau MCTO,niha pacicioB pacceiiBaHHH B atMocthcpc nH6pocoB (ao.ibi H ccpiiH-
CTWX raaofl) H3 rp\6 siehTpocraHnuft — TP\AH FFO, awn 172, 1965
15 FCHH. \OBnq E JI, Fpaqeaa B FI Ana.iHS jiionepcim ropHSOiiTa.ibiibix xoji«6a-
imfi nanpaB.iennn serpa — Tp\jbi FFO, Bbin 172, 1965
16 FopiiiH C M, 3 p a >K e B c K 11 i"i H M Ha^eniie oCTCkaima Moae.iefi pe.ib(J>a H
ropoacxofi aacrpoflxH B aapoaiiHaMimcchofl rpyBe — TPVAH FFO, sun 234, 1968
1T FopouiKO B B, naHtfeH-ioBa F A, .FH^bjcHChHO.Tbji P C., PHX-
rep B B PesyjJbiaTbi HaGnoAeimfi 3a 3arpH3HeHH6M aT^50c(pepH oKHC^iaMH 330-
Ta or xiiMHiecxoro 33Boaa — Tpyjbi FFO, Bbin 185, 1966
13 Fopouixo B B , 3 a H u e B \ C, Haaapi'HKo B H Bonpocw MCTOJIHKH H pe-
syibTaTbi jkc.ieaoBaHHH 3arpH3iieHiid aiMoc(J)epbi c noMombro BepTO.iera. —
FFO, Bbin 234, 1968
-------
itf arpasKCBCKHff tl iM, ,ZlopoiuenKo iB H, MCITIHK H T. HccjieaoBamie MHJI-
HHH pasjiimHbtx (JjopM pe.ibed>a na xapaKtepHCTjiKH BOsayuiHoro noroxa B aspoan-
HaMimecKOii TpyGe — Tpyjbi rfO. awn 207. 1968.
20 Kpa-ruepII A iK-iimar ropoaa HJI, M, 1958.
21 Meieopo.7ornsi H aTOMnaa SHeprHH riep. c a»rn HOA Pea E K Oefloposa HJI. M..
1969.
22 OmiKyjiP. H,naH e fl o p o B E K 'Hexoropbie JITODI BbinojineHHSi n^ana
Kenn5i Tevineparypbi B ca^y Mereo-
poflorHa H rajipoJiarHH, iNb 12, ,1965.
28 BerlyandM E To the Theory of the Industrial Emission Dispersion in the
Atmosphere of a .Coastal Zone The 11-th Pacific Sc Congr Symp on Air and
Water Pollution in the Pacific Area Tokyo, 1966 — Idojaras, v. 71. No 2. 1967.
29 BerlyandM E Theoretical Investigations of Transporting Aerosols in the
Boundary Layer of the Atmosphere General Assembly of the International
Union of Geodesy and Geophysics Symp on Tropospheric Aerosols and
Radioactive Tracers Lucerne, 1967
30 ChandlerT J Night-time temperature in relation to Leicaster's urban form —
Met Mag v 96, No ,M41. 1967
SI ClarkcJ T A Simple Diffusion Model lor Calculating Point Concentrations from
Multiple Sources — J APCA, v 14, No 9, 1964
''.I Frenkiel F, Sheppard P Atmospheric Diffusion and Air Pollution N Y
London, 1959
33 Dickson R R Meteorological Factors Affecting Particulate Air Pollution of
a City — Bull Am Met Soc v 42 No 8, 1961
3» KawamuraT Urban Climatology in Japan Tokyo — J of Climatology, v. 3,
N'1 I, '1966.
35 Kelenffy J. M6nk J, Varkonyi T Vertical Distribution of Town Air
Pollution — Idojaras, v 71, No 5, 1967.
36 Lucas D H. The Atmospheric Pollution in Cities Int Journ. Air and Water PoH
v, I, J958
J7 M 1 1 1 e r M t , H o 1 1 z w o r t h G C An Atmospheric Diffusion \Aodd fnr
Metropolitan Areas —Journ. APCA, v 17 No 1 1%7 for
™
39
Noi2 1966A Traeef StUdV °f D'Spersion over a CltV -Journ. APCA. T. 11.
41 Stern A Air Pollution v 1, N Y. London, 1962.
42 Symposium Air over Cities US Public Health Service —Tech Rep NoA62 5
No".
rco()jn3imccKa»
oficepaaropKi
5 VIII 1969
-------
DANGEROUS CONDITIONS OF POLLUTION OF THE ATMOSPHERE
BY INDUSTRIAL DISCHARGES
M. Ye. Berlyand
From Trudy, Glavnaya Geofiz. Observat. im. A. I. Voeykova, No. 185,
p. 15-25, (1966).
The initial ascent of impurities from smokestacks is studied as a function
of the coefficient of turbulent exchange and temperature gradient in the atmos-
phere. Conditions are indicated for which the initial ascent of the impurity in
the_presence of a_temperature inversion may be slight. In cases where such con-
ditions are associated with low wind velocities, an abnormally dangerous situation
arises in which the surface concentrations of the impurity reach very high values.
1. Introduction
The surface concentrations of impurities discharged through smoke-
stacks and air ducts substantially depend on the meteorological conditions.
For a constant discharge of these impurities, their concentration at a
given distance from the stacks may change by a factor of tens and even
hundreds depending on the wind velocity, stability of the atmosphere and
some other meteorological characteristics. Cases with the highest con-
centrations of noxious substances in the ground layer of air pertain to
dangerous conditions of atmospheric pollution.
Dangerous conditions are frequently related to the presence of layers
of an elevated temperature inversion. It is assumed that elevated inver-
sions were observed during periods of known "disasters" in London and
other places where considerable pollution of air was observed, associated
with human victims and a marked increase of the disease rate among the
population. The dangerous character of such stratification of the atmos-
phere is usually determined from qualitative considerations. It is
assumed that the layer of an elevated inversion is characterized by an
attenuated turbulent exchange hindering the transport of impurities to
higher levels. As a result, the bulk of the impurity mass concentrates
under the inversion layer near the earth's surface.
The published literature gives little experimental material on the
increase of the concentration under these conditions. Lowry [8] points
out that according to the observational data, when the inversions are
located above the source, the concentrations increase by a factor of up
to 20. The same results, without mention of the specific conditions of
observation, are cited in [5] and in some other studies. In the presence
of elevated inversions, an increase in concentration, but much smaller,
by a factor of approximately 1.5-2, is shown by the data of [11] and of
others. According to the existing classification of plume forms [5, etc.],
these cases pertain to conditions of "smoke pollution" of the ground layer
-------
of air. A series of attempts at a theoretical evaluation of the concen-
tration of an impurity from a source have been made under such conditions
(Bierly and Hewson [6], Holland [7], etc.)- However, the calculations
in these studies were based on very primitive assumptions of a uniform
vertical distribution of the impurity in the subinversion layer.
A more rigorous approach to this problem was developed in [1] and [2]
as a result of a numerical solution of the equation of turbulent diffusion
of the impurity from the source, solution which made it possible to take
into consideration the complex character of the variation of the exchange
coefficient with the height. The studies give calculations of the surface
concentration for cases in which the inversion layers are located at dif-
ferent heights above the level o:= the source. It was assumed that the
vertical component of the exchange coefficient in the inversion layer is
sharply attenuated. It was found that the increase of the impurity con-
centration substantially depends on the height of the lower inversion
boundary above the source and on the height of the source. This increase
is greater the closer the base of the inversion layer is to the source
and the lower the level of the source. If the layer of attenuated turbu-
lence is located at a sufficient distance (about 200 m or more) above
the source, the increase of the surface concentration is relatively slight,
and it is substantial only at very large distances. In cases where the
blocking layer begins immediately above the source, the increase in the
maximum of the ground concentration amounts to 50-70% and sometimes to
more than 100%.
Cases of elevated inversion pertain to abnormal stratification con-
ditions. Some cases of abnormal distribution of the wind velocity with
the height were examined in [4]. It was found that the presence of still
layers in the propagation zone of the impurity causes an increase of its
ground concentration. When a layer with an attenuated wind velocity is
located at a certain level, the lower this level, the stronger the influ-
ence of the given layer. According to the performed calculations, in the
presence of a surface calm up to a height of 30 m, the maximum concentra-
tion q from a source 100-150 m high increases by approximately 70% as
compared to the values of q in the absence of a calm.
The solution of the problem of determination of the highest concen-
trations under normal conditions, when the temperature decreases continu-
ously with the height and the wind velocity increases with the height in an
approximately logarithmic manner, was investigated in [3]. Such conditions
are usually characteristic of summer daytime in fair weather. Analysis of
this solution shows that the values of the highest concentrations depend
in a complex manner on the wind velocity. On the one hand, for a fixed
discharge, the maximum of the ground concentration increases with decreas-
ing wind velocity. On the other hand, an attenuation of the wind leads to
an increase in the initial ascent of the impurity, so that its surface
concentration decreases. Consequently, there exists some dangerous wind
velocity ^ at which the highest value of the surface concentration q is
-------
reached. We obtained a formula for determining um and q in [3].
In evaluating dangerous conditions, it is of interest to determine the
distribution of the air temperature in height, particularly the distribu-
tion of elevated inversions and wind velocity. The present paper is de-
voted to an analysis of this problem.
2. Influence of the Temperature Gradient on the
Initial Ascent of the Impurity.
In recent years, the initial ascent of an impurity caused by its over-
heating has been studied, and a number of works have been devoted to the
vertical escape velocity from the stack.
One of the most extensive investigations of the jet stream of a heated
gas in stationary air is due to Pristley and Boll [10]. Certain restrictions
in the application of their results to atmospheric problems are due to the
fact that they excluded the influence of turbulent exchange by introducing
a new, unknown parameter.
In [4] and [3] we dwelt on a rigorous formulation of this problem.
Certain difficulties in its solution pertain to the consideration of the
driving effect of the wind velocity. However, for many aspects which will
be discussed below, it is sufficient to obtain the solution for simpler
conditions of consideration of this effect.
We shall examine the Pristley-Boll problem by generalizing it with
respect to the consideration of the intensity of atmospheric turbulence.
As the initial conditions for an axisymmetric jet, we shall take the
following equation of motion
the equation of influx of heat
and the energy equation obtained directly from the equation of motion by
multiplying the latter by w
d lruvft\ , d (rw*\ „, d . dw
the above equations having been transformed by using the continuity equa-
tion in the cylindrical coordinate system (r being the radius and z the
height) . Here u is the radial and w the vertical component of the dis-
placement velocity; ~W±s the deviation of the temperature in the jet from
6, where 9 is the temperature of the surrounding atmosphere on the absolute
scale (9 is assumed to be a function of only the vertical coordinate z) ;
k is the turbulent exchange coefficient; g is the acceleration due to gravity.
-------
It is assumed that the changes of air density in the jet and also the
difference in the. densities of air in the jet and the surrounding medium
are slight.
According to the boundary conditions, on the axis of the jet (r = 0),
u, dw i and d0 "disappear, at a large distance from the axis (r->oo), w and 0
~W ~dT
turn to zero, and at z = 0, w and 6 take the assigned values.
We obtain for the axisymmetric jet
*"", iHf. (5)
where R is some effective radius of the jet, and the subscript m pertains
to values of w and 0 on the jet axis.
Integrating equations (1) , (3) and the boundary conditions with respect
to r from 0 to oo, we obtain the following system of ordinary equations
dz ™" 0 m m^ m ^°)
with the boundary condition
where w0 is the escape velocity of the stack gases from the stack and AT0
is the difference in the temperature of the gases and of the surrounding
atmosphere at the level of the mouth of the stack.
It follows from (6) and (7) that
~OT \Wm *" > i ~g~ ~fe- ~fa-(wmR)* =*= 0 (9)
or for a constant value of 09
dz
where A is the integration constant.
Further, determining dm from (10) and substituting into (9), we obtain
where B — — —
gA* dz
-------
From (6) and (8) it follows that
6k
Substituting (12) into (11) , we get
^(w«/?)a=-^-(l -BwWf. (13)
We introduce the substitution of variables
i
s = wmRB* (14)
and integrate (13) ; then
s
where t = 8A If B E is the integration constant found from initial data.
' 3kQ '
When the integration is carried out in (15) , it is necessary to
tinguish two cases depending on the sign of the temperature gradient dQ
ar dz'
For the case of a positive gradient po (hence, B > 0 also), i.e., in the
dz
presence of an inversion stratification, when the potential temperature
increases with the height, we introduce the substitution of variables
S2 = sin
-------
Let us now consider the problem of the exchange coefficient k.
In cases where the jet arises in a laminar medium, it becomes turbulent
under certain conditions. For these conditions, k is the turbulence coef-
ficient in the jet, and it should increase with the distance from the
source or with increasing radius of the jet, since the exchange includes
large-sized eddies, and also with increasing traveling speed in the jet.
When the jet propagates in a turbulent medium, these considerations
with respect to k are insufficient. Obviously, the exchange in the jet
and the turbulence in the surrounding medium should be closely related.
When the turbulent mixing in the medium is sufficiently vigorous, it will
practically determine the exchange in the jet as well, with the exception
of the region in the immediate vicinity of the origin of the jet. This is
usually the case with the conditions of propagation of discharges from
smokestacks, and k in the case under consideration may be taken to mean the
coefficient of turbulent exchange in the atmosphere, whereas above stack
level the turbulence is usually approximately isotropic, and the exchange
coefficient is approximately constant with the height, or is some function
of height z.
Substitution of (19) into (12) and integration of the relation obtained
gives the dependence of R on z, according to which
K _L JL "*
sin2 (//?2 + £)-^2=3fi4 f&fe, (21)
where R0 is the radius of the mouth of the stack (at z=0) .
Relations (19) and (21) define the dependence of w on z. We obtain
similar expressions for the case dQ ^ Q using the substitution of vari-
ables (17):
wm = rsin A2 (//?2 + ZT), (22}
__ \Af£f J
RB*
= 3B* \kdz. (23)
y
It can be readily seen that despite a certain similarity in the ex-
ternal appearance of formulas for cases of gradients of potential temper-
ature ^_with different signs, the character of the dependence of the jet
parameters on _g_ is substantially different in the two cases. The main
difference lies in the fact that when ®>Q at a height of z = z , when
-------
according to (18), wm = 0. In this case, the ascent of the jet
ceases, having reached a certain "ceiling".
The corresponding value of RC is given by the formula
*•_-!; («-£).
Substituting this value of RC into (21) at constant k, we obtain
sin
3fcBT0
We transform this expression by introducing a substitution of the
integration variables
Then
(24)
where
and
Values of !„ for different values of E, which in practice change
from -0.01 to 0.1, are given in Table 1. It is apparent that !„ undergoes
relatively little change.
Table 1
£ -0,01 0,001 0,01 0,1
tB -0,601 0,607 0,509 0,656'
There are errors in the determination of the level z , since as the
velocity falls off to 0, the temperature of the jet decreases indefinitely,
this being obviously due to the nonrigorousness of the formulation of the
problem [no account is taken of the vertical exchange and of certain other
secondary factors whose influence may be felt at the boundaries of propa-
gation of the jet; there are also limits to the application of the selected
relations (4), (5), etc.]. Therefore the conclusion that a "ceiling" is
present should be considered approximate.
-------
Similar conclusions were also reached in [9, 10, etc.], where it was
also proposed to determine the level zt, where 0m=0 __From the expression
for #m one can readily see that zfc can be found if ,p = _fL. In the general
£
case zt < zc. According to {10], zt^:0.7 zc- As follows from the formulas,
there is no "ceiling" when d& Q .
dz
Let us also consider some special cases resulting from the formulas
obtained. We shall first consider the solution of the given problem with-
out taking the influence ofj?0_into account, i.e., for the equilibrium
/ M \ dz
case (-^-—jOj The study of the general solution for this case is compli-
\ dz I
cated by the fact that it is necessary to achieve passage to the limit. It
is more convenient to prpceeji. f rom the intermediate formulas (10) and (11).
According to (10) , when <& _ n
~~u>
and from (13)
d (wmR)* _
~dIV 3*8"
Hence
and
/?2=—5 — , (jf.\
w — D \£v)
where D = ° — and M is the integration constant.
Substituting (26) into (25) and integrating the equation obtained,
we find an expression relating wm to the height z
where L is the integration constant.
The values of A0 , L and M are found directly from the initial data,
in particular,
and the value of L follows from (26) at z - 0. The working formulas for
the determination of wm then assume the following form:
\kdz
6 6 -,. (27)
-------
Thus far, k has been considered the exchange coefficient in the
atmosphere, excluding a small region of the jet close to its origin,
where k may depend on the parameters of the jet. For the latter region,
it may be postulated as the simplest assumption that k increases in pro-
portion to the traveling speed and scale of the jet (the jet scale being
related to the jet radius R) , setting
. (28)
Here c is the proportionality constant, the coefficient 1/3 being intro-
duced for convenience. From (12) and (28) it follows directly that
R = cz + R0. (29)
If a coordinate system is introduced such that R_ = 0, then in this
case (29) and other formulas for the determination of wm and ^1 practically
coincide with the results of Pristley and Boll [10].
It is true that k is not introduced directly in [10], but a turbulent
stress is assumed in the jet. Obviously, that such an assumption is equiv-
alent to selecting the dependence of the exchange coefficient in the form
(28), which results in an indefinite increase of k with z.
It is of interest to compare the formulas for calculating wm in cases
of constant k and indefinitely increasing z. One can readily ascertain
that according to (27), in the first case (k constant) for 2-»-op, the ver-
tical velocity w decreases to a certain constant value. In the second
case for z-»-o6 fe-*-oo and wm->-0 (at large heights tam~z~i>3).
From an evaluation of the terms of the energy equation, taking into
account the fact that in the case under consideration the quantity AQ pro-
portional to heat flux from the source remains constant, it may be con-
cluded that the value of kw^ for z -> oo remains constant. Hence, when z->-oo
m
in the case of constant k, wm remains constant, and in the case k~*-oo a>m-»-0.
It is evident that neither of the two schemes corresponds to actual con-
ditions in the range of large z values, and the two schemes are inapplicable
in this range. For small z values, the two schemes yield similar results
with suitable parameters.
It is useful to consider one special case, the propagation in a strati-
fied atmosphere of a cold jet whose initial temperature coincides with the
temperature of the surrounding medium_._ For the conditions of the equilib-
rium state of the atmosphere / d9 __Q\ , the solution of the problem is
\ dz I
considerably simplified, and for the accepted premises concerning the function
/Ol) in form (5), we obtain from (25) as a special case for Ag=0,
'tt>0R0, (30)
-------
and from (12), using (30), we obtain
3*
and consequently,
«>„ = •
It should be noted that a more rigorous solution may be obtained for
the case under consideration by introducing stream functions into the
initial equation and without specifying a definite dependence f(r\) in form
(5). In so doing, one can utilize the ready solutions of equations of a
given type given in known papers of hydrodynamics. Results thus ootained
essentially coincide with (31) and (32). If &*_ , Q is taken into con-
da •
sideration, the working formulas are found as a special case from the ones
obtained above after substituting in them the value of A corresponding to
the initial value of ATQ = 0.
3. Characteristics of Dangerous Conditions
The preceding section gives some results of an analysis of the change
of the vertical velocity in a warm and a cold jet. These results are re-
lated to the evaluation of the level of the initial ascent of the impurity
at which the ratio of the vertical ascent velocity to the wind velocity is
comparatively small, and further propagation of the impurity in the atmos-
phere is chiefly determined by processes of horizontal transport and turbu-
lent diffusion. The value of the ratio may be obtained, in particular,
from experimental data on the initial ascent of the jet for individual
cases. Subsequent generalization makes it possible to obtain the depen-
dence of the initial ascent, or as it is otherwise termed, the effective
ascent AH, from the initial parameters of the discharge and atmospheric
characteristics. Thus, in [3], the following formulas was obtained for
determining the initial ascent
= ~L1T~1(2'5"^ ' ILa" ° ) » (33)
where u is the wind velocity at the height of the vane, and the remaining
symbols are the same as above.
For inversion conditions it was found above that at some height z
the velocity WG turns to zero. It is understandable that the effective
height AH for the given conditions should be less than z . Differences
between zc and AH substantially depend on the wind velocity u. However,
in cases where zc is small, and the stack height sufficiently large, the
absolute differences between ZG and AH should not play an appreciable part
in the evaluation of the surface concentration from a high source. It is
of interest therefore to study cases where z is small. It is obvious that
-------
the problem of the difference between the ceiling heights z and zfc deter-
mined from the decrease of the vertical velocity wc to zero and from the
temperature difference is of no practical importance in these cases.
In order to determine the conditions for which z is small, we shall
cite the results of calculations for two characteristic examples, taking
in both cases the value of the inversion gradient to be dQ ~ IA^. 0, and
-=— = 10^ /M
£=1-5-5 M/sec> In tjje first example, we shall consider the condition of a
large heat source corresponding to discharges from stacks of thermal power
stations, setting t»o«10-5-20 M&etf-, /?o=2+3 M, A7o—100°. The second example
pertains to conditions characteristic of many chemical plants, with
w*~ 10 M/seq, /?0=0,5~ M, Ar0=20°.
It is easy to see that in both cases, according to (10), the expression
for A is simplified
and on the basis of (24) and Table 1
fw,£T R\
~W~ (34)
I/
\
In the first case for large thermal sources zc = 200-800 m, and in
the second, for comparatively cold discharges from stacks of moderate
diameter, z = 20-40 m. Consequently, in the second case z takes rela-
tively small values and it is obvious that for stacks over 50-100 m high,
the refinement of zc by taking the influence of the velocity into consider-
ation should not play an appreciable part. Thus, for the first approxima-
tion, the effective ascent can be determined from zc-
Formula (34) leads to the fully understandable result that the smaller
the capacity of a thermal source, i.e., the lower the amount of heat in
the_volume of air emerging from the stack (this amount is proportional to
tt>oA7y?jj), the smaller the height zc. The value of zc also decreases with
increasing inversion temperature gradient d9 xhe dependence on dQ is
62 ' dz dQ
complex in character, since the exchange coefficient k also depends on --
dz
It may be assumed that for a given thermal capacity of the source zc reaches
the minimum value at some inversions, not too deep, when ^ d9 is not too
6Q dZ
small (in layers of isotropic turbulence £_ — is proportional to the turbu-
lent heat flow). dz
It is understandable that on the basis of the above, the results ob-
tained relative to z should be regarded as very approximate. Furthermore,
-------
in cases where z is small, even with considerable errors in the determ-
ination of zc, the basic conclusion that the initial ascent of the impur-
ity from the stacks will be limited to small heights regardless of the
magnitude of the wind velocity seems convincing.
In regard to the influence of the temperature gradient in an unstable
atmosphere we shall note that according to the formulas of the preceding
section, this influence on the vertical traveling speed and hence on the
initial ascent of the impurity is chiefly determined via the quantity A.
On the basis of the above calculations, covering a wide range of condi-
tions of practicaLiaterest, the value of A is essentially relatively
independent of
This leads to the conclusion that the change of
dz
the velocity of the vertical motion and the initial ascent of the impurity
in an unstable state of the atmosphere also depend only slightly on the
magnitude of the temperature gradient.
Let us note that formula (33) was obtained for conditions of equilib-
rium and unstable stratification, for which the highest values of surface
concentrations could be expected. At the same time, the problem of the
influence of the temperature gradient was not considered. From the con-
clusions just reached it follows that this influence is slight, and this
permits a broader application of the formula obtained.
Let us now consider the problem of dangerous wind velocity. Accord-
ing to [3], the magnitude of the dangerous wind velocity u- at the level
of the vane for the discharge of an impurity from a stack of height H is
given approximately by the formula
a. = 0,65
_For large sources such as smokestacks of thermal power stations,
5 M/sec. For other types of sources, i^ may change appreciably.
Thus, for the *>bove example of comparatively cold discharges with parameters
pertaining to many chemical plants, etc., MM«l-r-2 M/Sec. It must be re-
membered that the lower the dangerous velocity, the higher the maximum con-
centration q, other things being equal. Indeed, according to [3], in the
very general case, q as a function of the discharge capacity Q, wind
velocity and source height H may be calculated from the formula
where c and g are some constants.
For u = um and suitable stratification conditions, q reaches its
H^?7SLVaJUe qm and tl?US' qm is lnversely proportional to «. However,
the highest concentration values usually are not reached at very low wind
velocities, since this causes a sharp increase of the effective source
-------
height H (sum of the stack height and initial ascent AH). In all of the
formulas for the determination of AH employed at the present time, and in
particular according to (33), as u decreases to zero, AH increases indef-
initely. In addition, from the results obtained above it follows, under
inversion conditions, that this does not occur, since some "ceiling" may
exist for the initial ascent of the impurity. It is understandable that
if the "ceiling" is located relatively low over the stack, the impurity
concentration can increase substantially in the presence of slight winds.
Thus, in the presence of an inversion above the stack and of a marked
attentuation of the wind, very unsafe conditions should arise in the surface
layer under certain conditions. This may explain the cases of particularly
high concentrations indicated in the beginning [5, etc.].
Such cases must be thoroughly investigated when analyzing the condi-
tions of atmospheric pollution by industrial discharges.
LITERATURE CITED
I B e p Ji si H A M. E [HAP] O sarpHsneHHH aTMOccpepw npoMbiuuieHHbiMH suopocaMH
npn aHoinaJibHbix VCJIOBHHX cTpaTH(pHKauHH Mereopojiormi H rHApoJioma, Nfc 8, 1963
2. EepJiHHflM E. [H up.]. MitcJieHHoe HcwreAonaHHe aTMoccpepHoft flH(py3HH npn Hop-
MajibHbix H aHOMajibHbix yoioBHHX cTpaTHcpHKauHH. TpyAu iTO, sun 158, 1964.
3Bep;isiHAM. E.reHHXOBHiE JI, OHHKyaP H O pacqere sarpHSHeHHJi
aTMoctpepbi su6pocaMH H3 AMMOBHX rpyfi 3fleKTpocT3HUHA. Tpyau ITO, sun 158,
1964
4, B e p Ji n H A M. E, FeHHxoBHi E JI, fleMbHHOBHq B. K. HexoTopue
aKtya^bHue sonpocw iiccneflOBaHHH atMoccpepHofi AH
-------
THEORY OF THE DEPENDENCE BETWEEN THE CONCENTRATION OF AEROSOLS
IN THE "ATMOSPHERE AND THEIR FLOW ONTO A HORIZONTAL BOARD
M. Ye. Berlyand, Ye. L. Genikhovich, and G. Ye. Maslova
From Trudy. Glavnaya Geofiz. Observat. im. A. I. Voeykova, No. 185,
p. 3-14, (1966).
A theoretical study of the deposition of aerosols from the atmosphere on horizontal
boards has been made. The surface of the board is assumed to be adhesive, causing the
impurity concentration of the incoming air flow to change. The process of concentration
change is described by the equation of turbulent diffusion. A theory of flow past the
board in a turbulent atmosphere is developed, and the field of the traveling speeds and
of the exchange coefficient in the boundary layer above the board is determined. The
equation of turbulent diffusion of the impurity, taking into account the field of the
traveling speeds and of the exchange coefficient, is solved by a numerical method.
Results of a calculation with an "Ural-4" computer are given. The dependence of the ratio
of the vertical flow of aerosols to their concentration on the wind velocity and exchange
coefficient in the incoming air flow and also on the size of the board and height at which
it is mounted above the underlying surface is established.
1. Introduction
The study of many problems of propagation of aerosols in the atmos-
phere involves the necessity to determine their concentration and flow
onto a horizontal board. One can mention here mainly the problem of
evaluating the degree of pollution of the ground layer of the atmosphere
with dust or ash from industrial plants. Usually, such an evaluation is
made from data on the concentration of dust in air. In addition, attempts
are made to use the results of measurement of the amount of dust deposit-
ing on a horizontal board. The use of the latter is much simpler than the
measurement of concentration, which requires a special apparatus, particu-
larly air flowers operating on line current, high-capacity batteries or
motors, etc.
Attempts al-jo are made to use the flow of aerosols onto a board for
the determination of their deposition on the surface of the ground. The
literature frequently provides data on the amount of dust settling on the
ground from industrial sources, the dust pollution of the snow cover, etc.
Some formulas for calculating the deposition of industrial aerosols are
known [9, etc.].
However, the practical application of the results obtained is very
difficult, since there are at present no indicators for the deposition of
aerosols on a horizontal surface analogous to maximum permissible concen-
tration values (MPC).
Considering the simplicity of obtaining board data, it appears
essential to relate these data to the concentration values, and then, using
MPC, to establish the pattern of unsafe pollution of an air layer and of
the earth's surface with aerosols.
-------
The solution of this problem is of major importance for the study of
the spatial distribution of aerosols originating from various sources and
of the background pollution on the earth's surface. It is well known that
at some points the collection of dust, including radioactive dust [1, 7, 8],
is made on adhesive boards, in vessels sometimes filled with water, etc.,
and that at other points the dust concentration is measured by means of
filtering-ventilating devices. As already noted, such devices are much more
complex than boards, and the number of places where they are used is consid-
erably smaller than the number of points where board measurements are made.
Switching from the readings of one set of observations to those of the other
set permits the establishment of a relationship between them and a consid-
erable expansion of the potential of the analysis of spatial aerosol distri-
bution.
Recently, in the USSR [2, 3] and abroad, studies of atmospheric dif-
fusion have made use of the method of dumping of fluorescent and luminescent
powders from towers. These powders are then collected on adhesive boards
located at various distances around the source. Such powders are relatively
easy to observe on the board, this being a definite advantage of the method.
In this case also, the problem of the relationship between the amount of
impurity deposited on the board and its concentration at the point where the
board is located is of essential importance. Such problems also arise in
the study of the effectiveness of aerosol methods of treatment of agricul-
tural crops for the purpose of controlling pests, in the collection of
aerosols [6], and also in the solution of other practical problems.
It is obvious that the empirical establishment of a relationship
between the concentration of aerosols and their vertical flow onto a board
is highly complex. This relationship is determined by a large number of
parameters (meteorological elements, characteristics of aerosols and of the
board). For this reason, despite numerous attempts to find this relationship
by processing experimental data, no satisfactory results have been obtained
thus far. Not enough attention has been given to the theoretical solution
of this problpm. We know of only one study of this nature, by V. I. Bekory-
ukov and M. L. Karol' [4], which dealt with the problem of the effectiveness
of trapping aerosols from a high source on adhesive boards placed on the
earth's surface. However, even here the problem of the relationship between
the concentration and the vertical flow of aerosols was not studied.
In addition to the fundamental importance of the problem of finding
the indicated relationship, of great practical importance is the study of
the effect of the size of the board, the height at which it is mounted,
and an evaluation of the influence of meteorological factors, etc. The
present paper is devoted to a study of these problems.
2. Statement of the Problem
Usually, the boards used for the collection of depositing dust are a
few tens of centimeters in size and are mostly placed at the height of one
-------
to several meters above the earth's surface in order to avoid their con-
tamination with dust rising from the ground. Boards with a specially
deposited adhesive coating and containers with water, since a water surface
almost completely absorbs the aerosol particles falling on it, may be con-
sidered to be absolutely absorbent. The aerosol concentration on their
surface is equal to zero. The remaining cases deal with a partial adhes-
iveness of the board; they include boards coated with gauze [1, 8] and
certain other materials designed for collecting aerosols. These cases
deal with the trapping coefficient O, which stands for the ratio of the
flows of deposited impurity above the board under consideration and above
an absolutely adhesive board under identical conditions.
The presence of the board causes a certain disturbance in the natural
distribution of aerosol particles in space, and differences arise in their
concentration on the board and in the surrounding medium. The process of
turbulent diffusion of aerosols above the board in the general case is
described by the differential equation
dl' W d W d dl' d
where q" is the concentration of the aerosol in air, u and w are the hori-
zontal and vertical components of the displacement velocity, and kx, ky, and
kz are the components of the exchange coefficient along axes x, y, and z
respectively. Axis x is parallel to the direction of the wind velocity,
axis y is oriented in a transverse direction along the horizontal, and axis
z is oriented along the vertical upward. We shall refer the level z=0 to
the plane of the board, and the level x=0 to its windward edge.
Usually, the effect of diffusion along the direction of the wind is
relatively slight and may be neglected. Considering the small size of the
board and the instability of the wind direction in the horizontal plane,
we can also neglect to a first approximation the term in (1) describing
the diffusion along axis y, assuming that the aerosol particles are uniform-
ly distributed along this axis. As the boundary conditions we shall take
at x = Q g' = qot (2)
at 2 = 0 ?' = 0, (3)
where qo is the aerosol concentration in the atmosphere.
The latter condition corresponds to the fact that the surface of the
board is assumed to be absolutely absorbent. For other surfaces, the con-
version of the aerosol flow is done by multiplying the corresponding quan-
tities obtained for absolutely absorbent boards by the trapping coefficient.
We shall assume that at sufficiently large distance from the board
(*-»•*>), the influence of the latter on the distribution of the impurity
concentration in the atmosphere dies out.
-------
In cases where the board is located at some height above the under-
lying surface, it has a disturbing influence not only on the process of
transport of the aerosol, but also on the nature of the incoming air
flow. As a result of the flow past the board, a boundary layer is cre-
ated in which the horizontal velocity u does not coincide with the wind
velocity, a vertical component of the motion of air arises, and the ex-
change coefficient differs substantially from the corresponding values
in the surrounding atmosphere. For this reason, in studying the diffusion
process, it is necessary to supplement equation (1) with a system of
equations and boundary conditions describing the flow of air past the
board. The solution of such a system should precede a study of the dif-
fusion of the impurity, and this solution is discussed in the next section
of the paper.
Since the size of the board is small, the thickness of the boundary
layer is also small. This size is also much smaller than the distance
from the sources of pollution to the board. This makes it possible to
consider the initial concentration in the atmosphere q0 to be constant in
height, something that is fulfilled with a great accuracy in practically
all of the cases of interest to us.
We shall now transform the initial equations and boundary condi:ions,
taking into account the indicated simplifications, and introduce the
notation
(4)
¥0
We then obtain
_ ^ •> a _
(5)
The flow of impurity onto the board PX at point x is given by the
formula
The total flow of impurity onto the board P of length L and unit width
is found by integrating Pv with respect to x, i.e.,
P
o
It is convenient to separate the turbulent flow
^rdx' (8)
-------
Then
3. Theory of Flow Past a Horizontal Board
in a Turbulent Atmosphere.
The theory of air flow past plates has been treated extensively in
the literature. Cases of laminar and turbulent boundary layers on plates
have been studied in detail. However, in both cases the studies per-
formed refer chiefly to conditions in which the plate is in a laminar
flow. In the study of flow past a board, it is essential to take into con-
sideration the turbulent exchange of the atmosphere. This is particularly
important in the determination of the turbulent flow Pt , which is directly
dependent on the exchange of coefficient k, bearing in mind that the values
of k in the boundary layer above the board and in the surrounding medium
differ appreciably. The exchange coefficient k changes from the value of
the molecular viscosity of air on the surface of the board to relatively
high values characteristic of the atmosphere at the level of the board out-
side the boundary layer.
The process of the flow of air past a board, taking the above simpli-
fications into consideration, may be described by the equation of motion
VQ * g ap ' vg
and by the continuity equation
The following boundary conditions at the board surface are taken:
at z = 0 u = 0, TO = WO, (ii)
which for convenience of presentation includes w0 , the gravity settling
rate of the aerosols.
It is assumed that at a sufficiently large height above the board,
the disturbing influence dies out, i.e., the traveling speed coincides
with the wind velocity and the turbulent flow of momentum disappears.
We solve the problem by familiar boundary layer methods.
We introduce the thickness of the boundary layer 6 and assume that
at 2 = 8 u*=V, (12)
where V is the wind velocity in the incoming flow.
-------
For an elevated board (usually more than 1 m above ground), consid-
ering a small thickness of the boundary layer corresponding to the small
size of the board, the quantity V may be considered constant with height.
Equation (9) will be transformed to the integral form. To this
end, considering (10), we reduce it to the form
-£rtt(V— «) + -^-w(V~u)= — k ~ (13)
ox '^
Integrating (13) with respect to z from 0 to 6 and taking the boundary
conditions into consideration, we obtain the so-called "equation of momenta"
(14)
In addition to the unknown traveling speeds, equations (13) and (14)
also contain the exchange coefficient, which itself depends on the travel-
ing speeds. From general considerations one can only state that at the
surface of the board, the coefficient k should reach the value of the mole-
cular viscosity v, and at a certain level 6-1 above the board it will prac-
tically take the value of the turbulent exchange coefficient in the incom-
ing flow K. The level 6-^ may not necessarily coincide with the hej.ght of
the boundary layer 6 for the displacement velocity u, i.e., # _8_ where
*~~ r'
The above equations are insufficient for finding k. Therefore, we
shall also introduce an energy equation. To do so, we multiply (9) by u and
exclude w on the basis of the continuity equation. The relation obtained is
then integrated with respect to z from 0 to 6. After some simple transform-
ations, we obtain an integral expression for the energy equation
fe. (15)
Equations (14) and (15) contain two unknown quantities, u and k. To
find them, one can make approximate use of known experimental facts, accord-
ing to which the displacement velocity in the boundary layer increases with
the height in an approximately logarithmic manner, whereas the exchange
coefficient increases linearly. Therefore, satisfying the boundary condi-
tions as well, we shall seek the solution of the problem by assuming
u
for z < 8 and « = V for z > 8 (16)
and
k = vi for 2<51 = -~and * = K fOT? z>\*\ (17)
where r\ = v + (K — v) -^ and T>1=='n 'r-i •
-------
elevated board, K like V may be assumed independent of the
Substituting (16) and (17) into (14) and (15), we obtain
For an
height.
and
These two equations contain two unknown quantities, 6 and r. Excluding
first, we obtain an algebraic equation for determining r.
where
Integrating equation (18), we then find a comparatively simple expres
sion for the height of the boundary layer
Since the function f(r) changes sign on passing through the root, to
find r we made use of the algorithm of division in half of the interval at
the ends of which f(r) takes values of different signs.
To find the roots r by this method, a program was written for the
"Ural-4" computer. The calculations cited showed that r changes relatively
little from the values of V and K, and its average value amounts to about
5.5. Thus, the boundary layer for the exchange coefficient is approximately
5.5 times thinner than the traveling speed.
After finding 6 and r, we determine u and k in accordance with (16) and
(17). Then, after integrating the continuity equation (10), w is found from
the value of u obtained. It is considered that w also includes the gravity
settling rate of aerosols wfl.
The results obtained relative to the field of velocities and exchange
coefficient for the flow past the plate may also be of interest in them-
selves. However, we shall not dwell on a detailed discussion of these re-
sults, since they are of an auxiliary nature in the present paper and are
-------
used here for the purpose of calculating the process of diffusion of
aerosols above the board.
4. Numerical Solution of the Problem and Analysis
of Results of the Calculation.
Since the coefficients of the equation of turbulent diffusion of an
aerosol (5) are complex functions of the coordinates, its solution was
carried out numerically. To this end, a standard program (SP) was written
for the solution on the "Ural-4" computer of a parabolic equation in the
form
where a, b, c, d, and e are functions of x and z with boundary conditions
in the general form:
at* = 0 q = F(z),
atz-*co
where A, B, C, D, E, and G are functions of x.
The solution of the corresponding difference equation was obtained by
the method of successive differences.
The SP program was written in symbolic addresses according to a system
of subroutines; in the solution of a specific problem, the use of this system
makes it necessary to prepare subroutines which are determined directly by
the equation being solved (for example, the subroutine for calculating the
coefficient of the equation) , and the program is then revised by means of
the component program (CSP). A more detailed description of the program will
be published separately.
The revised program used for the computations on the "Ural-4" computer
made it possible to store in the memory up to 400 points on a single x layer
(i.e., for a fixed x, up to 400 values of this solution could be stored).
Ratios for determining u, w and k obtained in the preceding section were
used as the working formulas in the subprogram for calculating the coefficients
of the equation.
Since in the vicinity of the surface of the board the impurity concen-
tration q changes markedly with the height because the exchange coefficient k
is close to zero at the boundary of the board, the "effective exchange coef-
ficient" was used to determine the coefficients of the difference equation,
as was done in (5).
-------
When the solution of equations of the boundary layer are used, dif-
ficulties are known to arise in the determination of the vertical velocity
at the upper boundary of the layer 6 since the corresponding boundary con-
ditions are not imposed here. In order to elucidate the possible error in
the calculations caused by this situation, the computations were made by
retaining for z > 6 the value of w reached when z = 6, and for a rapid de-
crease of w to zero for z > 6. It was found that such changes of the
values of w above the boundary layer had practically no effect on the mag-
nitude of the turbulent flow of the aerosol onto the board. This is quite
natural, since outside the boundary layer the concentration gradients are
close to zero, and the variations of w taking place here should not have
any appreciable influence on the distribution of the concentration at the
surface of the board.
The calculations were made at different values of the input para-
meters V, K, WQ , and L. It follows from the results of the calculations
that the turbulent flow of aerosol onto the board depends relatively little
on changes in the gravity settling rate of the aerosol w0 from P__to__0._l m/sec.
As an example, Table 1 lists the data of the calculation of v dq \ at
~*~te\*-*
V = 1 m/sec for WQ = 0 and w0 =0.1 m/sec.
Table 1
** ........ °.05 0.10 0,15 0,20 0,25 0,30 0,36 0,40 0,45 0,50
dq\ wo=>0 0,276 0,190 0,154 0,133 0,118 0,108 1,100 0,0930,0880,083
o«b-0,l 0,267 0,173 0,140 0,119 0,101 0,090 0,0820,0750,0690,065
The results of the calculations also show that the values of q are
practically independent of the variation of v, which is understandable,
since the coefficient of molecular viscosity of air V is many times smaller
than the exchange coefficient k.
Results of the numerical calculation show that power complexes may be
taken for the local flow Pfc as the approximating expressions in the range
of arguments K, V which is of interest to us.
On the basis of the results of the calculation, we can set
o
where 4>(x) is some function of x.
Fig. 1 and Table 2 are given as illustrations,
-------
Table 2
•*« ..... 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1,0
K ...... 1,989 1,992 1,993 1,994 1,995 1,996 1,997 1,997 1,998 1,998
Fig. 1 gives the quantity — K—jL\ ^ as a function of x for different
values of V and a fixed K. Table 2 shows the quantity R, equal to the
ratio of the flows _ „ dq | at velocities V = 1 m/sec and V = 4 m/sec.
According to (22) and (8) , the total turbulent flow of aerosol onto
a board of unit width may be represented by the formula
(23)
The values of the function qj(L) as a function of L are given graph-
ic ally _in_ Fig. 2. They are satisfactorily approximated by the function
-------
and the boundary conditions:
at
Hence we find that
— — and at
:-* oo q = 0.
- ,.(»-/*+*)•
i iii ii i i it i
Switching to the functions of a transform and determining the turbu-
lent flow onto the board, we find
•3-
V
1/~SJT
K irnr
-------
0,2
OJS
Fig. 2.
For the cases which are of practical interest to us, the second term
in (25) may be neglected, and in the third, it is sufficient to retain
the first term of the series expansion of the probability integral. Form-
ula (25) is thus simplified and
or at small wn values (usually up to 0.1-0.2 m/sec)
-K
dq
~
(26)
Hence, trie total turbulent flow onto the board determined in accord-
ance with (18) is
Formula (27) is somewhat analogous in structure to formula (23) ob-
tained above if one considers that
-------
ST
Fig. 3.
The result obtained appears natural. The use of constant K and V should
indeed result in values of the turbulent flow of impurity that are much
too high, since no account is taken of the abrupt decrease of the exchange
coefficient at the board surface and of the influence of vertical currents,
which reduce the transport of the impurity to the board.
LITERATURE CITED
1. AjieKcanapoB H H, KosaneHKo B. F., FtampHJioBa F. A. Conoctas-
flcHHe peayjibTaTOB HaGJuoaenufi aa aTMoapepnuM EwnaaeHHeM c noMombro paanHiHux
c6opn«KOB Tpyaw FFO, sun. 158,1964
2. AflexcaHflpoBa A. K. MctoaKKa HccTCflOBaHHS pacnpocrpaneKHst HCKyccTBeHHoro
aspoao^si B npnaeMHOM cnoe Boanyxa. C6. «H3yqeH»e norpaiiHMHoro cnon aTMoc^epw
c 300-MCTpoBofi MeteopoflorHMecKofi 6auiHH» Han AH CCCP, M., 1963.
3. AaeKcaHApOBa A. X., Bwaoaa H. /!., MauiKosa F. 6. Onuru no pac-
npocTpaneHHio ocawflawmeftcH nptmecH OT ToieMHoro HcroiHHKa B HHJKHCM cnoe
aTMOc^epu. C<5. «HccAeAOB8HHe muKHero 300-MeTpoBoro cnon atMOC^epu*. HSA. AH
CCCP, M., 1963.
-------
4. BCKopKDKOBB H, Kapojib M. JI TeopeiHiecKan oiiexKa acptpexTHBHocTH yjta-
MHBamisi aspoaojieii JIHHKHMH miamiieTaMH a npuaeMHOM cnoe aiMocepepu C6. «Bo-
npocbi Hflepnoft MereopoJiorHH* rocaTOMHSflar, M, 1962.
5. 5 e p Ji n H A M E. [H ap.J. HucneHHoe peuieHHe ypaBHeHHfl rypfiyjieHTHoft AH4>ipy3HH n
pacqei aarpnaHeunsi aTMOCtpepu B6JIH3H npoMuuiJieuHux npeanpHsiTHH. Tpyau rro.
sun. 138, 1963.
6. H y H c K H ft B. O. [H ap.]. OceAanne rpyfioAHcnepcHoro aapoaojra Ha noACTHJiaioutyio
nOBepXHOCTb 36MJ1H. CM. H3CT c6
7. C T u p o E. H. Bonpocu HAepnoft MCTeopojiorHH. AH J!HT. CCP, SHJIBHIOC, 1959
8. TOM COM H. H. CpauHHTCJibHan xapaKTepHCTHxa cnoco6oe or6opa npo6 JVM onpeae-
JICHHH paAHoaKTHBRocTH oceflaiomeft ntiflH TpyAw rro, aun. 168, 1964
9. Hawkins,Nonhebel. Stacks and diffusion of Smoke. J. inst. Fuel, vol. 28, No 178,
1955.
-------
METEOROLOGICAL OBSERVATIONS IN THE STUDY OF INDUSTRIAL POLLUTION
OF THE GROUND LAYER OF AIR
B. B. Goroshko, V. P. Gracheva, G. P- Rastorguyeva,
B. V. Rikhter, and G. A. Fedorova
From Trudy, Glavnaya Geofiz. Observat. im. A. I. Voeykova, No. 138,
p. 18-30, (1963).
The paper discusses the program of meteorological observations for the purpose of study-
ing the conditions of propagation of discharges from industrial enterprises, and also the
results of expeditionary studies in the region of the Shchekino State Regional Electric Power
Plant, carried out in the autumn of 1961. The program of observations consisted of gradient
measurements including measurements of the air temperature and huditity and wind velocity at
heights from 0.2 to 17 m, measurements of the soil temperature at depths up to 20 cm, and
also actinometnc observations. Average data on meteorological elements and results of cal-
culations of the turbulent exchange coefficient by various methods are given.
Experimental studies of atmospheric pollution by discharges of various
industrial enterprises are now being conducted in many areas. In determin-
ing the meteorological conditions, the authors usually confine themselves
to measurements of the wind velocity and sometimes of the air temperature
at the same level. However, the transport of impurities in the atmosphere
substantially depends on the turbulent exchange, the evaluation of which
requires a considerably more complete set of meteorological elements.
At the present time, the problem of an accurate determination of
meteorological conditions arises with particular urgency in connection
with the growth of large enterprises and electric power stations, around
which the danger of atmospheric pollution may increase substantially if
necessary preventive measures are not taken.
At the present time, special studies have been started to determine
the meteorological conditions around the Shchekino State Regional Electric
Power Plant (SREPP). The power plant is located in characteristic topo-
graphical conditions of the Middle Russian elevation in a river valley.
The valley follows a nearly meridional direction. In the immediate vicin-
ity of the station there is a water reservoir whose shores consist of
slopes of hills 30-50 m high. There are no forests in the immediate vicin-
ity of the power station, if some small groves are not considered. About
80% of the surface area of the hills has been plowed for field crops, and
only the low-lying parts are covered with grassy vegetation. The soil of
the region consists of chernozem, and in some areas, in low spots, clay and
outcrops of limestone are present. The region has only a few populated
areas, the closest of which are located 1-3 km to the southeast and south-
-------
In line with the formulated objective, most of the attention was
concentrated on obtaining the characteristics of turbulent exchange in
the boundary layer of the atmosphere. For this reason, as complete a
set of observations as possible was obtained, including stationary and
expeditionary measurements permitting an evaluation of the characteristics
of turbulent exchange by various methods and thus a check of the conclusions
reached.
The experimental work was carried out during the period when the ground
concentration of sulfur dioxide and dust was measured on the basis of the
plume from the smokestacks of the power station. It was done by the Main
Geophysical Observatory in collaboration with the Moscow Scientific Re-
search Institute of Hygiene, the All-Union Heat Engineering Institute,
and the Southern State Trust for the Organization and Efficiency of Electric
Power Plants. The present paper gives the results of meteorological obser-
vations only. These observations make it possible to determine the meteoro-
logical conditions of atmospheric pollution and to compare the results of
the calculation with the experimental data.
Fig. 1. General view of observational platform.
The station observations should yield the necessary meteorological
data for calculating the concentration of noxious impurities in the air
during the period when no direct measurements of this concentration were
made.
The observation platform was located in an open area on a hill, at the
foot of which a state regional electric power plant was located at a distance
-------
of approximately 1 km. The level difference was approximately 40 m.
During the period of the studies, in September-October, the platform
was covered with a sparse grass 3-5 cm high. The size of the platform on
which the instruments were set up was 70 x 20 m. Its overall appearance
is shown in Fig. 1.
During the period of the expedition, the program combined station-
based and expeditionary observations and included gradient and balance
measurements.
In line with the program of gradient observations, the wind velocity,
air temperature and humidity, and soil temperature at various depths were
measured.
The wind velocity was measured with contact anemometers set up at
heights of 0.25, 0.5, 1.0, 2.0, 5.2, 9.7, 11.7, 16.0; up to a height of
2.0 m the anemometers were mounted on poles, and above this height, on a
telescopic mast, where they were braced with brackets. The anemometer
readings were taken with electromagnetic counters.
To measure the air temperature and humidity, Assman psychrometers
were used, in which ordinary thermometers were replaced with resistance
thermometers. The positions of the thermometers was measured with a
Wheatstone bridge. The psychrometers were installed at levels of 0.5, 2.0,
4.8, 9.7 and 16.7 m; up to a height of 2.0 m they were mounted on poles,
and above, on a second telescopic mast. In addition, the air temperature
and humidity were measured simultaneously with ordinary Assman psychrometers
installed at heights of 0.25, 0.5 and 2.0 m. At levels of 0.25 and 0.5 m,
the psychrometers were mounted horizontally, and at 2.0 m, vertically.
The soil temperature was measured with resistance thermometers (of the
same design as for measuring the air temperature) , mounted at depths of 2,
5, 15, and 20 cm and on the surface of the ground. Simultaneously, at the
same depths, the temperature was measured with "Savin" thermometers, and
the temperature of the surface of the soil was also measured with a periodic
thermometer.
The balance observations included the determination of evaporation
from the soil and actinometric measurements.
The evaporation was measured by means of microevaporators, consisting
of cylindrical metal containers 11.2 cm in diameter and 7.0 cm high. Every
day, before each period of observations, the evaporators were filled with
soil, with the structure of the soil being left as undisturbed as possible.
The microevaporators were set out on the platform every hour, during which
gradient and balance observations were made.
For actinometric observations, thermoelectric instruments were installed
on the platform: an actinometer, a pyranometer and a balansometer, which
-------
were used to determine the radiation balance of the underlying surface,
the direct solar radiation on the perpendicular and horizontal surfaces,
the reflected shortwave radiation, scattered and total radiation, and also
the albedo and effective earth radiation.
The observations were made from 12 September through 12 October 1961.
In September, the observations were performed only in the daytime, mainly
during the period when the concentration was measured (at 9:30 - 10:30 AM,
11:30 AM - 12:30 PM, 1:30 PM - 2:30 PM, and 3:30 PM - 4:30 PM).
In order to calculate the atmospheric pollution under conditions when
no direct measurement of dust and gas concentrations were taken, round-the-
clock series of observations were performed at 1-hour intervals. They were
set up daily from 2 through 12 October and also on 25-26 September.
Upon completion of the work of the expedition, regular observations
were continued at the station. The program of these observations included
the determination of the wind direction by means of the M-12 recorder,
measurement of the wind velocity with contact anemometers at heights of
0.25, 0.5, 1.0, 2.0, and 5.0 m, measurement of the air temperature at
heights of 0.25, 0.5, and 2.0m (with Assraan psychrometers) and 0.5 and
2.0 m (with resistance thermometers), measurement of the soil temperature
on the surface and at depths of 2, 5, 10, 15 and 20 cm with resistance
thermometers, actinometric observations with the balansometer, pyranometer,
actinometer, and visual observations of the cloud cover, visibility, and
atmospheric phenomena.
During the first days of the work of the expedition (12-15 September)
the weather was primarily determined by the influence of a large cyclone
with a number of centers and fronts associated with this cyclone. Start-
ing on 15 September over the central part of the European territory of the
USSR, an anticyclonic transformation began, which resulted in the onset of
a general high-pressure band stretched from west to east.
Starting on 18 September, the slow moving high lost its independent
significance and changed into an extension of a newly formed, larger, and
developing anticyclone over the Scandanavian peninsula. However, up to
20 September, the region of the expedition remained under the influence
of passing diffuse fronts, which caused cloudy weather.
Later, the anticyclone from the Scandinavian peninsula, growing
stronger, began to move in a southeastern direction, and its influence
spread to the region of the expedition.
From 20 September to 12 October, the weather was determined by the
influence of this anticyclone (Fig. 2 and 3), and for this reason fair
weather prevailed, sometimes with a poor visibility.
-------
Fig. 2. Synoptic oap for 3 P.M., 1 October 19&1.
The weather was considered fair when during the observation period
the upper and middle cloud cover did not exceed 7 points, and the lower
cover, 3-4 points.
In the processing of actinometric observations, a recording of the
state of the solar disc (0°, O, 02. n) was taken into consideration.
For an upper and middle cloud cover of 8-10 points and a lower cloud cover
of over 6 points, the weather was considered overcast.
Results of all the observations were collected in a table and average
values for 1 hour of the measured meteorological elements were found.
Let us first examine the results of actinometric measurements, made
from 9 A.M. to ' P.M., i.e., during the period when the concentration of
impurities in the atmosphere was measured. The magnitude of total radia-
tion Q during these hours varied on overcast days from 0.05 to 0.40 calico?
min. , and on fair days, from 0.2 to 1.0 cal/cm2 min. At the same time,
the direct solar radiation on a horizontal surface S1 measured 0.1 to 0.9
cal/cm2 min, and the direct solar radiation on the perpendicular surface S,
from 0.3 to 1.3 cal/cm2 min. The radiation balance B of the underlying
surface measured from 0 to 0.9 cal/cm2 min. On the average, the radiation
balance amounted to 0.15 cal/cm2 min on overcast days and 0.30 cal/ctn2 min
on fair days.
The diurnal variation of the direct radiation, total radiation, and
radiation balance averaged over the period of the expedition and calculated
for overcast and fair days is shown in Fig. 4. It is apparent that the
maximum value of all these quantities is observed between 11 A.M. and 12 m.
-------
On overcast days, the total radiation and radiation balance undergo rela-
tively little change, and the amplitude of their value does not exceed
0.15 cal/cm2 min. On clear days, the amplitude and maximum values of Q
and B are 3 times the values on overcast days. According to the observa-
tional data, the albedo was equal to an average of 0.18, and in some
cases amounted to 0.14-0.30.
(,11 001 0* 09 0* OZ II 0! i»
Fig. 3. Synoptic map for 9 A.M., 14 October 1961.
On the average, the effective emission E for the period observed
changed from 0.08 to 0.18 cal/cm2 min. Since the effective emission in
the daytime was determined as the remainder term of the radiation balance
equation, it was useful to compare the values obtained with the results of
calculations of E. To this end, calculations of E were made according to
M. Ye. Berlyand and T. G. Berlyand [1], using data on the air temperature
and humidity at a height of 2.0 m and on the air - soil temperature differ-
ence. The results of the calculations and observations were in mutual
agreement.
Let us now consider the gradient observations. Average data for each
hour were used to plot the profiles of the air temperature and humidity,
soil temperature, and also wind velocity. All the profiles were obtained
on the semilog scale. The plotted profiles made it possible to check the
quality of the observations, analyze the material, and compare synchronous
observations of one and the same element by means of different instruments.
In addition, mean profiles of the values of meteorological elements were
plotted for cases of fair and overcast weather.
Fig. 5 and 6 show mean velocity profiles for fair and overcast days.
The profiles for overcast days for periods from 6 P.M. to 8 A.M. were
plotted on the basis of only two available serial observations. It is
-------
evident that, on the average, the wind velocity adequately follows a
logarithmic distribution. In the majority of cases, the deviation of
the values from straight lines does not exceed 0.1-0.2 m/sec, and only
in some isolated cases does the maximum deviation amount to 0.3 m/sec.
cal/co? min
>.or
N
0.5
V
\
_L
15
16 hours
1 - B0, 2 - Qol
"9 W 11 12 13 I"
Fig. A. JDiurnal variation of components of radiation balance in fair-
and in overcast_(Bp, Q0, So) weathejp.
3-^Q, 4- Q0, 5-S0, 6- SQ-
The mean values of the wind velocity in fair weather ranged from
0.5 to 2.5 m/sec at a height of 0.25 m and from 1.5 to 3.5 m/sec at a
height of 16.0 m; in cloudy weather, from 1.5 to 3.0 m/sec at the lower
level and from 3.0 to 5.0 m/sec at the upper level.
The highest wind velocity at the uppermost level (16.0 m) was 6.0
m/sec in fair weather and 11.0 m/sec in overcast weather based on periodic
observations.
Tables 1 and 2 list data on the daily variation of the wind velocity
averaged over fair and over overcast days at various heights. As is evi-
dent from the table, the maximum wind velocity at all heights was observed
during the period from 12 m. to 4 P.M. and the minimum, between 10 P.M.
and 12 P.M.
The mean amplitude of the daily variation of the wind velocity A on
fair days was approximately the same at all heights and equal 1.9-2.3 m/sec.
On overcast days an amplitude of about 2.0 m/seC is preserved up to a
height of 1.0 m, and starting at 2.0 m, it increases to 3.5 m/sec.
As was noted above, the measurement of temperature and humidity at
-------
thermometers simultaneously. Comparison of the temperature and humidity
values obtained with both instruments showed that the difference between
them was slight (Tables 3 and 4) and subsequently the material was treated
only by using data of the resistance thermometers.
* hours
9-25~l. U 23 tf 0* 0.6 0,4 1.6 f3 1,0
Fig. 5. Profile bf mean wind velocity for fair days.
hours
for 2 days
Fig. 6. Profile of mean wind velocity for overeat days.
-------
Table 1
Daily Variation of Wind Velocity (m/sec) in Fair Weather.
z m
0,25
0,5
1,0
2,0
5,2
9,7
16,0
Hours
10
1,3
1,4
1,7
1,8
1,7
1,7
1,8
1,8
12
2,2
2,4
2,7
30
3,1
33
3,1
3,4
14
2,3
2,6
29
3,2
2,9
3,1
3,2
16
1.9
2,1
2,3
2,5
2,9
33
34
3,4
18
0.4
0,6
0,9
1,2
1,4
1,8
1,9
2,1
20
0'.7
1,0
1,2
1,6
2.0
2,2
2,4
22
0,4
0,4
',0,6
0,9
1.2
1.4
1,4
1,4
0
1,0
1.0
1,2
1,4
1.9
2,3
2.3
2.4
4
0,9
1,0
1,3
1,5
2,0
2,3
2,3
2.3
6
1,0
1.1
1,3
1,5
1,9
2,3
2.5
2,5
8
1.5
1,7
i,y
2,1
2,4
2.8
2,9
3,1
A
1,9
2,2
2,i
2,3
1.9
1,9
2,0
2.0
Table 2
Daily Variation of Wind Velocity (m/sec) on Overcast Pays.
Hours
z m
0,25
0,5
1.0
2,0
5,2
9.7
11,7
16,0
10
2,7
3,1
36
4,0
4,6
5,1
5,2
5,2
12
3.1
3,4
4,0
4,5
5,2
5,7
5,7
5,8
14
3,2
3,5
4,0
4,6
5,4
6,0
6.0
6,0
16
2,9
3,2
3,8
4,3
5,0
5.6
5,8
6,1
18
2,0
2,2
2,5
2,8
3,6
4,2
4,4
4,7
20
1,7
1,8
2,2
2,5
3,2
3,9
4,1
4,3
0
1,3
1,3
1,6
1,7
2,0
2,6
2.8
3,0
4
1.4
l|4
1,7
18
2,4
2,9
31
3,4
6
1,7
1,7
2,0
2,2
2,6
3.1
3.4
3,6
8
2,0
2,\
2,4
2,4
2.8
3.3
3,4
3i6
Table 3
Average Distances of Air Temperature Values (deg.) Based on
the Resistance Thermometers and Assman Psychrometers
at Heights of 0.5 and 2.0 m.
0,5
2,0
Hours
10
0,1
0,0
12
0.5
0.2
14
0,5
0,0
16
0,1
0,0
18
0,3
0,1
20
0,0
0,0
22
0,0
- 0,2
0
—0,2
0,2
4
-0.2
0,0
6
-0,5
-0,3
8
'0,1
0,1
-------
Tnble
Mean Monthly Differences of Air Humidity Values (mb) Based on Resistance
Thermometers and Assman Psychrometers at Heights of 0.5 and 2.0 m.
z in
0.5
2,0
Hours
10
0,9
0,5
12
0,8
0,6
14
1,0
0,8
16
0,7
1,2
18
0,8
1.8
20
1.3
1,6
22
0,6
0,7
0
0,8
0,8
4
0,7
0.6
6
0.6
0.4
8
0.6
0,9
Table 5 lists data on the daily variation of the air temperature at
all the heights. Since on days of serial observations primarily fair
weather was observed, for overcast days it was possible to detect only a
diurnal temperature variation at the different heights. Data for over-
cast days are given in Table 6.
Table 5
Daily Variation of Air Tenroerature (deg. C.) in Fair Weather.
z m
0.5
2,0
4,8
9,7
16,7
Hours
10
7,7
6,9
7.0
7,0
6,9
12
13,1
11,9
11 0
11,5
11,3
14
13,7
12,1
11 5
11 5
11,2
16
12.7
12,0
11,8
11.8
11,4
18
8,8
9,6
9,9
10,2
10,3
20
6,3
7.1
7,6
8,2
8,7
22
4,9
5.2
5,6
6,0
6,6
0
4,4
4,8
5,3
5.6
6,0
4
2,7
3,4
4,5
4,3
4,2
6
2.1
2,6
3.0
3,2
3,3
8
3,4
3,4
3,6
3,7
3,8
A
11,6
10,3
9,0
y.a
9,0
As is evident from Table 5, the daily amplitude of the air temperature
on clear days at heights of 0.5 and 2.0 m is 10-12°C., and above 2.0 m,
9.0-9.5°C. The diurnal amplitude on overcast days is slight and amounts
to 2.0-2.5°C. at all the heights.
Table 6
Diurnal Variation of Air Temperature
P ^ nn flya-pna gf Days,
z m
0,5
2,0
4,8
9i?
1C ,7
Hours
10
10,0
9,4
9,4
9,4
9,1
12
12,0
11.3
11,2
11.2
10,8
14
12,7
11,9
11,8
11,7
11,6
16
12,2
11,8
11,6
11,6
11,3
-------
10
4
9,T
24-
fS hoars
10,0
12,0
12.7
12,2'
Fig. 7. Profile of mean air temperature" for
overcast days.
10 12 11
16
05
77 131 13.7 12.7&8 63 49 4* ~"?T 2J~ J. '
Fig. 8. Profile of mean air temperature in fair weather.
-------
The maximum value of the air temperature on fair and on overcast
alVeb8elVe\at I *•;• «d amounted to 13.7°C. at a height of 0 5 m
and 11.3 C. at a height of 16.7 m in fair weather. On overcast days
I 5 rSrTl^r"^^1^18'111^ 3nd amOUnted to 12-7°C' « * heit of
0.5 m and 11.6 C. at 16.7 m. The minimum in fair weather was observed
before sunrise at about 6 A.M. The temperature values averaged over the
overcast and over clear days were used to plot the profiles ?Fi« 7 JM *n
The illustrated figures show that on the average* to within £&. Se
temperature at all the heights falls on straight lines of a logarithm^
profile.
The temperature differences between the heights of 0.5 and 2 0 m
amounted to 0.8-1.2'C. during the day and -0.4, -0.8°C. at night during
fair weather and 0.2-0.4°C. during the day and 0.0-0.2°C. at night on over-
cast days.
The humidity observations were subjected to a similar treatment. The
data obtained show that the daily variation of the humidity is rather in-
distinct. Its average amplitude for clear days at all the heights ranges
from 0.5 to 1.5 mb.
The maximum value of the absolute humidity, based on the same averaged
data, is observed at a height of 0.5 m and amounts to 8.4 mb; it decreases
with increasing height.
On overcast days, the absolute humidity is higher than in fair weather
and equal to 11-12 mb in the daytime at a height of 0.5 m and to 9.5-10.5 mb
at a height of 9.7 m.
Vertical profiles were plotted from average values of the absolute
humidity for overcast and fair days (Fig. 9 and 10).
The roughness of the underlying surface was determined from the plotted
wind profiles under close-to-equilibrium conditions. The equilibrium con-
ditions taken were those in which the parameter &T"(where AT is the difference
"'
in air temperatures at the levels of 0.5 and 2.0 m and u^ is the wind velocity
at a height of 1m), characterizing the state of constancy, was equal to or
less than 0.05 in absolute value. The average value of the roughness parameter
of the underlying surface was found to be approximately equal to 1 cm.
Table 7
Mean Values of the Turbulence Coefficient (m?/sec) at
a Ht>ig)rh of _Ljj on Overcast Pays.
by M.
by D.
by M.
by the
I. Budyko's method
L. Laykhtman's method
P. Timofeyev's method
heat balance method
flours
10
0,13
0,14
0,12
12
0,14
0,11
0,12
0,12
14
0,14
0,17
0,11
0,05
16
0,12
0,12
0,10
0,06
-------
ra 10 12 it //? 18 20 22 0_ t 6 8 hoars
8,4 8,4 8$ 428$ 8,3 7.8 «J tO^ 8.3 8.1
Fig. 9. Profile of mean absolute humidity in fair weather.
10 . /.?. W # . hours
16.7
2,0
11,9
ft. 2mb
Fig. 10. Profile of mean absolute humidity on
overcast days.
-------
In order to characterize the turbulent exchange from observational
data for every hour in the daytime, values of the turbulence coefficients
were calculated at a height of 1 m by various methods based on gradient
observations, according to [2, 3, 5], and by the heat balance method
[3, 4]. Results of the calculations are shown in Tables 7 and 8, which
list mean values of the turbulence coefficient for overcast (Table 7) and
for fair (Table 8) weather.
The turbulence coefficient in fair weather ranges from 0.09 to
0.12 nrVsec according to the methods of M. I. Budyko and M. P. Timofeyev
and from 0.05 to 0.11 m2/sec according to the method of D. L. Laykhtman
and to the heat balance method.
Table 8
Mean Values of the Turbulence Coefficient (n^/seo) at
a Height of 1 m on Fair Days.
by M. 1. Budyko 's method
by D. L. Laykhtman 's method
by M. P. Timofeyev's method
by the heat balance method
Hours
10
0,11
0.05
0,10
12
0,09
0,10
0,11
0,10
14
0,12
0,05
0,11
0,11
16
0,10
O.OG
0,09
0,05
In overcast weather, the absolute values of the turbulence coefficient
as given by all the methods are higher than in clear weather, and range
from 0.05 to 0.12 m2/sec as given by the heat balance method and from 0.10
to 0.15 m2/sec according to all the other methods.
The fluctuations of the turbulence coefficient from noon to 4 P.M.,
calculated by using the heat balance method, are approximately the same
for fair and for overcast weather.
The turbulence coefficient for a 10-hour period could not be calculated
by using the heat balance method because of the lack of observations for an
eight-hour period or because of the slight difference (less than 0.1 cal/cm2
min) between the radiation balance values and the heat flow into the ground.
The results obtained make it possible to characterize the meteorological
conditions of the distribution of discharges from stacks in the ground layer
of air.
LITERATURE CITED
M E FIpeACKasaHnc M pcryvrnpcmaHife reruioBoro pe^KHMa
C.TOH aivtoccpepu Fn.apOMeTeon3.uaT, 71, 1956
2 B > j bi K o M M HcnapcHHc B ecrrecTBeHHbix ycjionnsix FnApoMereoH3AaT, Jl, 1948
3 BpeweHHue MercuimecMic yKasaHiin nupoMCTeopoJioni'iecKHM cTanuHHM, JSTs 5
MCTeomAaT, 71, 1961
4 Fop6>HOBa H F, CepoBa H B Ten;io
-------
CHARACTERISTICS OF THERMAL STABILITY IN
*
THE GROUND LAYER OF AIR
V. P. Gracheva
From Trudy, Glavnaya Geofiz. Observat. im. A. I. Voeykova, No. 238,
p. 153-179, (1969).
Richardson's number Is usually employed as a stability characteristic
allowing not only for thermal factors (temperature distribution in height)
but also for dynamic factors (change of the wind velocity with height).
This number is frequently replaced by a parameter that is readily determ-
inable and functionally associated with it, equal to the ratio of the
temperature difference at two levels to the square of the wind velocity at
a third level in the ground layer of air (B = Ak,T5T'. Parameter B has been
*h
widely employed in studies of the turbulent regime in the ground layer of
the atmosphere by M. I. Budyko, A. R. Konstantinova, D. L. Laykhtman,
T. A. Ogneva, M. P. Timofeyeva, and others.
Different levels are taken as 2^(i = 1, 2, 3), but in the Soviet
literature they are most frequently 0.5 and 2.0 m for the temperature and
1 m for the wind velocity. The use of gradient observations of heat bal-
ance stations makes it possible to identify certain characteristics of the
distribution of the stability parameter at different points of the terri-
tory of the Soviet Union at different times of the day and year and» in
addition, to elucidate the associated patterns of distribution of certain
characteristics of the turbulent exchange. B is essentially a basic para-
meter on which the turbulence factor depends in any of the schemes now in
use. Other characteristics of turbulent exchange dependent on the
meteorological conditions and associated, in particular, with the horizontal
components of uie mixing factor also are determined via this stability
parameter.
The knowledge of the characteristics of a turbulent regime is necessary
for studying the diffusion of impurities and air pollution under various
meteorological conditions [2, 3, 5, 15]. Thus, the study of the variation
of the stability parameter in different regions of the Soviet Union is of
considerable interest.
Despite the numerous studies where this parameter is used, the problem
of its distribution in space and time has been inadequately investigated.
Most significant in this connectio.n_is the study by T. A. Ogneva [10], where
the annual and daily variation of ** for various weather conditions is given
«i
on the basis of observations made at Koltushy (Leningrad Oblast') in 1947-
1951. Basically, however, we know from the literature the daily course of
-------
the stability parameter, determined on the basis of expeditionary data
collected for individual months in certain regions.
The object of our study was therefore to investigate the distribution
of the stability parameter in the ground layer of the atmosphere /_^f ^ over
<«? i
the territory of the USSR and in time. The distribution over the territory
was considered for the most unfavorable weather conditions of impurity propa-
gation from high smoke stacks, namely, in a developed turbulent exchange.
The latter was assumed to be most probable around midday in summer. As a
result, the material of gradient observations for the 1 P.M. period in July
was used. Data of 67 stations for the 1955-1967 period were treated.
To study the change of the stability parameter with time, its daily and
annual course was analyzed, as determined from a five-year interval (1963-
1967) for seven stations of different landscape zones. The calculation of
B was based on daily temperature difference data for each period at the 0.5
and 2 m levels (Atg. 5-2. 0^ and wind velocity data at a height of 1 m (u^) -
Since the wind velocity observations were made at the stations at the 0.5
and 2.0 m levels only, the wind velocity at the height of 1 m was taken as
the average of these levels (corresponding to the logarithmic law of wind
velocity distribution with height). In the absence of data on the wind
velocity at the height of 0.5 m, or since its value was dubious, UT was taken
equal to 0.85 U£ (in accordance with the logarithmic law for a roughness of
the underlying surface of 2 cm) or taken with another factor according to the
table in Ref. [6] for other values of the roughness of the underlying sur-
face. According to the instructions, no gradient observations were made
during a strong wind, precipitation, fog, dust storm or thunderstorm. The
air temperature was measured only at the height of 2 m (in a psychrometer
booth), and the gradient was assumed equal to zero in these cases. In the
calculation of average monthly values of B, such cases were not considered,
since a gradient equal to zero does not always correspond to the reality,
especially in the summertime in the southern regions. Cases with a wind
velocity of less than 1 m/sec at a height of 1 m were also discarded.
It_should be noted that the relative errors in the calculation of para-
meter Af-t with the existing accuracy of temperature and wind velocity measure-
«?
ment at the stations (6(At) = 0.2° and 6u = 0.1 + 0.03 u) are generally very
large. For example, when |At|<0.6° and u ranges from 0.5 to 10 m/sec,
>40%. It was therefore impossible to use only cases with an error of,
say, less than 40%. Consequently, one would have to reevaluate an excessive
amount of data, and in the central and northern regions, almost all of the
data, since At>0.6° is seldom observed in these regions. Averages for the
period 1955-1967 were then calculated from the computed average monthly values
of B for the individual years.
Because of a change in the method of temperature observations in 1965
(vertical mounting of the psychrometer at the 2-meter level replaced by
-------
horizontal mounting) the mean monthly values of -^- were analyzed separately
"i
before and after. 1965 in individual years. At a number of stations, the
values of the stability parameter during the first period (up to 1965) were
systematically lower than after 1965. At all of these stations except three
(Aydarly, Fort Shevchenko and Artema Island), the wind velocities in summer
in the daytime were usually below 3 m/sec. However, according to Ref. [11,
12], the divergences in gradients due to different modes of suspension of
the psychrometers are insignificant. Gradient values which were too low
(a vertically suspended psychrometer gave a diurnal air temperature that
was too high) could have been given only at those stations under considera-
tion where the wind velocity was above 3 m/sec, namely, at the stations of
Aydarly, Fort Shevchenko and Artema Island, and this should be taken into
account in the analysis of the data.
At 56 stations,l^4 were calculated for a period of over 2 years, and at
«T
the remaining 11 stations, for 1-2 years. Series of observations obtained
at different stations were highly heterogeneous. As a result, in accordance
with Ref. [6], successive averaging over all the longer periods was used,
beginning with the last year of observations, then the data of averaging
over short periods were compared with data for the longest periods of
averaging at a series of stations. As a result, the values of B were found
to depend very strongly on the period of averaging. Comparison of averages
for an 8-13 year period at 24 stations with averages for the 1-3 year period
showed that at only nine stations during the three year period did the rela-
tive error fall below or at 30%, whereas it exceeded 30% at 15 stations.
Therefore, in order to obtain more comparable values, it was necessary to
average the most uniform data for the latest years beginning with 1961.
At 39 stations, average B were calculated for a 5-7 year period. Values
of B determined during short periods were treated as tentative. Average
values of B in July at 1 P.M. for the 1961-1967 period for different regions
of the USSR are shown in Table 1. The last column of Table 1 gives the
average wind velocity at a height of 1 m, calculated for the same period as ~ .
U1
To analyze the results obtained, we shall use L. S. Berg's scheme for classi-
fying the territory of the USSR according to landscape zones [1]. By natural
landscape zones Berg meant areas similar in predominant character of the
relief, climate, and plant and soil covers. Eight landscape zones are con-
sidered. It should be noted that some stations located in zones of semideserts,
deserts, desert sands and mountain landscapes do not characterize these zones
as such, since the stations are usually located on floodplains, river shores,
and also irrigated sites (oases). Therefore, in the indicated zones, the
quantity J£ substantially depends on the water facilities situated in the
«?
vicinity and on the degree of sheltering of the stations. Thus, at the Artema
Island and Fort Shevchenko stations, located on the Caspian Sea coast in an
open area, the values of the stability parameter are small because of the
slight temperature differences in height and high wind velocities. (According
-------
to the data for 1965-1967, the values of ^, are 0.09 and 0.11 respectively).
At the Dushanbe and Fergana stations, located in oases, where the wind
velocities are small and At substantial, the values of M_ are higher.
Analysis of the material presented in Table 1 indicates that for summer
conditions at noon, the highest values of the stability parameter (above 0.20)
are characteristic of submontane regions or foothills, Central Asia, south
of the European territory of the USSR, and central regions of eastern Siberia.
In the southeast of the European territory, in Kazakhstan and Central Asia,
as a result of high wind velocities in the summertime and somewhat decreased
values of the radiation balance in the desert zone [11, 12], and hence, of
At, the stability parameter has a lower value, of the order of 0.15-0.20.
In the central and northern regions of the European territory and Ukraine,
and Baltic regions, _AJ is of the order of 0.10.
The stability parameter values obtained are somewhat higher than those
given in Ref. [7]. The divergences are explained by different methods of
calculation. In the present study the stability parameter is calculated
from daily initial data for At and uj_, whereas in Ref. [7] it was based on
mean monthly values.
It is well known that average values alone do not constitute a complete
characterization. Therefore, for a number of stations located in different
landscape zones, the frequency of the different limits of A-Lwas calculated
under the following weather conditions: clear, overcast, variable cloudiness
and average conditions. The weather conditions were determined from the de-
gree of masking of the solar disc by the clouds. Daily values of B in July
for the 1 P.M. period during 1955-1967 were used for this purpose. Table 2
gives the frequency of the different limits of A* under different weather
u
conditions in percent of the number of cases under corresponding weather
conditions, and also the frequency in percent of the total number of cases
under average conditions and when the wind velocity at a height of 1 m was
less than 1 m/sec or when the gradient observations were not made.
The data of Table 2 show what kind of weather conditions predominated
in any given landscape zone during the period considered, and also what
changes of the stability parameter were the most probable under different
weather conditions. At all stations with the exception of Smolensk, Fergana,
Chardzhou, Tamdy, and Dushanbe, weather conditions with variable cloudiness
predominated (in approximately 40-60% of all cases).
The remaining cases were distributed with the predominance of clear
weather at the stations of Siberia and south of the European territory of
the USSR and overcast weather in the northern and central regions of the
European territory.
-------
Table 1
Average Values of ££ in Different Landscape Zones of the USSR. July, 1 P.M.
Zone
No.
1
2
3
4
Zone
Tundra
Tayga
Tayga with an ad-
mixture of broad-
leaved species
Forest steope
Chernozem steppe
Station
Kotkino^-
Srednekolymsk1
Khibiny
Arkhangelsk
Ust'-Vym1
Petrozavodsk
Kargopol1
Nolinsk
Tura
Turukhansk
Verkhoyansk
Oymyakon
Yakutsk
Aldan
Kostroma
Nikola ye vskoye
Riga
Tiyrikoyya
Pinsk
Smolensk
Torzhok
Toropets
im. Nebol'sin
Khomirtovo
Kushnarenkovo
Pavelets
Sovetsk
Cheben'ki
Beregovo
Ogurtsovo
Solyanka
Khakasskaya
Kuybyshev1 (flood-
plain of Volga River)
Borispol1
Poltava
Kamennaya Step1
Chita
Mangut
Askaniya-Nova1
Period of
Observation
1964-1967
1966—1967
1963—1967
1964—1967
1964—1967
1962-1967
1964-1967
1661—1967
1965-1966
1965—1967
1964—1966
1967
1961-1967
1966-1967
1961—1967
1961—1667
1961-1967
1961-1967
1963—1967
1961-1967
1961-1967
1961—1967
1961-1967
1963—1967
1963-1967
1961-1967
1965—1967
1967
1961—1967
1961—1963,
1965-1966
1961—1967
1965-1967
1961,
1963—1967
1961-1967
1961 — 1967
1963-1967
1964—1967
1966-1967
1961—1967
T
0.09
0,14
0.15
0,10
0,10
0,06
0,10
0,09
0,12
0,18
0,26
O.H
0,29
0,27
0,13
0,15
0,09
0,07
0,13
0,08
0,07
0,09
0,09
0,30
0,10
0,12
0,12
0,21
0,16
' 0,12
0.23
0,37
0,19
o.n
0,10
0,05
0,26
0,17
0,21
tii
2,6
2,1
2,4
2,2
2,4
2,6
2.2
2,5
2.1
2,4
2.Z
1,7
2,1
1,9
2.3
2,5
_ 2,5
" 2,7
2.3
2,7
2,5
2.7
2,4
2,0
2,4
2,t>
3,5
3.2
1,9
2,9
2,3
1.7
2,1
2,8
2,9
2.8
2.?
2,2
2.7
-------
Table 1 (Cont'd)
Zone
No.
5
6
7
8
1 £
Zone
Dry steppe
Semi-desert
Desert
Sands of desert
zone
Mountain
landscape
Lowlands of the
Amur and Ussuri
border with Man-
churian-type
forests
Station
Rudnyy
Tselinograd
Yershov
Astrakhan' (Flood- .
plain of Volga River)
Gigant
Nakhichevan'
Kalmyk ovo
Balkhash
Karasuat
Telavi
Fort Shevchenko
Churuk
Aydarly
Beki-Bent
Fergana (Oasis)
Ak-Molla
Chardzhou
Tamdy
Frunze
Artema Island
Nikitskiy Sad
Dushanbe (Oas s)
lermez (Oasis)
Skovorodino
Kyzyl
Bomnak
Primoraksya
Tolstovka
>tations located in transition zones
Period of .•*'.
Observation , J 1 «-i
J- 1
1962—1907
1963—1967
1966—1967
1963—1967
1961-1967
1966—1967
1961—1966
1965-1967
1965-1967
1962—1967
1961-1964.
1966—1967
1966-1967
1965-1967
1962—1967
1961—1967
1964-1967
1961,
1963-1967
1963—1967
1961—1967
1964—1967
1961—1967
1965-1967
1966-1967
1961-1967
1964—1967
1962—1967
1965—1967
' 1967
1965-1967
1966—1967
0,16
0,17
0,14
0,24
0,22
0,24
0,16
0,14
0,15
0,17
0,01
0,11
0,21
0,25
0,29
0.34
0,11
0,10
0,17
0.43
0,07
0,09
0.37
0.48 '
0,22
0.15
0,19
0,23
0,11
0,15
3.3
3,6
3.4
2,4
2.3
2.4
3,5
4.2
3,8
1,9
4,8
4.8
3.3
3.0
2.6
1,2
3.7
2,0
3.7
l.b
4.7
4,3
1.7
1.2
2,5
1,9
1.8
1.1
2.9
2,5
-------
Frequency of Different Limits
St&trioii
Clear
ooi'o->
-0,100<0,0
8
I
o
I
0
5
I 0,201-0,300
8
o
i
o"
1
A
% I^
-------
<*«»
July, 1 P.M.
Table Z
Variable cloudiness
8
V
o
a
V
g
o"
8
o"
i
g
c»i
o"
1
o
o"
o
1
8.
o
o"
1
1
s
V
o"
A
•fc
n
f S
o
° 0 U]
§r-l 0)
II?
2
4
2
1
3
18
13
46
52
57
23
28
19
14
11
26
26
5
8
8
16
18
13
2
5
6
8
12
2
4
29
19
59
88
178
31
128
48
48
58
50
52
broad-leaved species
1
1
2
20
12
10
22
3
38
48
54
50
35
18
17
24
19
11
11
8
5
2
7
3
5
4
3
11
9
9
3
2
33
198
211
178
115
67
59
57
48
31
43
steppe
3
2
28
45
25
27
9
9
10
6
25
11
64
170
52
55
steppe
1
9
2
1
46
32
41
29
25
20
6
11
11
5
11
11
4
19 '
16
97
47
177
63
51
52
steppe
55
20
12
95
51
desert
2
2
1
29
49
25
29
15
12
9
5
18
4
130
77
47
41
ert
8
1
4
26
34
8
27
22
17
16
12
17
15
7
8
15
21
42
74
100
12
48
40
10
desert zone
7
57
61
15
13
14
7
5
7
14
14
79
23
32
landscapes
I
4
11
13
35
13
24
8
5
15
4
51
17
39
78
16
50
-------
Station
Nolinsk
Yakutsk
Gigant
Kalmykovo
Fergana (oasis)
Dushanbe ( oasis} . ....
Skovorodino ....
Overcast
§
°
V
3
2
2
3
1
5
4
3
5
50
7
o
0
V
8
7
26
16
22
24
14
?5
6
9
19
9
9
11
11
4
5
33
36
ooi'o-o'o]
66
51
60
50
41
59
53
6G
65
55
56
60
62
57
58
71
53
56
75
71
100
82
50
39
I 0,101-0,200
17
14
9
24
10
17
8
5
28
" 1
21 ;
9'
i I7
13
11
17
28
19
4
f 0,201 -0.300
4
3
5
17
12
2
4
8
4
11
13
4
5
9
9
6
8
25
33
9
7
5
G
4
12
TE
1
3
I
6
2
7
5
4
34
A
s
3
4
33
9
tiga wi
4
1
12
4
4
9
5
9
3
IA
h 0)
fen
5
29
35
55
6
42
;h admi
68
83
88
130
18
16
53
C
21
23
55
28
36
25
4
21
3
,!
2
28
0 tl
(DrH C
O C0 C
a>t>«H
a.E-i o
T
23
19
18
10
17
jcture
20
22
24
35
12
Fore
13
17
hernozi
13.5
25
16
D
15
Sei
13
14
D
2
9
3
Sands ..(
3
4
Mounta.
1
18
ay
of
st
2m
py
ni
iS
rf
In
-------
Table
-------
Values of the limits of B with the highest frequency under average
weather conditions are in fairly good agreement with the average values
of the stability-parameter listed in Table 1, thus confirming the validity
of the calculated average values.
As already noted, in order to study the characteristics of the varia-
tion of the stability parameter with time, the annual and daily course or
variations of this parameter were analyzed at a number of stations located
in the zones of the tayga, forest steppe, and mountain landscapes. The
calculations were made for the five-year period from 1963 to 1967 with the
exception of the winter months of 1963 and 1964, when no gradient observa-
tions were made at the stations.
The annual course was determined for the 1 A.M. and 1 P.M. periods,
and the daily course, for January and July. However, since no observations
were made in winter at 7 and 10 A.M. and 4 and 7 P.M., the daily course in
January is given only in terms of two periods of observations, nocturnal
and diurnal. Results of the calculations are shown in Tables 3 and 4.
In analyzing the daily variation of J±L , one must note first of all
its large amplitudes in summer and small amplitudes in winter, the increase
in amplitudes from the tayga zone to the steppe zone, and a considerably
larger daily amplitude on the Asian territory of the Soviet Union, particu-
larly in the summer period as compared to the daily amplitude on the
European territory of the USSR.
The maximum values of the stability parameter in the daily course in
summer were observed between 10 A.M. and 1 P.M. and the minimum values,
primarily at night. The transition from negative to positive values in
the morning and from positive to negative values in the evening is difficult
to determine because of large gaps between the periods of observations, but
it may be stated, nevertheless, that in summer during the 7 A.M. period
the average values of B are positive for all stations, and negative during
the 7 P.M. period. In winter, the values of B are positive at noon for
the southern regions and negative for the northern and mountainous regions;
during the 1 P.M. period, the B values are negative for both the northern
and southern regions.
In the annual course of the stability parameter, the maximum values at
noon are observed in the summer months, and the minimum values in the winter
period. In the forest steppe and steppe zone, the average values of the
stability parameter are positive the year round during the 1 P.M. period,
and negatiye_in the zone of the tayga and mountainous landscapes. The annual
course of A£. is different for different landscape zones at night. Whereas
"i
for the dry steppe regions (Askaniya-Nova and Poltava) the annual course of-JT.
at night is the opposite of the one during the day, i.e., with a maximum in
winter and a minimum in summer, in the more humid regions of the tayga zone
-------
.All
Annual Course of B2 (196J-1967)
Table 3
Hour
1 A.M.
1-P.l.
1 A.M.
1 1 A.M.
-j 1-P.M.
1
1 A.M.
1 P.M.
1 A.M.
1 P.M.
1 A.M.
1 P.M.
1 A.M.
I- P.M.
I
II
III
IV
V
VI
VII
VIM
IX
X
XI
XII
Amplitude
Torzhok
(layga with admixture of oroad-leaved species)
-0.112
—0.070
-0.140
0.036
—0,146
—0,038
-0.094
-0,007
-0.071
—0,006
—0,252 1 —0322 1 —0,226
0,015 [ 0,054 1 0,097
-0.129
0.076
—0.219
0.106
-0,219
0,100
Solyank
-0,199
0,179
-0.093
0.084
a (forest
—0,291
0.282
—0.124
0,072
steppe)
—0.236
0,210
—0,102
0,058
—0,265
0,201
-0.049
0.025
—0,113
0,119
—0.101
0,042
—0.080
0,030
—0,025
-0.034
—0.148
0,015
0,107
0,170
0,242
0,267
Askaniya-Nova (steppe)
—0.016
0.021
—0,040
0,020
—0,052
0,071
—0,101
0,083
—0,136
0,119
-0,121
0,150
—0.162
0.222
—0,111
0.171
—0,117
0,201
—0,090
0,127
-0,027
0,084
—0.033
0,029
0,146
0,202
Tiffiflffiansk (tayga)
—
—
-0.174
—0,032
—0.144
0,040
—0,208
0,012
-0,127
0,003
—0,170
0,107
—0,309
0,185
-O.203
0,173
-0,085
0,079
—0,015
0,001
—0,079
—0,039
-0,103
0,230-
0,288
Poltava (steppe)'
—0,022
0,011
_
—0.048
—0.067
0,011
-0.031
—0,054
0,002
—0,288
0,035
—0,111
-0,211
0,067 0,075
—0.312
0.111
SI
-0.133
0,121
—0,194
0,113
covorodino
0,114
-0,193
0,111
( mount ai
0,184
—0,165
0,123
n landsoa;
0,146
—0,144
0,072
xO
0,127
—0,119
0.072
0.114
—0,074
0,028
-0,024
—0,035
0,000
0,189
0,123
- 1 -
—0,017 1 0.232
Beresovo' (1963-1964, 1966-1967) (forest steppe)
-0,028
0,004
—0.094
0,029
—0,109
0.070
-0,200
0,102
-0,205
0,126
—0,268
0,206
—0,163
0,238
—0,175
0.206
—0,074
0.181
- 0,099
0.072
-0,051
0,077
-0,006 1 0.274
-------
Table 4
Daily Course of ~V (1963-19&7)
Moni/h
i
VII
I
VII
VII
Hour
1
lorzl
-0,112
-0,093
-0,016
-0,293
-0,140
-0,162
7
lok (Tayga
0,088
0,179
0,099
10
with adnu
0,148
Solyanka
0,286
Asksniya-
0,198
13
ixtnre of bro
-0,070
0,084
(forest s€ip
0,021
0,282
•N-ova (steppe
0,036
0.222
Turukhansk (tayp5
I
VII
I
VII
I
VII
—0,309
—0,022
-0.193
—
0,106
0,042
Skov
0,106
0,221
Poltax
0,109
orodiito (i
0,206
0,185
ra (steppe)
0,011
0,111
lountain latid
-0,048
0,184
16
ad-leaved
0,044
pe)
0,162
)
0.111
0,152
0.060
soape)
0,040
19
spe efes)
-0,102
-0,041
-0.048
-0.034
-0,072
0,153
Amplitude
0,042
0,250
0.037
0,577
0,176
0,384
0,530
0,033
0,304
—
Beregovo (1963—1964, 1966—1967) (forest steppe)
I
VII
-0,028
-0,163
0,097
0,197
0,004,
0,238
0,105
-0,055
0,032
0,401
(Turukhansk and Torzhok) a second minimum is observed at the end of winter
or in spring, and a second maximum is observed at the end of spring and
beginning of summer. This maximum is apparently due to a decrease in the
effective radiation in these regions at night, because of a higher air
humidity, a lower temperature of the underlying surface, a heavier clpudiness,
etc. At night during the entire year, '.AL are negative at all the stations
«3
considered. The annual amplitude, like the daily amplitude, is larger on the
Asian territory of the USSR in both the daytime and nightime.
We shall nov* consider the validation of the use of the commonly employed
climatological method of calculation of the stability parameter in the ground
layer of the atmosphere from daily initial data. To this end, we shall
-------
examine the dependence between the initial parameters: the temperature
difference at two levels and the wind velocity. The establishment of this
relationship may also be significant in solving a number of other problems.
For example, problems connected with the determination of climatic charac-
teristics of the heat balance (determination of the heat flow, calculation
of the vertical component of the mixing factor, etc.) through the use of
various formulas expressing the dependences of the indicated characteristics
on the values of other meteorological quantities measured directly.
The relationship between the wind velocity and the magnitude of the
temperature gradient (temperature difference at two levels) in the ground
layer has been pointed out in a number of studies. Thus, according to the
data of Best [16], during the superadiabatic period (noonday period), the
temperature gradients decrease as the wind velocity increases above 4 m/sec
(wind velocity at a height of 13.4 m above the surface). At a wind velocity
below 4 m/sec, the temperature gradient falls as the wind weakens. In the
presence of inversions, the temperature gradient decreases very rapidly
(in absolute value) as the wind grows stronger from zero to 2 m/sec (z =
13.4 m) , and at high velocities a strengthening of the wind has a slight
influence. The studies of A. S. Monin [8, 9] also note strong temperature
inversions in the presence of a weak wind, the weakening of inversions as
the wind increases in strength, and the absence of calms in the presence of
a strong instability.
The conclusions of the indicated studies were based on materials of
observations at a single point (Porton in England in the case of Best, and
the region of the 1959 Tsimlyansk expedition of the Atmospheric Physics
Institute, in the case of Monin). These studies made use of data for a
very brief period: March and June, 1932 and 1933; and July and August, 1959.
It was of interest, therefore, to study this dependence by using data
of a large number of points of different landscape zones. To this end, the
dependences of the temperature difference between the 0.5 and 2.0 m levels
(At) on the wind velocity at a height of 1 m (u^) were sought on the basis
of gradient observations of a series of heat balance stations for 1967.
Graphs of the relationship were plotted for each month during all the hours
of observation. Envelope curves were drawn through the points obtained on
each graph.
Individual points that protruded sharply were neglected.
As an example, Fig. 1 shows a graph of the temperature difference
At0 5-2 o versus the wind velocity ux for the Nebol'sin station for July, 1967,
and*Fig! 2 gives the envelope curves for the superadiabatic and inversion
states for all months of 1967 at the Rudnyy station.
The maximum monthly values of the temperature difference (positive and
negative) and the corresponding wind velocities, taken from the curves, are
shown in Tables 5 and 6. The tables also list data for a number of points
where observations were made during the expedition of the Main Geophysical
-------
Maximum Values of A *o,6—2,0 under Superadiabatic Conditions and
Station
Yea*
I
ii
in
IV
V
lay
«richang«l«*f • •
Kargtflibfk
JJolinsfc. ....
TUrWcIianSkr< . . .
XafcntsJf
1967
1967
1967
1967
1966
1965
1967
0,1
2.3-3
0,6
o.o
0,0
3
0.0
1-4
0,6
5.8
0.3
0,3
0,1
1,0; 4,0
0,5
2,0
0,9
2,4-4,3
0.3
0,2-0,8; 5,6
1,6
0,5
0.7
1,3
0,8
1,1
0,6
1A 1,6
0,6
1,3; 2,7
1.0
3,8
1,7
1.6-3,1
0,9
1,5; 4,3
0,9
3,2
1.3
2?, 3.0
Ta^ga «ith admixtuj»J
VoyeykSVB ....
RiAr
fjyrak-ay& . . .
Pinsk
Smolensk ....
im. Nebol'sin
Pavelets ....
.
Sovetek • • • •
1967
1966
1967
1967
1966
1967
1967
1967
1967
1967
1966
0.4
2,0
0,2
0.0; 0,9; 1,6
0,2
* 0,2-3,6
0.3
0,0
0,5
3,4
0,5
3,2; 3,8
0,1
0,6; 1,4; 3,6
0,2
2,1
0,1
0,6-3,3
0,1
2.2
0,7
3.4
0.8
5,3, 6,1
0,4
2.4
0,4
2.6; 5,8
1.0
1,5-2.4
0.3
2,6; 5.2
0.2
2,3
0,8
4,3
1,2
5,0
1,3
2,6
1,5
2,0; 2,8
0,8
1,2; 3,6
1,2
2,0
1,0
2,2; 4.0
1.1
Ift 3.7
1.4
3.5; 3,8
1,4
2,8
1,1
1,8
Forest
1,9
3,0
-------
Table 5
- *
are obwvnd (First ,l,ai
"i
\ll
aaA l>. Coound limo ui)~
VIII
IX
X
XI
XII
•
ga
1.1
1,8
2.2
1.8
2.5
1.0
2.5
1.6
0.8
1,6-2,0
1.6
1.4
1.4
1.2
2.0
1.2
1.5
1.2
2.1
1.4
2,0
3.1
0.8
1,7
1,0
1.2
1.8
0.4
1.3
1,6
1,6
2,6
0,8
3,1
0.9
1.6
1.2
2.0
1,2
3,4
1,0
3,1
0.3
1.0-2,6
0.5
1,5
1.1
1.8
0,8
1,6
0.2
1,3-2.3
0.3
1,3; 2,3
0.5
0,7
0,1
1.3
0.1
0.6
0.0
1,0; 1,8
0,1
2.8
0.2
0,6; 1.8
-------
Station
Year
1965
19651
19631
1967
I
1,0
2,2
ll
0,3
0,2-2,4
III
0,6
5,2
1,1
1,2
IV
1,7
4,7
0,6
2,3
1.5
4,2
V
1,5
3,4
1,8
3,8
•
st
Kanennaya Step1"
Askaniya-Nova . .
Dnestrovsk . . .
Balakovo ....
Karasqafe ....
Sovkhoz "30 teb
Oktyabrya*
Uakhtaly ....
Aydarly ....
Beki-Bent ....
Telavi
Kyzyl . . , ,
1967
1967
19671
1967
1966
19671
1967
1967
1966
1965
19521
19591
1967
1967
1967
1967
0.7
2,9
0.5
2,1-2,9
0,4
4.9
0,7
4,2
0,8
4.3
0.4
0.6; 1,8, 4,7
1.9
1,7
1,0
1,&
0,5
1.9
0.7
2.4; 4,8
0,7
3,4
0,5
3,0
1,3
1,8; 4,2
0,6
1.0, 1.5; 1,9
1.6
1.5; 1,8, 3.2
0.8
1.1
1.3
0.8, 5,5
1,0
2,0
0,3
2,4-4,4
1,6
4,7
1.4
3,6
2.2
1.4
1.2
4,0
1.5
2,8
1,7
3,0
1,3
2,1
2,0
3,4
1,6
2.2
2.2
1,5
1,7
5.9
1.7
2,5
2,2
2,8
1.0
0,5, 3.0; 4.0
Semi
2,2
3.8
Des
2,2
1,9
2,6
4,2
Mountain
1In these years, the observations were carried out by an expedition Of the
Main Geophysical Observatory.
-------
Table 5 (Cont'd)
vt
1,5
2.8
1,7
1.8
eppe
1.4
1,0
1,4
2,0; 2.2
1.4
3.0
2,1
2,4
desert -
2.9
4.2
VII
1.8
2.3
2.2
2.2
VIII
1.3
3,8
1.6
2.2; 2,6, 3,7
1.8
3.9
IX'
1.5
2.1
X
0.8
3,8, 5,0
1.6
1.8
XI
0,5
0.0; 4,0
0.3
0.0
XII
0,3
33
0.3
2.1
1,4
2.0
1.7
1,8; 3,1; 3,8
1.6
3.2
2.2
0,8
2,4
5.2
2.7
2.8
2.4
2.4; 4,0
2.1
3,0
2,0
3,0
1.2
2,5
2.3
1,6
2.1
4.0
2,1
2,8
1,5
2.3
2,3
3,8
1.3
3.6
2.1
2,6
1.8
3,1
1.5
2,7
2.0
4,2
2,1
3,6
1,6
2,2; 2,6
0,9
1,8
1.6
2,8
i.o
2,3; 3,8
1.2
4,2
1.5
2.6
1.1
3.0
0,7
1,8
0,7
2.0; 2,7
1.0
3.6
0,4
2,8; 4,2
0.4
2,5
0.3
0.6-3.7
landscape
ere
2.3
1.7
2,7
2.0
2.6
5.1
2.2
2,2
2.5
2.2
2.0
5,8
2.2
2,0
2,1
2,4
1,8
2.2, 4,2
1.1
1,4-3,9
1.6
1.4; 4,2
1.1
1,2
0,9
3,6. 4.0
1.2
1.3-3,4
0,9
1,2; 3.2
-------
Largest Negative Values of A70i5_2>o in tne Presence of Inversion and Wind Velocity uj
Station
Year
I
11
in
IV
V
lay
Arkhangelsk
Kargopol1 . .
Nolinsk . . .
Turukhansk • •
Yakutsk. . . .
1967
1967
1967
1967
1965
1965
1967
-0,7
0,9
-0,9
1,9
-1,5
0.0
-0,3
0,2-3.6
-1.0
1,1
-0.9
1,0
-0,5
1,6
-0,5
2,4
-0,9
0.0
-0.8
0,4; 1,3
-1,7
0,6; 1,0
-0.6
i °'2
1
-,
.-2,1
0.4
-1,4
1,1
t
^
-2,7
0.0
-0,5 (
0.5 |
-2,6 ;
0,0
-1,1
1
i
0,3 j
-1.3
0,7
-2,0
OA 0,7
i
i
TSyga with admixture
Voyeykovo. . .
Riea
Tiyrikoyya . .
Pinsk ....
Smolensk . .
im. Hebol'sin
Pavelets . •
Sovetsk . . .
1967
1966
1967
1967
1966
1967
1967
1967
1967
1967
1966
•^1,5
1.4
-1,0
0,0
-2.1
0,0
-0,4
0,9
-0,8
1,4
-0,2
1.2
-1,0
0,7
-0.3
0,2-3,5
i
-0,3
2,8
-0,6
3,4
-1,7
1,3
-0,4
2,1
-1,1
1,0
—0,6
0,0
-0,4
1,5; 2,6; 3,8
-0.2
0.0-4,0
-0,2
3,7
-0.2
S.2; 3,4; 3.6
—1.4'
0,2; 0.6
-1,0
1.0
-0,9
0,8
-1,1
0,0
•
—0,7
2,2
-0,9
0,4
-1.4
1,6
-1.6
0,8
-2.2
0,0
-1.8
0.0
Fores
-1.9
0,0
b
-------
Table 6
at a Height of 1m at which the Largest Negative VaKes are Observed
_Second. Line.
« — _ _._ _j
VI
VII
VIII
IX
X
M
XII
ga
—1.7
0,3
—2.2
o.o
-1.6
0.8
-2,1
1.0
-1,4
0,0
-1,6
0,0
-1,6
0,0
—1,6
0.5
-2,0
0,5
-2,9
0,8
—1.4
0,0
—1,4
0.2
-1.8
0,6
-2.4
0.0
-2.0
0.5
-2.1
0.0
04
*t*
0,0
-1.2
0.4
—1.9
0,0
-1.5
0.5
-2,6
0.2
-0.8
0.0
—0,6
2,3
-0.5
0,8
—1.3
0,0
-0,3
0.2, 2,0
—0.5
1,7
-0,8
o.o
-1,3
0.9
_0,4 —0,3
1.4
-1.0
2.2
1.8
-0.5
2,8
of broad-leaved species
-1,1
0.7
-1,0
1.4
—2,7
0,0
—2.0
0.0
Steppe
-1,3
0,0
-1.0
0.3
—0,8
0,8
—3.5
0.0
-0,9
0.5
-0,7
0,6
-2,7
0,0
-2.5
o.o
-2.0
0,0
-1,3
0,0
-0.7
0.2
-1,0
1,0
-rl.3
0,2
—0.5
1,5
—1.4
0,0
-2.9
0.0
-1,2
0,0
-0,9
1,7
-i.o
1,0
-4,2
0,0
-2,6
0,0
-1,4
0,0
-0,6
5,8
-1.1
0.8
—2,3
0.0
-1.5
0.6
-1.0
0,6
-1,2
0,0; 0,8; 1,0
—0,3
1,9, 4,1
-0.7
1.6
-0.7
1.9
-0,5
0,0, 2,2-3,1
-0,8
-0,3
2,0
-0,7
0,7
-2.3
0,3
-2.0
0.6
-0,3
2,4
-0,6
-0,1
1.4; 1.7
-------
Station
Sovetsk . . .
Solyanka . . .
Year
1965
1967
l
-1,2
0,9
II
-1,3
0,2; 1,2
HI
-0,8
1.1
-1.4
0,2
IV
-0,9
1.9, 2,4
-1,1
0.5, 2,0
V
-1,1
1,2
-2.4
1,1
st
Kamennaya Step
Askaniya-Nova
Rudnyy . . .
1967
1967
1967
1966
-2,4
0,0
-0,5
1.8
-0.4
1,7
-0,4
0,4; 2,2; 4,4
—3,0
0,0
-0,5
1.5
-0,5
0,0; 0,4; 1,2
-0,9
2.0
-2.5
1,3
-0,5
2,6
-0,3
0,0; 1,8, 2,6
-1.3
0.2
—1,2
1,3, 2,9
-1.3
0.2
-1.9
0.3
-2.4
0.4
-3.5
0,0
-2,3
0.3
-1.0
0.4
-1,1
0.8
Balkhash .
Karasuat
Semi
1967
1967
1966
1965
—0,5
0.3
-0,4
3,0
-0,7
2,0
-1.9
0.0
-1,4
0,6
Des
Aydarly . . .
Beki-Bent . •
1967
1967
-4,3
1.2
-2,5
0.9
-0.7
0.0
-2.2
1.2
-2.3
0.2
-1.2
1.6
-1.4
0,8
-1.2
0.7
-1.4
2,8
-0,8
1.7
Mountain
Telavi ....
Kyzyl ....
1967
1967
0.0
0.5; 1,0
-------
Table 6 (Cont'd)
VI
-1,6
1.4
-2.9
0.8
VII
-1.4
0,0
-2.5
0,2
VIII
-1,1
0,2
-1.1
0,6
IX
-2.4
1.0
X
-0,8
1,2
-1.6
0,7
i
XI
—0.4
3.8
-0,7
0,8
eppe
—3,6
0,0
-1.8
0,4
-0,7
2.3
-1.9
0,4
-2,7
0,0
-1.7
0.8
-1.0
0,0
-1,5
0.7
-2,0
0,0, 0,5
-1,1
0,6
-1.9
0.6
-1,5
0.4
—3,9
0,0
-2.2
0,0
-2,5
0,4
-2,6
0,3
-3,4
0,8
-1.8
0,0; 0,3
—1,5
0,7
-1,6
1,0
-0.5
1.8
—0.6
0,4
-0,4
XII
-0,1
5.5
-0,9
0,0; 1.0
-0.3
0,7, 1,8; 2.8
—0,2
3,4
-0,5
0,0 0,5
desert
-1,1
0,8
—1.0
1,2
-1,5
0,2
-1.6
0,0
-2.7
0.6
-1,1
0.6
-1.7
1,0
-2.0
0,0
-U
0.5
ert
-1,9
0.3, 2.0
-0.9
06
-1.6
1.4
-0,6
2.7
-3.4
0,5
-0.6
1,6
-1,8
0,5
-1,1
1,6
landscape
-1,6
1,8
-2,4
0,5
-2.3
1,0
-0.8
1,7; 2,0
-1,6
1.2
—3.2
0.7
-3,2
0,9
-------
Al
-i
* t
» 2
x 3
e 4
A 5
of
Fig. 1. Temperature difference in the 0.5 and 2.0 m layer.as a
Observatory. Analysis of the plotted graphs and data of Tables 5 and 6 shows
that there is indeed a correlation between At and Ul. In the great majority
of cases, the largest values of the temperature differences in the presence
of inversions are usually observed in a calm or in a slight wind, and the
-------
^nu.n
Fig. 2. Envelope curves of the dependence of At on Ui during superadiabatic
and inversion state. Rudnyy, January-December 19&?.
1 - January, 2 - February, 3 - March, 4 - April, 5 - May, 6 - June,
7 - July, 8 - August, 9 - September, 10 - October, 11 - November,
12 - December.
largest positive gradients, in the presence of a moderate wind, and in some
months, at a number of stations, even in the presence of a stronger wind
-------
temperature differences are observed changes within certain limits from
one station to another and from month to month. It is most frequently
below 1.0-1.5 ntfsec, or close to calm for the largest negative gradients
in the presence of inversions and amounts to 2-3 m/sec for the largest
Table 7
"d-in. the_Prasenee fl£ Easitive Temperature_Gra_dients _
Station
Rudnyy
Rudnyy
Kamennaya
Step'
Balkhash
Aydarly
Beki-Bent
Yakutsk
Turukhansk
Year
1967
1966
1967
1967
1967
1967
1967
1967
I
5
2
2
2
4
8
_
H
2
0
0
1
3
2
_
0
III
1
2
0
0
10
2
_
1
IV
4
2
1
6
2
4
1
0
V
3
2
0
1
11
1
1
0
VI
6
0
6
1
1
1
4
2
VJl
6
3
2
0
6
2
8
11
vin
1
2
4
1
3
0
5
7
IX
1
3
2
3
7
1
9
3
X
3
3
3
1
—
6
3
1
XI
2
0
0
2
1
6
_
1
XII
0
3
0
_
5
7
_
1
Total num-
ber of
Observations
in each Month
90-93
IV— X
28—31
XI— III
90—93
IV— X
28—31
XI— III
90-93
V-X
28-31
XI— IV
90—93
IV— X
28—31
XI— III
90-93 •
III— X
28-31
XI— II
84-93
I— XII
90-93
VI— IX
28—31
X-V
90-93
VI— IX
28—31
X— V
positive gradients under superadiabatic conditions. Occasionally, mainly
during the cold period, some deviations from this relationship are observed:
the most negative gradients take place in the presence of a moderate wind
and conversely, the largest positive differences occur during a calm and a
slight wind. The number of such cases, particularly under superadiabatic
conditions, is small. After reaching its maximum value at a certain wind
velocity, the superadiabatic gradient begins to decrease as the wind velocity
increases further. The rate of this decrease varies at different stations
and in different seasons of the year. In some cases this decrease occurs
very rapidly (Fig. 2, August) and in others, very slowly (Fig. 2, April).
When the role of turbulent exchange is decisive in the formation of
the daily variation of the wind velocity, it may be assumed that during the
-------
Table 8
Values fl£_the Temperature Difference a;b High Wind Velocities
Station
Sovetsk
Kamennaya Step1
Rudttyy
-
Rega
Voyeykovo
Month
VI
VI
VI
Vll
VIII
IV
IV
V
V
III
IV
V
V
V
V
V
V
V
VI
VI
VI
VII
IX
IV
V
«l
6,0
6.8
7,1
6,2
6,0
6,8
8,0
6,0
5.9
6.5
6,0
6,0
6,3
6,4
7,4
7,8
7,4
7,6
6,8
7,6
6,8
6,2
6,2
6,3
6.1
At
0,7
0.8
0,5
0,5
0,9
0.7
1.2
1,4
0.8
0.5
0.5
1.2
1,3
1,1
1,3
1,2
0.5
0,5
1,3
0,8
1,2
0,0
0.8
0.7
0,7
Station
Aydarly
Ba'Jchasli
Beki-Bent
1
Month
IV
IV
IV
V
V
VII
V
VI
VII
VII
VII
VII
VII
VII
VIII
VIII
III
V
V
V
VI
VI
VIII
VIII
IX
IX
IX
«l
6,0
6,2
6,8
6,0
6,2
5,9
6,6
5,9
6.2
5,9
6.1
6.6
6.8
6.8
6.2
6.6
6,5
6.1
7.6
7.4
6.5
5.9
6.2
6.3
6.0
6.6
6.4
i/
0,7
0,9
1,1
1,8
0,6
1,5
C,5
0,8
0.5
C.5
0.8
0,8
1,5
1.8
1.4
0.9
0,6
0.5
1.4
2.1
0.8
1,3
1.1
0,9
1,1
1,0
0.9
periods of 10 A.M., 1 P.M. and 4 P.M., when the largest positive temperature
differences can be observed, calms and slight winds occur very seldom.This
is confirmed by Table 7, which lists the frequency of cases of slight wind
(Ul«l m/sec), based on data of several stations, during the 10 A.M., 1 P.M.
and 4 P.M. periods at positive and zero values of the temperature gradients.
It is evident from Table 7 that a slight wind is observed in no more
than 10% of che cases during the warm period and the 10 A.M., 1 P.M. and
4 P.M. periods.
As the wind increases from a calm to 1.5-2.5 m/sec, the negative temp-
erature gradient decreases very rapidly in most cases, and slowly as the
-------
wind strength increases further. In some cases, an irregular rate of
decline of the negative temperature gradient was observed in the entire
range of wind velocity, and even t completely insignificant dependence
of the gradient on the wind velocity was observed. If the largest negative
gradient takes place in the presence of a moderate wind, then both a
strengthening and a weakening of the wind cause a regular decrease of the
temperature gradient.
The largest values of the gradients, both positive and negative, are
observed during the warm period of the year, with the exception of sta-
tions in the desert zone, where the largest gradient values in the pres-
ence of inversions are observed during the winter months. At the same time,,
the largest positive values in summer are several times greater than the
winter values, whereas the largest negative temperature differences change
insignificantly from season to season.
It is usually assumed that at a high wind velocity (for example, above
10 m/sec as given by the wind vane), the temperature differences at two
levels in the ground layer of the atmosphere is slight. This is not always
so, however. At such wind velocities, the positive temperature gradients
may still be relatively considerable. They are sometimes even greater than
one degree, especially in summer in the southern regions. This is confirmed
by single values of temperature differences at high wind velocities, given
in Table 8 (wind velocity at a height of 1 m, temperature difference in
the 0.5-2.0 m layer). The values of the gradients (largest positive and
negative ones) and wind velocities at which they are observed will of
course change from year to year according to the synoptic situation. Never-
theless, the general nature of the dependence of the temperature gradient
on the wind velocity, judging from many graphs, plotted on the basis of
observational data for several years, remains unchanged.
-------
LITERATURE CITED
1 E e p r JI C OnuT pa3«e^enna CHOHDH H TypKecraHa na ;ianAUja(pTHbie H \:op-
i na rypoy-iciiT-
iiocTb Cf> «ATMOC(J)cpn;iH rypCyjieiiTiiocTb it p.iciipocTp.iiiciuie p;i;uiono.Ti:» Tp MC/KAV-
nap hOJuiOKBiiyMa HSA «Hayip-
CyjieHTHoro noioKa renjia no AauHbiM Ha6jiioAenHfl ua CCTH. Tp. TFO, sun. 174, 1965
12 l"IepHfM< 3 l\, ripoKO^bCBa Jl. H, Crpyaep JI P HexoTopue SKCIIC-
HTd.ibui.ic ,iannL,ie o ciiCTeMainiecKiix norpeuiHocrnx rpaAHeHTiiux HaojiiOAeiniii Tp
, nun 160, 1964
13 npiiCTJiii C. X. B TypoyjieiiTHbifi nepeHoc B npHseMHOM c-noe aTMOc({iepbi
reoiiSAaT, JI , 1964
14 CanowKHKoaa C A Ten^onoa 6
-------
BASIC PRINCIPLES OF ORGANIZATION OF THE SURVEY OF
ATMOSPHERIC POLLUTION IN CITIES
B. B. Goroshko and T. A. Ogneva
From Trudy, Glavnaya Geofiz. Observat. im. A. I. Voeykova, No. 238,
p. 123-135, (1969)
The main sources of air pollution in cities are large industrial enter-
prises and motor transport, which discharge large amounts of noxious ingred-
ients into the atmosphere. The most common are sulfur dioxide, carbon mon-
oxide, nitrogen dioxide, phenol, hydrogen sulfide, carbon disulfide, soot,
etc.
The degree of air pollution is characterized by the single concentration
of the impurity if the sampling is carried out for 20-30 min, and by corres-
ponding average values of the concentrations for sampling lasting longer per-
iods (days, months, a year). According to observational data, in a number
of cities with an extensive industry, there are frequent cases where the con-
centrations of noxious impurities considerably exceed the maximum permissible
ones (MPC), and the frequency of such values amounts to up to 50% of the days
per year. As was noted in [3, 6, 9, 13, 17, 18, etc.], the concentrations
increase particularly under unfavorable meteorological conditions, at "danger-
ous" wind velocities, in the presence of a temperature inversion, fog, etc.
The basic principles of organization of surveys of atmospheric pollution
in cities result from the characteristics of propagation of the impurities.
The basis for a survey of the pollution of a city air reservoir is the
determination of the concentrations of noxious substances in its various
parts under different meteorological conditions, the measurement of meteoro-
logical elements determining the dispersal of impurities, the collection of
quantitative characteristics of discharge of noxious impurities into the
atmosphere, and various kinds of medical-biological data.
The existing methods of air sample collection are divided into four
types: at stationary observation points, in the area of individual industrial
enterprises, and itinerary and episodic observations. The choice of the type
of observation is determined by the size and character of the built-up area
of the city, the capacity and number of the pollution sources and their rela-
tive location with respect to the residential districts. In the majority of
cases, these methods of sampling are combined and mutually supplement each
other.
The organization of stationary observation points is particularly impor-
-------
prises there are many low and small pollution sources spread over the whole
area of the city. Such an arrangement of the sources of discharges pro-
duces a heavily polluted general background. The creation of a large net-
work of stationary points involves a large economic investment, and for
this reason, when a more detailed survey is to be made and the linear
dimensions of the city are greater, it is desirable to use a specially
equipped motor car as a moving point. The latter makes it possible to
select representative points of sampling, to determine the zones of maximum
concentrations and thus to obtain a more detailed picture of the concentra-
tion field.
Analysis of the observational material shows that the concentrations
of noxious substances undergo a marked change in different parts of the
city, and for this reason, in order to obtain a complete characterization
of the degree of pollution of a city, it is necessary to organize one
sampling area of 10-20 km2 on flat terrain and one area of 5-10 km2 on
rugged terrain. After conducting a detailed survey, it is sufficient to
take regular measurements at 3 to 4 of the most representative points
located in different sections of the city, so that it will be possible to
estimate variations in the degree of atmospheric pollution. If, however,
the territory of the city has large pollution sources, which play a def-
inite part in atmospheric pollution, and particularly if they are concen-
trated on a single industrial site, then in addition to stationary points
it is desirable to set up the collection of samples at various distances
from the center of the site under the axis of the visible plume.
Stationary points of sample collection are organized by taking into
account the planning and layout of the city districts, the location of air
pollution sources, the topography of the area, etc., in order that the
selected samples characterize not the local, but the general pollution of
the air reservoir, determined by the action of turbulent diffusion, in all
sections of the city. A suitable distribution of the collection points is
very important, since it largely determines the concentration values. In
[19] it is shown that the concentrations of noxious substances in the vicin-
ity of highways are much higher than the average pollution background. This
is where the maximum number of pedestrians and passengers in private and
public motor transport are concentrated, where they are subjected to the
influence of these higher concentrations. In addition, high concentrations
also act on residents in apartments located in buildings along the highways.
In organizing the studies, it is indispensable to consider the laws of dis-
tribution of impurities as a function of the meteorological conditions. In
the absence of unorganized discharges near a high source, the concentration
is zero, then along the direction of the wind, it increases, reaches its
highest values at distances equal to 10-40 stack heights, then gradually
diminishes to zero. In [2, 4, 8, 16, etc.] it is shown that the maximum
value depends on the capacity of the discharge, stack height, temperature
and velocity of the ejected gases, and also to a considerable extent on the
weather conditions. The higher the source, the more the impurity is dispersed
in the atmosphere before the noxious substances reach the underlying surface.
-------
The dispersing capacity of the atmosphere depends primarily on the wind
velocity and vertical distribution of the temperature. If a temperature
drop with the height is observed, an unstable state of the atmosphere is
established, and conditions of intense turbulent exchange are created, which
on dry land are mostly observed during the summer in the daytime. Under
such conditions at the earth's surface, under the plumes of high sources
with a gas ejection temperature of about 100°C, maximum concentrations are
observed [4, 8, 16], and their large fluctuations with time are possible.
If the temperature increases with the height in the ground layer of air,
i. e., an inversion is observed, the eddy motions and the impurity dispersal
become considerably attenuated. Under these conditions, high concentrations
.are produced at the surface of the ground as a result of discharges from low
sources, and, conversely, low concentrations are observed due to discharges
from high sources. For this reason, in the presence of large and lasting
ground inverstions and in the presence of low or random discharges, the
concentrations of noxious substances may rise sharply on industrial sites
and in adjacent areas.
The magnitude of the ground concentration in the presence of elevated
inversions will 'substantially depend on the relative positions of the lower
inversion boundary and source o*f the discharge: if the inverstion boundary
is located above the source and prevents the penetration of noxious sub-
stances into the upper layers of the atmosphere, the bulk of the impurity
will concentrate near the ground and this will result in high concentrations.
These most unfavorable conditions for impurity dispersal are produced
during the spring period, when the stable ground inverstion breaks down,
and under certain synoptic conditions, and sometimes also in the course of
a brief period in the mornings during the warm part of the year.
The wind velocity also has different effects on the field of concen-
trations near the ground, depending on the method of discharge of the
noxious substances [1, 6, 8]. When the discharges are low and not organized,
low wind velocities result in the formation of stagnant situations and in an
increase of the concentration. When the discharges are high, the concentra-
tions near the ground decrease as a result of an increase in the ascent of
the plume and the transport of the impurity upward, particularly when the
discharge is strongly overheated.« At high wind velocities, the initial as-
cent of the impurity decreases, but because of an increase in the transport
velocity of the impurity, the ground concentration decreases. For this
reason, the maximum concentrations are observed at a certain wind velocity
(3-6 m/sec), called the'tiangerous" wind velocity.
An instability of the wind direction promotes an increase of the dis-
persal along the horizontal. Large areas are thus subjected to the influ-
ence of lower concentrations.
In the presence of fog, a stronger influence of pollution is observed,
because on the one hand, water solutions of certain ingredients such as
sulfur dioxide are more toxic; and on the other hand, the meteorological
conditions associated with fogs promote the accumulation of discharged im-
purities in the ground layer of air. Sometimes smogs are produced, which
-------
are considered to be associated with known cases [10] of sharp increase of
illness among the population and in some cases with large numbers of victims.
As shown by theoretical theses and studies made in wind tunnels [5, 7],
under conditions of a rugged terrain and above a city with modern buildings,
there takes place a disturbance of the air stream leading to an increase in
concentrations in certain situations. In some forms of the relief, for
example in basins, the stagnation of air causes the accumulation of noxious
substances near the underlying surface, particularly in the presence of a
temperature inversion and low sources of discharge. On the whole, in the
presence of roughness of the terrain, the maximum of the ground concentra-
tion is usually higher than above an even area. The dispersal of impuri-
ties under urban conditions is substantially affected by the layout of the
streets, their width and direction, height of the buildings, presence of
green tracts, water reservoirs, and even the planning and location of indi-
vidual buildings [15], since these factors create an irregular surface,
form different types of obstacles to an airstream and produce special
meteorological conditions i.e. , the microclimate of a city.
The concentrations of noxious substances undergo considerable changes
in space. They are observed when the territory of the city contains a rel-
atively large number of points of sample collection. Therefore, in order
to obtain a detailed and complete picture of the pollution of a city area
and to identify the sections with maximum and minimum concentrations of any
ingredients, it is necessary to set up an extensive observational network.
Treatment of the experimental data obtained makes it possible to draw
a. number of important conclusions. Table 1 gives the mean monthly and
maximum concentrations of sulfur dioxide and nitrogen oxides for two cities
located under identical climatic conditions but with different arrangements
of the main sources of pollution.
The uppp-^ part of the table characterizes the atmospheric pollution
of a city where the industry is scattered over the entire territory, and
the lower part, a city in which the industrial complex is located on a
single industrial site. In each city there were eight points of sample
collection. In Figs. 1-4 these results are represented in graphical form
for clarity. A detailed examination of the data given in the table and in
the figures show that the concentrations undergo considerable changes from
one point to another and from one month to the next, i.e., in space and
time. At the same time, there is a pronounced tendency on the part of the
variations of mean monthly and maximum monthly concentrations to coincide in
time at all points, i.e., over the entire area of the city. This tendency
is particularly characteristic of a city where the industry is distributed
over its entire area, and less pronounced in a city where the atmospheric
pollution sources are located on the same industrial site. This is due to
the fact that when the sources of discharge are distributed over the en-
tire territory of the city,* a city "cap", i.e., a general pollution of the
-------
Values of Concentrations Based on Data for 19&7 on the Territory
f
VjUlt
OjJO
tt
. 1
"Jiv. "
Max.
II
Av.
Max.
in
~AvT
" I "Max".
IV ' V i
"Av.
"Max" j Av. Max.
1 _ i . i
Sources of Pollution
Sulfur
1
2
3
4
5
6
7
8
0,20
0,17
0,14
0,15
0,13
0,15
0,11
0,14
0,30
0,30
0,30
0,30
0,20
0,30
0.20
0,20
0,20
0,18
0,15
0,15
0.13
0,14
0,11
0,10
0,30
0,28
0,20
0,30
0,20
0,27
0,17
0,17
0,23
0,24
0,18
0,18
0,20
0,19
0,16
0,16
0,41
0,40
0,30
0,40
0,53
0,40
0,40
0,40
0,24
0,26
0,13
0.40
0.41
0,38
0.18 i 0.35
0,19
0,19
0,17
0,22
0,34
OJ5
0,42
0.43
0.14
0.17
0,14
0.11
0,12
0,15
0,11
0,19
0,30
0,30
0,30
0,30
0,30
0,30
0,30
0,30
Nitrogen
I
2
3
4
5
6
7
8
0,12
0,14
0,13
0,10
0,10
0,10
0,10
0,10
0,30
0,30
0,30
0,30
0,20
0,30
0.20
0,20
0,16
0,15
0,lf>
0,14
0,12
0.13
0.10
0,10
0,37
0,39
0,34
0,33
0,27
0,33
0,12
0,28
0,16
0,15
0,14
0,13
0,16
0,15
0,12
0,13
0,36
0,29
0,31
0,29
0,45
0,40
0,23
0,28
0,13
0,22
0,17
0,15
0,19
0.17
0,12
0,19
0,58
0,49
0,35
0,36
0,47
0,41
0,31
0,35
0,26
0,23
019
0,19
0,20
0,15
0,17
0.17
0,55
0,60
0,58
0,52
0,42
0.39
0.42
0,50
Pollution Sources Concentrated on a
Phen
1
2
3
4
5
6
7
8
0,26
0,32
0,22
0,29
0,27
0,35
0,26
0,26
0,96
0,72
0,72
0,78
0.96
1,20
0,60
0,54
0,30
0,36
0,34
0,40
0,49
0,53
0,32
0,36
0,72
0,84
0,84
1,08
1,04
1,08
0,72
0,76
0,40
0,31
0,15
0,35
0,26
0,47
0,33
0,32
1,40
1,00
0,60
0,90
0,72
1.08
0,72
0,72
0,39
0,36
0.31
0.54
0,66
0.79
0,38
0,42
0,73
1,20
1,00
0,93
1,66
1,09
0,84
0,73
0,32
0.41
0,46
0^4
0,56
0,77
0,51
0,44
0,80
1,06
1,07
1,20
1,07
1,46
1,90
0,93
Carbon
I
2
3
4
5
6
7
8
13
19
11
18
15
16
13
7
20
30
20
30
30
30
20
10
14
10
12
6
12
8
15
10
30
20
20
30
40
30
30
40
25
10
9
12
11
16
16
13
35
20
25
20
28
35
28
20
15
7
1
6
9
12
7
10
29
10
10
15
20
38
15
20
12
10
10
17
12
11
8
8
30
30
40
45
45
27
40
40
-------
Table
of Two Cities with Different Arrangements of Sources of Pollution
VI
AV.
Max.
VIII
Av.
Max.
IX
Av.
Max.
X
Av.
Max.
XI
Av.
Max.
XII
Av.
Max.
Scattered all over the City
Dioxide
0,17
0,12
0.21
0,21
0.23
0.15
0,19
0.21
0,30
0,37
0,44
0,44
0,51
0,44
0,38
0,37
0,22
0,18
0,22
0,16
0,17
0,17
0,21
0,20
0,30
0,29
0,37
0,22
0,22
0,22
0,40
0,37
0,17
0.19
0,17
0,10
0,18
0,24
0,19
0,07
0,88
0.81
0.29
0,88
0,29
0.80
0,84
0,89
0,35
0,45
0,45
0,41
0,56
0,34
0,39
0,41
0,80
0,70
0,68
1,06
1,55
1,20
0,87
0,94
0,35
0,28
0,27
0,30
0,26
0,27
0,26
0,29
0,75
0,08
0,75
0,57
0,35
0,58
0,38
0.49
0,18
0.15
0,15
0,16
0,29
0,16
0,18
0,18
0,48
(MO
0,33
0,27
0,75
0,34
0,39
0,61
Oxides
0,18
0,16
0,18
0,19
0,18
0,10
0.14
0,15
0,43
0,40
0,38
0,43
0,37
0.31
0,38
0,50
0,17
0,15
0,15
014
0,U
0,14
0.14
0,15
0,31
0,25
0,31
0,19
0,32
0,25
0.25
0,25
0,15
0,14
0,14
0,13
0,10
0,17
0,08
0,11
0.59
0.29
0,38
0,31
0,37
0,48
0,30
0,27
0,22
0,25
0,23
0,31
0,20
0,20
0,23
0,23
0,61
0,90
0,86
1,14
0,50
0,50
0,72
0,82
0,20
0,16
0,13
0,15
0,13
0,16
0,14
0,15
0,66
0,42
0,20
0,35
0,30
0,29
0,28
0,35
0,26
020
0,22
0,20
0,23
0,21
0,18
0,19
0,58
0,36
0,47
0,29
0.58
0,47
0,34
0,32
Single Industrial Site
ols
0,29
0,40
0,44
0,57
0.43
0,49
0.31
0,19
0,53
133
1,55
1,55
0,80
1,55
0,66
0,63
0,37
0,41
0,39
0,40
0.38
0.39
0.29
0,37
0.53
1.30
0,80
0,67
0,53
0,67
0,53
0,67
0,50
0,48
0,47
0,50
0,50
0,46
0,49
0,45
0,77
0,77
0,80
0,65
0,65
0,79
0,76
0,67
0,37
0,32
0,34
0,43
0.52
0,66
0,24
0,37
0,85
0,83
0,74
0,56
1,71
1,93
0,59
1,08
0,16
0,32
0,33
0,56
0,41
0,46
0.37
0,36
0,51
1.63
1,23
1,42
1.04
1,25
1,25
1,00
0,45
0,52
0,57
0.55
0,68
0,73
0,73
0,40
0.90
1,20
1,22
1,03
1,22
1,20
1,23
1.13
Monoxide
12
13
15
16
18
14
10
9
40
45
50
50
41
35
35
25
12
11
7
11
11
20
6
11
20
40
20
23
22
30
17
25
13
7
3
8
12
17
8
3
31
14
10
20
23
27
16
8
12
8
6
13
12
15
3
2
38
15
31
31
24
26
19
12
6
10
11
12
15
15
13
7
12
20
32
32
37
42
44
18
9
11
7
10
10
14
6
7
23
23
16
14
28
28
15
30
-------
atmosphere up to several hundred meters, is formed above it. Because of
exchange, noxious substances migrate from the cap into the ground layer.
Thus, the city cap .becomes a kind of storage space for the noxious sub-
stances, which then spread with the wind over the entire territory of the
city and far beyond its limits. This cap is a volume source of pollution.
0,8
o,e
0,2
"a)
< 1
2
.-«—.—« 3
x X * <•• 4
,-. S
—x—x-6
_o — o— 7
8
b)
••*„
J I
I
II HI IV
VI VII VIII IX
XI XII
Fig. 1. Annual variation of maximum (A) and mean monthly (B) concentrations
of nitrogen oxides over the territory of a city at eight points when the
pollution sources are scattered over the entire area of the city.
1-8 - points of sample collection.
As a result, high concentrations are observed even in areas where obvious
pollution sources are absent. It is for this reason that samples taken in
different parts of the city show a good correlation. The correlation is
poorer in cities where the industry is concentrated on a single site, since
-------
0,8
0.6
0,2
M
0,6
0.2
(b)
VII I/Ill I If X XI XII
Fig. 2. Annual variation of maximum,(a) and mean monthly (b) sulfur
dioxide concentrations over the territory of a city a.t eight points
when the pollution sources are scattered over the entire area of the
city.
For the designation of each of the eight sampling points, see Fig. 1.
-------
in this case the concentration is determined by the location of the point
under the plumes of the sources.
SO
40
30
W
o)
20
10
III IV
VI VII VIII IX
XI XH
Fig. 3. Annual variation of maximum (a) and mean monthly (b) carbon
monoxide concentrations over the territory of a city at eight points
when the main pollution sources are located on the same industrial site.
For the designation of each of the eight sampling points, see Fig. 1.
Major sources of discharge of noxious substances make a significant
contribution to atmospheric pollution, and it may be expected that maximum
concentrations will be observed under their plumes, directed toward the
-------
1.6 r
',2
1.0
0.8
0.6
\
J - 1 - 1
1 - 1 - ' ' '
0.8
0.6
0,4
0,2
JL
JL
JL
J_
/// IV
VI
7
Ifll VIII IX
XI XII
Fig. 4. Annual variation of maximum (a) and mean monthly (b) phenol
concentrations over the territory of a city at eight points when the
main pollution sources are located on the same industrial site.
For the designation of each of the eight sampling''points, see Fig. 1.
-------
residential districts. Therefore, in order to obtain the maximum possible
concentrations in residential districts, observations are conducted under
the plumes of the.sources at different distances from the center of the
discharge.
If the sources of noxious substances are concentrated on a single
industrial site, it is useful to organize the determination of the concen-
trations under the general plume at various distances from this site, and
particularly when the plume is directed toward the city. At the same time,
it is also necessary to conduct observations at one to two points on the
territory of the city to determine the background pollution.
In surveying the atmospheric pollution of a city, it is necessary to
study the change of the concentrations with time in the neighborhood of
individual sources as well. It is necessary to investigate fluctuations in
the daily and annual course of air pollution. It must be kept in mind that
the pollution level is determined by many factors, including changes in
meteorological elements, which determine the dispersal of noxious substances
in the atmosphere. Of major importance is the schedule of operations of
enterprises and other sources of discharges of noxious substances. For
example, motor transport and small enterprises discharging a considerable
part of noxious substances into the atmosphere operate mainly from 8 A.M.
to 4-6 P.M., and the heating system, during the cold part of the year. It
is essential also to consider the arrangement of the sources over the terri-
tory of the city and the method of discharge of noxious substances (low and
unorganized discharges or discharges through high stacks).
Several studies [12, 15] indicate the presence of a concentration
maximum in the annual variation in spring, or two maxima in spring and
autumn [10].
In [8] it is shown that as noxious substances spread from a single
source (SREPP), an increase in concentration from the morning to the daytime
period takes place in the effective range of this plume, followed by a de-
crease toward the evening, this being due to an increase of turbulent ex-
change in the daytime. The daily variation of concentrations under city
conditions is more complex in character and shows differences in different
seasons of the year, as is evident from the example of a region in southern
Ukraine [6]. During the spring and autumn periods, a maximum is observed
during morning and evening hours, and a minimum at 1-2 P.M. In summer, a
small maximum is observed during the day; in winter, there is first a decrease
of the concentrations from morning to midday, then the concentration increases
again and reaches maximum values at 4-5 P.M. with subsequent decrease by
9 P.M.
These changes are accounted for by the nature of the discharge of noxious
substances and by the variation of meteorological elements. In modern cities
and in the majority of industrial centers, there are large individual high
sources and many small ones. In spring and autumn, the morning and evening
maxima are due to discharges mainly from small sources in the presence of a
-------
slight turbulent exchange and the accumulation of noxious substances in
the lowest layer of the atmosphere. In the daytime, these areas (southern
Ukraine) are characterized by an increased turbulent exchange, causing
the elimination of stagnant zones in the ground layer. However, the turbu-
lence is still insufficient and does not promote the transport of noxious
substances from high sources which lead to aft increase of the concentrations.
Such conditions are observed in summer, when the pollution level undergoes
little change during the day, since in the mornings and evenings in the
presence of weaker turbulence, this level is determined primarily by dis-
charges from low sources, and as a strong turbulent exchange develops, a
more substantial role is played by high sources. Besides meteorological
factors, an additional contribution of discharges from furnaces probably
determines the winter maxima in the morning and at 4-5 P.M.
Considerable attention should be given to meteorological observations
in surveys of atmospheric pollution in cities. In particular, it is necessary
to consider the general synoptic situation, which is determined by macro-
circulation processes. The nearest weather bureau is used for this purpose.
In analyzing the pollution, use should be made of indices of the type of
weather and synoptic situation.
Data on the structure of the boundary layer of the atmosphere, mainly
in regard to the distribution of temperature and wind up to heights of
2-3 km, are more local in character. Their collection requires the organiz-
ation of special observations at a specific point of the survey. They should
include data on the distribution of air temperature and wind, obtainable by
means of a set of aerological observations, observations at heights and from
television towers, by means of helicopter and airplane sounding, and also
gradient observations in the ground layer in the area of the weather station.
It is essential to obtain a set of measurements that supplement each other
and provide an adequate representation of the structure of the atmosphere
from the standpoint of mixing of the impurities. Aerological observations
can provide a general idea of the thermal stratification and wind character-
istics of a city district. At the same time, the lowest 100-meter layer
has been described in insufficient detail, and it is difficult to separate
the influence of the built-up area of the city as the active surface on the
structure of the air current. This makes it possible to emphasize aircraft,
or better, special helicopter sounding, on the basis of which one can study
not only the vertical structure but also the spatial variations under the
influence of the city's building^. 1
A more detailed distribution of the meteorological elements in the 100-
200 meter layer of the atmosphere, where the chief sources of impurities
are concentrated, can be obtained by setting up special observations at
heights and on television towers.
Gradient observations in the ground layer make ilf possible to calculate
the values of the turbulence coefficient which make up the basis for calcula-
tions of the dispersal of impurities. Also needed are data on the general
meteorological characteristics of the region being surveyed, which can be
-------
readily obtained from current observations of the weather station, and if
the latter is absent, observations conforming to its program should be
specially organized* for the period of the survey.
Since the city's buildings and the city as such lead to the formation
of a particular and complex active surface, a substantial change also takes
place in the meteorological regime, particularly in regard to the character-
istics of the temperature and wind. Methods of studying the meteorological
regime of the city itself include observations at points of collection of
air samples. They provide data on deviations of meteorological characteris-
tics (during the period of determination of the chemical composition of air)
in various parts of the city as compared to macro- and mesoconditions, which
provide synoptic maps and observational data at meteorological stations.
One can also plan special micrometeorological surveys: they provide the
spatial distribution of the air temperature and wind direction and velocity
for the city and the ground layer. This makes it possible to compile, for
the territory of the city, the characteristic of air currents caused both
by the direct deflection of the main air current under the influence of
buildings and layout of the city and by the possible generation of local
currents [11, etc.].
In the set of special observations involved in the surveys of cities,
one should also include comparative actinometric observations of shortwave
solar radiation (within the city in the most polluted part and outise the
city when no significant pollution is present). The evaluation of the
degree of attenuation of radiation in the city may serve as an index of the
general state of pollution of air.
-------
LITERATURE CITED
I 5esyrjian3 JO. K onpeAejiemno noteHUfia.ia aarpnaHeiiHsi sosAyxa. Tp. FFO,
n. 234, 1968.
2 BepJiiiHA M' E. [n Ap] Hiicjioimoe peiuemie ypaBneimn TypSyjienTiioft A
3:111 u pac-iet aarpaaiieHiiH aiMoccpepu n6jiii3n npOMbiuuieiiHbrx npeAnpiiHTHfi Tp.
nun. 138. 1963
3 EepJiflHAM E [HAP] Miic^ennoe iicoieaonaHHe arMoccpcpnoft A"y3>itj npn
HopManbiiux H aiiOMajibiihix ycjiOBHHX CTpaTii(|)iiKauHii. Tp Fro, own 158, 1964
4 B c p .1 n u A M. E., O H it K y .1 P M 4>ii3iiKcncoa,
10 .'lyiic A/K. B .IT ran 3,irpii3iicinioc ncCo FIoA pc;i E. II. TCIIUDOIICKOIO. Hw.
->, M , 1967.
II PacToprycua T. 11, OCOOCIIIIOCTH TcpviH-iccKoro POKHMJ ropoAon. CM
IIJCT. CO.
12 Co M bK n n ."I. P , P a 3 dc ra CD a E A , Tc pcx o B a K. M K uonpocy o MC-
TCopoioni'ici-KoiT ooyc.ioujiciiiiocTif aarpsiaiicimn ooiAyxa nan lopoAaSiw Tp ITO
ni.ni 183, 1966
13. Co u bK n n JI P. Anajiiu MCTcopoJioriinecKHx yc^oDiifi onaciioio 3.irpndiiciui»
iioiAyia u ropOAax Tp ITO, BMII 234, 1968.
14. C o n bK n 11 JI P, M a Ji M KOB JI B 06 o6paooTKe n aiiajuiac iiaC^iowinin 3a
jnipnanciiiicM BO3,iy\a n ropOAax Tp ITO, nun 207, 1968.
15 TOM co n II. N CaiiiiTapnan oxpaiia aT.\ioc([)epnoro no3Ayxa or aarpsnnciiiin.
^\p,ini3, 1959
16. yK.i3.uniH no pac'iery p
-------
ORGANIZATION AND METHOD OF OPERATION OF
. ATMOSPHERIC POLLUTION OBSERVATION POSTS
I. A. Yankovskiy, A. A. Gorchlyev, and D. R. Monaselidze
From Trudy, Glavnaya Geofiz. Observat. im. A. I. Voeykova, No. 238,
p. 222-228, (1969).
Recently, considerable efforts have been made toward further improve-
ments of observations of atmospheric pollution. This has been measureably
promoted by steps taken to improve the methods and technical principles
of collection of air samples, their chemical analysis, and the treatment
of measurement data.
It should be noted that some success has now been achieved in the
organization of sound observations of the content of noxious impurities in
the atmosphere taking into account weather conditions. Some use has been
made of materials of a series of seminars conducted by the A. I. Voyeykov
Main Geophysical Observatory in close collaboration with the territorial
offices of the Hydrometeorological Service (OHMS).
This article will deal with the most essential steps taken by the OHMS
that can be applied in the practical operation of stationary posts and
hydrochemical laboratories.
Of prime interest in this connection is an improvement, suggested by
P- N. Zaytsev of the Far Eastern OHMS, of a device developed earlier at
the Kamchatka OHMS [1] for automobile rotation of the intake tube acted
upon by the wind, which is analyzed for gaseous ingredients.
The advantage of the proposed device (Fig. 1) is that it causes the
rotation of the intake tube against the forward flow, at different velocities
of the latter. In addition, it permits the determination of the wind direc-
tion without having to step out of the pavilion in which the air sampling
is being carried out.
In this device, in contrast to the one developed at the Kamchatka OHMS,
the moving tube is somewhat longer, and of such length that its lower end
protrudes 4-5 cm from the stationary tube passing through the ceiling and
roof of the pavilion. A metal arrow showing the direction of the wind is
firmly attached to this protruding part of the tube.
To determine the wind direction, a metal disc with a hole at the center
is mounted under the ceiling of the pavilion. The moving intake tube passes
through this hole. Divisions giving the 16 main bearings are marked on the
disc. The disc is oriented toward the points of the compass. Its diameter
is about 40 cm.
-------
As the vane is turned by the wind, the upper L-shaped end of the
moving intake tube is set against the wind and its direction, and hence
the direction of the wind is recorded by the arrow (i.e., by the bearing
against which it has come to rest).
The device under consideration has a relatively high sensitivity
at a wind velocity of 0.5 m/sec, the vane turns the intake tube freely.
The use of this device provides more reliable data on the concentration of
the ingredients being measured. This is apparent from a large number of
observations of sulfur dioxide and nitrogen dioxide at stationary posts in
the city of Khabarovsk. The concentrations of the indicated ingredients at
points equipped with the automatic device were found to be 25-50% higher
than the concentrations obtained in observations at the same points but
without this device. The results of measurements by means of the automatic
device are in good agreement with data of sanitary epidemiological stations.
In discussing the equipment of stationary
posts, we should point out the successful
solution of this problem in the Georgian SSR.
Here the construction of pavilions for obser-
vations at stationary posts is carried out
by using the resources and facilities of
industrial enterprises in accordance with
the objectives set down by the OHMS. A
general view of such a pavilion is shown
in Fig. 2. The framework, facing and inner
paneling consist of boards. The roof is
covered with roofing iron. The entire
exterior of the pavilion is coated with
multicolored oil paints.
In Leningrad, Dzerzhinsk, Saransk,
Belgorod and other cities, the construction
of pavilions has also been organized within
the framewor1' of industrial enterprises.
Air
Fig. 1. Improved device for
automatic rotation of intake
tube acted upon by the wind and
for the determination of the
wind direction.
In many offices of the Hydrometeorolog-
ical Service (OHMS of the Uzbek, Tajik,
Turkmen, Azerbaijan, Ukrainian and White
Russian SSR), pavilions built by the Tashkent
hydrometeorological instrument plant have been introduced (Fig. 3) The main
element of the design of this pavilion is a GR-70 hydrometric booth (control
cabin). Inside, the pavilion walls are covered with a heat-insulating layer
and lined with plywood, and outside they are faced with a wood laminated
plastic. The pavilion is illuminated in the daytime and aerated by means of
a window in its front wall. Heating in winter is done with electrically
operated oil-burning* heaters.
* Russian description refers to "maslenymi elektronagrev-atelyami".
-------
In addition, a new group of pavilions will be equipped with the automatic
device for rotating the intake tube.
In an experiment at the OHMS of the Uzbek SSR in Tashkent, at one of
the points of observation of atmospheric pollution, a rotating pavilion
was installed in which the holes for the intake of air samples to be
analyzed for both gaseous and mechanical impurities were located in one
of its walls. During the collection of samples, the observer rotates the
pavilion so that the wall with the holes faces in the direction from which
the wind is blowing.
Experience has shown that the proposed design of the pavilion cannot
be widely applied in a hydrometeorological network, especially during the
cold period of the year.
Fig. 2. General view of the pavilion of the Georgian OHMS.
The practical experience with various types of observations at the
OHMS of the Kazakh SSR deserves some attention. In determining the dust
concentration by means of an automobile aspirator, AFA-V-18 filters are
used instead of filters prepared on the spot out of FPP-15 cloth.
However, since the AFA-V-18 filters are smaller in size (working
area) than the filters specified by the design of the automobile aspirator
[2], it was necessary to alter the filter holder to some extent by increas-
ing the width of the ring of the holder on which the filter was placed by
the amount that the working area of the first filter exceeded that of the
second, i.e., 18 cm^.
These changes in the filter holder required a new calibration of the
set of rheometers.
-------
The use of ready-made standard filters at the OHMS of the Kazah SSR
chiefly saves considerable time, which the chemical laboratory staff
would otherwise have to spend preparing the filters, and also permits a
certain improvement in the quality of the dust observations.
At the present time, simultaneously with the collection of air
samples, meteorological observations are invariably made on the direction
and velocity of the wind, temperature and humidity of air, and condition
of the weather and underlying surface. Certain difficulties arise in
the implementation of this observation program. Various instruments are
used to measure the wind velocity and direction: an 8-Yu-Ol-M wind meter,
a manual anemometer, wind vanes, and even a pendant. Obviously, their
accuracies are different. When a wind meter is used, relatively little
time is required for its assembly, mounting, and dismantling, particularly
in itinerary and under-the-plume observations.
In this connection, G. M. Imam-Aliyev and Yu. V. Manukoyan of the OHMS
of the Azerbaijan SSR have prepared experimental models of meteorological
field instruments: a wind indicator, a barothermohygrometer, and a weather
station. A brief description of the instruments follows.
Fia 3 General view of the pavilion of the Tashkent
" plant of hydrometeorolosical instruments.
The wind indicator is designed for measuring the wind velocity and
direction under field conditions. The velocity indicator used is a manual
ARI-49 induction anemometer. A manual anemometer of a different type can
also be employed. A small weather vane was used to determine the wind
direction.
-------
In general, the design of the wind gauge was as follows. A two-stage
ebonite rod is mounted on the screw thread of the ARI-49 anemometer in
place of the crank! The first stage is 22 mm high and 8 mm in diameter,
and the second, respectively 10 and 27 mm with a lower groove of 7 x 25 mm.
To this rod is attached, by means of an M-5 thread, the main brass staged
rod on which a small-sized, light vane rotates on two ball bearings and
on which are mounted 8 rods (pins) designed for the determination of the
wind direction.
On the rod corresponding to the northern bearing is mounted a "KIM"
type compass, and on the opposite side, a long rod acts as a stopping
lever by means of which one can orient the vane according to the compass
and fasten it on its axis. This instrument can be assembled and disas-
sembled, and is therefore convenient for operation under field conditions.
It is being successfully used in itinerary observations of atmospheric
pollution in Baku.
The wind indicator proposed by G. M. Imam-Aliyev and Yu. V. Manukoyan
differs advantageously from the 8-Yu-Ol-M wind gauge in that it is portable
and accurately determines the wind velocity. It can be set up relatively
fast and does not require a special support. It is mounted on the hood
of an automobile. The wind gauge can be successfully used at stationary
posts as well, but in this case a special stand is required to set it up
at a given height.
The barothermohygrometer is an instrument measuring the pressure,
temperature, and air humidity under field conditions. All three instruments
are mounted on a single board. The pressure is measured with an MD-19 aneroid
barometer, the air humidity with an MVK-hygrometer, and the temperature with
a TP-1 type mercury thermometer. The housing of the instrument has ventila-
tion holes on both sides, in the base, and in the back cover. The weight
of the instrument is 5.4 kg.
In the course of operation of these instruments during itinerary obser-
vations in Baku, it was found desirable and possible to combine them into
a single unit in the form of a field weather station.
The field weather station (Fig. 4) consists of a wind indicator and a
barothermohygrometer. During the observations, it is set up on a special
support. The weather station as a whole and its individual instruments
were tested under laboratory and field conditions, and yielded satisfactory
technical and performance data.
In organizing observations of the chemical composition of atmospheric
air, it was found necessary to supply systematic information on the state
of urban atmospheric pollution to interested organizations in order to
take steps to reduce and eliminate noxious discharges into the atmosphere.
At the present time, all the offices of the hydrometeorological service
are carrying out extensive work in this direction. According to data from
regular observations, every month information bulletins on the state of
-------
atmospheric air pollution are issued and sent out to the consumers.
The Volga OHMS has accumulated
experience in issuing a daily report.
The latter gives information on the
concentration of noxious impurities
in the atmosphere based on observa-
tions during the preceding day
(for 2-3 periods). The observational
data are plotted on a schematic map
of the city, as a weather report is
plotted on a synoptic map. In addi-
tion, on the basis of the predicted
weather conditions, a qualitative
assessment of the expected state of
air pollution for the following day
is given.
At the Upper Volga office of the
hydrometeorological service, the
enterprises of Dzerzhinsk are system-
atically informed on impending weather
conditions that may give rise to danger-
ous pollution levels as a result of nox-
ious discharges. On the basis of
this information, the enterprises take
appropriate measures to alter the
operating schedule of the enterprises,
primarily by decreasing random
(unorganized) discharges into the
atmosphere.
We should point out the efforts
of the Ural OHMS toward improving
information on the state of pollu-
tion. To this end, monthly informa-
tion reports include a calendar (table)
of air pollution in addition to the
usual information. For each day and
observation post and successively for
each ingredient, this table lists the
general characteristics of pollution
levels, using certain conventional
symbols.
Fig. 4. Small sized field weather
station.
The calendar provides a picture of air pollution in the different
districts of the city for a month, and also reveals districts with high
and with low pollution levels.
One of the main conditions for obtaining reliable data on the con-
centration of noxious substances in the atmosphere is an adequate
-------
organization of the collection of air samples for chemical analysis.
We shall consider, as an example, the organization of the operation
of stationary posts in the city of Sverdlovsk.
AB indicated elsewhere [1], stationary posts provided with stone
pavilions are adequately heated in the wintertime. To ensure a smooth
operation, two electric aspirators collecting samples for analysis of gas-
eous ingredients are set up in each pavilion. One of them is in operation,
while the other is kept in reserve and is used only in case the first
breaks down. The working aspirator is periodically checked and in case of
any irregularity sent for repairs.
In this connection, we should mention the adequacy of the operation
of instruments used at the posts and the efficiency with which the observers
carry out their instructions.
In order to prevent errors in the collection of samples and in execution
of meteorological observations, the operation of stationary posts is checked
regularly. This checking is performed by the staff of the hydrochemical
laboratory, and by meteorological engineers and other specialists of the
OHMS.
In conclusion it should be noted that many territorial offices (north-
western, Krasnoyarsk, Irkutsk, Primor'ye, Murmansk and also the OHMS of the
Kazakh, Kirgiz, Georgian, Turkmen SSR, etc.) have taken a number of major
steps toward further extending the investigations of this problem. These
offices of the hydrometeorological service have prepared comprehensive sit-
uation reports on the state of air pollution and have obtained the adoption
of a special resolution concerning the protection of atmospheric air from
pollution. Unfortunately, some OHMS are still giving insufficient attention
to this problem.
The scope of the problems connected with the operation of stationary
posts is much broader than discussed here. However, the contents of the
article point to the unquestionable urgency and importance of these problems
in stationary observations of atmospheric pollution.
LITERATURE CITED
I. flHKOBCKHfi H. Ai K o6o6meHmo onuTa pa6oTH UOCTOB Hagjuoaemift sa XHMU-
iccKHM cocTauoM aTMOcfcepHOro B03Ayxa ropoaon. Tp. ITO, awn. 234. 1968
2. ABToMo6HJibHbifi aciwpaTop mm or6opa npofi aosAyxa M3 PCOCP, STH, 1967.
-------
USE OF STATISTICAL METHODS FOR THE TREATMENT OF
OBSERVATIONAL DATA ON AIR POLLUTION
E. Yu. Bezuglaya
From Trudy, Glavnaya Geofiz. Observat. im. A. I. Voeykova, No. 238
p. 42-47, (1969).
Stationary points for monitoring the pollution of the urban air
reservoir yield information on the content of noxious impurities in air
for individual areas and for the country as a whole« The organization of
such posts and the classification and correlation of the enormous amount
of data collected pose a number of questions: Are the measurement data
sufficient and can they render the actual picture of the air pollution?
How much more strictly should the material be treated and in what form
should the results of the correlation be presented? Statistical methods
can aid in the solution of these problems.
On the basis of some general considerations [3, 4, 5], the probability
distribution of an impurity in air may be assumed to be described by the
lognormal law
where f (q) is the density function of the impurity concentration q and s
and m are parameters of the lognormal distribution. To determine s and m,
the measured values of the impurity concentration q are subdivided into
gradations, and the accumulated frequencies p^ are determined for each
gradation.
A graph is then plotted with the impurity concentrations laid off on
the log scalp along the ordinate axis and with values of the argument z^ of
the integral function of the normal (Gaussian) distribution ^(z^), which
coincides with our accumulated frequency $(zk = pfc) , laid off along the
abscissa axis. Tables of the normal distribution are given in [1, 2].
If the impurity concentrations are distributed in accordance with the
lognormal law, all the points are grouped near a straight line whose slope
is equal to s and whose median value (corresponding to the argument zk = 0)
is equal to m. In the plotting of such a probability graph, instead of
values of zk, the abscissa axis usually carries values of the integral
probability corresponding to them in the normal distribution. A detailed
derivation of these relations is given in [1, 3].
An example of such a calculation for the concentrations of nitrogen
oxides in Kursk in 1968 is shown in Table 1 and Fig. 1 (curve 1).
-------
Table 1
Determination of
Gradations, mg/m*
frequencyf-#
Accumulated fre-
quency Pk, %
0,41-0,50
1
1
-2,37
0,31-0,40
3
4
-1,75
0,21-0,30
14
18
-0,92
0,11-0,20
42
60
—0,25
0,00-0.10
40
100
3,90
We carried out the treatment and plotted graphs for the distribution
of the concentrations of sulfur dioxide, nitrogen dioxide, dust, carbon
monoxide, and soot (104 distributions) for seven cities. In 75% of the
cases, the points fall in the vicinity of a straight line on the probability
graphs. This suggests that, as a rule, the impurity concentrations obtained
by taking air samples at different
times are distributed in accordance
with the lognormal law.
Analysis of air pollution data
by means of probability graphs leads
to a number of essential conclusions
with regard to the quality of the
observations and the spreading of
noxious impurities in cities.
In many cases, the points cor-
responding to low impurity concen-
trations deviate from the straight
line on the probability graphs.
Since low values are measured with
a low accuracy, these deviations
from the straight line may be assumed
to be due to measurement errors.
Analysis of the position of the
points on graphs plotted on the basis
of measurements of different impur-
ities permits an evaluation of the
limit to which the concentrations of
noxious substances are measured with
satisfactory accuracy. Thus, this
will be 2-3 mg/m3 for carbon monoxide, 0.2 mg/m3 for dust, 0.1 mg/m3 for
sulfur dioxide and nitrogen dioxide, and 0.02 mg/m3 for carbon disulfide.
The probability graph can be used as a vivid illustration of the manner
in which a certain subjectivity is manifested in the determination of sulfur
dioxide concentrations. For example, in cases where the chemical analysis of
the air samples did not involve the use of a photoelectrocolorimeter, but of
several scales - test tubes corresponding to a series of concentration
values. As a result, the frequency of high impurity concentrations may be
too low, and the frequency of low concentrations may be too high (or vice
0,01
0.3 t
Fig. 1. Distribution of concentrations of
nitrogen oxides and soot in winter.
-------
versa). The points on the graph will then deviate on both sides of the
curve.
In cases where there is a great scatter of points, i.e., a deviation
from the lognormal distribution is observed, it is necessary to explain
the causes of this situation. The latter may be frequently related to a
low quality of the selection or analysis of the samples, and sometimes to
an abrupt change in the discharge parameters of one or several basic sources
of air pollution.
According to the results of treatment of data on sulfur dioxide measure-
ments at one of the points in the city of Kuybyshev in 1967, two independent
distributions appeared to be present: one with high, and the other with low
impurity concentrations. Analysis of the data and a determination of the
causes have shown that the observational data for a certain period were
unreliable in this case.
The distribution of soot concentrations in the city of Irkutsk was
also studied. During the warm half of the year, when the air pollution is
minimal, all the points satisfactorily fall on a straight line. During
the winter, all the values above the maximum permissible concentration
(MFC) are actually measured with a large error and, according to the above,
urban distribution (Fig. 1, curve 2) , very high values are probable. This
indicates that the soot concentrations which we obtained in winter were too
high.
Thus, the use of the method of analysis of the impurity concentration
distribution on a probability graph permits the determination of reliable
measurement data and range of the most reliable values. In observations
carried out with sufficient precision, sharp changes in the nature of the
discharges of the main sources can be revealed.
Another possible use of the method under consideration consists in
evaluating the correctness of the choice of the area for setting up the
points of sample collection. In selecting areas for urban points it is
important that the climatic characteristics of the air pollution differ
most substantially. At the same time, not only the average values, but
also the distribution of the impurity concentrations obtained from data
of observations at individual points should be sufficiently different.
Hence, the statistical parameters m and s at the different points should
also differ from each other. Therefore, in approaching the selection of
areas for stationary posts, it is useful first to carry out itinerary
observations at a number of urban points. On the basis of the data ob-
tained, using a statistical treatment, it is necessary to determine the
parameters s and m and to select points which permit an effective monitor-
ing of the air pollution.
We have carried out such a treatment of observational data for expe-
ditions in the areas of the Moldavskaya and Shchekino SKEPP (State
Regional Electric Power Plant) and the Krasnoyarsk and Cherkassk synthetic
fiber plants. The data were used only in cases where there were 50 obser-
-------
vations at each distance from the source of discharges (0.5, 1.0, 2.0 km
etc.), i.e., the frequency of a single observation would be no less than
0.5%. For a smaller amount of data, the points of probability distribu-
tion of the different gradations of impurity concentrations may deviate
considerably from a straight line. As a result, it was found that the
distribution of impurity concentrations at different distances from the
same source of air pollution also follows the lognormal law. An example
is given in Fig. 2. The curve of distribution of the impurity concentra-
tion for each distance has a definite slope (value of s) , which initially
increases with the distance from the source and then begins to fall off
after reaching a certain maximum.
If in observations at different distances from a plant the distance
at which the highest impurity concentration was observed in a given period
is recorded, then the probability of occurrence of the highest value is
calculated for each distance, then by applying a suitable treatment, as
shown in Table 1, the distribution of the maximum values as a function of
the distance can be obtained. On the probability graph, the distance from
the source of the discharges in this case is laid off on the log scale
along the ordinate axis [4].
Q mg/m?
0,6 r
0,2
0,10
0.06
0,04
0.02
0,0 tO
0,005
0,001
-l-J 1 LJ_U 1 II I III!
J L
0305 I 2345 10 1620 30 40506070 SO 90 95
99p
Fig. 2. Distribution of Concentration of Sulfur Dioxide in the
Region of Moldavskaya SREPP
From the source at a distance: l) 0.5 km; 2) 1 km; 3) 2 km-
^0 3 km; 5) 5 km. '
-------
In cases where the distribution of the values is lognormal, by using
(1) one can obtain analytical expressions for the average value of the
concentration q, its variance 02, the variation coefficient V etc •
The probability of occurrence of a value of q above some value qA is
(2)
(3)
(4)
(5)
As was noted at the beginning of this article, low values of the
impurity concentration are measured with large errors, which necessarily
causes errors in the determination of the average values. The higher the
frequency of low impurity concentrations, the larger the error. This may
be avoided by calculating the average concentration q" from formula (2),
Such calculations were carried out for four cities for nitrogen dioxide,
sulfur dioxide, carbon monoxide, and dust (Table 2). In Table 2, qav is
the average arithmetic value of the impurity concentration; V is the vari-
ation coefficient calculated from (4) , and Ve is the variation coefficient
obtained from observational data.
Table 2
Average Impurity Concentrations (mg/m3).
Year
Minsk
Alma-Ati
Dushanbe
Kursk
Period
1967
1968
1967
1968
1968
1967
1967
Winter 1957
Summer jggg
Winter 1968
Summer 1968
1968
Summer 1968
Impurity 1 lav
Nitrogen Dioxide
Dust
Nitrogen- Dioxide
Sulfur Dioxide
Carbon Monoxide
Nitrogen Dioxide
Carbon Monoxide
Nitrogen Dioxide
n M
Sulfur Dioxide
n n
Carbon Monoxide
Dust
0,11
0,4
0,19
0,15
7
0,20
8
0,05
0,12
0,07
0,07
6
1,8
q
0,13
0,5
0,29
0,24
11
0,27
14
0,09
0,18
0,13
o.r>
8
2,7
V
0.8
0,6
1,3
1.4
1,2
0,9
0.8
0.5
0,4
0,8
0,3
0,8
1.4
Ve
0,8
0,6
0,0
1.1
0,6
-------
As is evident from Table 2, in all cases the calculated averages
are higher than the arithmetic mean values, although in some cases this
excess is slight. Expression (5) was used to calculate the probabili-
ties of concentrations of nitrogen dioxide and carbon monoxide above
10 MFC F(q>10 MFC). Comparison of the data obtained by direct treatment
shows a good agreement (Table 3).
Of definite interest is the study of the variance of the different
impurities. On the basis of observations in several cities for 1968,
the mean square deviations were calculated for each month for dust, carbon
monoxide, nitrogen dioxide, and sulfur dioxide. Fig. 3 shows the depend-
ence of a on qav. The data indicate that the mean square deviation is
close to the average, as also follows from the lognormal law. Formula (4)
was also used to calculate V, the values of the variation coefficient
(Table 2). The latter ranges from 0.4 to 1.9. The calculated and actual
values of Ve are rather close. The calculated values of V show that they
are lower during the warm half of the year than in winter, this being
due to a greater influence of changes in the discharges and weather con-
ditions during the cold half of the year.
At the present time, average (qav) and maximum (qm) concentrations
are used as air pollution characteristics in the correlation of informa-
tion. It follows from the above that the quantity qav + a has some
advantages over qm, since a is less dependent than qm on the observation
period treated.
0,02 -
0,02 0,06 0,10
Fig. 3. Average (gav) vs« "lean square (fj) deviation
of the impurity concentrations.
In each specific case, the error in the determination of maximum
values is related to the number of observations of a given selection.
Since this number varies in cities, the maximum values also frequently
turn out to be incomparable. A maximum with any probability of being
exceeded can be obtained by transforming expression (5). When F(a>a«'> =
0.001, H H0'
to-* me* (6)
-------
Table 3 lists the maximum values from observational data on qm and
values obtained from (6).
The author expresses his sincere appreciation to Ye. L. Genikhovich
for his useful comments.
Table 5
Impurity
Point
,*
F (q 10 MPC$
«M
*
Alma-Ata
Nitrogen Dioxide
n «
n n
Sulfur Dioxide
Nitrogen Dioxide
Carbon Monoxide
Dust
2
3
4
4
2, 3, 4, 5
2, 3, 4, 5
2, 3, 4, 5
5
10
4,5
0,3
Kuft3f
0,1
0,9
15
o,5
11
5,5
0,22
0.1
1,6
13
1.50
2,30
3,40
1,17
0,42
22
9,1
1,50
1,93
4,07
1,19
0,45
27
14.1
LITERATURE CITED
1. B p y K c K. n Kapyaepc H npHMetietme cTaTiicTimecKMx MCTOAOB B Mereopo-
.loruii FiiflpoMeTeoHaflaT, 1963
2 BcHTuejibE C TeopHH ueponTHOCTefi 4>H3MairH3, 1958
3 TyMCejibS CrarHCTHXa aKcrpeMaJibHbix sHaieHufi MSA «MHp», 1965
4.Alb«rt J. Elschant. Messungen Staub und gasformiger Luftverunrcmiguu-
Rpn in der Umgebung ernes Isoliert liegen den Kraftwerks. Staub, 25, Nr II, I961)
5. Charles E Zimmer and Ralph J Larsen Calculating Air Quality
and its Control. Air Poll Contr. Assoc, v 15, No. 12, 1965.
-------
STATISTICAL ANALYSIS OF DATA ON AIR POLLUTION IN CITIES
BY MEANS OF NATURAL FUNCTIONS
»
N. G. Vavilova, Ye. L. Genikhovich and L. R. Son'kin
From Trudy, Glavnaya Geofiz. Observat. itn. A. I. Voeykova, No. 238,
p. 27-32, (1969).
As a result of observations of atmospheric pollution in many cities,
a considerable amount of information has now accumulated on actual concen-
trations of noxious impurities in the atmosphere and the accompanying
meteorological conditions. Statistical analysis of this material aims
primarily at obtaining a sufficiently complete picture of the atmospheric
pollution existing in cities. In addition, as the factual data accumulate,
it becomes possible to use them for the statistical forecasting of the level
of atmospheric pollution in cities and industrial areas. An important stage
in the solution of this major problem involves the statistical analysis of
the initial information. For forecasting purposes it is essential that the
statistical analysis filter out the noise, i.e., the reliability increases,
and the volume of information processed in the forecasting decreases. At
the same time, the statistical relationships established in such an analysis
are of interest in themseJves, since they reflect rules actually existing
(although difficult to identify) and governing the processes of dispersal
of impurities in cities and major industrial districts.
An essential part of the statistical analysis of initial data is the
analysis of measured quantities of ground concentrations of various noxious
impurities in air.
The concentration of each individual ingredient is a random function
q(x, t) of a point of space x and instant of observation t. At the present
time, in analyzing this type of random fields, efficient use is made of the
method of expansion in natural functions (in a statistically orthogonal
system), which is one of the methods of the analysis of variance [1-6].
This method involves the plotting of a system of random orthogonal functions
^K(X)) such that a segment of the Fourier series 2o|c(t)
-------
Actually, observations of the air pollution level in a city are made
at a number of fixed points, so that at each specific instant of time we
have a set of N measured quantities q^, q«» • ••» qjj corresponding to the
observation points with numbers from 1 to N. In this case, the natural
function Kl, ..., (f)^, pertaining
to the corresponding observation points. The natural functions <$>K are the
eigenvectors of the correlation matrix R "(r^) of the concentration field,
the elements of this matrix being found from the formula
— 1, 2, ... , N. (2)
Here q^ is the concentration at the 1-th point, the bar indicating averag-
ing in time. In other words, the natural functions satisfy the equation
(3)
We shall assume hereinafter that the natural functions K are numbered
in decreasing order of the eigenvalues AK corresponding to them.
The expansion coefficients 0^(0 are found from the formula
N
i«»i
wtere (^ are components of vector (t>K. Let us note that the ratio of the
variance described by expression (1) to the total variance of the process is
At
N
Hence it is evident that if all the N natural functions are used in
the expansion, a complete description of the concentration field will be
obtained. More essential, however, is the fact that even if only a few of
the first natural functions are used, the main part of the variability of
the quantity studied will be identified. Discarding of the subsequent terms
of the expansion (higher harmonics) will make it possible to filter out the
noise and thus increase the reliability of the initial information. Thus,
instead of a set of concentration values at the observation points, the
impurity concentration field is described by the first few expansion coef-
ficients. Also important is the fact that because of the orthogonality of
the natural functions, the expansion coefficients are statistically inde-
pendent, so that their forecasting is made independently of one another.
Let us examine the results of a statistical analysis of data on atmos-
pheric pollution at seven points in the city of Sverdlovsk. The calcula-
tions carried out on a "Ural-4" computer, utilized data on concentrations
of nitrogen oxides and sulfur dioxide for the period from December 1966 to
-------
February 1967. As follows from the results of calculation for nitrogen
oxides, the first.term of the expansion describes 70%, and the first two
terms, 90% of the total variability; for sulfur dioxide, the first term
of the expansion describes 48%, and the first two terms, 70% of the total
variability. The fields of the first and second natural functions are
shown in Fig. 1. Attention is drawn to the conformity of the fields of
the first natural functions. Indeed, their maximum and minimum values
are observed in the same parts of the city.
SOi
Fig. 1. Field? of the first (a) and second (b) natural functions.
The first natural function characterizes the basic characteristics of
the spatial behavior of the impurity concentration field. Indeed, if in
expansion (1) we confine ourselves only to one term, we obtain the approx-
imate equality q(x^ t) ^ ^ (t) ^ (x)i (5)
whose accuracy depends on which part of the total spatial variability is
described by the first natural function. It follows from (5) that the
first term of the expansion pertains to simultaneous changes of the concen-
tration over the entire city. Here coefficient a^ characterizes the level
of total atmospheric pollution in the city, and its change with time shows
by what factor this general air pollution has increased or decreased.
-------
The proportions of the components of vector ^ show the relationships of
the concentrations at different observation points during a simultaneous
change of the general level of atmospheric pollution.
Thus, the closeness of the first natural functions for different
ingredients signifies that as the level of the general atmospheric pollution
of the city changes, the relationships between the concentrations of nitrogen
oxides and sulfur dioxide at different points turn out to be relatively simi-
lar. This probably indicates a similarity in the spatial distribution of
sources of pollution of the atmosphere with sulfur dioxide and nitrogen
oxides.
The second natural functions given in Fig. 1 b characterize the ten-
dencies of the spatial variability of that component of the concentration
fields of these impurities which is associated with the basic deviations
from simultaneous pollution of the city's atmosphere (for example, as a
result of a directed transport of the impurity, etc.). As is evident from
the figure, an increase of the concentration in one part of the city is
associated with its decrease in another part.
Let us note that the first term of the expansion for sulfur dioxide
contains less information than for nitrogen oxides. This could be explained
by saying that in the atmospheric pollution of cities with impurities dis-
charged from high sources (including 862) , a more essential role is played by
processes associated with a nonsimultaneous change of the atmospheric pollu-
tion in the city, which are largely determined by the second term of the
expansion.
However, the S02 concentrations in cities are usually low, so that large
errors are possible during their measurement. For this reason, the preced-
ing conclusion requires further confirmation.
Fig. 2 shows the time dependence of the first and second expansion
coefficients. As already noted, for the first expansion coefficient such a
dependence describes the change of the total level of the city's pollution
with time, i.e., the change of the concentrations at all the observation
points simultaneously. It follows from Fig. 2 a that for different impurities
there exists a common physical and meteorological mechanism causing a simul-
taneous change of the impurity concentration over the entire city, since the
expansion coefficients for S02 and N205 change with time in similar fashion
(the correlation factor between them is equal to 0.46).
Obvipusly, the values of ai characterizing the air pollution over the
city as a whole contain less accidental information than single measurements
of concentration at the observation points and should be more closely related
to the meteorological situation. Thus, parameter a^ is useful in the study
of the relationship between impurity concentrations in air and weather con-
ditions and, in the final analysis, in the statistical forecasting of air
pollution. High values of coefficient d]_, corresponding to a heavier atmos-
pheric pollution, should be reached under conditions most unfavorable for
-------
the dispersal of impurities. Indeed, the highest values of coefficients ai
during the period un.der consideration were noted on 2-5 January 1967 (see
Fig. 2 a). On these days, the city was situated in a slow-moving pressure
crest. Definite conditions of air stagnation prevailed (average velocity at
the surface of the ground was 0.1 m/sec, and at a height of 500 m, 3.8 m/sec)
The average lapse rate in the layer up to 500 m was 9.2 deg/100 m, i.e., a
strong inversion was observed in this layer.
The time dependence of coefficient 0.2 is shown in Fig. 2 b. As is evi-
dent from the figure, most of the time the values of these coefficients for
S02 and N205 are different (the correlation factor between them is equal to
-0.36). Hence one can conclude that for different impurities there exists
a single mechanism leading to deviations from a simultaneous change of the
general level of atmospheric pollution (apparently related to certain wind
directions). However, at fixed observation points, the sign of these devia-
tions for sulfur dioxide and nitrogen oxide is different.
Fig. 2. Time dependence of the first (a) and second (b) exnansion
coefficients. «I«»*UB
1 - for S02, 2 - for M205.
-------
The present article gives the first results of an analysis of inform-
ation on air pollution obtained by the method of expansion in natural com-
ponents. In the future, the development of the research should take several
directions. By obtaining natural functions for a number of cities, it will
be possible to analyze the existing state of air pollution and the causes
of its formation. The expansion coefficients of the first terms can be used
in working out a prediction scheme. In addition, these quantities will
further be used for the qualitative study of the relationship between air
pollution and meteorological conditions.
LITERATURE CITED
1 Anflepcon T. B BoeAeHHe B MHoroMepuufi CTarncTHMecKHH aHa;iH3 mecKHx
MSB AH CCCP, cep reo(pH3. M! 3, 1960
~ ~ 06
4PyxoBeuJI. B O6 ontHMa^bHbix npeAcraBflCHHox BepTHKajibHwx
HHfl HexoTopbix MereopojiornHecKHx 3JieMCHTOB Has. AH CCCP, cep reo 4, 1963
5 VHJIKCC MareMaTHqecKan CTarucTHKa HSA «HayKa», 1967
6. K>AHH M. H. O6 H3yMCHHM (J)aKTOpOB, o6yCAOBAHBaK>U(HX HCCTaUHOHapHOCTb 06-
u(eft uupxy^flUHH aTMOc4>epbi. C6. «AHuaMHxa KpynHOMacuira6Hbix aTMOc^epuux npouec-
COB» HSA.
------- |