Post on 11-Feb-2017
Vehicular Emission and Economic Activities:
A Study in Kolkata
ASISH KUMAR PAL
A Thesis Submitted to the University of Burdwan,
For Partial Fulfillment of the Requirements for the Award of the
Degree of
Doctor of Philosophy
Under the supervision
Of
Dr. Atanu Sengupta
The University of Burdwan,
Burdwan - 713104
West Bengal, India
Dr. Atanu Sengupta
Associate Professor,
Dept of Economics
The University of Burdwan
Golapbag, Burdwan-713104
West Bengal (INDIA)
Tel 91-342-2558554 Ext 438
(Office)
919832174249 (Mobile)
Date: ___________________
Certificate
This is to certify that Mr. Asish Kumar Pal has prepared his thesis paper entitled
“Vehicular Emission and Economic Activities: A Study in Kolkata” under my
supervision and guidance. He has fulfilled the requirements relating to the nature,
period of research and presentation of seminar talks etc.
It is also being certified that the research work brings to light the results of an original
investigation made by Asish Kumar Pal. The thesis being submitted now is in partial
fulfillment of the requirements of the Ph.D degree in Arts (Economics) of the University
of Burdwan. It is further to certify that it has not been presented anywhere for any
degree whatsoever either by him or anyone else.
Dr. Atanu Sengupta,
Preface
In the modern world pollution is the buzz word. Any modern city suffers from vehicular
emission that seeks to pollute its skyline. Air pollution is a result of vehicular emission
causes serious health hazard problem having wide socio-economic implication. The
present study wishes to unravel vehicular emission in Kolkata ― An Indian metropolis.
We have considered both macro and micro aspects of vehicular emission. Vehicular
emission has its own dynamics. It is related to the pace and spread of urbanization. It is
related to the vehicular population and road congestion. It also has seasonal fluctuation
pattern. Our macro study wishes to cover all these three aspects. For this we require data
from secondary sources (Census Report, WBPCB Report, Ministry of Petroleum and
Natural Gas, Ministry of Road Transport and Highways etc.). However, it is the
individual who is tormented by vehicular emission. We have selected a particular group
― the traffic police whose duties link them to the day to day pollution in the city. This is
our macro endeavor. Primary data has been collected from the regarding their
assessment, awareness and health conditions. In all we humbly aim to some of the
niceties of vehicular emission that is omnipotent in the city of joy.
Acknowledgements
This thesis has been submitted for partial fulfillment of requirements for the Ph.D degree
in Economics from the University of Burdwan.
I am grateful to various persons at the time of preparation of this thesis for their
valuable guidance and assistance.
I take the opportunity to sincerely acknowledge Dr. Atanu Sengupta, Associate
Professor, Department of Economics, Burdwan University, Burdwan, for his valuable
guidance and encouragement. Unless his kind cooperation and guidance it is not possible
to prepare this thesis.
I am also grateful to other teachers of our department for their spontaneous
cooperation and valuable suggestions.
I wish to thank my friends and fellow research scholars for their help and
cooperation. Special thanks are due to Dr. Krishanu Nath for his cooperation to prepare
the thesis.
I express my gratitude to the traffic police personnel of Kolkata for giving
information and kind cooperation to complete my research work.
I also express my thanks to all of my family members.
However the usual disclaimer applies.
Date:
Place:
----------------------------
Asish Kumar Pal
Department of Economics,
Burdwan University, Burdwan,
West Bengal, India
Contents
Chapter Page No.
1. Introduction..................................................................................................... 1― 7
1.1 Introduction .................................................................................................. 1 - 2
1.2 Vehicular emission in Kolkata ..................................................................... 3 - 4
1.3 Plan of the work ........................................................................................... 4 - 7
2. Literature Review ........................................................................................ 9 ― 22
2.1 Vehicular Emission in General .................................................................. 9 - 13
2.2 Vehicular Emission in Developing countries ........................................... 13 - 18
2.3 Vehicular Emission in Kolkata ................................................................ 18 - 22
3. Data and Methodology of the Study.......................................................... 23― 28
3.1 Introduction ..................................................................................................... 23
3.2 Data description ...................................................................................... 24 – 26
3.3 Methodology ........................................................................................... 26 – 28
4. Urbanisation and Vehicular Population .................................................. 29 ― 49
4.1 Introduction ............................................................................................. 29 – 30
4.2 Data and Methodology ............................................................................ 30 – 31
4.3 Urbanisation - its dynamics with special emphasis on West Bengal ...... 32 – 37
4.4 Urbanisation and Vehicular Ownership .................................................. 37 – 44
4.5 Towards a non – parametric Analysis ..................................................... 45 – 48
4.6 Conclusion ....................................................................................................... 49
5. Urban Vehicular Problems: Some Issues ................................................ 50 ― 69
5.1 Introduction ............................................................................................. 50 – 51
5.2 Urban Conditions in India and West Bengal ........................................... 52 – 53
5.3 Vehicular Population ............................................................................... 54 – 58
5.4 Roadway Congestion in India and West Bengal ..................................... 58 – 59
5.5 Vehicular Pollution ................................................................................. 59 – 62
5.6 Vehicular Ownership Pattern .................................................................. 62 – 67
5.7 Public Transport: Some Issues ................................................................ 67 – 68
5.8 Conclusion ....................................................................................................... 69
6. Analysis of Vehicular Emission in Kolkata ............................................... 70 – 92
6.1 Preliminary View ............................................................................................ 70
6.1.1 Introduction .................................................................................... 70 – 71
6.1.2 Trend in Vehicular Emission ......................................................... 71 – 72
6.1.3 Seasonal Fluctuation in Vehicular Emission .................................. 73 – 77
6.1.4 Status of Other Vehicular Pollutants ...................................................... 78
6.1.5 Seasonal Fluctuation of Air Pollution Level ................................... 78 - 80
6.1.6 Conclusion .............................................................................................. 80
6.2 Spectral Analysis ...................................................................................... 81 - 92
6.2.1 Introduction ............................................................................................ 81
6.2.2 An Overview on Spectral Analysis ................................................ 82 – 83
6.2.3 Data Used ....................................................................................... 83 – 84
6.2.4 Result of the Study ......................................................................... 84 – 91
6.2.5 Conclusion .............................................................................................. 92
7. Effects of Vehicular Emission – A Study of traffic police in Kolkata ... 93 – 117
7.1 Introduction ............................................................................................. 93 – 95
7.2 Rational of the Study ............................................................................... 95 – 97
7.3 Theoretical Framework ......................................................................... 97 – 100
7.4 Data Description ................................................................................... 100 - 110
7.5 Empirical Findings .............................................................................. 109 – 119
7.6 Conclusion ..................................................................................................... 120
8. Conclusion ................................................................................................ 121 – 125
9. Reference .................................................................................................. 126 – 140
Appendix ........................................................................................................ 141 - 167
1
Chapter – 1
Introduction
1.1 Introduction:
“Anna chai, pran chai, alo chai, chai mukto baiu,
chai bal, chai swasthya, ananda ujjwal paramaiu”
―Rabindranath Tagore
“We want food, we want vitality, we want light and the free air
We want strength, health and a meaningful life of love and enjoyment"
Pollution is a staple problem of today’s world. Rich or poor, planed or market
oriented, democratic or autocratic, modern or traditional ― nation all over are steeped in
pollution. While the facets of pollution changes from country to country or region to region,
it is undeniable that we can not live without it. In a way, it is the price of the riches that the
science and technology bestowed upon us. Like a thrown of a rose or the catered pillar from
which the butterfly is brown, pollution is embedded if we are to enjoy the magnificence of
human triumph in science and technology.
Air pollution is one of the very important sources of pollution. It is more prevalent in
the cities and urban conglomerates of the developed and developing world. A major source
2
of air pollution in the city proper is the spurt of vehicles. Development is almost
synonymous with a movement from mechanical vehicles (such as bicycles or rickshaws) to
the fuel consuming two wheelers and four wheelers. Vehicles are necessary for transit of
both man and goods across the length and breadth of a city.
However, the vehicular population could have been checked if there would have
been an emphasis on efficient public transit system (such as light-railways, metro-railways
etc.). However, the problem is accentuated both by and increasing inefficiency of public
transport (use of backdated fuel consumption technology) and also the rise in the privately
own two wheelers and four wheelers. Added to this inadequate road space, dilapidated and
ill- maintained road ways, the increase in number of vehicles per kilometer of roads and the
slow average movement of vehicles flare up the problem.
In Delhi, the data shows that of the total 3,000 metric tones of pollutants belched out
everyday, close to two third (66%) is from vehicles. Similarly the contribution of vehicles to
urban air pollution is 52% in Bombay and close to one-third in Kolkata.
The ill effects of vehicular emission are well noted by the scientists (Cochrane et.al,
1978; Wag staff, 1986; Dardanai & Wag staff, 1987; Leu & Gerfin, 1992). Respiratory
problem, incident of air born disease and other sorts of ailment become widely prevalent. It
is thus an urgent task of the policy makers to frame out an efficient way of combating these
demons without hampering the development proper. A study in vehicular emission is very
crucial from the point of view.
3
1.2 Vehicular Emission in Kolkata:
The main focus of our study is vehicular emission of Kolkata. Kolkata is a centre of
commerce and trade of West Bengal. It is an important state of business activities in India.
Huge people of rural areas come to Kolkata daily and complete their necessary works. To
complete their activities they require vehicles to move from one area to another area
throughout Kolkata. The people use different types of vehicles like two-wheelers and four
wheelers. These lead to huge amount of vehicle emission. Vehicular emission contributes
30% share of total atmosphere pollution there. The extreme use of vehicles creates pressure
on transport system in our study area. The number of motor vehicles in Kolkata has risen
dramatically over the years. The CPCB report, 1988-89, attributes the pollution problem in
Kolkata to the rise in the number of vehicles. Due to inadequate and narrow road space,
traffic congestion also happens in peak hours. Traffic problem is very serious there. “Due to
this traffic problem vehicles move slowly. These slow moving cars emit a large amount of
smoking” (‘S.M Ghosh’ quoted in Banerjee et. al. 2002). The total road length has not
increased significantly in Kolkata during the last decade and has remained as low as 6% of
the total city area (Banerjee et. al. 2002). Also many vehicles in Kolkata are backdated,
older and made by outmoded technology. This is also another cause of vehicular emission.
Vehicular emission includes various types of pollutants. These pollutants are
suspected particulate matters, respiratory particulate matters, NO2, SO2, CO and lead etc.
Again the level of these pollutants varies according to seasons, it is well known that the
pollution level increases in the winter season and is the lowest in the rainy season. In
4
Kolkata however, ‘a festive season’ occurring September, October and early November also
have significant impact. Thus pollution levels are molded not only by natural factors but also
by social factors. The scenario and fluctuation of the pollutants level of the important traffic
points of our study area are shown in the chapter five.
As it is well known, vehicular emission has a long run negative health consequences
for those who are continuously exposed to it. This is a rather broad group including almost
all the citizens of metropolis. However, the impact varies and it depends on the direct
contact with vehicular emission. The traffic police personells of the city are a very
vulnerable group in this regard. Standing long time in the city junctions facing the vehicles
directly has a direct negative effect on them. However, the standard economic theory
prescribes that agents take risk according to the returns. This prescription becomes
problematic when the risks themselves are uncertain. Moreover, the environment conscious
is very low in developing countries such as India. Pollution hazard rarely come into the
‘rational calculi of the traffic men. Moreover the policemen’s duty is divided between –“on
the road” and “off the road”. Obviously the former involves a far greater health hazard with
direct exposure to the vehicular emissions. These duty allocations are determined by the
higher authorities without little choice of the policemen themselves. In such a scenario the
traffic men can only mitigate their health losses if they take some actions to prevent the
detrimental effect of environmental pollution. Our work deals with all the social and
economic calculation that an agent facing pollution hazard has to undergo.
5
The objectives of our study are three.
• To understand the nature and causes of vehicular emission in various urban
centers
• To unravel the dynamics of vehicular emission
• To understand the impact of vehicular emission on a particular group of people
― traffic police who are directly exposed to it
1.3 Plan of Chapter:
Our plan of work is arranged by the various chapters in thesis. We wish to carry out
this work from both micro and macro perspective.
In the first chapter we wish to give a brief introduction to our problem. We will also
spell out the relevance of such a study in an underdeveloped economy like India. The
rational of the work will emanate from this elementary discussion.
The second chapter is on literature review. Various studies concerned with
vehicular pollution are discussed in this section. We also critically analyses the earlier
studies and find out the research gap.
6
In the third chapter, there are various types of data available on the pollution
parameters. Various national and international agencies publish regular data on pollution
aspects that might be relevant for our study. However, a primary survey is essential when we
deal with some specific issues. Our work tries to use both the secondary and primary data
for the purpose of analysis.
The fourth chapter gives a brief panoramic view of the city of Kolkata with
emphasize on urban development and pollution problems. The problem in Kolkata in
automobile emission is then discussed. A comparative discussion with the other rural areas
of West Bengal is also introduced in this chapter.
In chapter five, we considered the issues related to urban vehicular problems of all
India level as well as West Bengal. This chapter also deals with the macro dynamics of
vehicular pollution in the city of Kolkata. The various aspects of this problem are discussed
in this chapter. Issues such as the alarming increase in vehicular pollution, seasonality and
periodicity of the pollution parameters etc. are dealt in this chapter and the underlying
pattern is sought to be identify.
In the sixth chapter, spectral analysis is the analysis of time series in the frequency
domain. Air pollutant level shows considerable amount of seasonality. Spectral analysis aids
us to confront this seasonality in a systematic manner. Unlike the seasonal auto regressive
schemes (SARIMA), the length of season is not fixed a priori. This is determined by the data
in the spectral analysis.
7
In order to analyse the health hazard effect on the traffic police on Kolkata, we have
collected primary data on a set of traffic personnel at the various junctions of the city. We
then use the subjective evaluation to study the impact of health hazard. The health
production function approach is used for this purpose in chapter seven.
In chapter eight, we give a synoptic view of the entire analysis. The links between
the chapters are clarified. We are drawn towards the conclusion through a linkage of the
summaries of several chapters.
8
Chapter -2
LITERATURE REVIEW:
2.1 Introduction:
There is a vast literature on vehicular emission and its impact on environment. A
large number of these studies also relate themselves to the questions of socio-economic
causation. It is not our intention to give an encyclopedic view of the entire literature in this
chapter. Our aim is rather modest. We wish to concentrate only on a few studies that are
relevant for the exercise we have undertaken in this dissertation. We divide our review into
three subsections ― a) a synoptic view of the literature of vehicular emission in general b)
vehicular emission in developing countries and c) vehicular emission in Kolkata.
2.2 Vehicular Emission in General:
Faucet and Sevingny (1998) have argued at the same conclusion that inefficient
transportation is the major culprit of air pollution accounting for over 80% of total air
pollutants. This is a clear indication that vehicle emissions are a major source of ambient air
pollution. The form of urban growth in most developing countries has tended to increase the
use of motorized transport, particularly road transport, which leads to increase
environmental impacts.
9
The composition of traffic on city roads also affects emission.” The future trend in
vehicle growth can have serious implications and fuel consumption patterns,” warns VSN
Shrinivasan, an energy expert from TERI (Tata Energy Research Institute). A large number
of private vehicles for example, are two-wheelers, which are cheap and reliable but also high
on emissions.
Different studies have been done in the field of motor vehicular emissions in the
different regions of the world, especially to establish the level of air pollution from the
operation of motor vehicles and the general urban air quality as a whole. Three of such
studies which have relevance to this study are: the vehicle activity study in Nairobi, Kenya,
conducted in March 2001 by the U.S. EPA, CE-CERT5, and GSSR6 the evaluation of
evaporative emissions from gasoline powered motor vehicles under South African
conditions, conducted in 2003 by Van des Westhuisena et al. (2004); and the impact of
automobile emissions on the level of platinum and lead in Accra, Ghana conducted in 2001
by Kylander et al. (2003). All of these find strong correlates between air pollution and
vehicular efficiency (in the fuel use).
Ambient temperature and local meteorology influences the concentration and
location of vehicle-emitted pollutants. For example, elevated sulphur dioxide levels are
typically reported in the winter and elevated ground-ozone levels in the summer (Goldberg
et al. 2001; Rainham et al. 2005).
10
Cold weather can result in higher levels of pollutants in ambient air due to reduced
atmospheric dispersion and degradation reactions. The genotoxic effects of PM2.5 and
PM10 have also been found to be greater in the winter months (Abou Chakra et al. 2007).
Dispersion of pollutants is also affected by other meteorological factors like humidity, wind
speed and direction and general atmospheric turbulence.
Many studies have found that chronic exposure to high levels of traffic noise
significantly increases the risk for cardiovascular diseases and death by myocardial
infarction (Babisch 2000, Fogari 1994 and Davies 2005). A study in Denmark of 28,744
men with lung cancer found an increased risk among taxi drivers and truck drivers when
compared with other employees, after adjustment for socioeconomic factors (Hansen et al.
1998). Other studies have found similar effects for lung cancer in taxi, truck, and bus drivers
(Borgia et al. 1994, Guberan et al. 1992, Jakobsson et al. 1997, Steenland et al.1990).
A similar study confirms that there is a prevalence of chronic bronchitis and asthma
in street cleaners exposed to vehicle pollutants in concentrations higher than WHO
recommended guidelines, thus leading to significant increase in respiratory problems
Rachou (1995). Other studies in Ethiopia, Mozambique, and Kenya found significantly
higher prevalence of asthma in urban school children exposed to traffic pollution compared
to rural child (Bekele 1997, Mavale-Manuel 2004 and Ng'ang'a 1998).
In 2004, Toronto Public Health released a study that calculated the burden of illness
associated with ambient (outdoor) levels of air pollution in Toronto. The study estimated
11
that smog-related pollutants from all sources contributed to about 1,700 premature deaths
and 6,000 hospitalizations each year in Toronto. The study indicated that these deaths would
not have occurred when they did without chronic exposure to air pollution at the levels
experienced in Toronto. TPH stuff used the Air Quality Benefits to determine the burden of
illness and economic impact from traffic related air pollution.
In 2008, Erica Moen conducted a survey on “Vehicle Emissions and Health Impacts
in Abuja, Nigeria” and put a question on the seasonal variation of pollution and fond that the
greatest percentage of respondents (42%) reported that their symptoms are more severe in
the dry season. This implies that documented concentrations may actually be lower than
what is observed in the dry season, which is supported by similar monitoring studies.
Several studies reported significant health risks and increased morbidity and
mortality rates and hospital admissions because of cardio respiratory diseases, oxidative
stress, and an increase in the incidence of cancer among the urban population (Maynard
1999).
Aside from exposures while traveling inside a vehicle, a significant proportion of the
population are exposed through occupations that lead to extended periods of time on or near
roads and highways or close to traffic like asphalt workers (Randem et al. 2004), traffic
officers (de Paula et al. 2005; Dragonieri et al. 2006; Tamura et al. 2003, Tomao et al. 2002,
Tomei et al. 2001), street cleaners (Raachou-Nielsen et al. 1995), street vendors, and
tollbooth workers. Health impacts are greater for these groups who work close to traffic than
12
for those that are not occupationally exposed. The ill-effects of this pollution are mainly by
the people who came close contract with it (such as constables, street venders, shop-keepers
etc.).
A study in Copenhagen found that street cleaners had a greater risk for chronic
bronchitis and asthma when compared with cemetery workers (Raaschou-Nielsen et al.
1995). It has been reported that traffic policemen present with airway inflammation and
chronic respiratory symptoms at higher rates than in non-exposed groups (Dragonieri et al.
2006 and Tamura et al. 2003). Asphalt workers have also been reported to have an increased
risk of respiratory symptoms including lung function decline, and chronic obstructive
pulmonary disease (COPD) as compared with other construction workers (Randem et al.
2004).
Individuals living close to major roads are at increased risk of exposure to traffic
related pollution and related health effects. In fact, residential proximity to a major road has
been associated with a mortality rate advancement period of 2.5 years (Finkelstein et al.
2004). Of particular concern are communities close to border crossings, where traffic levels
are high and include a large proportion of transport trucks. For example, individuals living
close to the Peace Bridge, one of the busiest US-Canada crossing points, show a clustering
of increased respiratory symptoms, particularly asthma (Lwebuga-Mukasa et al. 2005;
Oyana et al. 2004 and Oyana et al. 2005).
13
Moreover, street vendors frequent high traffic intersections, working both on the
sidewalk and walking through the intersections during slow traffic. Street vendors, therefore,
may be at high risk of developing health effects, which has been documented by a study in
India (Chantanakul 2006).
Traffic wardens, part of the Federal Capital Territory Area Command (FCTAC),
were included in the study to link measured pollutant concentrations with health impacts.
Wardens are the highest exposure group because they stand in intersections and direct
traffic, so they are directly and frequently exposed to vehicle emissions. Thus, it is
reasonable to expect their health status to directly reflect that level of exposure. There are
roughly 300 active traffic wardens in Abuja who work on average 8 hours per day between
the hours of 7am and 8pm, 5-7 days per week, with 2 weeks of vacation per year (Akoni
2008).
2.3 Vehicular Emission in Developing Countries:
A Case Study on Beijing is done by Hao and Wang (2005) urban air pollution is one
of the major environmental issues. Air pollution problems are induced by high-speed
urbanization, rapid economic growth, and explosive motorization. Chinese cities pose a
direct threat to long-term economic sustainability and social benefit. Air pollution problems
in Chinese cities are serious, especially in large cities.
14
From the CPCB report of 2010, air pollution is one of the serious environmental concerns
of the urban Asian cities including India where majority of the population is exposed to poor
air quality. Most of the Indian Cities are also experiencing rapid urbanization and the
majority of the country’s population is expected to be living in cities within a span of next
two decades. Since poor ambient air quality is largely an urban problem this will directly
affect millions of the dwellers in the cities. The rapid urbanization in India has also resulted
in a tremendous increase the number of motor vehicles. The vehicle fleets have even
doubled in some cities in the last one decade. This increased mobility, however, come with a
high price. As the number of vehicles continues to grow and the consequent congestion
increases, vehicles are now becoming the main source of air pollution in urban India.
The developing countries suffer more than the developed countries from air pollution
which happens from vehicle emissions. High levels of lead, primarily from vehicle
emissions, have been identified as the greatest environmental danger in a number of large
cities in the developing world. For them the problem is becoming more acute as the numbers
of motor vehicles are growing rapidly. Delhi’s inhabitants inhale the most polluted air in the
country and vehicular pollution is responsible for 64 percent of the pollutants which make it
so (Bhattacharyya et. al. 2002).
Transport is a vital component of any vibrant city. With the case study of
Ahmedabad, the primary reasons for pollution and lack of management in respect of
transport are lack of integration between land use and transport planning, concentration of
economic and other activities in the core of the city, lack of scientific design of road
15
networks, high rate of growth of vehicles, mixed traffic on roads, and low quality and
adulterated fuels (Brar 2004).
The pressures on transport systems are increasing in most developing countries, as
part of the process of growth. It is even worst in urban areas where population densities are
higher Motor vehicle ownership and use are growing even faster than population, with
vehicle ownership growth rates of 15 to 20 % per year common in some developing
countries (World Bank 1995).
Shariff (2012) expressed his concern that private vehicles today have become the
main means of travel of urban living in developing countries. Consistent economic growth,
rising incomes, and urbanization have led to rapid growth in vehicle ownership and usage.
Private vehicle ownership is also associated with externalities such as traffic congestions,
accidents, inadequate parking spaces and pollutions. Rising vehicle congestion and slower
travel speeds are the most obvious impact of rapid motorization. With the increase in vehicle
ownership, it has been emphasized that the demand for travel to central city areas would
grow far beyond the capacity of the road network. Hence air pollution and other
environmental hazards are another important concern.
According to WHO report published in 1994 the Indian capital is the fourth most
polluted city in the world. And no wonder, the amount of pollutants the transport sector
pumps into Delhi more than the sum of vehicular pollutants emitted in Mumbai, Bangalore,
and Kolkata. However, the Delhi has taken stringent steps recently that brought down the
16
vehicular pollution to a considerable extent. India is moving on a fast track with the increase
in GDP by 2.5 times, industrial pollution by 8 times over the last two decades (Pattanaik and
Pattanaik 2002).
Noise, air and water pollution are all serious problems in Indian cities, and transport
sources contribute all three kinds. Heavy transportation in metropolitan cities is a major
contributor to environmental pollution in addition to industrial and commercial activities.
Indian cities face a transport crisis characterized by levels of congestion, noise pollution,
traffic facilities and injuries, and inequality far exceeding those in most European and North
American cities. India’s transport crisis has been exacerbated by the extremely rapid growth
of India’s largest cities in a context of low incomes, limited and outdated transport
infrastructure, rampant suburban sprawl, sharply rising motor vehicle ownership and use,
deteriorating bus services, a wide range of motorized and non-motorized transport modes
sharing roadways, and inadequate as well as uncoordinated land use and transport planning
(Pucher et. al. 2004).
According to India Development Report of 1997 the reasons for deterioration of
urban air quality throughout the Indian cities are growing industrialization without any
priority for pollution abatement, and rising number of motor vehicles especially of poorly
maintained vehicles that used leaded fuel. The rate of generation of solid waste in urban
centers has outpaced population growth in recent years with the wastes normally disposed in
low-lying areas of the city’s outskirts (India: State of the Environment 2001).
17
Vehicles are a major source of pollutants in cities and towns. A three-fold increase in
the number of motor vehicles has been found in India in the last decade. The concentration
of ambient air pollutants in the metro-politan cities of India as well as many of the Indian
cities is high enough to cause increased mortality. The life of the urban dwellers of India
may become more miserable which may be the cause of health hazards and worst
devastation. In all the four metro cities SPM was found highest along with the problem of
solid wastes. The noise pollution was noticed more than the prescribed standard in all the
four metro cities. Five and more person residing in a room was faced by more than one
fourth population of Mumbai followed by a little less than one fifth population of Kolkata
and about 10% population of Delhi and Chennai both. India’s urban future is grave (Maity
and Agarwal 2005).
The total pollution load from transport sector has increased from 0.15 million tones
in 1947 to 10.3 million tones in 1997 (TERI 1997). Like many other parts of the world, air
pollution from motor vehicles is one of the most serious and rapidly growing problems in
urban centers of India (UNEP/WHO, 1992). In India, the number of motor vehicles has
grown from 0.3 million in 1951 to approximately 50 million in 2000, of which, two
wheelers (mainly driven by two stroke engines) accounts for 70% of the total vehicular
population.
According to UNEP-WHO report, 1992 and the World Development Report 1992
vehicles are nowhere the principal cause but its relative importance is growing rapidly over
time. In World Bank’s report of 1992 the Bank expressed the premonition that such
18
economic growth will be associated with applying environmental damage. Degraded air
quality would be one of such impending damage that may cripple the developing country.
According to the report of CPCB in 2010, air pollution is a major environmental risk
to health and is estimated to cause approximately 2 million premature deaths worldwide per
year. The high level of pollutants are mainly responsible for respiratory and other air
pollution related ailments including lung cancer, asthma etc., which is significantly higher
than the national average (CSE, 2001 and CPCB, 2002).
2.4 Vehicular Emission in Kolkata:
Now coming to Kolkata, condition is not at all better. Traffic problem in Kolkata
seems to be on its way to accruing the dimension of cities like Bangkok, where commuters
more or less live in their vehicles, bathing and getting dressed for works while inching their
way through nightmarish traffic jams. Kolkata’s citizens already spend hours gulping down
fumes while stuck in traffic jams. During the last decade the road length has not increased
significantly in Kolkata. So Traffic jam occurs. As Traffic moves slowly, slow moving
vehicles emit more carbon monoxide, the problem is more surmounted for the kolkata by the
lack of adequate data availabity. This co-effect is toxic to human’s body. It reacts with the
hemoglobin of blood and affects oxygen supply to the brain. Thus it causes death (Banerjee
and Bhattacharjee 2001). They have also observed that the air of Kolkata is polluted
differently in different seasons. During the monsoon season (July-October) the air is
19
comparatively clean due to heavy rainfall and during the summer (April-June) high winds
blow away the pollutants. But air quality is worse during winter months (Nov.-Feb.)
The population residing in the vicinity of the city, daily commuters, and business
people are always exposed to the traffic air containing particulate matters, inorganic gases,
and volatile and semi volatile organic compounds (Chattopadhyay et. al. 2007).
One of the major causes for road congestion and therefore, vehicular emission is the
massive increase in the vehicular pollution plying in and around the Kolkata city area.
According to the report published in the recent Telegraph 7.12.99, the aggregate registered
vehicular population in Kolkata has increased from 6, 34,835 in 1997 to 9, 50,000 by the
end of 1997, in 2006-07, it has increased massively. Due to this huge vehicular pollution
growth, the energy demand (both diesel and petrol) increased manifold. One of the major
factors that determine vehicular emission is the speed of the vehicles. According to the
eminent scientist there exists a critical speed at which the emission is less of the vehicles. If
the speed is well below or well above that critical one then the emission will rise
significantly. So vehicle pollution is one of the important components in air pollution. Citing
a study by NEERI they found that the main concentration levels for various atmospheric
pollutants have increased in all the major cities. According to the World Resource Report
(1996-97) the motor vehicles are responsible for 90% emission of (CO), 85% of sulfur
dioxide (SO2) and 37% of (SPM) in Delhi. According to the same report, automobiles
caused about 52% of the total (NO2) emissions, 5% (SO2) emissions and 24% of SPM
emission in Mumbai in 1992. In the year 1993-94, the share of automobiles in total pollutant
20
load was as high as 64% in Delhi and 52% in Mumbai. The share was 30% in Kolkata in
1988-89. According to a report of the National Commission on urbanization 1988, the
increase in projected travel demand per day in Kolkata in 2001 will be 63% compared to the
1981 base (Haldar 1997).
The vehicular pollution is severe in the Kolkata Municipality Corporation. Due to the
huge population growth, the estimated energy demand (both diesel and petrol) increases
from 352 thousand tones in 1990-91 to 1603 thousand tones in 2000-01, nearly five times
increase, It is also estimated to be 30 percent of the total pollution, (Agarwal 1996).
According to NEERI study in 1973-34, 176.7millions tones of CO emitted in Kolkata’s
atmosphere every day and transports sector contributes 138.7metric tones. Most of the
emission flows from four wheelers (approx. 78%) and two & three wheelers 22% of the total
emission.
A huge number of motor vehicles in Kolkata have been increased over the recent
years. From the CPCB report we know the pollution in Kolkata raises due to large number
of vehicles, the inadequate and narrow road network. In Kolkata, we see a fair number of
rickshaws and bicycles, which reduce the average traffic speed and as a result emission
increases (Chakraborty 1997). According to a report of the National Commission on
urbanization 1988, the increase in projected travel demand per day in Kolkata in 2001 will
be 63% compared to the 1981 base (Haldar 1997).
21
To investigate the health impact of vehicular pollution of Kolkata (former Calcutta),
a cross-sectional study by Lahiri et. al. (2002) was carried out during 2007-2010 among 932
male non-smoking residents of the city and 812-age- and gender-matched rural subjects as
control. The urban group included 460 men who were occupationally exposed to vehicular
pollution: 56 traffic policemen, 188 street hawkers, 82 auto rickshaw drivers, 78 bus drivers
and 56 motor mechanics. Remaining 472 participants from the city were office employees.
Compared with control, urban subjects had increased prevalence of respiratory symptoms,
asthma, headache and reduced lung function, chronic obstructive pulmonary disease and
hypertension. Lahiri et. al. (2002) also compared prevalence of respiratory symptoms such
as breathing problem of individual residing in Kolkata and individuals residing in rural parts
of West Bengal where air pollution Level is not alarming. The study also reported whereas
45 percent of rural individuals complained about such problems, 75 percent of urban
households in Kolkata exhibited respiratory problems.
Majumdar (2007) studies the health cost of air pollution in Kolkata. He argued that
there is a direct link between health and air quality. Deterioration in health quality leads to
the greater vulnerability towards diseases. In order to understand the impact of air quality in
health, Majumdar (2007) collected data on 600 samples KKM from 10 locations in Kolkata.
He collected a number of quantitative and qualitative information from his samples. He
specially collected data on the socioeconomic variables, preponderance of illness and the
health cost of incurred. It was found that the vulnerable people having chronic element and
smoking habits, as well as the minority communities susceptible towards health risks. It was
22
also found that citizen of Kolkata have to bear a significant health cost due to airborne
diseases.
However, a very important dimension of vehicular pollution is the damaged that is
caused due to constant exposure of deadly gases by the different sections of people whose
occupations are directly or indirectly linked with such exposure (such as traffic police ,
street vendors , peddlers etc). Study (Sinha 1993) pointed out to various health elements that
affects these people.
2.5 Conclusion:
The above studies clearly reveal that pollution have a numerous effect on the life of
man. However, pollution itself is a result of a complex socio-economic process. Our aim in
the dissertation is to understand some of the aspects of vehicular emission in the context of
Kolkata. We start our journey with the brief data description.
23
Chapter – 3
Data and Methodology of the Study
3.1 Introduction:
Air pollution has both macro and micro dimensions. On the macro level it increase
the overall toxicity of the air. At the micro level we consider the individual’s reaction while
facing such toxic environment. Our analysis thus required both macro and micro data. The
macro data is culled from various secondary sources-both published and unpublished. The
micro data however required primary data collection. This chapter discusses briefly the data
sources that are used by us. This chapter is divided into two sections. In section 3.2 we
provide the data description. This section is also divided into two subsections. In section
3.2.1 we describe the secondary data collected from various sources like Census data,
District Statistical Hand Book, West Bengal Pollution Control Board report, Ministry of
Road Transport and Highways, Ministry of Petroleum and Natural gas etc. As for primary
data, the affected traffic police personnel were interrogated for a host of information. This is
discussed in subsection 3.2.2. In the last our basic approach to the study is discussed.
24
3.2 Data Description:
3.2.1 Secondary data:
For the secondary data, we have used both published and unpublished documents.
Since our requirements were multifarious, different sources had to be utilized. Some of the
sources are quite general.
(a) Census Data, 2001:
India census gives us a rich data at various levels of desegregation. From the census
we have gathered three types of data. First, the information on urbanization and its related
features are taken from the census. Secondly, the household amenity data is utilized to
collect information on various types of vehicle owning households is collected. Information
is also collected on the literacy level of the household.
(b) Report of WBPCB: 2001-2008
In our study we examine the fluctuation and trend of the different vehicular
particulates and or emitants on the basis of season of Kolkata throughout the year, which are
collected from the report of WBPCB. The West Bengal pollution control Board has
collected these types of data daily by the regular monitoring from the various traffic points
in Kolkata.
25
(c) Ministry of Road Transport and Highways, 1999, 2000.2003:
To examine the growth rate of different types of vehicles over the years we used data
from the Ministry of Road Transport and Highways, 1999, 2000.2003. Data gives
information on the volume of vehicles. The information is necessary to asses the gamut of
vehicular population.
(d) Ministry of Petroleum and Natural Gas, 2002:
In our study to show the actual picture of Air pollution level the largest Indian cities
like Delhi, Mumbai, Kolkata, Chennai and Bangalore etc.we collected data regarding the air
pollution particulates like SPM, RPM, (SO)x , (NO)x, CO and lead etc. from the Ministry
of Petroleum and Natural Gas.
(e) Others:
However, all the official information is incomplete. Thus miscellaneous other
sources were utilized by us. This includes the research carried on by other scholars, District
Statistical Hand Books and some unpublished official information.
26
3.2.2 Primary data:
For primary data we concentrated on a particularly affected group of people-traffic
police. The information is collected through direct questionnaire method. A detailed
questionnaire regarding their occupation, illness, exposure to vehicular pollution etc. was to
be prepared. We surveyed 98 traffic police personnel who are scattered and continuously
exposed to pollution at the time of their duty throughout Kolkata.
3.3 Methodology:
To analyse these data, we also deal with some statistical operations like OLS, Whit’s
Heteroskedastic Consistent Regression, Multiple Regression, Tobit Regression, Non-
Parametric analysis, Spectral analysis, various charts and diagrams in our study.
3.3.1 OLS, Multiple Regression, and Non-Parametric analysis:
In the fourth chapter we have used the multiple regressions for analyzing the
relationship between literacy rate, urban population, and urban vehicle holding household.
In this chapter we have also taken a non parametric analysis to show the pollution of urban
area is higher than the rural area. The non parametric analysis is based in the concept
of ‘distance’ between two independent distribution. Instead of testing the differences at
a particular point, As the regression considers only “average points” non - parametric
method take into consideration all the feasible points of the distribution.
27
3.3.2 Tobit Regression and White’s Heteroskedastic Consistent Regression
The primary data is analysed using the Tobit egression. Since our data are qualitative
in character, Tobit regression is very helpful tool for analyzing the data. Using the
framework of health capital formation, the appropriate structure is built up. The structure
was then utilized as a basis of our Tobit regression. Also, White’s Heteroskedastic
Consistent regression estimate has been utilized for this exercise.
3.3.3 Spectral analysis:
Under the frequency domain analysis this technique enables us to analyse the
relationship between any meaningful components of a pair of time series. If there are some
cyclical components of two time series, then this analysis is also necessary for
distinguishing between the short-term relationship and the long-term relationship has been
recognized in many fields of economics. The parameters, which this new statistical
technique estimate, in regard to the relationship are the closeness of the relationship, the
regression co-efficient and the lead or lag between each pair of components.
The time series χt is expressed as the sum of independently varying cosine and sine
curves with random amplitudes. Thus χt can be exactly fitted by a finite Fourier series, be
xt = a 0 + Σ[ ak cos (λkt) + bk sin (λkt)] (for k = 1 to q)
28
Where ak’s and bk’s are uncorrelated random variables with zero expectations and λ (lambda)
is the frequency expressed in terms of radians per unit time and σ2 is variances. The
frequencies are equally spaced and separated by a small interval. The purpose of the analysis
is to see how the variance of χt is distributed among oscillations of various frequencies.
Fourier series reveals that there are few, if any persistent sinusoidal components in
the data. Nevertheless, the oscillations in this series may be described in sinusoidal terms by
spectrum analysis. This is a method that describes the tendency for oscillations of a given
frequency to appear in the data, rather than the oscillations themselves.
In order to show the periodic fluctuation and to compare the oscillation of the series
of pollutants at different periods, we use the technique Cross S Spectral analysis.
After dealing with the data description, in now turn to the main study. The next three
chapters deal with the macro aspect of vehicular emission.
29
♦♦♦♦
Chapter – 4
Urbanisation and Vehicular Population
4.1 Introduction:
Protection of the global environment is in the interest of all of us living in the lonely
planet. All over the world planners are expressing increasing concern about the control of
air pollution. Every now and then the environment protection agencies announce policies
that are intended to cut down on the level of air pollution. The World Development Report
(2007) has identified three specific sources of air pollution (a) emission from industry (b)
transport and (c) domestic emission.
All these components of air pollution are important by themselves. However, in this
paper, we wish to concentrate on vehicular pollution. The vehicular pollution is important
from many dimensions. First, we live in the era of globalization that is closely linked with
the industrialization and the concomitant urbanization, urban growth is associated with a
growth of vehicular population. Secondly, globalization has led to the emergence and
expansion of the middle income classes. This has boosted consumption in various
consumers’ durables – vehicle is one of the important components of it. In any modern
Indian city one can see the spurt of privately own vehicles in the recent years. Thirdly,
increased privatization has led to dismantling of the public transport system and its
♦
This chapter draws heavily on a published paper: ‘Vehicular Pollution in West Bengal’ (Sengupta and Pal),
Environment & Ecology 30 (1): 130-132, January – March, 2012.
30
replacement by private operators. In all these there is a surge of automobile demand that is
going to contribute heavily towards vehicular population.
This chapter concentrates on the last problem. Specially, we focus on the ownership
and pattern of vehicular population in the urban areas, its features and determining factors. It
is divided into six sections. In section 2 we provide a brief description of the data used and
the methodology utilized for our study. Section 3 describes the evolution of urbanization
with special reference to West Bengal. Section 4 analyses the relation between possession of
polluting vehicles and some other factors. Section 5 studies the relationship in a non-
parametric way. The paper is concluded in section 6.
4.2 Data and Methodology:
In order to understand the nexus between urbanization and vehicular population we
have considered data from secondary sources. The largest body of data available of
urbanization in India is the census data. Census documents the growth of municipalities,
towns and cities for more than 100 years. In the census we get detailed records of the urban
population, its size and composition and its various socio-economic features. All these are
helpful to understand the multifaceted feature of urban development. More over, the 2001
census give us a detailed recording of the various types of assets of the households. Among
the assets we get a picture of the various types of vehicles held by the households. This data
covers information about the ownership of bi-cycles and all type of two wheelers and four
wheelers. These vehicles can be broadly classified as oil consuming (i.e. polluting and non-
31
oil consuming) vehicles. The pollution created by the oil consuming vehicles is very high as
compared to the non fuel consuming vehicles. The census data gives us an opportunity to
understand the possession of the fuel consuming vehicles and several factors that may be
important in this respect. One of the important factors is undoubtedly urbanization. With the
growth of towns and the rise in middle class, there is created a demand for these vehicles.
However the relationship and its intensity are far from obvious. In the case of planned
urbanization with the rising public transport system, the demand may be mitigated to a
certain extent. However if urbanization is unplanned with a languishing (or non-existing)
public transport system, the spurt in private vehicles is often the only logical solution.
In order to stress the relationship, we first depict the pattern of urbanization in the
next section. These give us an idea of the terrain that we have to traverse. Having
understood some of its features, we next move to the econometric analysis of the
relationship between urbanization and the possession of polluting vehicles. However
econometric analysis assumes certain structures- the result that we get may be driven by
such structure. This is the reason why we turn to the non-parametric techniques in order to
corroborate our findings based on econometric analysis. Comparison between these two
techniques may help us to unravel the relationship between urbanization and the possession
of polluting vehicles.
32
4.3 Urbanization - its dynamics with special emphasis to West Bengal.
Kingsley Davis has explained urbanization as process (Davis 1962) of switch from
spread out pattern of human settlements to one of concentration in urban centres. It is a finite
process --- a cycle through which a nation passes as they evolve from agrarian to industrial
society (Davis and Golden 1954).
The onset of modern and universal process of urbanization is relatively a recent
phenomenon and is closely related with industrial revolution and associated economic
development. As industrial revolution started in Western Europe, United Kingdom was the
initiator of industrial revolution. Historical evidence suggests that urbanization process is
inevitable and universal. Currently developed countries are characterized by high level of
urbanization and some of them are in final stage of urbanization process and experiencing
slowing down of urbanization due to host of factors (Brokerhoff; 1999, and Bremman;
1998).
A majority of the developing countries, on the other hand started experiencing
urbanization only since the middle of 20th century. The pattern of urbanization in India is
peculiar. The lopsided dynamics of sectoral composition where the primary sector is
increasingly replaced by an expanding service sector with the slow growth in secondary
sector had its impact on the process of urbanization. It is characterized by continuous
concentration of population and activities in large cities. Kingsley Davis and Golden
(1954) used the term “over - urbanisation” to characterize this process. They argued
33
“over – urbanization” where urban misery and rural poverty exist side by side with
the result city can hardly be called dynamic”. Another commentator observed the
Indian urbanization as a process whereby inefficient, unproductive informal sector
becomes increasingly apparent (Kundu and Basu, 1998). Yet another scholar (Breese,
1969) depicts urbanization in India as pseudo urbanization where in people arrive in
cities not due to urban pull but due to rural push. Reza and Kundu (1978) expressed
opinion of dysfunctional urbanization as well as urban accretion which results in a
concentration of population in a few large cities without a corresponding increase in
their economic base. In short, India’s urbanization is followed by some basic
problems in the field of inadequate housing, unprecedented growth of slums,
inefficient transport system, problems in water supply and sanitation, water pollution
and insufficient provision for social infrastructure (school, hospital, etc.). It is this
lopsided growth that adds to the problem of vehicular pollution.
The developing countries suffer more than the developed countries from air pollution
(Bhattacharya and Banerjee 2002) that happens from vehicle emissions. High levels of lead,
primarily from vehicle emissions, have been identified as the greatest environmental danger
in a number of large cities in the developing world. For them the problem is becoming more
acute as the numbers of motor vehicles are growing rapidly due to rapid urbanization.
One of the interesting studies is done by Bhattacharya and Banerjee in
chapter - 3 of their book – ‘Air pollution and willingness to pay’. They gave a brief
review of the urban air quality and the intensity and spread of vehicular pollution.
34
Citing a study by NEERI they found that the main concentration level for
various atmospheric pollutants have increased in all the major cities. Bhattacharya
and Banerjee identified various causes of this urban pollution - namely - (a) traffic
composition, (b) lack of public transports and (c) insufficient road space.
From this all – India background, we move to West Bengal. First we observe the
pattern of urbanization in West Bengal. The table below shows the growth of census towns
in West Bengal. From the table below and the figure we see that between 1901 and 1951
there was a relatively slow growth in the number of towns in this state. The real breakpoint
is form 1961 after which there is an explosion in the number of urban centers. In recent
census 2001 the growth rate of town was negative (-1.83).
35
Table 4.3.1: Growth of census towns in West Bengal
Census Years Number of Towns Rate of growth overthe previous Census
1901 78 -----1911 81 3.851921 89 9.881931 94 5.621941 105 11.701951 120 14.281961 184 53.331971 223 21.201981 291 30.491991 382 31.272001 375 -1.83
Source: Census Data: 2001
Census year wise number of towns are shown by the following bar diagram.
Fig-4.1: Number of Towns in West Bengal (1901-2001)
Number of Towns in West Bengal (1901-2001)
050
100150200
250300350400450
1901 1911 1921 1931 1941 1951 1961 1971 1981 1991 2001
Number of Towns
Source: Census Data: 2001
We have discussed the evolution of urbanization by the part which is the trend of
urbanization
36
Table: 4.3.2 Trend of Urbanisation
Source: Census Data: 2001
[Trend of urbanization in any particular period = {(% of urban population of the particular year ― % of urban
population of the previous year) / % of urban population of the previous year}* 10}].
In the pre independence period the trend of urbanization was negative in case of Jalpaigunri
and Malda districts in 1911. In 1921 the negative trend was seen in Malda and Nadia. In
1931 it was also negative in Purulia. In post independence period, there were also negative
trends in Coachbihar in 1961, in Darjeeling, Coachbihar, Murshdabad, Bannkura and
Midnapur in 1971. In 1991 the urbanization rate was negative only in 24 pgs (north &
south).The last census year shows the negative trend of urbanization in Westdsinajpur,
Birbhum, Nadia, and Bankura. As in Kolkata the rate of urban population is always cent
percent, the trend is zero percent.
District/Year 1911 1921 1931 1941 1951 1961 1971 1981 1991 2001Darjiling 0.91 1.11 3.42 1.80 3.73 0.91 -0.04 1.95 1.06 0.61
Jalpaigunri -0.60 1.84 2.19 2.80 12.02 2.62 0.54 4.63 1.64 0.90Coachbihar 0.74 0.94 0.47 3.72 7.86 -0.66 -0.24 0.10 1.31 1.66
Westrdinajpur -------- -------- -------- 0.52 41.55 7.41 2.49 1.97 1.94 -0.66Malda -1.03 -0.05 0.62 1.78 1.65 1.07 0.16 1.01 5.24 0.35
Murshidabad 0.81 0.48 0.30 0.96 0.71 0.85 -0.10 1.08 1.14 1.98Birbhum -------- -------- 13.41 2.07 1.24 0.78 0.08 1.79 0.85 -0.46
Bardhaman -------- -------- 2.80 3.81 2.85 2.01 2.52 2.90 1.94 0.53Nadia 0.02 -1.25 3.17 1.64 3.14 0.12 0.18 1.52 0.48 -0.6024 Pgs 1.94 0.64 0.75 1.32 2.10 1.66 1.05 1.04 -1.10 0.01
Hooghly 0.61 0.36 0.89 1.21 2.28 0.55 0.20 1.16 0.56 0.32Bankura 0.52 1.89 0.13 1.79 0.06 0.24 -1.08 1.65 0.87 -1.11Purulia -------- -------- -0.05 3.67 1.92 0.13 0.54 2.55 0.49 0.67
Midnapur 1.21 0.06 3.64 1.90 2.78 0.23 -0.09 1.13 1.60 0.40Howrah 0.53 0.05 0.60 2.41 1.24 2.49 0.36 0.76 0.99 0.16Kolkata 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
37
The long run growth of urbanization is shown below by the following figure.
Table: 4.3.3 Long run Growth of Urbanization
L ong run growth in urbanisation
00.10.20.30.40.50.60.70.80.9
1
Darjiling
J alpaigunri
Coachbihar
Westr
dinajpurMalda
Murshidabad
Birbhum
Bardhaman
Nadia
24 Pgs
Hooghly
Bankura
Purulia
Midnapur
Howrah
Kolkata
Source: Census Data: 2001
From the above figure we see the long run growth of urbanization was the highest position
(0.94) in the district of Murshidabad and it was the lowest (0.01%) in Howrah. This may be
due to the fact that Murshidabad is relatively ruralised district with enough scope for
urbanization. On the other contrary, Howrah is historically an urbanized district.
38
4.4: Urbanization and Vehicular Ownership:
It is argued that urbanization leads to a rise in the middle classes and the
concomitant rise in the ownership of vehicles. We now consider this argument using
the West Bengal data.
Table 4.4.1: Ownership of vehicles (West Bengal-2001)
West Bengal Total no of
Households
Bicycle
Scoter,
Motorcycle,
Moped
Car, Jeep, Van
Total 1,57,15,915 82,64,357 7,90,322 2,97,634
Rural 1,11,61,870 60,59,054 3,41,227 1,35,005
Urban 45,54,045 22,05,305 4,49,095 1,62,629
(Source: Census data, 2001)
In table : 4.4.1 we have shown the different vehicles holding households in
total west Bengal. Then we have divided state by two sectors as rural and urban and
analysed the several vehicles holdings households. There is a clear dichotomy in the
ownership pattern among the rural and urban households. Though the total number of
vehicle owning households is far higher in the rural sector as compared to the urban
sector, most of the vehicle owners in the rural area owns bicycle. In fact, though
bicycle owners in the rural area are much higher, for the other two categories urban
owners far exceed the rural owners. Since bicycle is non-fuel consuming and hence (at
39
least directly) less polluting than the other two categories, it is clear that urbanization
leads to a higher ownership of polluting vehicles.
Table 4.4.2: Ownership of vehicles (% of total)-West Bengal 2001
(Source: Census data 2001)
The percentage figures in table 4.4.2 merely corroborate our earlier findings. From
the above Table, we realize that the total percentage of polluting vehicles in urban
area is greater than rural area in West Bengal. But in case of non polluting vehicles
it is lower in the urban area. Obviously it is clear that the pollution in the urban area
is higher then the rural area. In percentage term, the ownership of fuel consuming
vehicles is more than three times higher in the urban areas.
To fix the idea we consider the district wise distribution of vehicles both in
rural and urban areas of West Bengal. An analysis of the Census data reveals that
in the rural area of Hooghly (rural) we get the highest percentage of polluting
vehicles (8.35%) and Dakshindinajpur is the lowest percentage of polluting vehicles
(2.05%) district. Undoubtedly rural Hooghly is well linked to the urban sector and
West Bengal % of bi-cycle
holding
% of Scoter, Motor
cycle, Moped %Car, Jeep, Van
Total52.59 5.03 1.89
Rural54.28 3.06 1.21
Urban48.43 9.86 3.57
40
shows a high degree of semi-urban or pre-urban features. As for urban areas, we get the
highest percentage of two wheeler & four wheeler vehicles in the district of Barddhaman
(21.44%)-and the least percentage of two wheeler & four wheeler vehicles holding
district is Nadia (7.78%). Here Kolkata is pulled down to the second position. However
the composition of polluting vehicles show a higher percentage of fuel consuming two-
wheelers in Burdwan (18.44%) vis-à-vis Kolkata (8.70%) while Kolkata (6.87%) outstrips
Burdwan (3%) in the percentage distribution of four wheelers. In fact Darjeeling (4.59%)
and 24-Parganas (North) (3.5%) also overwhelm Barddhaman in this matter.
Table 4.4.3: percentage of Polluted vehicles owners (rural and urban) and
percentage of urban Households
District% of polluted
vehicles% urban
households% urbanpopulation
Darjiling 7.8 31.17 32.34Jalpaigunri 5.35 18.49 17.84Coachbihar 3.25 9.18 9.10
Uttardinajpur 3.01 11.39 12.06Dakshindinajpur 2.96 12.07 13.10
Malda 3.37 7.88 7.32Murshidabad 3.26 12.23 12.49
Birbhum 4.73 8.93 8.57Bardhaman 12.17 36.94 36.94
Nadia 3.96 22.18 21.2724 Pgs (North) 8.1 55.54 54.30
Hooghly 9.62 34.95 33.47Bankura 5.81 7.71 7.37Purulia 5.81 10.29 10.07
Medinipur 6.16 10.68 10.24Howrah 8.42 52.09 50.36
24 Pgs (South) 4.63 17.62 15.73Kolkata 15.57 100.00 100.00
Source: census data of India 2001
41
Next we consider the percentage distribution of the owners of polluting
vehicles (urban and rural) and the rate of urbanization in the following table. It is
seen that largely true that districts with the higher degree of urbanization indicates a
higher degree of ownership of pollting vehicles. Kolkata-the fully urbanized district
shows the highest percentage of ownership of polluting vehicles and lowest in
Dakshin Dinajpur.
We now try to evaluate the relationship between the ownership of polluting
vehicles and the factors associated with urbanization. It is generally conjectured that
if the proportion of urban households increases, the probability of polluting vehicles
holdings also increases. There are 100% urban households in Kolkata and we find
there is a highest percentage (15.57%) of polluting vehicles. Another determining
factor is the literacy rate. Conventional wisdom delineates a negative relation between
literacy rate and pollution. In general, if the literacy rate increases, the consciousness
increases leading to a fall in pollution. We want to see whether this factor is important
in determining the ownership of fuel consuming vehicles.
42
Table 4.4.4: Regression results (Urban Households)
Dependent Variable: % of households having polluting vehicles
Regression
Statistics
Multiple R 0.88
R Square 0.78
Adjusted R
Square 0.75
Standard Error 1.72
Observations 18.00
ANOVA
Df SS MS F
Significance
F
Regression 2.00 158.41 79.20 26.87 0.00
Residual 15.00 44.21 2.95
Total 17.00 202.62
Coefficients
Standard
Error t Stat
P-
value
Intercept -1.65 3.20 -0.52 0.61
% urban
households 0.08 0.03 2.26 0.04
lit rate 0.10 0.07 1.53 0.15
43
The regression results show that vehicular population is highly correlated with the
distribution of urban households. Literacy rate has some positive effect on possession of
polluting households (significant at 15% level). This type of perverse relation between
literacy rate and possession of polluting vehicles is quite indicative. High education is often
associated with high income leading to arise in the possession of polluting vehicles.
Similarly we have tried to find out the percentage of polluted vehicles, urban
population and literacy rate in our next regression. The regression results show that
vehicular population is highly correlated with the distribution of urban population. Literacy
rate has some positive effect on possession of polluting households (significant at 14%
level). Again the perverse relation between literacy rate and possession of polluting
vehicles is a result of strong associated of literacy rate with high income leading to arise in
the possession of polluting vehicles.
44
Table 4.4.5: Regression results (Urban Population)
Dependent Variable: % of households having polluting vehicles
Regression Statistics
Multiple R 0.88
R Square 0.78
Adjusted R
Square 0.75
Standard Error 1.71
Observations 18.00
ANOVA
Df SS MS F
Significance
F
Regression 2.00 158.60 79.30 27.02 0.00
Residual 15.00 44.02 2.93
Total 17.00 202.62
Coefficients
Standard
Error t Stat
P-
value
Intercept -1.65 3.18 -0.52 0.61
lit rate 0.11 0.07 1.54 0.14
% urban
population 0.08 0.03 2.28 0.04
45
4.5 Towards a Non-Parametric Analysis:
The above methodologies are a parametric in nature. They depend on a number of
assumptions. The non parametric analysis is based in the concept of ‘distance’ between two
independent distributions. Instead of testing the differences at a particular point, as the
regression does non-parametric method test into consideration all the feasible points of the
distribution.
For non-parametric evaluation suppose A represents the event of “possession of
polluting vehicles”, R –“the rural residence” and U-“the urban residence”. Our proposition
is that the probability of polluting vehicles by rural residence is less than the probability of
polluting vehicles by urban residence, i.e. P (A/R) < P (A/U).
And assume P (A/R) = P (A∩R)/P (R)
P (A/U) = P (A∩U)/P (U)
We measure the probabilities by their observed proportions.
P (A∩R) = No of households in the rural area having polluted vehicles/ No of
households in the rural area
P (A∩U) = No of households in the urban area having polluted vehicles/ No of
households in the urban area
P(R) = proportion of households living in the rural area/total households
P (U) = proportion of households living in the urban area/total households
And P(R) = 1- P (U).
46
Here we have to test whether the probability of polluted vehicles holding by rural
residence is equal to or less than the probability of polluted vehicles holding by urban
resident.
Our first test is mean test.
Table 4.5.1: Mean Test
T - Test: Paired Two Sample for Means
P(A/R) P(A/U)Mean 0.041117 0.131811111
Variance 0.000501 0.001593407Observations 18 18
Pearson Correlation 0.336864Hypothesized Mean
Difference 0Df 17
t Stat -9.95971P ( T<=t ) one – tail 8.21E-09t Critical one – tail 1.739606P ( T<=t ) two – tail 1.64E-08t Critical two – tail 2.109819
Analyzing the data from the above table we find that the t-coefficient is significant
both for one-tailed and two-tailed tests. Hence the mean test justifies that rural people have
lesser probability of holding polluting vehicles.
This test judges the conditional probability of the events to occur between two
distributions at the mean. We test whether the conditional probabilities for households
owning polluting vehicles is high if it is located in the urban area. In contrast to whether it is
located in the rural area.
47
Our null hypothesis is
Ho: mean (P (A/R)) = mean (P (A/U)) and
H: mean (P (A/ R)) < mean (P (A/ U))
However, mere testing of mean does not reveal much about the underlying
distribution.For this we have to compare the distributions of two related variables. The
appropriate test to use depends on the type of data. We first use the non-parametric
Kolgomorov-Smirnov test for testing the difference between the distributions of the two
variables. Our results are given below. From the results below, we see only P (A/U)> P
(A/R) holds the positive ranks and in other cases holds negative ranks and ties. Therefore we
can state that rural people have lesser probability of holding polluting vehicles than urban
people.
Table 4.5.2: Kolgomorov - Smirnov Test
Ranks
N Mean Rank Sum of RanksP (A/U) > P(A/R) Negative Ranks 0a .00 .00
Positive Ranks 18b 9.50 171.00Ties 0c
Total 18a. P(A/U) < P(A/R )
b. P (A/U) > P(A/R)
C. P (A/U) = P(A/R)
Next we consider the Wilcoxon Signed Rank Test and the Sign Test. This test can
be used if the data are continuous. The sign test computes the differences between the two
variables for all cases and classifies the differences as positive, negative, or tied. If the two
variables are similarlydistributed, the number of positive and negative differences will not
differ significantly. The Wilcoxon signed-rank test considers information about both the
48
sign of the differences and the magnitude of the differences between pairs. Because the
Wilcoxon signed-rank test incorporates more information about the data, it is more powerful
than the sign test. In our case however the results from various non-parametric tests
converge. They all clearly demonstrates that P (A/R) - P (A/U) <0.
Table 4.5.2: Test Statistics
Test Statisticsb
P(A/R) – P(A/U)
Z -3.724a
Asymp . Sig. (2- tailed) .000
a. Based on negative ranks.
b. Wilcoxon Signed Ranks Test
Table 4.5.3: Sign Test
Frequencies
N
P (A/U) > P(A/R) Negative Differencesa 0
Positive Differencesb 18
Tiesc 0
Total 18
a. P (A/U) < P(A/R)
B P (A/U) > P(A/R)
c. P (A/U) = P (A/R).
Table 4.5.4: Test Statistics
Test Statistics
P (A/U) > P(A/R)Exact Sig.(2 – tailed) .000a
a. Binomial distribution used.
b. Sign Test
49
4.6. Conclusion:
In this chapter our attempt is to us to test the relationship between urbanization and
vehicular population using the West Bengal data. For our analysis we concentrated on
various secondary sources particularly the census data. The relationship is an expected;
urbanization raises the possession of fuel consuming vehicles.
This is significantly positive both using econometric as well as non-parametric
techniques. Even the literacy rate is perversely affecting the possession of polluting vehicles,
the reason is that higher literacy implies higher human capital and enhances income. The
picture is not bright. Unless proper steps are taken toward efficient use a public transport
system and planned urbanization, there is a little chance in abating the fleet of polluting
vehicles.
From our above discussion we clearly understand that the probability of polluting
vehicles holding is higher in the urban area than the rural area. It has been proved by the
various statistical tests. Therefore we can obviously state that the pollution in the urban area
is more than the rural area in West Bengal. Urbanization leads to a conglomeration of
polluting vehicles and boosting up pollution.
This chapter clears the relationship between urbanization and vehicular population,
also that of polluting vehicles. However, in the large urban conglomerates, there are some
specific issues of vehicular problems. This is the study we turn to our next chapter.
50
Chapter – 5
Urban Vehicular Problems: Some Issues
5.1 Introduction:
Urbanisation is a fact of life in underdeveloped countries. With urbanization comes
the problem of urban transport and mobility. Urbanisation increases the demand for urban
transport manifold. A part of it is made by public transport system. Private owners also ply
vehicles a significant extent. There is also a rising demand of personally own vehicles due to
the expansion of urban middle class.
There is however some serious gaps in the urban transit system. Firstly, the supply of
urban infrastructure and services often lag behind the demand (Pucher, Korattyswaropam, &
Mittal; 2005). This not only leads to a overcrowding of buses and other public vehicles but
also a congestion of the vehicles. The possibility of pollution increases sharply as a
consequence. A much related problem is the inequity in the urban transit system. Side by
side the overcrowded buses and public vehicles, we see cars and private four-wheelers
carrying only one passenger thereby leading to an underutilization of the transport space.
51
In this chapter, we are concerned with three basic issues that have wide implications.
First, the question of equity is considered. This reflected in the ownership of transport and
subsequently its use. With wide ranging poverty in most of the urban cities of India,
ownership of vehicles is scanty. Furthermore, the conditions of road and the flux of traffic
greatly constraints the use of bicycle- the most common form of private transport. This leads
to overdependence on public transport. Second, the problem of road congestion is
considered. The flow of urban transport often outstrips the availability of urban roads. This
leads to slow mobility often cropping up in pollution. Third, we consider the issue of air
pollution. It is shown that a few cities virtually dominate the secretion of polluting gases. All
these problems are interrelated and they often reinforce upon one another. The analysis was
made both with reference to India as well as West Bengal- a major Indian state.
West Bengal has two cities (Kolkata & Asansol) of more than one million
populations according to the 2001 census. There are however a large number of cities and
towns with huge population. The problem of Kolkata is in the serious one. Suffering from a
largely unplanned growth and the proliferations of slums and an unsustainable population
density, Kolkata needs a special mention. The problem becomes serious because it is
economic, political, cultural and educational capital of West Bengal. The preponderance of
all these utilities with a single city makes its really vulnerable. Though recent attempts have
been made in developing Satellite Township around Kolkata, the pressure have not
delimitated to any significant extent. We have compared West Bengal with India to get a
proper perspective of the problem.
52
5.2 Urban Conditions in India and West Bengal:
Most of the developing countries are suffering from rapid urbanization, huge
population growth and rising vehicle. The urbanization in India as well as West Bengal
increased three decades ago. In 1971 it raised from 109 million in 1981 and 217 million in
1991. In 2001 it also up to 285 million increased (office of the Registrar general of India,
2001a; Padam and Singh, 2001).
The speedy growth of India’s cities and the towns of West Bengal have created a
correspondingly a large growth of travel demand. The rapidly increasing levels of motor
vehicle ownership and use have resulted in deep levels of congestion, air and noise pollution.
We have explained the urban situation specifically by the two economic ways. One is
distribution of slum population in the mega cities of India and the main towns of West Bengal.
And another is poverty.
In the urban centers poor people generally live in overcrowded, unhygienic habitat
often termed as slums. Slums are integral part of any large urban centre in the world. In India,
the conglomeration in slums is not insignificant. Now if we follow the picture of slum
population in the mega cities of India, the Census 2001 reveals that there is near 48.88% slum
population in Greater Mumbai. In Kolkata, the relevant figure is 32.55% Chennai 25.60% is
not far away from Kolkata. The situation of Delhi (18.89%) and Bangalore (8.04%) is
comparatively better (Sivaramakrishnan, Kundu & Singh, 2005). These slums are hart bingers
53
of a lot of pollution. The smoke that arises than cooking in overcrowded, mostly places creates
a lot of air pollution.
Another correlates of depreciation is the poverty. Poor people are deprived descent
mean by urban transport. There are lots of measures of poverty. The problem with the official
figure that considers only to bare subsistence. However there are a lot of people who may be
able to meet their minimum subsistence need and yet vingering massive deprivation. The
distribution of asset less may be to more ideal for this purpose. According to the 2001 census
data 18.36% of the urban people of India have no asset. The figure varies from as high as
32.18% in Bihar to a low 6.64% in Chandigarh. In West Bengal the percentage is 20.39%
which is above the national figure. It ranks rather high in the incidents of asset less people,
being one of the poorest owning states in India.
The district wise profile in urban West Bengal again shows the wide variation. The
figure is high 30.97% in Murshidabad and a low 16.7% in undivided Medinipur. Kolkata the
state capital is the second. Thus the problem of poverty is in respect to asset ownership is quite
pronounced in the state. Since the asset list contents the set of two-wheelers. These people
have to depend on public transport for their movement. The pressure on it is enhanced.
54
5.3 Vehicular Population:
Now let us turn to the total fleet population in the country as well as West Bengal. The
figure of given table -1 is below. Due to rising of income of the people, the socio-economic
status changes year after year. So the private vehicles increase more than the increase of
population.
Table-5.3.1: No of vehicles and population in India
Year Actual vehicle (crores)
1971 0.34
1975 0.40
1981 0.79
1985 1.19
1991 2.53
1992 2.73
1993 2.90
1994 3.08
1995 3.31
1996 3.59
1997 3.92
1998 4.28
1999 4.48
2000 4.81
2001 5.35
(Source: Census data, 2001)
55
During the 30 years of period in India, there is about 15.74 fold increase in the total
vehicular population. This is enormous increase from its initial levels. The government also
increases huge public buses due to rapid urbanization.
Table-5.3.2: No of vehicles and population in West Bengal
Year
Actual vehicle
(crores)
1971 0.51
1975 0.74
1981 0.87
1985 1.35
1991 2.05
1994 2.28
2001 3.26
2006 3.58
(Source: Census data, 2001)
Similarly the above figure states that the actual (per-capita) number of vehicles also
increases over the years for the same reason. In the state of West Bengal private vehicles
increase very rapidly. The various types of fuel consuming vehicles also increase year after
year. This is about 7 fold increase in the vehicular population in West Bengal. Apparently this
is somewhat lower than the all India figure. However given the small land area of West
Bengal that accounts for only 2.71% of land area, the difference is not really significant.
56
Next we consider the increase in the types of vehicles. If we follow the growth of
vehicle type in perspective of India the motorcycle ownership increased 15.74 fold increased
between 1981 and 2002. Private car ownership increased almost 7 fold during the same
period. All fuel consuming vehicles increased from 1991 to 2002.
Fig: 5.3.2.1 Growth of India’s motor vehicle fleet by type of vehicles
Grow th of India's motor vehicle f leet by type of vehicle
0
500000000
1000000000
1500000000
2000000000
2500000000
3000000000
3500000000
4000000000
4500000000
1981 1986 1991 1996 2002
Year
No
of v
ehic
le
Goods VehicleMotor car/JeepMotorcycle/scooterother motorised vehicleBuses
(Source: Ministry of Road Transport and Highways, 1999, 2000.2003)
(Note: “others” includes tractors, trailers, motorized three wheelers and passenger vehicles)
57
The low-density development around Indian cities has made cars and motorcycle
ownership increasingly affordable. Rising incomes among the Indian middle and upper classes
have made car and motorcycle ownership increasingly affordable. Fig-1 shows the number of
motorcycle/scooter has increased upward. This upward trend indicates from 1981-2002 the
ownership of motorcycle/scooter has increased in huge amount due to raising income as well
as standard of living of people in India. Other goods vehicles, motorized vehicles and buses
also increased during this period.
Fig: 5.3.2.2 Growth of West Bengal’s motor vehicles fleet by type of vehicles
Grow th of West Bengal's motor vehicles fleet by type of vehicles
0
500000
1000000
1500000
2000000
2500000
1981 1986 1991 1995 2001 2007 2008 2009
Year
No
of v
ehic
le
Goods Vehicle
Motor car/Jeep
Motorcycle/scooter
Taxi/Contact carriage
Mini bus/Stagecarriage
Auto Rickshow
Tractor/Tailors
Otherss
(Source: Ministry of Road Transport and Highways, 1999, 2000.2005, 2010)
(Note: other miscellaneous vehicles that are not separately classified)
58
And then we consider the case of West Bengal. This figure: 2 show the extremely
rapidly growth of motorcycle ownership and other private and public fuel consuming vehicles.
The motorcycle vehicles increase a large proportion from 1981-2008 and after that decreased.
Goods vehicle and motor car/jeep increased rapidly from 2001-2008. At the same time all
other typical polluted vehicles increased.
However the crucial factor in the transport use is not only availability of transport but
also the congestion in the urban transport. For if the vehicles can not move at a significant
space, then the possibility of pollution increases. Also it creates a further disutility to the
travelers. Since their movement is slow and requires a lot of time.
5.4 Roadway Congestion in India and West Bengal:
Traffic congestion is probably the most troubling feature in the cities of developing
countries. It affects all modes of transportation and all socio-economic groups. Average
roadway speeds for motor vehicles in Mumbai fell by half from 1962 to 1963, from 38 km/h
to only 15-20 km/h (Gakenheimer, 2002). In Delhi, the average vehicular speed fell from 20-
27 km/h in 1997 to only 15 km/h in 2002 (Times of India, 2002). Moreover the periods of
congestion in Delhi now last 5h: from 8.30 to 10.30 in the morning and from 4.30 to 7.30 in
the evening in Chennai, 10 to 15 km/h overall but falls to only 7 km/h in the centre (Times of
India, 2003), (World Bank, 2002).
59
The cause of congestion is the rapid increase in travel demand. According to the World
Bank, 2002, the average annual rate of growth of travel demand has been 2.2% in Kolkata,
4.6% in Mumbai, 9.5% in Delhi and 6.9% in Chennai.
Congestion also happens if the total road space increases smaller than the raising of
total vehicles. As a result of this, usable road density decreases. Here we have tried to show
the actual congestion picture of urban West Bengal.
There are some discernable features of congestion. First there is a high degree of road
congestion in the city of Kolkata. The figure is substantially higher than that of the other urban
centers in the state of particular maintained can be made of the district of Howrah. 24 Pgs (N),
Nadia and Burdwan that have congestion rate lower than that of Kolkata but higher than the
most of the other districts. However, in north, Darjeeling shows a surprising high congestion
rate as compared to the other surrounding districts. This may be probably due to the high
tourist concentration in the district. In urban Jalpaigunri and urban Malda also, the congestion
is high.
5. 5 Vehicular Pollution:
Air pollution is the serious problem in Indian cities as well as West Bengal towns. As
shown in the following diagrams. The levels of air pollution concentrations are the
compositions of suspended particulate matters (SPM), respiratory particulates matters (RPM),
NOx and SOx according to the World Health of Organisation (WHO). CO and Lead pollutants
60
have dramatically fallen over the past decade from 1995 to 2000 (Ministry of Petrolium and
Natural Gas, 2002). In 1997 Tata energy Research Institute (TERI) found an important source
of air pollution remains the large and mostly old fleet of two-wheelers (motorcyclecles and
scooters) and three-wheelers (auto-rickshaws) with highly inefficient, poorly maintained, very
polluting 2-stroke engines. The Indian Government has already notified to reduce particulate
pollution by mandating conversion of all buses, auto-rickshaws, and taxies in Delhi to CNG
fuel by January 2001 (Urban Transport Crisis, 2005).
Fig: 5.5.1 Air Pollution Levels in the largest Indian Cities
Air Pollution levels in the largestIndian cities
050
100150200250300350400450
Mumba
i
Kolkata Delh
i
Chenn
ai
Banga
lore
Hydera
bad
City
Con
cent
ratio
n in
pcm
(mic
rora
ms) SPM
RPMSO2NO2
(Source: Ministry of Petroleum and Natural Gas, 2002)
61
The all India scenario clearly shows that the three metros Mumbai, Kolkata and
Delhi are the culprits for emitting SPM and RPM. Among them Kolkata is the largest
emitter of SPM while Delhi of RPM. For other metros the emission rate is not very
significant. However, the emission of SO2 and NO2 are almost similar across the metros.
Fig: 5.5.2 Air Pollution Levels in the Towns of West Bengal
Air Pollution Levels in the largesttowns of West Bengal
050
100150200250300350400
Haldia
Kolkata
Howrah
Durgap
ur
Asans
ol
Towns
Con
cent
ratio
n in
pcm
(mic
rogr
ams) SPM
RPMSO2NO2
(Source: Ambient air quality report of W.B, 2006)
Now coming to the West Bengal figure Kolkata remains the almost sole culprit in
emitting all these positions gasses (SPM, RPM, SO2 and NO2). So Kolkata , the capital city
of West Bengal is in the dangerous situation. Other urban Conglomerates are almost
62
insignificant in this regard. This is in spite of the fact that appears in the million plus city in
2001 census.
5.6 Vehicular Ownership Pattern:
Another important aspect of urban transport system is the possession of vehicles. To
facilitate our analysis we distinguish between two types of vehicle – fuel consuming or
polluting (Scooter, motorcycle, moped, car, jeep and van etc.) and non-polluting vehicles
(bicycle).
We first consider the distribution of polluting and non-polluting vehicles across the
states of India. It reveals a unique region divide. Eastern region lags for behind the others
zone in the distribution of polluting vehicles. The states of the northern region are
dominating. As per the distribution of non-polluting vehicles, the eastern region is well-
endow. The gini-coefficient shows that there is a higher inequality in the interstate
distribution of polluting vehicles as compare to the non-polluting vehicles. Now the state-
wise distribution of percentage of vehicular ownership per household is given in the
following table.
63
Table: 5.6.1 State-wise distribution of percentage of vehicular ownership per household
Region State
%Ownership ofpolluting
vehicles/household
%Ownership ofnon-polluting
vehicles/householdNorth
Chandigarh 58.62 68.34Delhi 40.99 37.58
Haryana 23.24 50.05Himachal 10.03 9.10
J & K 10.93 12.78Punjab 37.40 71.76
Rajasthan 15.62 36.23Uttar Pradesh 12.62 69.48Uttaranchal 14.61 30.88
EastAssam 7.26 46.39Bihar 4.55 40.64
Jharkhand 10.88 50.32Orissa 8.94 51.96
West Bengal 6.92 52.59West
Chhattisgarh 12.17 59.83Goa 2.94 4.38
Gujrat 24.53 37.31Madhya Pradesh 13.86 42.80
Maharastra 16.56 30.07South
Andhra Pradesh 11.30 32.83Karnataka 17.51 30.14
Kerala 13.80 18.69Tamilnadu 18.28 42.44
North East 7.55 20.48Group of Union
territories 23.64 39.78Value of Gini Co-
efficient 0.35 0.25
(Source: Census data, 2001)
Next we consider per household distribution of polluting vehicles across the Indian
state in the table 5.6.2. It is clearly evident that for almost all the states the percentage of
households having polluting vehicles is much higher in the urban areas than the rural areas.
64
In some of the states the difference is almost five times or more. The lack of proper
roads, inadequacy of fuel fill-up stations etc might be a contributory factor to these
distortions.
Table: 5.6.2 State-Wise distribution of ownership per Household polluting vehicles all over India
Region State
Ownership ofpolluting
vehicles/household(rural)
Ownership of urbanpolluting
vehicles/household(urban)
NorthChandigarh 23.33 62.79
Delhi 27.99 41.91Haryana 15.68 41.93Himachal 7.96 25.88
J & K 5.32 27.64Punjab 30.69 49.89
Rajasthan 8.94 37.5Uttar Pradesh 8.17 30.32Uttaranchal 7.41 40.3
EastAssam 4.47 23.7Bihar 3.10 18.43
Jharkhand 4.55 3.49Orissa 5.14 32.64
West Bengal 4.27 13.43West
Chhattisgarh 6.44 36.41Goa 40.03 58.78
Gujrat 13.66 41.57Madhya Pradesh 6.89 34.13
Maharastra 9.61 26.02South
Andhra Pradesh 5.66 28.4Karnataka 8.65 34.14
Kerala 10.25 25.14Tamilnadu 11.81 27.35
North East 8.74 18.74Group of Union
territories 8.36 37.79Value of Gini Co-
efficient0.38 0.22
(Source: Census data, 2001)
65
Coming now to the inter-zonal disparity, the north-zone seems to have a very high
percentage of households in the ownership category. On the other hand the figure is very low
for the eastern-region. The situation is medium in the western states. The picture is more or
less same for the urban households.
The picture for non-polluting ownership is completely reverse. Here, the rural areas
outperform the urban areas for almost all the states. Zone-wise, the eastern region is
dominating, closely followed by southern and western states. In all it is seen that the
percentage of polluting vehicles ownership is high in the urban areas and low in the rural
areas. This points out to the stark reality of the urban centers. The state-Wise distribution of
ownership per household non- polluting vehicles all over India is shown in table 5.6.3 in the
next page.
66
Table: 5.6.3 State-Wise distribution of ownership per household non- polluting vehicles all overIndia
Region State
Ownership of ruralnon-polluting
vehicles/household(rural)
Ownership of urbannon- polluting
vehicles/household(urban)
NorthChandigarh 76.67 37.21
Delhi 72.01 58.09Haryana 84.32 58.07Himachal 92.04 74.12
J & K 94.68 72.36Punjab 69.31 50.11
Rajasthan 91.06 62.5Uttar Pradesh 91.83 69.68Uttaranchal 92.59 59.7
EastAssam 95.53 76.3Bihar 96.9 81.57
Jharkhand 95.45 96.51Orissa 94.86 67.36
West Bengal 95.73 86.57West
Chhattisgarh 93.56 63.59Goa 59.97 41.22
Gujrat 86.34 58.43Madhya Pradesh 93.11 65.87
Maharastra 90.39 73.98South
Andhra Pradesh 94.34 71.6Karnataka 91.35 65.86
Kerala 89.75 74.86Tamilnadu 88.19 72.65
North East 91.26 81.26Group of Union
territories 91.64 62.21Value of Gini Co-
efficient0.05 0.11
(Source: Census data, 2001)
As per the question of equity is concerned, the ownership pattern is highly
skewed as the census data reveals. As for the distribution of polluting (fuel consuming)
67
vehicles the gini coefficient is 0.38 in rural and 0.22 in urban. However, for the non
polluting vehicles the corresponding figures are 0.05 (rural) and 0.11 (urban). Thus we
see that there is a greater equality in the distribution of non – polluting vehicles, at least
in rural India. So far as the polluting vehicle is concerned, its distribution is highly
skewed.
5.7 Public Transport: some issues
The inadequacy of private transport pressurizes the public transport mechanism. Due
to the faulty urban transport policies, “the cost of travel” especially for the poor has been
increasing considerably. The Indian cities vary considerably in terms of the population, area,
urban form, topography and economic activities. Thus the requirement of public transport
mechanism varies from one city to another. In the metropolises, over crowded public buses
hopelessly stuck on congestive road ways with an average speed of 6-10 km/h
(Gakenheimer and Zegras, 2003). However even this meager public transport system are
beyond affordability of most the Indian poors. Due to the recent phenomenon inflation and
rise in the cost of living there is an enormous pressure on the poor man’s budget. Kolkata
had India’s only underground metro system 16.5 km while Delhi is constructing 62.5 km
more extensive metro. In contrast, Chennai has an hybrid system of both surface and
elevated metro extending up to 19.8 km suggested (Puchar, Korattyswaropam, & Mittal,
2005). Also, Kolkata has the India’s only reaming tram system (6.8 km double track
network) though old and seriously deteriorating tax and vehicles.
68
From the cost side, Kolkata imposes least on its passengers – covering 42% of the
revenue through passenger’s fares. This may be one of the reason why public transport
system (Marwah, Sibal and Sawant, 2001) is popular among the urban poor. The use of
public transport varies across the cities. It is highest in Kolkata (80% of the trips) followed
by Mumbai (60%), Chennai and Delhi (42%). It is even lower in all other cities.
Debates emanate on the relative efficiency of public and private transport systems. It
is argued that the private operators carry a much larger intake of passengers per bus per day,
earning more revenue and requiring less staff than the public system. Thus it enjoys a higher
profitability and lower cost. On the contrary, the employees in the private secure face lower
wages, less job securities and less job securities and less fringe benefits viz., pension and
Health Insurance (Pucher, Korattyswaroopam & Ittyerah, 2004).
However, the scarcity of public funds and the inefficiency of the public system lend
its support to the issue of privatization. Equally important is the use of new technologies and
replacement of overuse buses, so as to reduce air pollution to a considerable extent.
69
5.8 Conclusion:
Our urban centers show increasing demand for transport. There is however a number
of problems associated with urban transportation. In this paper we hope to deal with some of
these problems. These problems are multifaceted and often entangled with one another. Here
we propose to cover some of the issues with the respect of both India as well as West
Bengal. The inequalities in private urban transit systems are documented. Though there is
arise in the total number of vehicles, there is a sizable portion of the urban families who
have no vehicles. The problem is complicated by the lack of adequate road space and
congestion that prevents the use of bicycles. Severe problems of air pollution are also noted.
Kolkata has one of the cheapest public transport systems among the Indian metropolis.
However even this ‘cheap’ transit system is beyond the reach of the very poor in this city.
Further more, the system is also inefficient with waste of resources. An urban transport
planner has a tough task. He has to balance between equity, efficiency and sustainability of
transit system. This requires long run planning by the urban planners.
Having delt in details about the relationship between urbanization, vehicular
population and pollution in chapter 4 and 5. Now turn our attention to Kolkata – the site for
our study. The next three chapters deal with these problems. In chapter ― 6 we introduce
some aspects of vehicular emission as revealed from the macro data. Chapter ― 7 considers
the seasonal fluctuation of emission. In chapter ― 8 is a macro study of the effect of
emission on the traffic police personnel in Kolkata.
70
Chapter – 6
Analysis of Vehicular Emission in Kolkata
6.1 Preliminary View
6.1.1 Introduction:
Air pollution has been promoted by developments that typically occur as countries
become industrialized: growing cities, increasing traffic, rapid economic development and
industrialization, and higher levels of energy consumption. The high density of population to
urban areas, increase in consumption patterns and industrial development have led to the
problem of air pollution. Recently, in India, air pollution is widespread in urban areas where
vehicles are the major contributors and in a few other areas with a high concentration of
industries and thermal power plants. Vehicular emissions are of particular concern since
these are ground level sources and thus have the maximum impact on the general
population. Also, vehicles contribute significantly to the total air pollution load in many
urban areas.
The direct impact of a growth in various causal factors/pressures is the increase in
the emission loads of various pollutants, which has led to deterioration in the air quality. In
India, there is no systematic time series data available related to air pollutant emission loads
71
and trends. The availability of emission factors for Indian conditions in another issue that
has not been given due attention so far.
TERI (1998) provides some broad estimates of the increase in pollution load from
various sectors in India. The total estimated pollution load from the transport sector
increased from 0.15 million tones in 1947 to 10.3 million tones in 1997.
6.1.2 Trend in vehicular emission:
The terrific increase in number of vehicles has also resulted in a significant increase
in the emission load of various pollutants. The quantum of vehicular pollutants emitted is
highest in Delhi followed by Mumbai, Bangalore, Calcutta and Ahmedabad. Here we have
wanted to focus the vehicular pollution in Kolkata.
Apart from the concentration of vehicles in urban areas, other reasons for increasing
vehicular pollution are the types of engines used, age of vehicles, congested traffic, poor
road conditions, and outdated automotive technologies and traffic management systems.
Vehicles are a major source of pollutants in metropolitan cities.
Under the National Ambient Air Quality Monitoring (NAAQM) network, three
criteria air pollutants, namely, SPM, SO2, and NO2 have been identified for regular
monitoring at all the 290 stations spread across the country. By the WBPCB report we have
shown the trends of four air pollutants mainly SPM, RPM, SO2 and NO2.
72
The emission level has shown downward trend in recent years that is shown in the following
table.
Table 6.1.2.1: Trend of Vehicular Emission
Year SPM RPM SO2 NO22001 2178.07 2178.07 105.29 764.572002 1954.79 1006.14 64.97 749.502003 2502.36 1186.79 65.07 673.002004 2855.73 1005.76 105.68 678.012005 2706.11 1297.81 102.65 936.542006 2569.85 1253.76 92.36 764.022007 2221.64 1044.36 66.57 716.17
(Source: WBPCB Report)
Next we have shown the trend of major vehicular pollution in Kolkata by the following
diagram.
Figure: 6.1.2.2 Trend of Major Vehicular Pollutants in Kolkata
Trend of Major Vehicular Pollutantsin Kolkata
0.00500.00
1000.001500.002000.002500.003000.00
20012002
20032004
20052006
2007
Year
Po
lluta
nts
(Mic
rog
ram
s)
SPMRPMSO2NO2
(Source: WBPCB Report)
73
From the above figure we can clearly understand level of SPM & RPM are very
high. In recent years level of various kinds of pollutants are downward because the major
intervention came in the form of substituting fuel in vehicles, building of flyovers to check
road congestion, etc. However the level is still high.
6.1.3 Seasonal Fluctuation in Vehicular Emission:
The area of study is mainly in Kolkata, the eastern Gateway of India, the capital city
of West Bengal, and one of the most populous cities in the country, is a centre of commerce,
trade and industry in east and north east region. The extent of the city is longitudinal,
running from north to south. The geographical area of the city of Kolkata had undergone
wide changes in the last three centuries.
The government has taken a number of vehicular emission control measures, pollution
prevention technologies; action plan for problem areas, development of environmental
awareness. Yet despite all these measures, vehicular pollution still remains one of the major
environmental problems. At the same time there have been success stories as well such the
reduction of ambient lead levels (due to introduction of unleaded petrol) and comparatively
lower SO2 level (due to progressive reduction of sulphar content in fuel).
74
6.1.3.1 Variation in SPM:
Suspended particulate matter is one of the most critical air pollutants in most of the
urban areas in the country and permissible standards are frequently violated several
monitored locations.
Figure: 6.1.3.1(a) Season-wise SPM Level in Kolkata
Season wise SPM Level in Kolkata
0200400600800
10001200
2001
2002
2003
2004
2005
2006
2007
Year
SP
M (M
icro
gram
s)
WinterMonsoonSummerFestival
(Source: WBPCB Report)
The above diagram is certain pattern in the SPM level in Kolkata. First on average
there has been a dramatic draw in the SPM level in the recent year (2007). This is a
significant drop, given the near static picture over four years 2003, 2004, 2005 and 2006.
These drops may be a result of the introduction of fuel efficient vehicles with improved
technology. Some stringency in the government effort, given the pressure of international
norms may be omnipotent. Also there is a wide season – wise variation in the SPM level. It
is high in the winter season and low in the monsoon. The festive season shows a significant
75
disposal of SPM into the city’s atmosphere. The social factors play an important role in the
environmental degradation along with the natural factors. Thus environmental problem
causes to be mainly a natural phenomenon, at least so far as the air pollution is concerned.
6.1.3.2 Variation in RPM:
WBPCB report reveals that Respiratory particulate matter is also much higher than
the national standard in residential areas. The situation worsens during winter month that is
shown in next figure.
Figure: 6.1.3.2(b) Season-wise RPM Level in Kolkta
Season wise RPM Level in Kolkata
0100200300400500600700
2001
2002
2003
2004
2005
2006
2007
Year
RPM
(Mic
rogr
ams)
WinterMonsoonSummerFestival
(Source: WBPCB Report)
76
A similar season wise is also observed in the emission of RPM. The social factors
could not be ignored in any meaningful analysis of urban air pollution. RPM does not fall
much as SPM in 2007. But the level of RPM has comparatively fallen in 2007 than the
previous years.
6.1.3.3 Variation in SO2:
In comparison to SPM, RPM, and NO2, SO2 is low. And it is also low enough to
have any significant health effect. In case of SO2 the seasonal variation is also pronounced.
Again the social phenomenon of festivity plays a dominant role.
Figure: 6.1.3.3(c) Season-wise SO2 Level in Kolkta
Season wise SO2 Level in Kolkata
0100200300400500600700
2001
2002
2003
2004
2005
2006
2007
Year
SOx
(Mic
rogr
ams)
WinterMonsoonSummerFestival
(Source: WBPCB Report)
From the above figure we see it is low in 2007 than the previous years.
77
6.1.3.4 Variation in NO2:
This is also another contributor element of vehicular pollution.
Figure: 6.1.3.4(d) Season-wise NO2 Level in Kolkta
Season wise NO2 Level in Kolkata
050
100150200250300350
2001
2002
2003
2004
2005
2006
2007
Year
NO
x (M
icro
gram
s)
WinterMonsoonSummerFestival
(Source: WBPCB Report)
It also reveals a wide seasonal fluctuation and the role of the local festivity. In case
of NO2 we can not see a lower position of NO2 level in 2007. This was more or less over the
years. We realize that levels of pollutants were high in 2001.Then over the years these have
decreased. This is because of the various measures taken by government to mitigate
emissions from transport sector.
78
6.1.4 Status of other Vehicular Pollutants:
There are some other air pollutants viz lead (Pb) and Carbon monoxide (CO). The
salient results of these additional parameters at some stations in the metropolitan city of
Kolkata in the respective years. The lead and carbon monoxide levels at most locations were
much higher than the prescribed permissible limit.This is because of high traffic density and
large number of motor vehicles operating on the roads.
6.1.5 Seasonal Fluctuation of Air Pollution level:
The methodological condition and turbulence in the atmosphere are the primary
factors affecting pollutant distribution and dispersion pattern and producing seasonal
variations. There are wide fluctuations in seasonal conditions within the country as the
seasonal conditions within the country as the seasons are not uniform throughout the country
due to diversity in physical and climatic conditions.
During monsoon (June to August), frequent rains wash down the air born particulates
and other gaseous pollutants. Therefore, the period between June to mid September is the
cleanest period in the year and frequent rain does not allow pollutants to build up to higher
concentration in ambient air though the pollution generating sources remain the same
throughout. The winter months of December to February are relatively much calm
conditions facilitate more stability to atmosphere and consequently slow dispersion of
pollutants generated and help in build up of pollutants generated and help in build up of
79
pollutants in vicinity of pollution sources. The general pollutant levels in terms of
percentage violation of standards increase considerably during winter basically due to lower
ambient temperature, calm conditions, lower mixing depth, pollution inversion and high
traffic density on the roads. Frequent change in wind direction in the atmosphere during
March to May months create turbulent conditions. Local disturbances in environment causes
frequent dust storm and hazy condition. Moreover, the winds from Thar desert area brings
dusty winds from arid and semi arid region, building up high particulate matter levels in
ambient air in these months, mostly contributing soil borne particles. In the festival season
there is huge conglomeration of people in Kolkata. Festive-shoppings, Pandel-hoppings and
increase business activities are some of the factors that increase the flux of urban traffic and
increase trips per vehicle. Since there is no concomitant rise in the road space, the huge flow
of traffic leads to congestion of road space, low vehicle speed and accident proneness-all
adds to the plight of urban people. It is this up search in social activity centering the festivity
that adds to the vehicular pollution.
Thus we find that Kolkata’s air quality due to vehicular emission is conditioned both
by natural and social factors. The festive season by itself is largely responsible for falling air
quality during their time when the natural factors may not be so omnipotent. Thus the
analyzing pollution pattern, Policy makers have to give a due consideration to this point.
At last we have shown the categorization of air quality on the basis of seasonal trend
that is represented by the following table.
80
Table 6.1.5.1: Seasonal Trend Based categorization of air Quality
(Source: WBPCB Report)
6.1.6 Conclusion:
The chapter is tentative. It brings out the important aspect of air pollution. For the
each of study, we have segregated the entire time period into four seasons-winter, summer,
monsoon and festival. These are of them are quite concern in the environmental literature.
The fourth season is tropical of Kolkata – arising mainly due to social causes. The data
reveals that this so called new season ranks very high in the disposal of environmental
waste. Thus an interesting suggestion of the paper is the role of environmental fallen in
increasing pollution and climate hazard. The point is clearly brought out by our analysis.
Category Period Critical Air Pollutants Feature
Moderate
pollution
March to
May
Particulate Matter
Low humidity, high turbulence,
frequent change in wind speed
Low
pollution
June to
August --------
Cleaner period due to high
humidity, rains and monsoon month
High
pollution
December to
February Particulate Matter
Low inversion, calm conditions,
unfavorable meteorological conditions
High-
medium
pollution
September to
November Particulate Matter
High conglomeration of people, high
emission, low vehicle speed
81
6.2 Spectral Analysis:
6.2.1 Introduction:
Rapid economic development of Kolkata has resulted in the rise of vehicles. The
percent growth of vehicles is higher than the growth of population. There is high vehicle
ownership, inefficient public transport, mixed on roads and the widely used adulterated fuel
has lead to the traffic congestion on roads and in turn the vehicular pollution. The West
Bengal pollution control board monitors ambient air quality of major urban centers
regularly. Air quality at respirable height is also monitored in some important traffic in the
city of Kolkata. In case of vehicular emission suspended particulate matter, respiratory
particulate matter, sulphur dioxide and nitrogen dioxide are measured in some major
sections in there. In studied years we see wide fluctuations in across months. It is low in the
monsoon season and high in the winter period.
The standard time domain analysis may fail to catch these seasonal fluctuations.
However cyclical pattern is an essential feature of environment parameters. We have divided
our chapter into 5 sections. First is introduction. Literature review is pointed in second then
methodology in third chapter. We have designed result of study in chapter 4 and it is
concluded in last chapter.
82
6.2.2 An overview on spectral analysis:
For a single time series, spectral analysis decomposes the overall variance of the
observations into components at different frequencies, producing what is called a spectral
density function. The estimated spectral density function of the series has large peak at
frequencies of one cycle and two cycles per year.
There are general approaches to time series analysis first, analysis in the domain
where we have auto and cross relation analysis regression, and the fitting of time domain
model at described by ‘Box and Genkins’, time series analysis, forecasting and control,
(Holder-Day, 1970). Secondly analysis in the frequency domain where we have auto and
cross spectral analysis. In most cases there is no special peak but rather power is
concentrated at these low frequencies and the amplitude of these long term component
decrease smoothly with decreasing period, (Granger; 1967, Farely and Hitch; 1969)
Spectral method proposed by (Hannan) will always be asymptotically efficient; they
are in frequently used because these components demanding at every large sample are
presumably required (Engel and Gardener; 1974).
Now coming to the angle of cross spectral analysis, it is a technique for examining
the relationship between two time series at various frequencies. The technique may be used
for time series while “arise in a similar footing” (Kins and watts; 1968)
83
Cross spectral analysis is recognized as a powerful analytical tool for the analysis of
time varying behaviour in a number of disciplines. (Barksdle, Hilliard and Guffy; 1974)
suggest that to explain ‘cross spectral analysis and illustrate the use of the new technique
studying the interaction between advertising and sales
The theory of spectral analysis is based on notion that a time-series is a sample
record of a stationary, stochastic process and the assumption that these measurements can be
used to estimate the characteristic of the process generating the data. The estimated idea of
spectral analysis is that a time series may be decomposed by component each of which
associate with frequency.
6.2.3 Data Used:
The data are purely collected from West Bengal Pollution Control Board. WBPCB
collected SPM, RPM, SO2 and NO2 from the some important locations in Kolkata at the
peak time when huge vehicles moved throughout the area. The particular time is at 8 a.m. to
11 a.m. and 4 p.m. to 7 p.m. At these times a lot of passenger cars, Buses, two-wheelers
moves extensively. The Board collected data on lead (Pb) and carbon-monxide (CO)
emission also. But these data are scattered, inadequate and a little and not covered for all
other important locations in recent years. These studied locations continuously face traffic
congestion. Actually the transport system in Kolkata is unique and consists of buses, local
trains, metro-rail, trams, taxies, and auto-rickshaws, coupled with slow moving traffic like
rickshaws, pulled by human beings, bi-cycles and walking. It is also interesting that despite
84
this varied mode of transport, the entire system is under severe strain due to congestion. Due
to such congestion the vehicles emit a large quantity of smoke. WBPCB monitors the
ambient air quality of major sections in Kolkata every day in every year. These vehicular
pollution data are undertaken from 14 major locations during the year from 2001 to 2007.
From these data we primarily observe a season wise clear trend in every year.
In this chapter, first we have shown the monthly fluctuation of the different
pollutants and then focused seasonal trend in the monthly level of pollution created by
different pollutants in different areas of Kolkata. This statistical tool analyses the deviation
of the series as a whole into periodic components of different frequencies and periods.
Smooth series have stronger periodic components at low frequencies. In our study we
calculate periodogram value and spectral density for individual frequency component for
univariate data as well as bivariate data. We also calculate cross spectral density, cross
periodogram, coherency and phase spectrum. All these results for univariate and bivariate
data are plotted under each frequency.
6.2.5 Result of the Study:
In order to understand whether there is any fluctuation in the different pollutants we
first plot the monthly distribution of pollutant level for different years (2001-2008). As a
sample, we produce the picture on monthly distribution of SPM in 2001. Since all figures
are similar, we relegate the similar figures for other pollutants and for other years in our
appendices.
85
Our analysis clearly shows that there is a discernable pattern of fluctuations in the
pollutant levels. For example the monthly level of SPM in 2001 reaches a very high level
between December to February with march showing a declining trend. The level
continuously drops up to August. Thence forth it again rises towards the pick level. Thus the
maximum pollution occurs in winter season while it is low in the monsoon season. More or
less similar picture are seen for all the pollutants considered by us. We now see whether
spectral analysis helps us to unravel this cyclical fluctuation.
Figure: 6.2.5.a Monthly Fluctuation of SPM 2001
(Source: WBPCB Report)
In our study the method of spectral analysis is applied to analyse the seasonal trend
of the pollutants in different traffic points of Kolkata. The spectral plots procedure is used to
identify periodic performance in time series. Instead of analyzing the variation from one
time point to the next, it analyses the deviation of the series as a whole into periodic
components of different frequencies. Smooth series have stronger periodic components at
low frequencies.
Monthly Fluctuation of SPM 2001
0100200300400500600
Apr MayJu
ne July
Aug Sep Oct Nov Dec Jan
Feb Mar
Month
Pollu
tion
86
The sine and cosine transforms periodogram value and spectral density estimate for
each frequency or period component. When bivariate analysis is selected then cross-
periodogram, squared coherency and phase spectrum are estimated for each frequency or
period component. For univariate and bivariate analysis we plot periodogram and spectral
density for each frequency and period and we also plot squared coherency and, phase
spectrum and gain for each frequency and period for every paired series for bivariate
analysis.
We start by looking month-wise data of pollutants in Kolkata. The data show the
obvious trend, since the periodogram1 represents a sequence of peaks with the lowest
frequency.
Figure: 6.2.5.b Spectral Periodogram of SPM 2001
Spectral Periodogram of SPM 2001
0.000000
20000.000000
40000.000000
60000.000000
1 2 3 4 5 6
Fr e que nc y
(Source: WBPCB Report)
1 A very few of the spectral diagrams are shown in the text. The others are relegated to the appendix because ofthe paucity of space.
87
Figure: 6.2.5.c Spectral Density of SPM 2001
Spectral Density of SPM 2001
0.000000
10000.000000
20000.000000
30000.000000
40000.000000
1 2 3 4 5 6
Fr e que nc y
(Source: WBPCB Report)
The plot of the periodogram shows a sequence of peaks that stand out from the
background noise (random), with the lowest frequency peak at a frequency 0.0833 and also
at the period 12.
Spectrum analysis will also identify the correlation of sine and cosine functions of
different frequency with the observed data. If a large correlation (sine or cosine coefficient)
is identified, one can conclude that there is a strong periodicity of the respective frequency
(or period) in the data. The periodogram values can be interpreted in terms of variance (sum
of squares) of the data at the respective frequency or period.
The largest values in the periodogram column of all series occur at a frequency of
0.0833 implies the strong periodicities in data, precisely what you expect to find if there is
an annual periodic component. Therefore, this information confirms the identification of the
lowest frequency peak with an annual periodic component.
88
But the other peaks at higher frequencies are best to analyse with the spectral density
function, which is simply a smoothed version of the periodogram. The spectral density
consists a distinct peak that appears not to be equally spaced but the lowest frequency peak
simply represents the smoothed version of the peak at 0.0833. Smoothing provides a means
of eliminating the background noise from a periodogram, allowing the underlying structure
to be more clearly isolated. To understand the significance of the higher frequency peaks,
remember that the periodogram is calculated by modeling the time series as the some of
cosine and sine functions. Periodic components that have the shape of a sine or cosine
function (sinusoidal) show up in the periodogram of each series as single peaks, with the
lowest frequency peak in the series occurring at the frequency of the periodic component.
Hence we have now accounted for all of the discernible structure in the spectral density plot
and conclude that the data contain a single periodic component of 12 months. Using the
spectral plots procedure you have confirmed the existence of an annual periodic component
of a time series and you have verified that no other significant periodicities are present. The
spectral density was seen to be more useful that the periodogram for uncovering the
underlying structure because the spectral density smoothes out the fluctuations that are
caused by the non-periodic component of the data.
Now to uncover the correlations between two series at different frequencies the
cross-spectrum analysis is indispensable. For two covariance-stationary series of equal
length and comparable intervals, it is possible to estimate the cross power spectrum, or
simply the cross-spectrum, and compute certain spectral statistics which provide information
on the relationships between the frequency components of the two series. Just as the auto
89
spectrum of a single series is the Fourier transform of the auto-covariance, the cross
spectrum is the Fourier transform of the cross covariance. The cross-spectrum is a complex
quantity composed of a real (in-phase) element called the co-spectrum. Under this analysis,
cross-spectrum measures the strength of association between the components of two series
at each frequency by revolving the crossed series. Under this bi-variate analysis we obtain
the real parts of cross peridogram, co-spectral, cross amplitude, squared coherency, phase
spectrum, and gain for each frequency.
Figure: 6.2.5.d Cross Periodogram of SPM 2001 Figure: 6.2.5.f Cross Density of SPM 2001
Cross Periodogram of SPM 2001
-10000.000000
0.000000
10000.000000
20000.000000
30000.000000
40000.000000
50000.000000
1 2 3 4 5 6
Fr equency
Cross Density of SPM 2001
0.000000
10000.000000
20000.000000
30000.000000
1 2 3 4 5 6
Frequency
(Source: WBPCB Report) (Source: WBPCB Report)
Figure: 6.2.5.g Cross Amplitude of SPM 2001
Cross Amplitude of SPM 2001
0.000000
10000.000000
20000.000000
30000.000000
1 2 3 4 5 6
Fr e que nc y
(Source: WBPCB Report)
90
However, as shown above the series were created so that they would contain two
strong correlated periodicities. Here the cross-spectrum consists of real numbers. These can
be smoothed to obtain the cross-density estimates. Looking at the results of cross-
periodogram, we see the strong periodicity at frequency 0.0833. Likewise we get the similar
results of cross-density. The square root of the sum of the squared cross-density and quad-
density values is called the cross-amplitude. The cross amplitude can be interpreted as a
measure of covariance between the respective frequency components in the two series.
One can standardize the cross-amplitude values by squaring them and dividing by
the product of the spectrum density estimates for each series. The result is called the squared
coherency, which can be interpreted similar to the squared correlation coefficient that is the
coherency value is the squared correlation between the cyclical components in the series at
the respective frequencies.
Phase describes the angular shift if the crossed series relative to the base series. The
phase estimates may be used to define the lead or lag relationship between the two
processes. In our study the positive value of phase implies the angular shift of the crossed
series relative to the base series illuminates that the base series leads the crossed series.
Conversely the negative value of phase implies that the base series lags the crossed series.
Another statistical value termed as gain may be described as the ratio between the
amplitude of values in the crossed series and the amplitude of values in the base series. The
91
gain value is computed by dividing the cross-amplitude value by the spectrum density
estimates for one of the two series in the analysis.
Figure: 6.2.5.h Spectral Coherency of SPM 2001
Squared Coherncy of SPM 2001
0.000000
0.200000
0.400000
0.600000
0.800000
1.000000
1.200000
1 2 3 4 5 6
Fr e que nc y
(Source: WBPCB Report)
Figure: 6.2.5.i Phase Spectrum of SPM 2001
Phase Spectrum of SPM 2001
-1.000000
-0.500000
0.000000
0.500000
1.000000
1 2 3 4 5 6
Fr e que nc y
(Source: WBPCB Report)
Figure: 6.2.5.j
Gain of SPM 2001
0.000000
0.500000
1.000000
1.500000
1 2 3 4 5 6
Fr e que nc y
(Source: WBPCB Report)
92
6.2.6 Conclusion:
The main purpose of this study is to demonstrate the seasonal trends (Periodicity) of
the different pollutants in Kolkata. This new technique is justified since the above discussion
exhibits the significant results of spectral estimates, which really represents the seasonal
periodicities of the pollutants under equal time intervals. We also understand correlation
between two series at different frequencies, using cross-spectral analysis. The strong
correlated periodicities among different series of pollutants under different years can be
analysed by the value of coherence, phase and also gain value etc. These totally indicate that
the model has a great implication of predicting long run trends.
Having briefly survey the macro aspects of vehicular emission, it is now our turn to
peep into the macro features. Since it is the people who have to live and thrive in the
polluted environment with high risks of health hazard, their position is crucial to tackle the
entire problem.
93
Chapter ― 7
Effects of Vehicular Emission ― A Study of traffic police in Kolkata
7.1 Introduction
Environmental economists are concerned about the effects of environmental
degradation on the life of people who may be adversely affected by them (Agarwal; 1996,
and Ghosh; 1998). In the tradition of ecological economics various methods (Contingent
valuation etc.) have been developed to address the economic cost that is levied on the
affected people.
However, these effects may be of two origins - place of residence and place of work.
An individual or a family may be exposed to toxic environment in their very place of living
and similarly an individual may face toxicity in his/her workplace environment (Sinha;
1993, Faiz et.al; 1996, Chakraborty; 1997, WHO-UNDP Report; 1988, and Bhattacharryya
& Banerjee; 2002). Since, workplace occupy a significant amount of the individual daily
schedule, the threats cannot be ignored.
Neo-classical economists argue that rational agents self-select themselves into jobs
according to their risk proneness. Numerous studies have approached the problems of job
hazard pay and worker risk; (Viscusi; 1978, Buhai and Cottoni; 2011,, Rosen; 1974, 1986,
94
Hersch and Viscusi; 1990, 2001). However in many situations, a person’s choice of the
appropriate job (in parlance with the risk and return) is severely restricted2. This is
particularly true in the developing economies with scanty knowledge of environmental
hazard, asymmetry in information and weak enforcement mechanism. In such cases workers
often observe a decline in their working capacity. It is the worker’s interest then to prevent
(or at least reduce) the negative impact on working capacity.
We see whether better consciousness about the environmental hazard of their jobs
can shield a person against the odds or at least mitigate the riskiness. This type of awareness
building would then seem very essential for the worker’s welfare.
In the current chapter we deal with the empirical study of the traffic police personnel
in Kolkata, India. According to the existing rules, any incumbent policemen under the
Kolkata police has to perform the job of traffic police for a certain period of their entire
tenure. However there are significant differences among the traffic polices regarding the
environmental health hazard regarding their jobs. Our aim is to understand whether such
awareness is any way helpful towards maintaining their working capacity. We also try to
find out regarding their assessment of jobs using a number of parameters.
Our chapter is divided into five sections. In section 2 the rationale of our study is
provided. Section 3 provides a theoretical underpinning of our chapter. In section 4 we
2 Long years ago in 1776, Adam Smith opined that hazardous jobs are associated with higher payment. Heconstructed the theory of compensating differentials in wage payments.
95
discuss the data collection procedure with a short description of data. Section 5 provides our
empirical findings. The conclusion is given in the final section.
7.2 Rationale of the study:
In order to assess the impact of continuous exposure to motor vehicle emissions on
human beings, a pioneering word is done by Sinha (1993) on a group of 100 traffic police
constables in Jaipur, India. These constables were posted at the road intersections with
considerable amount of exposure to deadly gasses (CO, HC & SO2) for about six hours or
more daily. He found that a majority of the constables suffer from some short of physical
disorder respiratory difficulties and digestive problems. There was a high incidence of
tubercolerosis among the younger constables (20-30 years of age). Coming to Kolkata, the
condition is not at all better. There are the traffic police personnel who are continuously
monitoring traffic regulations at the busy intersections in city. They are always at greater
risk to the exposure of air pollution. These personnel comprising nearly 130 male traffic
police constables aged from 25 to 55 years serving at busy roads/intersections within KMA
for a period of 3 moths to 16 years. Many of them complained of frequent headache,
irritation of eyes, disturb sleeps, increased cough and respiratory problem and most of them
have indigestion problem (Basu and Brama; 2000).
Huge theoretical structure exists on job hazard and worker’s attitude towards it. The
literature starts from Smith who argued that people self-select them towards particular jobs
according to the risk and hazard attached with them. This work has been extended by
96
number of studies (see paragraph 1). However the basic assumption of these approaches is
that the incumbent worker is aware of the risk or hazard i.e. associated with their prospective
jobs.
The same cannot be maintained when such hazards arises from exposure to
environmental risk. People are often not aware of exposure to such risks. Some of its
symptoms (such as lead poising, breathing problem etc.) may take years to develop.
Moreover the environmental factors are multifaceted and complex. Hence the causal effect
cannot be usually asserted. The problem is more pronounced in a developing economy. The
only way to mitigate this situation is the development of genuine environmental awareness.
There may be two ways in which this information incompleteness may develop in
the case of environmental risks. Firstly, a particular job may involve number of duties –
some of which may be (environmental) risk-prone, some less risky while some others with
virtually no risk attached with them. Thus, instead of simply depicting jobs are risky and
riskless, it is very likely that the jobs that the incumbent chooses may have varying degrees
of risk according to the nature of duties attached.
Secondly, there may be a genuine lack of awareness about the extent of danger due
to environmental pollution. The environmental theorists opine that the environment is a
highly non-linear dynamic structure often leading to chaotic fluctuation. As such, it is often
impossible fully stress out the impact of environmental hazards. Moreover there is often a
lack of environmental awareness among the common people. The common people are
97
generally myopic. Long run effects are often outside their provided. Hence, job decisions
based on possible environmental hazard are often incomplete and based on shaky
foundations.
7.3 Theoretical Framework:
The economists have viewed health as an important component of human capital
(Grossman 1972 a, b, 2000; Case and Deaton, 2005; Cropper, 1977; Muuerinen and Le
Grand, 1985; Becker, 1964; Mushkin, 1962; Fuchs, 1966. A simple argument in the
literature is that a person’s working capacity depends on his current health. Health is thus an
asset on which the individual’s make purposeful decision of investment and maintenance.
However like all other capital assets health also as depreciation in the form of using up of
health capacities. The health investment production function has developed by ‘Grossman’
is an easy way of capturing the complex process of health generation and maintenance.
There are several theoreticians (Borjas, 2004; Boskin 1974; who are developed the
occupational choice model based on this human capital framework.
There have been a number of controversies regarding the Grossman’s mode on
health as a form of human capital. The most important criticism, as acknowledged by a
Grossman himself, is that of Ehrilch and Chuma, (1990). These main argument is that in the
model Grossman does not specify and optimum health threshold, instead he assumes that
individual instantly adjust themselves to the optimum health investment. This brings a very
serious empirical relation of a positive relationship between health expenditure and health.
98
However a number of studies (Cochrane et.al, 1978; Wag staff, 1986; Dardanai & Wag
staff, 1987; Leu & Gerfin, 1992) opined that the relationship between health care and health
level is negative. This limitation is removed by a health threshold model provided by
(Galama & Kepteyn, 2009). In the new model spaces are opened up between health level
and health attainment. Empirically it is then found that blue colored workers allow their
health to fall to a much lower level than the white colored workers. Similarly rooms are
made of temporary medical innless that does not itself relate to the long run health
prospects.
The most important extension for our purpose is the avenues that the new approach
opens up for the analysis of occupational health. (Hurley, Kliebenstein and Orazem (1999),
in their approach present health condition of an individual depend on the human capital that
he is endowed with the behaviors that positively and negatively affect health and also the
occupational variables. Following Hurley, Kliebenstein and Orazem (1999), we can write
the health production process as follows.
hit = ƒ(Hit, Oit, µ it) ---------------------------- (1)
Where hit is the measure of health for an individual i at time t, Hit is the human capital
variable that an individual has accumulated over his life time. Here Hit includes both the
effects of positive investment (e.g. nutrition, good health, habits and healthy exercises etc.).
In this function µit is the original individual specific health endowment.
The occupational variables are captured by Oit. These variables measure the presence
of an intensity of exposure to environmental hazards that decreases the individual’s current
99
health attributes. For the estimation procedure, generally a linearised form of the above
production process is taken into account. The Hurley et.al (1999) study uses a number of
dummy and proxy variables to capture the occupational hazards that are associated with the
individual’s occupational choice.
In our model we try to prove deeper into the components of yt. Our simple argument
is that there are many aspects of occupational hazards that can be properly mitigated if
sufficient steps are taken in the right direction. However it is the level of awareness to the
occupational risk that become an important determinant towards adoption appropriate
protective measures.
We may reformulate the Farely et.al health production function as follow.
hit = ƒ(Hit, ψ (Oit), µ it)------------------------------- (2)
Where ψ (Oit) is awareness function of Oit. We assume that ψ'>0 and ψ''>0. It simply
means that awareness increases with the level of pollution and at an increasing rate.
The introduction of the awareness function has a very serious implication for hit. Consider
two identical individuals for whom the only difference is the level of awareness.
Taking a linear form as (Hurley, Kliebenstein and Orazem, 1999) we then get
― ― ― ―hpt = α Hpt + β ψ p(Opt) + γ µpt)------------------------- (3)
― ― ― ―hp’t = α Hpt + β ψ p’ (Opt) + γ µpt) ------------------------ (4)
Now, from equation (3) ― (4)
― ― ― ―hpt ― hp’t = ψ p(Opt) ― ψ p’ (Opt)
100
This simply implies that the differential health level of an individual depends upon
on the awareness differential of occupational hazards. This brings out a testable hypothesis
that we can empirically verify. This is our task in the rest of the chapter.
7.4 Data description:
From the above theoretical model it is clear that there are some important socio-
psychological determinants of a person’s occupational health. These data are micro theoretic
in nature and not available from any large scale macro survey. This necessitates a detailed
survey of the micro responses of the traffic polices in our sample city. Our selection of
Kolkata is purposive. The data on vehicular emission3 clearly indicates Kolkata is one of
the most polluting cities in India. Increasing population pressure, inadequate road space,
existence of old fashion polluting vehicles, overcrowding of slow moving vehicles adds to
its woe. The thick air is inhaled by the city’s pedestrians and the traffic polices are the worst
victim.
As noted early in this chapter Kolkata police have no separate traffic police cadres.
All recruitments to the lower cadre of the Kolkata Police are exposed to the environment
risk of the polluted air space in the city. Their job is however divided into two types of
duties – in off-roads and on- roads. The former is relatively less risky with regards to
environmental hazards. However once again there is no choice between the types of duties.
It is prefixed.
3 See: figure in page 60.
101
Kolkata has a track record of air pollution and pollution hazard. Numerous studies by
the environmental scientists have documented the deterioration in air quality and the health
conditions of the inhabitants in the city. (Chattopadhya et.al; 2007, Samanta et.al; 1998,
Chatterjee et.al; 1988). These studies reveal that environmental hazard is wide spread near
the main junctions in the city. However there are few studies that relate vehicular pollution
to the perception and utility of those who are directly exposed to it. The theoretical exercise,
given above, sets up a structure whereby these issues could be properly addressed. It is now
time to look at it empirically.
The West Bengal Pollution Control Board (WBPCB) gives us detailed information4
regarding the air pollution level of selected 14 junctions of the city. This data show that
there is a great intensity of air pollution at these junctions. Also it is observed there is a large
amount of seasonal fluctuation in the air pollution level. The seasonality pattern across the
junctions is almost identical. It is highest during the winter season and lowest in monsoon
season. This chapter wishes to observe whether the traffic police personnel5 posted at these
junctions are well aware of the situation and the protection they take to combat this.
4 See WBPCB Report for various years.5 Undoubtedly, they are one of the worst suffer of automobile pollution being constantly exposed to thepolluted air over long stretches over time.
102
Fig- 7.4.1.a Monthly fluctuation at 14 junctions (RPM)-20016
Fig- 7.4.1.b Monthly fluctuation at 14 junctions (NO2)-2007
(Source: WBPCB Report)
For this we collected information from 98 traffic police personnel who are having
regular duties in these junctions. These traffic police are recruited from the general police
staff. Hence, their choice is very limited regarding the nature of job. However, they are
provided a number of protective of protective gears to salvage themselves from the
environmental ill-effects among the protective gears that are provided to the traffic police
are helmet, mask, umbrella, sunglass, glucose tablets, regular medical check-up etc.
6 We give year only two of the fluctuation diagrams. The remaining are in appendix.
Monthly Fluctuation of RPM 2001
0
100
200
300
400
Apr May June July Aug Sep Oct Nov Dec Jan Feb Mar
Mont h
Monthly Fluctuation of NO2 2007
0
50
100
150
Apr May June July Aug Sep Oct Nov Dec Jan Feb Mar
Month
103
However, the traffic police personnel are not always aware about the usefulness of these
protective gears. Hence, they may fail to use these protective gears properly. As the
consequence they expose themselves to greater environmental risk and health hazards. This
adversely affects their productive capacities and also adds to their medical expenditure.
Hence by proper utilization of these gears such unwanted cost might be reduced. This
chapter wishes to unravel these irrational tendencies that arise due to low information about
environmental risk and low awareness.
In the present study we concentrate on the traffic police at 14 selected junctions of
the city. A detailed questionnaire was prepared to elicit a wide amount of information about
these policemen. The questionnaire may be sub-grouped under several headings-(a) familial
information (b) assessment of work condition (c) health and fitness (d) environment
consciousness in general (e) environment consciousness about their job and (f) the
preventive measures taken.
We culled information about the policemen socio-economic features – their parental
information and their familial history. We also collected information about their years of
service, their income, their duty hours, their fitness, and other such relevant information
regarding health and job.
A number of subjective information is also gathered. We have questions regarding
their job satisfaction, preference for duties, and likeness of the job etc. These questions are
designed in order to elicit various information regarding the nature of the problems faced by
104
the policemen, the possible solutions, suggestions for improvement and similar other
parameters. The data on the use of protective mechanism, training, types of protective
mechanism and the frequency of use were also generated in these interviews.
First we consider some basic socio-economic features of our sample policemen.
About 42.86% of our policemen are between 20 – 35 years of age while the 53.06% are in
the higher age group. However, out of them (only 4.08%) are above 50. In case of family
size, 59.18% holds 4 – 8 family members, 29.59% is in small scale family while 11.22% of
the sample policemen are in the larger family group. Considering the job experience of the
traffic police personnel in our study area, we see only 14.29% have above 15 years of job
experience, 51.02% have 1 - 8 years of job experience while some others (11.22%) have 9 –
15 years. The average years of job experience is roughly 10 years (9.96 to be precise). Thus
our sample largely covers the policemen who should have their full working capacity at the
time of sampling.
Table 7.4.2 (a) Age of the sample policemen
AgePercentage oftraffic police
20-35 42.8636-50 53.06
above 50 4.08
(Source: Author’s survey)
105
Table 7.4.2 (b) Family size of the traffic police
Average Familysize
Percentage oftraffic police
1-4 29.594-8 59.18
Above 8 11.22
(Source: Author’s survey)
Table 7.4.2 (c) Job experience of the traffic police personnel
Years of jobPercentage oftraffic police
1-8 51.029-15 34.69
above 15 14.29
(Source: Author’s survey)
Now, we come to education. 84.69% of the policemen have 8 – 12 years of
schooling. The higher education is rare (only 15.31%). However, if we compare it with their
father’s schooling. A substantial positive mobility is observed. Though, 34.69% of the
fathers had below 8 years of schooling, this has disappeared in their next generation.
However, in the upper echelon, there is stagnancy. In both the generation only 15.31% have
higher education. Also, in 81.63% of family of the traffic police personnel have a maximum
education of graduation or at least matriculation.
Table 7.4.2(d) Educational status of traffic policemen
Years ofschooling
percentage oftraffic police
8-12 84.6912-15 13.27
Above 15 2.04
(Source: Author’s survey)
106
Table 7.4.2(e) Educational status of the father’s of the traffic policemen
Years of father’sschooling
percentage oftraffic police
Below 8 34.698-12 50.00
Above 12 15.31
(Source: Author’s survey)
Table 7.4.2(f) Educational status of the family of the traffic policemen
(Source: Author’s survey)
As to the economic background, most of them come from agricultural family
(41.84%). The next big chunk is the government service (25.51%) while business account
for 15.31%. Thus most of the policemen could have no idea about the hazards that are
involved in the life of an urban traffic policeman.
Table 7.4.2(g) Ancestral family occupation of the traffic police
Father’s OccupationPercentage oftraffic police
Agriculture 41.84Rural-non farm 9.18Urban-non farm 2.04
Government service 25.51Private 6.12
Business 15.31
(Source: Author’s survey)
Maximum schoolingin family
percentage oftraffic police
1-9 4.0810-15 81.63
Above 15 14.29
107
In order to test their awareness about the environmental hazards of their job, a series
of question were constructed. We used some well known facts of vehicular pollution and
found out whether the police personnel were aware of them. The intensity of awareness is
measured by the proportion of right answers. For example only 36.73% of the policemen
could identify winter season is ‘most polluted’. On the contrary, 55.09% thinks that the
summer and monsoon season are ‘most polluted’. This is in complete contradiction with the
real data published by WBPCB7. 20% declare their complete ignorance in this matter. Also
detailed information regarding their assessment of the problems of vehicular emissions and
its causes were also asked.
Table 7.4.3 Perception about the most polluted season
Most polluted season Most Polluted Month %Police PersonnelFestive September-November 16.33Winter December-February 36.73
Summer March-May 48.98Monsoon June-August 6.12No idea ----------------------- 2.04
(Source: Author’s survey)
Before proving into the main analysis we first concentrate on some simple features
of the sample police personnel. The sample police personnel complain that they suffer from
number of disease headache, skin and eye problem, lumbago, hernia, hydroceal, asthma,
bronchitis, breathing problem and others etc. Many of these are linked to non pollutants
disease – such as lumbago and hernia may be linked with continuous standing at a particular
7 According to WBPCB reports, average figures for 2001-2007 reveals winter season to be most pollutedfollowed by the festive season. The least polluted is monsoon.
108
point over a long stretch of tine. However a majority of disease (headache, skin and eye
problem, asthma, hydroceal, bronchitis, and breathing problem etc) are airborne caused by
continuous exposure to the polluted air. Many of these diseases could have been prevented
had the police personnel taken proper care in using the protective gears that are necessary.
Thus in a direct physical way lack of awareness leads to a erosion of health. This adversely
affects his purse because of the cumulating medical expenditure.
Table 7.4.4 Affected police personnel by the different diseases
Disease % of police personnel affectedSkin 50.00
Headache 41.84Eye 73.47
Knee 55.10Lumbago 7.14
Hernia 8.16Hydroceal 9.18
Asthma 6.12Bronchitis 8.16
Breathing Problem 20.41
(Source: Author’s survey)
We also consider some other related health variables. The first is fitness. Roughly
57% of the police personnel are fit according to their own assessment. However, only
32.65% undertake regular medical check-up.
Our next query is regarding the utilization of the protective gears that are provided to
the police personnel. The percentage of users of the various protective gears is abysmally
low. It is seen that the highest percentage of users are that of helmet (0.80%). This might be
109
due to tropical climate on duty on the road. Next the percentages of users are that of
umbrella (0.60%) and sunglass (0.59%) respectively. These might be due to heavy rain and
the scorching heat they have to face in the tropical atmosphere where they operate.
However, one of the most important protective gears (masks) is used by only o.26% of the
police men. Only 0.06% uses all the protective gears. These non-optimal uses of protective
mechanism may have a direct link to the high percentage of morbidity among them.
Table 7.4.5 Percentage users of protective units
Protective units used % of usersHelmet 0.80
Umbrella 0.60Sunglass 0.59Rain coat 0.42
Masks 0.26Anklet 0.08
Glucose 0.18All 0.06
(Source: Author’s survey)
(* Note: The percentage figure exceeds 100 because of multiple uses)
These data reveals the precarious condition in which the police personnel find
themselves. They have very little choice of the nature of their duties. This is couple with
their non-optimal use of the protective gears available to them. As a consequence they are
under fire ― facing serious health problems.
In our survey we collected rather exhaustive data regarding subjective and objective
conditions of the traffic police and their awareness of the pollution hazard. We feel that the
110
data set can be usefully exploited for analyzing the problems at our hand. In the next section
we turn to this question.
7.5 Empirical Findings:
The main purpose of the chapter is to relate the extent of the knowledge of the police
personnel about hazard related to the job and the subjective and objective conditions in
which their job places them at the present. Among the objective features, our emphasis on
the productive capacity i.e. directly related to health. This has two consequences. On the one
hand, a deterioration of health increases medical expenditure of the police personnel. This
has negative dent on his earnings. Directly the person’s earnings capacity is also related to
their productiveness thus there are two ways in which health has an impact on the general
well-being of the policemen.
For the analysis we have used a number of composite indices. Each index is a
weighted average of a number of attributes.
In our study we have considered two indicators for this purpose fitness and health.
The question of health is complex involving the suffering from any chronic disease,
morbidity, medical problems and medical history. For fitness our concern is narrow. Here
we concentrated on the present capability in performing the traffic duty. These two concepts
are closely related but not same. Health refers to a wider age appropriate scenario. However
fitness is capability to adjust with the duty. A healthy person who is at the end of his carrier
111
may not be fit for this job. Similarly a person at a young age may also become unfit due to
some physical or mental constraints with he is born.
Next we consider subject wise aspect – job likeness and job satisfaction. Again these
concepts though related are not the same. Likeness is related to the initial decision making
about the job with limited a priori information. On the other hand, job satisfaction refers to
the preference towards the job after some experiences have been gathered. It is possible that
initial affinity may turn into disillusion. Also possible that the person is slowly adjusting
himself to the job and becoming satisfied through initially he/she may not like the job.
A number of independent variable is considered for analysis. These variables can be
broken up into three subgroups ─ a) family related b) subjective evaluation of the job c)
awareness of environment hazard and d) some demographic features. Under family features
we have two variables. One of them is an indexisation of the family history containing such
features as past experience of jobs by any family members, occupation, residents, education
etc. The next important feature is the number of children. The job evaluation indices include
job likeness, job satisfaction (except in the regression where it is dependent variable), duty
hours, health and fitness (except in the regression where there are dependent variables), the
pollution awareness has been constructed through a detailed set of questions. We consider
both pollution awareness in general and awareness of vehicle pollution. For measuring
pollution awareness we tested their knowledge regarding their nature and extent of pollution
(viz. most polluted season and most polluted time in the day). They are marked according to
the correct answer. A similar exercise conducted on the awareness of vehicular pollution.
112
These are two important determinant of pollution. To our surprise there are many who
scored poorly in terms of this text. This confirms our idea that these people are under fire –
their unaware of the risk they are facing while carrying out their duties.
It is clear that all the explanatory variables are not equally important. Further they are might
be mutual dependence between the so-called independent variables resulting in
multicollinearity. In order to remove such discrepancy, we run a step wise regression using
all the dependent variables for each of our four dependent variables. This step regression
helps us to identify the variables that are more important in each of the independent cases.
First we state in brief the basic statistical properties of our variables in table 1. From
table 1 it is clear that most of the quantitative variables lie within a certain range (roughly 0
― 2.5 most of the cases while 0 ― 1) in a few cases. The mean and the variance of the
variables are at a comparable level. The quantitative variables (age and number of children)
are also within the comparable stratum. This vindicates the quantity of variables and their
comparability for statistical and analytical purposes.
We have used both the ordinary least square and Tobit regression technique in the
analytical part. Since, this is a cross-sectional analysis, the problem of heteroscadisticityis
very much important. To tackle this problem we have used this White’s heteroscadisticity
consistent estimates instead of using the standard OLS structure. Such heteroscadisticity
consistent regression would yield consistent estimate of the regression coefficients. In order
to avoid the problem of multicolinearity the step regression technique was used to select the
appropriate variables.
113
In our exercise there is still problem in using OLS. This arises because the dependent
variables are truncated and qualitative in nature. The Tobit estimation may be more useful to
deal with such problems. These Tobit estimates are juxtaposed vis-à-vis the OLS
coefficients in order to bring out their comparability and reliability. It is seen that for all our
Tobit regressions the standard criteria for model selection (Akaiki, Schwarz and Hannan-
Quinn) are quite high. This proves that Tobit may not be an appropriate regression technique
in our case.
We now consider the health. In the table below we give the ordinary least square,
heteroscadisticity consistent regression and the Tobit. Our analysis clearly indicates their
health parameter is negatively related with age. This clearly shows that health decreases
with age – a very natural conclusion. The family parameter, an indicator of initial level of
human capital in health is also significant under all the three types of regressions. Similarly
the fitness parameter is also positively related with health.
. Again, the increase in pollution awareness helps to maintain better health. Generally
traffic police personnel can take several protective measures to mitigate the continuous
exposure to vehicular pollution and all resulting fatigue that strains on muscles and nerve
(Banerjee and Das; 2001). Protective head gears, murex, anklet, umbrella, sunglass, glucose-
powders and several other estimates are available. Most of these are supplied by their
employers. However, the police personnel in Kolkata rarely used all these protective
devices. This is an indication of their toxity in assessing the environmental risk. Proper use
of these protective mechanisms might help a lot in mainating health that is crucial for their
114
working performance. Our regression clearly indicates that from the policy point of view,
mere supply of the protective gears is not a pathway to a better heath. Arousing
environmental awareness is a sure way to enhance the production of health.
Second we come to fitness. It is clear that fitness is positively related with job
likeness and health. These are in the expected direction. However there exist a perverse
negative relation between job satisfaction and fitness and pollution awareness and fitness. In
the case of job satisfaction the perverse relationship may be a result of over working with
increase enthusiasm. However the negative correlation pollution awareness is tricky. It may
be the case that the traffic policemen become aware of pollution hazard through their
continuous exposure to polluted environment. Such a type of ‘learning’ process may be a
painful exercise. If the government has taken other appropriate training programme or any
other type of information dissemination process, then this type of ‘harsh’ learning procedure
would have been procedure.
Our third dependent variable is the job-likeness. Surprisingly pollution awareness
has no role in determining these variables. It clearly indicates that when jobseekers search
for job, environment consideration rarely enters into their pictures. This is true even when
the job has a considerable amount of environmental risk. The family features, as usual, are
an important determinant of job-likeness as the fitness. People feel that they can carry a job
of police personnel only if they are fit enough. The result is usual and not surprising.
However the total insensitivity to environment risk while selecting job is an alarming
feature. This reflects a general lack of environmental consciousness in the society.
115
Our fourth variable is job satisfaction, unlike job likeness; it determines the
assessment after indulging in the job. Strangely enough even here pollution awareness has
no role to play. Rather the nature of duty and the family burden (as captured by the variable
children) are important. However some effects of pollution awareness may be filtered into
the variable duties itself. These are two types of job for the police personnel – on the street
and off the street. On the street job is required to continuous exposure to environmental
pollutants. Hence, the medical risk is very high. Off the street duty is including some of
office works that are generally far away from the busy streets of the city. Here the pollution
risk is almost zero. Again experience is negatively related with job satisfaction (though not
significant).
By studying the various regressions we find a strange phenomenon. In none of the
cases awareness of vehicular pollution is stepped in. Though there is general pollution
awareness a specific knowledge about vehicular pollution is almost absent. However there is
clear dichotomy between the health variable and normative assessment of the job. It is true
that environmental awareness enters into the determination of health variable. However this
factor does not enter into the subjective assessment of the job. The result reflects a high
insensitivity towards environmental risk and pollution hazard. The policemen’s behavior
appears myopic. Drain in wealth due to the risk imposed to health has a long run
consequence that the policeman could not fathom. Only the direct possibilities of income
profile looms large in our policemen’s perception of the job.
116
Table 7.5.1: Descriptive statistics of the variables under studies
StatisticsVariables Mean variance Max MinFit 1.29 0.543645 2.27551 0Fam 0.40 0.012171 0.673469 0.040816Jobl 1.38 0.14065 1.653061 0.734694Duty 1.29 0.165403 1.836735 0.459184Hlth 0.34 0.037527 0.644898 0Polaw 1.19 0.184906 2.377551 0.331633Awvp 1.46 23.44753 48.7449 0.110204Jobs 1.76 0.43025 2.857143 0.581633Age 37.67 47.10877 54 24Children 1.63 1.925521 8 0
Table 7.5.2.1A: Results of regression analysis (Dependent Variable: hlth)
Ols Hetcov Tobit
Variable Coefficient t-ratio Variable coefficient t-ratio Variable Coefficient z-statAge -0.0076827 2.913 * Age -7.68E-03 -3.142 * Age -0.00835 -2.426 *Fam -0.56304 -3.268 * Fam -0.56304 -3.43 * Fam -0.58253 -3.1068 *Fit 0.073982 2.802 * Fit 7.40E-02 3.049 * Fit 0.082122 2.7179 *
Polaw 0.11806 2.75 * Polaw 0.11806 2.472 * Polaw 0.130444 3.2053 *
(Note: *- 1٪ level of significance and **- 5٪ level of significance)
Table 7.5.2.1B: Regression statistics ((Dependent Variable: hlth)
Ols-hetcov StructureR-SQUARE = 0.3364, R-SQUARE ADJUSTED = 0.3003VARIANCE OF THE ESTIMATE-SIGMA**2 = 0.38181
STANDARD ERROR OF THE ESTIMATE-SIGMA = 0.61791SUM OF SQUARED ERRORS-SSE= 35.127
MEAN OF DEPENDENT VARIABLE = 1.2928LOG OF THE LIKELIHOOD FUNCTION = -88.7817
Table 7.5.2.1C: Tobit Regression (Dependent Variable: hlth)
Tobit structureMean dependent var 0.335881 S.D. dependent var 0.193720Censored obs 10 Sigma 0.185746Log-likelihood 11.17109 Akaike criterion -10.34219Schwarz criterion 5.167618 Hannan-Quinn -4.068791
117
Table 7.5.2.2A: Results of regression analysis (Dependent Variable: fit)
Ols Hetcov TobitVariable coefficient t-ratio variable coefficient t-ratio variable coefficient z-stat
Hlth 0.81109 2.376 * * Hlth 0.81109 -2.336 * * hlth 0.906835 2.5189 * *Jobl 0.66626 3.911 * Jobl 0.66626 3.522 * jobl 0.676391 3.6333 *
Jobs -0.21924 -2.213 * * Jobs -0.21924 -2.323 * * jobs -0.23175-2.1716 *
*Age -0.024103 -2.543 * * Age -2.41E-02 -2.319 * * age 0.028122 -2.6516 *
Polaw -0.44319 -2.925 * Polaw -0.44319 -3.392 * polaw -0.48615 -2.481 * *
(Note: *- 1٪ level of significance and **- 5٪ level of significance)
Table 7.5.2.2B: Regression statistics (Dependent Variable: fit)
Ols-hetcov StructureR-SQUARE = 0.2268, R-SQUARE ADJUSTED = 0.1935
VARIANCE OF THE ESTIMATE-SIGMA**2 = 0.29867E-01STANDARD ERROR OF THE ESTIMATE-SIGMA = 0.17282
SUM OF SQUARED ERRORS-SSE= 2.7776MEAN OF DEPENDENT VARIABLE = 0.33643
LOG OF THE LIKELIHOOD FUNCTION = 35.5499
Table 7.5.2.2C: Tobit Regression (Dependent Variable: fit)
Tobit structureMean dependent var 1.293524 S.D. dependent var 0.737323Censored obs 8 Sigma 0.641131Log-likelihood -96.59263 Akaike criterion 207.1853Schwarz criterion 225.2800 Hannan-Quinn 214.5042
118
Table 7.5.2.3A: Results of regression analysis (Dependent Variable: jobl)
Ols Hetcov TobitVariable coefficient t-ratio variable coefficient t-ratio variable coefficient z-stat
Age 0.010159 1.874 age 1.02E-02 1.772 age 0.010145 1.615Fam 0.85376 4.246 * fam 0.85376 3.828 * fam 0.843702 1.6928Fit 0.20358 2.575 * * fit 0.20358 2.926 * fit 0.20517 2.1339 * *
(Note: *- 1٪ level of significance and **- 5٪ level of significance)
Table 7.5.2.3B: Regression statistics (Dependent Variable: jobl)
Table 7.5.2.3C: Tobit Regression (Dependent Variable: jobl)
Tobit structureMean dependent var 1.383174 S.D. dependent var 0.375034Censored obs 0 Sigma 0.329570Log-likelihood -30.27915 Akaike criterion 70.55831Schwarz criterion 83.48314 Hannan-Quinn 75.78614
Ols-hetcov structureR-SQUARE = 0.2189, R-SQUARE ADJUSTED = 0.1939VARIANCE OF THE ESTIMATE-SIGMA**2 = 0.11258
STANDARD ERROR OF THE ESTIMATE-SIGMA = 0.33552SUM OF SQUARED ERRORS-SSE= 10.582
MEAN OF DEPENDENT VARIABLE = 1.3812LOG OF THE LIKELIHOOD FUNCTION = -29.9919
119
Table 7.5.2.4A: Results of regression analysis (Dependent Variable: jobs)
Ols Hetcov Tobitvariable Coefficient t-ratio Variable coefficient t-ratio variable coefficient z-stat
Age -0.014127 -1.477 Age -0.014127 -1.567 age -0.0141639 -1.4389Duty 0.47843 3.088 * Duty 0.47843 3.165 * duty 0.478971 3.0491 *
children 0.094556 1.998 * children 0.094556 2.349 * * children 0.0944038 1.7017
(Note: *- 1٪ level of significance and **- 5٪ level of significance)
Table 7.5.2.4B: Regression statistics (Dependent Variable: jobs)
Ols-hetcov structureR-SQUARE = 0.1320, R-SQUARE ADJUSTED = 0.1043VARIANCE OF THE ESTIMATE-SIGMA**2 = 0.38637STANDARD ERROR OF THE ESTIMATE-SIGMA = 0.62159SUM OF SQUARED ERRORS-SSE = 36.319MEAN OF DEPENDENT VARIABLE = 1.7580LOG OF THE LIKELIHOOD FUNCTION = -90.4170
Table 7.5.2.4C: Tobit Regression (Dependent Variable: jobs)
Tobit structureMean dependent var 1.757028 S.D. dependent var 0.655934Censored obs 0 Sigma 0.607878Log-likelihood -90.27343 Akaike criterion 190.5469Schwarz criterion 203.4717 Hannan-Quinn 195.7747
120
7.6 Conclusion:
It has long been predicted by ‘Adam Smith’ that hazardous job should get risk
premium that equate it with the low hazard (Zero hazard) jobs. However if the workers are
unconscious of the hazard, the smithian risk premium story may go astray. This is the case
with our sample traffic police in Kolkata to whom the environmental risk does not loom
large in their assessment of the job. Thus raising pollution awareness among these who face
environmental hazard in their day to day activities is of utmost necessary. Unless their
awareness levels of workers increase, welfare of them cannot be ushered in the economy.
121
Chapter―8
Conclusion:
“Give us back those forests, take the cities”
(Rabindranath Tagore)
We live in an age of vehicles. Vehicles have raised our mobility, increased our
efficiency and have added to a substantial loss of the cost in transition. However like all the
gifts of modern science increase in vehicular population has its negative effect. Emission
from vehicles has deteriorated our environment. Of course, the faulty road structure and
congestion have fueled the problem.
Our study was initiated by the problems of vehicular emission inflicting heavy losses
in the environment. This proves to be a great burden for the humanity. We have
concentrated on the urban conglomerate of Kolkata where our study was directed.
It has two parts. In the first part, the objective assessment was made regarding the
spate and volume of vehicular emission. The problems of urban transit system and its
fragility discussed. Issues and problems were raised.
In the second part, we consider asset of people who are directly exposed to the
vehicular emission. They are the traffic police personnel working all day and night at the
122
busy junctions of the city. We assess their perception of the environmental pollution. The
steps they have taken to mitigate the problems are raised. An objective evaluation was made
regarding their health and fitness status. They are linked with their awareness and the little
they can do while surrounded by the five that is ready to engulf them.
In this study, we have briefly discussed the main conclusion under each chapter.
Here, we just collect them so that a better view can be approached.
The first chapter is a brief introduction of our study including the plan of the work
and objectives.
The second chapter is a literature survey. Thre types of study are surveyed in this
chapter. First, a general view of the vehicular emission is given. Secondly, we contextualize
it for the third world countries. Thirdly, we concentrate on India and finally, in Kolkata.
In the third chapter, the nature of data was discussed. We used both primary and
secondary data. Secondary data was collected from census and other official publications.
The secondary data was interview based. These two types of data are combined to get a
comprehensive view of the topic.
Fourth chapter attempts to test the relationship between urbanization and
vehicular population using the West Bengal data. For our analysis we concentrated on
123
various secondary sources particularly the census data. The relationship is an expected;
urbanization raises the possession of fuel consuming vehicles.
This is significantly positive both using econometric as well as non-parametric
techniques. Even the literacy rate is perversely affecting the possession of polluting vehicles,
the reason is that higher literacy implies higher human capital and enhances income. The
picture is not bright. Unless proper steps are taken toward efficient use a public transport
system and planned urbanization, there is a little chance in abating the fleet of polluting
vehicles.
From our above discussion we clearly understand that the probability of polluting
vehicles holding is higher in the urban area than the rural area. It has been proved by the
various statistical tests. Therefore we can obviously state that the pollution in the urban area
is more than the rural area in West Bengal. Urbanization leads to a conglomeration of
polluting vehicles and boosting up pollution.
Chapter five is concerned with the relation between urbanization and traffic
problems. Our urban centers show increasing demand for transport. There is however a
number of problems associated with urban transportation. In this paper we hope to deal with
some of these problems. These problems are multifaceted and often entangled with one
another. Here we propose to cover some of the issues with the respect of both India as well
as West Bengal. The inequalities in private urban transit systems are documented. Though
there is a rise in the total number of vehicles, still a sizable portion of the urban families who
124
have no vehicles. The problem is complicated by the lack of adequate road space and
congestion that prevents the use of bicycles. Severe problems of air pollution are also noted.
Kolkata has one of the cheapest public transport systems among the Indian metropolis.
However even this ‘cheap’ transit system is beyond the reach of the very poor in this city.
Further more, the system is also inefficient with waste of resources. An urban transport
planner has a tough task. He has to balance between equity, efficiency and sustainability of
transit system. This requires long run planning by the urban planners.
The chapter six has two parts. First, it brings out the important aspect of air
pollution. For the each of study, we have segregated the entire time period into four seasons-
winter, summer, monsoon and festival. These are of them are quite concern in the
environmental literature. The fourth season is tropical of Kolkata – arising mainly due to
social causes. The data reveals that this so called new season ranks very high in the disposal
of environmental waste. Thus an interesting suggestion of the paper is the role of
environmental fallen in increasing pollution and climate hazard. The point is clearly brought
out by our analysis.
The second part of this chapter demonstrates the seasonal trends (Periodicity) of the
different pollutants in Kolkata. This new technique is justified since the above discussion
exhibits the significant results of spectral estimates, which really represents the seasonal
periodicities of the pollutants under equal time intervals. We also understand correlation
between two series at different frequencies, using cross-spectral analysis. The strong
correlated periodicities among different series of pollutants under different years can be
125
analysed by the value of coherence, phase and also gain value etc. These totally indicate that
the model has a great implication of predicting long run trends.
The seventh chapter takes up the micro issue ― impact of vehicular emission on
traffic police. It has long been predicted by Adam Smith that hazardous job should get risk
premium that equate it with the low hazard (Zero hazard) jobs. However if the workers are
unconscious of the hazard, the Smithian risk premium story may go astray. This is the case
with our sample traffic police in Kolkata to whom the environmental risk does not loom
large in their assessment of the job. Thus raising pollution awareness among these who face
environmental hazard in their day to day activities is of utmost necessary. Unless their
awareness levels of workers increase, welfare of them cannot be ushered in the economy.
At the end of our journey we feel that the problems of vehicular emission are
multifaceted. It has both macro and micro dimension. The dynamics reveals the pattern and
volume of vehicular emission. However, the micro aspect study is the impact at individual
level. It covers the people to face it. Thus vehicular emission is not only environmental
phenomenon. It is both a cause and effect of the air pollution due to vehicles. Appropriate
measures are urgently needed to save our beautiful planet from an untimely death.
126
Reference:
Agarwal, A.A. (1996), Slow Murder: The Deadly story of Vehicular Pollution in
India, CSE/Delhi.
Association of State Road Transport Undertakings, 2002a. The performance of
STUs at a Glance for the period 1989-90 to 2000-01. ARSTU, Delhi, India.
Association of State Road Transport Undertakings, 2002b. STU Policy Issue.
ARSTU, Delhi, India.
Badami, M., Tiwari, G., Mohan, D., 2004. Access and Mobility for the urban poor
in India: bridging the gap between policy and needs.
Banerjee, S; (2001), Chapter 4 in R. N Bhattacharyya (-ed) Environmental
Economics : An Indian Perspective, (Chapter-4) Oxford University Press,
New Delhi.
Banerjee, S and R.L. Das. (2001) ’Vehicular Pollution in Calcutta: An Assesment
of intensity, in B. Moitra and R. Acharya (-ed) effects of Globalisation on industry
and environment, The Lancer Books, New Delhi.
Barsky, R.B., F. T. Juster, M. S. Kimball, and M. D. Shapiro, “Preference
parameters and Behavioral Heterogeneity: An Experimental approach in the
Health and Retirement Study,” Quarterly Journal of Economics, 112 (1997), 537-
580.
Basu and Brahma, (2000), ‘Studies of the Occupational Exposure to Hazards of the
Calcutta Police Personnel at traffic intersections’, WBPCB Report.
127
Becker, G.S. (1964), Human Capital (Columbia University Press for the National
Bureau of Economic Research, New York).
Bertraud, A., 2002. The economic impact of land and urban planning regulations
in India.
Bhagat, R. B.(1992) Components of Urban Growth in India with Reference to
Haryana : Findings from Recent Censuses Nagarlok, Vol. 25, No. 3, pp.10-14
Bloomfield,P. (1976) Fourier Analysis of Time Series; An Introduction. Wiley.
Chatfield, C. - Introduction to the Analysis of Time Series.
Bonin, H., T. Dohmen, Thomas, A. Falk. D. Huffman, and U. Sunde, “Cross-
sectional earnings risk and occupational sorting: The role of risk attitudes,”
Labour Economics, 14(6) (2007), 926-937.
Borjas G.J. 2004. Labor economics. New York, NY: McGraw-Hill
Bose A.: India’s Urbanisation: 1901-2001, Tata McGrow Hill Publishing Co. Ltd.,
New Delhi (1978)
Bose, R. K., 1998. Automobiles and environmental sustainability: issues and
options for developing countries. Asian Transport Journal, December 1998, pp.
13.1-13.16
Bose, Ranjan K. "Energy Demand and Environmental Implications in Urban
Transport - Case of Delhi." Atmospheric Environment 30 (1996): 403-412.14 Nov.
2006.
Boskin M.J. 1974. A Conditional Logit Model of Occupational Choice, Journal of
Political Economy, 82(2): 389-398.
128
Bowden, R. J., and Martin, V. L. (1993), "Reference Cycles in the Time and
Frequency Domains: Duality Aspects of the Business Cycle," in Models, Methods
and Applications in Econometrics: Essays in Honor of Rex Bergstrom, eds. P. C.
B. Phillips and V. B. Hall, London: Allen and Unwin, pp. 201-219.
Brandes, O., Farley, J., Hinich, M. & Zackrisson, U.: The time domain and the
frequency domain in time series analysis. Swedish Journal of Econo-mics 70, 25-
42, 1968.
Breese, G. (1969): Urbanisation in Newly Developing Countries, Prentice Hall,
New Delhi
Brockerhoff, M. (1999) Urban Growth in Developing Countries: A review of
Projections and Predictions, Population and development Review, Vol 25. No4,
PP 757-778.
Brown, C. (1980), “Equalising differences in the labour market”, Quarterly
Journal of E Campbell et.al. A Report on Toronto Public Health. Air Pollution
Burden of Illness from Traffic in Toronto – Problems and Solutions. November
2007. Toronto, Canada.conomics, 94,113-34.
Case, A., and A. Deaton (2005), “Broken Down by Work and Sex: How our Health
Declines”, in: David A. Wise (ed.) (2005), Analyses in the Economics of Aging,
The University of Chicago Press, Chicago, pp. 185-212.
Central Intelligence Agency, 2002. The CIA World Fact Book. CIA, Washington
DC.
Chakraborty, D. (1997), Kolkata in pollution perspective ’, World transport policy
and practice .Vol. 3 No .3.
129
Chakraborty, D. (1997), ‘Kolkata in Pollution Perspective,’ World Transport
Policy and Practice. Vol.3 No.3.
Chatterjee, S; S, Dipali; BP, Chatterjee; (1988), “Pulmonary function studies of
healthy non smoking men of Calcutta” Annal of human biology.
Chattopadhyay, B.P., A.K. Mukherjee, K. Mukherjee, A Roychowdhury (2007),
“Exposure to vehicular pollution and assessment of respiratory function in urban
inhabitants” Environment and occupational lung disease.
Clean Air Initiative, 2004. CNG Buses in Delhi. Clean Technologies Information
Pool. 2004.
Cropper, M.L. (1977), “Health, investment in health, and occupational choice”,
Journal of Political Economy 85:273-1294.
Dardanoni, V. and A. Wagstaff (1987), “Uncertainty, inequalities in health and the
demand for health”, Journal of Health Economics 6:283-290.
Dardanoni, V. and A. Wagstaff (1990), “Uncertainty and the demand for medical
care”, Journal of Health Economics 9:23-38.
Davis Kingsley (1962): "Urbanisation in India – Past and Future", in Turner, R.
(ed.) India's Urban Future, University of California Press, and Berkley.
Davis, K. (1965). The urbanization of the human population. Scientific American,
213(3), 41 – 53.
Delhi Metro Railway Corporation Ltd., 2003a. History of the Delhi Metro. Delhi
Metro Rail Corporation, New Delhi, India.
130
Delhi Metro Rail Corporation Rail Corporation Ltd., 2003b. Project Update. Delhi
Metro Rail Corporation, New Delhi, India.
Dorfman and Dorfman (ed) , Controlling Urban Air Pollution : A benefit
cost assessment ,’ Economics of the environment. 1993.
Ehrlich, I. and H. Chuma (1990), “A model of the demand for longevity and the
value of life extensions”, Journal of Political Economy 98:761-782.
Erica Moen, (2008), Vehicle Emissions and Health Impacts in Abuja, Nigeria A
study report at Watson Institute for International Studies.
Environment News India., 2001. India may pull old vehicles off roads. September
10, 2001.
Evaluation and use of epidemiological evidence for environmental health risk
assessment. Copenhagen, WHO Regional Office for Europe, 2000 (document
EUR/00/5020369) (http://www.euro.who.int/document/e68940.pdf.accessed 13
May 2004).
Faiz, A., Weaver. C. S and d Walsh. M.P., ‘Air Pollution from Motor Vehicles,
copy 1 & 2, World Bank Publication, 1996.
Fishman, G. S.: Spectral methods in econometrics. Harvard University Press,
Cambridge, 1969.
Fuchs, V.R. (1966), “The contribution of health services to the American
economy”, Milbank Memorial Fund Quarterly 44:65-102.
Gakenheimer, R., 1999. Urban mobility in the developing world Transportation
Research 33A, 671-690.
131
Gakenheimer, R., 2002. Planning Transportation and Land Use for Cities in India.
Massachusetts Institute of Technology, Cambridge, MA.
Gakenheimer, R., Zegras, C., 2003. Travel Demand Drivers: Mumbai and
Chennai, India. World Business Council for Sustainable Development, Geneva,
Switzerland.
Galama, T.J., Kapteyn, A., Fonseca, F., Michaud, P.C. (2009), “Grossman’s
Health Threshold and Retirement.”, RAND Working Paper, WR-658.
Ghosh, S.M. (1998), Health Hazards by Automobile Pollution: a Publicity Leaflet.
Granger, C. W. J, in association with Hatanaka, M.: Spectral analysis of economic
time series. Princeton Uni-versity Press, Princeton, 1964.
Granger, C.W. J, - Spectral Analysis- An Economic approach – Academic Press -
1966.
Granger, C. W. J., and A. O. Hughes. "Spectral Analysis of Short Series." J.
Roy. Statistics, Soc. (A).131 (1968):83-99.
Granger, C. W. J. and H. J. B. Rees. "Spectral Analysis of the Term Structure of
Interest Rates," Review of Economic Studies, 35 (January 1968), 67-76.
Granger, C. W. J.: The typical spectral shape of an economic variable. Eco-
nometrica 34, 150-161, 1966.
Grether, D. M. & Nerlove, M.: Some properties of 'optimal' seasonal adjust-ment.
Econometrica 38, 682-703, 1970.
Griliches, Zvi (ed.), Price Index and Quality change Cambridge: Harvard
University Press, 1971).
132
Grossman, M. (1972a), “On the concept of health capital and the demand for
health”, Journal of Political Economy 80:223-255.
Grossman, M. (1972b), The Demand for Health: A Theoretical and Empirical
Investigation (Columbia University Press for the National Bureau of Economic
Research, New York).
Grossman, M. (2000), “The Human Capital Model”, pp. 347-408, in A.J. Culyer
and J.P. New house (Eds.), Handbook of Health Economics, Volume 1, Elsevier
Science.
Government of Ghana. (1995). Ghana - Vision 2020. Presidential Report to
Parliament on Coordinated Programme of Economic and Social Development
Policies, Accra: Government of Ghana.
Goyal, P., and Sidhartha. "Present Scenario of Air Quality in Delhi: a Case Study
of CNG Implementation." Atmospheric Environment 37 (2003): 5423-5431. 12
Nov, 2006.
Goyal, S. K., S. V. Ghatge, P. Nema, and S. M. Tamhane. "Understanding Urban
Vehicular Pollution Problem Vis-a-Vis Ambient Air Quality - Case Study of a
Megacity (Delhi, India." Environmental Monitoring and Assessment 119 (2005):
557-569.12Nov.2006.
Gudmundsson, G. "Time-Series Analysis of Imports, Exports and Other Economic
Variables," Journal of the Royal Statistical Society, Series A, 134 (1971), 383-412.
Gwillian, K., 2003. “Urban transport in Developing countries”. Transport
Reviews 23 (2), 197-216.
133
Hannan, E: J.: "Regression for Time Series," in Time Series Analysis, ed. by M.
Rosenblatt. New York: Wiley, 1963.
Hao and Wang; 2005, Improving Urban Air Quality in China: Beijing Case Study,
Journal of Air & Waste Manage. Assoc. 55:1298–1305.
Hause, J. C.: Spectral analysis and the detection of lead-lag relations. Ameri-can
Economic Review 61, 213-217, 1971.
Health aspects of air pollution – answers to follow-up questions from CAFE.
Report on a WHO working group. Copenhagen, WHO Regional Office for Europe,
2004(documentEUR/04/5046026)(http://www.euro.who.int/document/E82790.pdf,
accessed 13 May 2004).
Hurley, T.H., P.F. Orazem, and J. Kliebenstein. “An Analysis of Occupational
Health in Pork Production.” Amer. J. Agr. Econ. 81(February 1999):144–63.
Jenkins, G. M. & Watts, D. G.: Spectral analysis and its applications. Holden-
Day, San Francisco, 1968.
Jenkins, G. M., and D. G. Watts. Spectral Analysis and Its Applications. San
Francisco: Holden-Day, 1969.
Jenkins, G.M. and Watts, D.G. (1968) Spectral Analysis and its Applications.
Holden-Day Diggle, P. - Time Series, A Biostatistical Introduction- Oxford
University Press 1990.
Kandlikar, M., Ramchandran, G., 2000. The causes and consequences of
particulate air pollution in urban India: A synthesis of the science. Annual
Reviews of Energy and the Environment 25 (1), 629-694.
134
Kathuria, Vinish. "Vehicular Pollution Control in Delhi." Transport and
Environment (2002): 373-387.11Nov.2006.
Kingsley and Golden H. H. (1954): “Urbanisation and development in pre-
Industrial. Areas", Economic Development and Cultural Change, Vol.3 no1.
Krupnic, A. J ,’ Urban Air Pollution in developing countries , : Problem &
Policies in Dasgupta . P & G Maler (eds ); The Environment and development
issue, ’ Vol 2 Oxford University Press 1997.
Kundu, A (1983): "Theories of City Size Distribution and Indian Urban Structure –
A Reappraisal", Economic and Political weekly, 18 (3).
Kundu , A. (1994) : "Pattern of Urbanisation with Special Reference to Small
and Medium Towns in India" in Chadha, G. K., Sectoral Issues in the
Indian Economy, Har – Anand Publications, New Delhi - (1997).
Labys, W. C., and C. W. J. Granger. Speculation, Hedging and Commodity Price
Forecasts. Lexington: D. C. Heath, 1970.
Larsen, J. T. Seasonality of the Cattle Market. USDA ERS-468, Jan. 1971.
Lipsey, Robert, “Comments on the value of saving a Life: Evidence from the
Labour Market,” in N. Terleckyz (ed.), Household Production and Consumption,
NBER Studies in Income and Wealth no. 40 (New York: Columbia University
Press, 1976), 301-302.
Low, N., Banerjee-Guha, S., 2003. The global tyranny of roads: observations from
Mumbai and Melbourne. World Transport Policy and Practice 9 (2), 5-17.
135
Maity and Agarwall; (2005) Environmental Degradation in the Context of
Growing Urbanization: A Focus on the Metropolitan Cities of India, Journal of
Human Ecology, 17(4): 277-28.
Maitra A.K.: Managing Urban Environment in India - Workshop on Role of Cities,
Volume II - Times Research Foundation, Kolkata (1993).
Maitra AK.: Urban Environment in Crisis, New Age International Publishers,
Delhi (2000).
Marwah, B.R., Sibal, V.K., Sawant, S., 2001. Bus transport in Delhi. In: Morris, S.
(Ed.), India Infrastructure Report 2001. Oxford University Press, New Delhi, India.
Metro Railway Kolkata., 2003. Status of Ongoing Projets. Metro Railway Kolkata,
Kolkata, India.
Ministry of Finance., 2002. Economic Survey 2001-2002. Government of India,
Ministry of Finance, Delhi, India.
Ministry of Petroleum and Natural Gas, 2002. Auto Fuel Policy Report. Ministry
of Petroleum and Natural Gas, Delhi, India.
Ministry of Finance., 2004. Economic Survey 2002-2003. Government of India,
Ministry of Finance, Delhi, India.
Ministry of Road Transport and Highways, 1999, 2000, 2003. Handbook on
Transport.
Ministry of Urban Affairs and Poverty Alleviation, 2004. National Urban
Transport Policy. Ministry of Urban Affairs and Poverty Alleviation, Delhi, India.
136
Mohan, D., 2001 “Planning for public transport: integrating safety, environment
and economic issues”. Paper presented at workshop on Transport, Land Use and
the Environment, Pune, India, Dec. 3-4, 2001.
Mohan R.: State of the Environment: India [Report No.1999EE45, Submitted to
United Nations Environment Programme (2001) National Institute of Urban
Affairs: Report of the National Commission on Urbanization. Vol. 2. (1988).
Mohan R.: Urbanisation in India: Patterns and Emerging Policy Issues in the
Urban Transformation of the Developing World. Josef Gugler (Ed.). Oxford
University Press, Oxford (1996).
Mumbai Metropolitan Region development Authority, 2003. Mumbai Urban
Transport Project. MMRDA, Mumbai, India.
Mushkin, Selma J.’Health as an Investment." J.P.E. 70, no. 2, suppl. (October 1962
1: 129-57.
Muurinen, Jaana-Marja, and Le Grand, Julian (1985), “The Economic Analysis of
Inequalities in Health”, Soc. Sci. Med, Vol. 20, No. 10, pp. 1029-1035.
Nerlove, M.: Spectral analysis of seasonal adjustment procedures. Econometrica
32, 241-268, 1964.
Nerlove, M. "Spectral Analysis of Seasonal Adjustment Procedures,
"Econometrica, 32 (October1 964), 241--86.
Newton, J.H. – Times Lab: A Time Series Analysis Laboratory-Wadsworth &
Brooks 1988.
Office of the Registrar General of India, 2001a. Rural-urban distribution of
population in The Census of India, 2001. New Delhi, India.
137
Office of the Registrar General of India, 2001b. Urban agglomerations/ cities
having population of more than one million in 2001 in The Census of India,. 2001.
New Delhi, India.
Oi, Walter, “An essay on Workmen’s compensations and Industrial Safety,” in
Supplemental Studies for the National Commission on State Workmen’s
Compensation Laws (Washington, D.C,: U.S Government Printing Office, 1973),
41-106.
Ojolo. et. al. (2007) A survey on the effects of vehicle emissions on human health
in Nigeria, Journal of Rural and Tropical Public Health 6: 16-23.
Padam, S., Singh, S.K., 2001. Urbanisation and Urban Transport in India: The
Sketch for a Policy. Transport Asia Project Workshop, Pune, India.
Parsons, L. J. and W. A. Henry. "Comparison of Time-Series Data using Spectral
Analysis." Journal of Marketilg Research, 9 (November 1972), 391-5. 101.
Pucher, J., Lefevre, C., 1996. Urban Transport Crisis in Europe and North
America. Macmillan Press, London, England.
Pucher, J., Korattyswaroopam, N., Ittyerah, N., 2004. The crisis of public transport
in India: overwhelming needs but limited resources. Journal of Public
Transportation 7 (3), 95-113.
Pune Municipal Corporation, Pune, 2004. Comprehensive Traffic and
Transportation Study for the city of Pune. Pune Municipal Corporation, Pune,
India.
138
Raj, D., 2001. Ban on old Vehicles hurt autorickshaw drivers. Environmental
Bulletin India.
Ramachandran, R., 1989, Urbanisation and Urban System in India. Oxford
University Press, Oxford, England. Statistics in India Transport Research Office,
Ministry of Road Transport and Highways, Delhi, India.
Reder, Melvin, “Wage Differentials: Theory and Measurement,” in Aspects of
Labour Economics (Princeton: Princeton University Press, 1962), 257-311.
Samanta, G (1998), “Air Pollution in Calcutta during winter” – A three years
study, Current Science, 75.
Sharma A. & Roy chowdhury A. (1996). Slow Murder: The deadly story of
vehicular pollution in India. State of the Environment Series. Centre for Science
and Environment.
Shariff; (2012), Private Vehicle Ownership and Transportation Planning in
Malaysia, IPCSIT vol. 26.
Sibal, V., Sachdeva, Y., 2001. Urban transport scenarios in India and its linkages
with energy and environment. Urban Transport Journal 2 (1), 12-20.
Smith, Adam, The Occupational Safety and Health Act: Its Goals and Its
Achievements (Washington, D.C,: American Enterprise Institute, 1976).
Sreedharan, E., 2003. Need for Urban Mass Transport System for our cities. Press
Information Bureau, Government of India, New Delhi, India.
Tata Energy Research Institute, 1997. Anon., 1997. Environmental Aspects of
Energy use in Large Indian Metropolises. Tata Energy Research Institute, Delhi,
India.
139
Thaler, Richard, and Sherwin Rosen, “The Value of Saving a Life: Evidence from
the Labour Market,” in N.Terleckyz (ed.), Household Production and
Consumption, NBER Studies in Income and Wealth no 40 (New York: Colombia
University Press, 1976), 265-298.
Times of India, 2002. Delhi roads have reached saturation point. November 30,
2002. Web edition, accessed on September 14, 2004.
Tiwari, G., Mohan, D., 1999. Sustainable transport systems: linkages between
environmental issues, public transport, non-motorized transport, and safety.
Economic and Political weekly 34 (25), 1589-1596.
Tiwari, G., 2001. Pedestrian infrastructure in the city transport system: Delhi case
study. World Transport Policy and Practice 7 (4), 13-18.
"Trends and Structure of Employment in the 1990s : Implications for Urban
Growth", Economic and Political Weekly, 32(24).
U.S Bureau of Labour Statistics, Handbook of Labour Statistics (Washington, D.C.
U.S Government Printing Office, 1972).
Van der Westhuisen H. (1998). Evaporative emissions and fuel loss in South
Africa, Unpublished B.Eng. Project Report, Centre for Automotive Engineering,
University of Stellenbosch, South Africa.
Vasconcellos, E., 2001. Urban Transport, Environment and Equity: The case for
Developing. Earth scan Press, London, UK.
Viscusi, W. K., Employment Hazards: An Investigation of Market Performance,
Ph.D. dissertation, Harvard University, 1976.
140
Viscusi W. K., “Wealth Effects and Earning Premiums for Job Hazards,” ‘The
Review of Economics and Statistics, Vol. 60, No. 3. (Aug., 1978), pp. 408-416.
Weaver, C., 2004. Challenges and Success of Delhi’s CNG Program: Lessons for
other Cities. Presented at: The Leap Frog factor: Towards Clean Air in Asian
Cities. New Delhi, India.
Weiss, Yoram, “The Wealth Effect in Occupational Choice,” International
Economic Review 17 (June 1976), 292-307.
Whitelegg, J., Williams, N., 2000. Non-motorised transport and sustainable
development: evidence from Calcutta. Local Environment 5 (1), 7-18.
WHO-UNEP Report (1988), Urban Air Pollution in the Megacities of the World.
World Health Organization/United Nations Environment Programme. (1992).
Urban Air Pollution in Mega Cities of the World. Oxford: Basil Blackwell.
World Bank, 2002. Anon., 2002. India’s Transport Sector. The Challenges Ahead.
The World Bank, Washington, DC.
Wynne, G. 2005. Bogota’s BRT achieves farebox coverage of operating costs.
Passenger Transport, Nov. 5, 2001. American Public Transportation Association,
Washington, DC.
141
Appendix (Figures of chapter- 6):
R P M P e r i o d o g r a m o f 2 0 0 1
0 . 0 0 0 0 0 0
2 0 0 0 0 . 0 0 0 0 0 0
4 0 0 0 0 . 0 0 0 0 0 0
6 0 0 0 0 . 0 0 0 0 0 0
1 2 3 4 5 6
F r e q u e n c y
R P M S p e c t r a l D e n s i t y o f 2 0 0 1
0.000000
10000.000000
20000.000000
30000.000000
40000.000000
1 2 3 4 5 6
F r e q u e n c y
C r o ss P e r i o d o g r a m o f R P M 2 0 0 1
-10000.000000
0.00000010000.000000
20000.000000
30000.00000040000.000000
50000.000000
1 2 3 4 5 6
Fr e que nc y
C r o s s D e ns i t y o f R P M 2 0 0 1
0.0000002000.0000004000.0000006000.0000008000.00000010000.00000012000.00000014000.000000
1 2 3 4 5 6
F r e q u e n c y
S qua r e d C ohe r nc y o f R P M 2 0 0 1
0.0000000.2000000.4000000.6000000.8000001.0000001.200000
1 2 3 4 5 6
Fr e que nc y
C r o ss A mp l i t ud e o f R P M 2 0 0 1
0.0000002000.0000004000.0000006000.0000008000.00000010000.00000012000.00000014000.000000
1 2 3 4 5 6
Fr e que nc y
P ha s e S p e c t r um o f R P M 2 0 0 1
- 1.000000
- 0.500000
0.000000
0.500000
1.000000
1 2 3 4 5 6
F r e q u e n c y
Ga i n of RP M 2 0 0 1
0.000000
0.500000
1.000000
1.500000
2.000000
2.500000
1 2 3 4 5 6
Fr equency
M o nt hly F luct uat io n o f R PM 2 0 0 1
050100150200250300350
Apr May June July Aug Sep Oct Nov Dec Jan Feb Mar
M ont h
142
Densi t y of SO2 2001
0.00000050.000000100.000000
150.000000200.000000250.000000
1 2 3 4 5 6
F r e q u e n c y
SO2 Cross Periodogram 2001
-100.000000
0.000000
100.000000
200.000000
300.000000
1 2 3 4 5 6
Fr equency
S O2 C r oss D e nsi t y 2 0 0 1
-50.000000
0.000000
50.000000
100.000000
150.000000
1 2 3 4 5 6
Fr equency
SO2 Squared Coherncy 2001
0.000000
0.2000000.400000
0.6000000.800000
1.0000001.200000
1 2 3 4 5 6
Fr equency
SO2 Cross Amplitude 2001
0.000000
50.000000
100.000000
150.000000
1 2 3 4 5 6
Frequency
SO2 Phase Spect rum 2001
-4.000000
-2.000000
0.000000
2.000000
4.000000
1 2 3 4 5 6
Fr equency
G a i n o f S O 2 2 0 0 1
0. 000000
1. 000000
2. 000000
3. 000000
4. 000000
5. 000000
1 2 3 4 5 6
F r e q u e n c y
M ont hl y Fl uc t ua t i on of S O2
0
10
20
30
40
Apr May June July Aug Sep Oct Nov Dec Jan Feb Mar
M ont h
Pe r io d o g r am o f SO2 2001
0 .000000
100 .000000
200 .000000
300 .000000
400 .000000
500 .000000
1 2 3 4 5 6
F r e q u e n c y
143
M o n t h l y F l u c t u a t i o n o f N O 2
02 04 06 08 0
1 0 01 2 01 4 01 6 0
Ap
r
Ma
y
Ju
ne
Ju
ly
Au
gS
ep
Oc t
No
vD
ec
Ja
nF
eb
Ma
r
M o n t h
Po
llu
tan
t
P er i odogr am of NO2 2001
0.000000
1000.000000
2000.000000
3000.000000
4000.000000
5000.000000
1 2 3 4 5 6
F r e q u e n c y
S pe c t r a l De nsi t y of NO2 2 0 0 1
0.000000
1000.000000
2000.000000
3000.000000
4000.000000
1 2 3 4 5 6
F r e q u e n c y
C r o s s P e r i o d o g r a m o f N O 2 2 0 0 1
-2000. 000000
-1000. 000000
0. 000000
1000. 000000
2000. 000000
3000. 000000
4000. 000000
5000. 000000
1 2 3 4 5 6
F r e q u e n c y
Cross Densit y of NO2 2001
-1000.000000
0.000000
1000.000000
2000.000000
3000.000000
4000.000000
1 2 3 4 5 6
Fr equency
NO2 Cross amplitude 2001
0.000000
1000.000000
2000.000000
3000.000000
4000.000000
1 2 3 4 5 6
Fr e que nc y
Squared Coherncy of NO2 2001
0.000000
0.500000
1.000000
1.500000
1 2 3 4 5 6
Fr e que nc y
N O 2 P h a s e s p e c t r u m 2 0 0 1
- 3 .0 0 0 0 0 0
- 2 .0 0 0 0 0 0
- 1.0 0 0 0 0 0
0 .0 0 0 0 0 0
1.0 0 0 0 0 0
2 .0 0 0 0 0 0
1 2 3 4 5 6
F r e q u e n c y
G a in o f NO 2 2 0 0 1
0 .0 0 0 0 0 00 .5 0 0 0 0 01.0 0 0 0 0 0
1.5 0 0 0 0 02 .0 0 0 0 0 02 .5 0 0 0 0 0
1 2 3 4 5 6
F r e q u e n c y
144
S P M S p e c t r a l P e r i o d o g r a m 2 0 0 2
0.000000
10000.000000
20000.000000
30000.000000
40000.000000
50000.000000
60000.000000
1 2 3 4 5 6
F r e q u e n c y
SPM Sp ect ral D ensit y 2 0 0 2
0.0000005000.00000010000.00000015000.00000020000.00000025000.00000030000.00000035000.000000
1 2 3 4 5 6
Fr equency
S P M Cr oss P e r i odogr a m 2 0 0 2
-10000.000000
0.000000
10000.000000
20000.000000
30000.000000
40000.000000
50000.000000
1 2 3 4 5 6Fr equency
S P M c r oss De nsi t y 2 0 0 2
0.000000
10000.000000
20000.000000
30000.000000
40000.000000
1 2 3 4 5 6
Fr equency
Cr oss Ampl i t ude of SPM 2002
0.000000
10000.000000
20000.000000
30000.000000
40000.000000
1 2 3 4 5 6
Fr equency
SPM co herncy 2 0 0 2
0.000000
0.2000000.400000
0.600000
0.8000001.000000
1.200000
1 2 3 4 5 6
Fr equency
G a i n o f S P M 2 0 0 2
0 . 0 0 0 0 0 0
1 . 0 0 0 0 0 0
2 . 0 0 0 0 0 0
3 . 0 0 0 0 0 0
1 2 3 4 5 6
F r e q u e n c y
Monthly Fluctuation of SPM 2002
0100
200300
400500
Apr May June July Aug Sep Oct Nov Dec Jan Feb Mar
M ont h
SPM Phase 2002
-1.000000
-0.500000
0.000000
0.500000
1.000000
1 2 3 4 5 6
Fr equency
145
S pe c t r a l P e r i odogr a m of RP M 2 0 0 2
0.000000
5000.000000
10000.000000
15000.000000
20000.000000
1 2 3 4 5 6
Fr equency
S pe c t r a l De nsi t y of RP M 2 0 0 2
0.000000
5000.000000
10000.000000
15000.000000
20000.000000
1 2 3 4 5 6
Fr equency
Cr oss P e r i odogr a m of RP M 2 0 0 2
-5000.000000
0.000000
5000.000000
10000.000000
15000.000000
20000.000000
1 2 3 4 5 6
Fr equency
Cross Density of RPM 2002
0.000000
5000.000000
10000.000000
15000.000000
1 2 3 4 5 6
Fr e que nc y
Cross Amplitude of RPM 2002
0.000000
5000.000000
10000.000000
15000.000000
1 2 3 4 5 6
Fr e que nc y
Sq uared C o herncy o f R PM 2 0 0 2
0.000000
0.200000
0.400000
0.600000
0.800000
1.000000
1.200000
1 2 3 4 5 6
Fr equency
P hase S pect r um of RP M 2002
-1.000000
-0.500000
0.000000
0.500000
1.000000
1.500000
1 2 3 4 5 6
Fr equency
Gain of RPM 2002
0.000000
0.500000
1.000000
1.500000
2.000000
1 2 3 4 5 6
F r e q u e n c y
Monthly Fuctuation of RPM 2002
0100
200300
400500
Apr May June July Aug Sep Oct Nov Dec Jan Feb Mar
M ont h
146
P e r i odogr a m of S O2 2 0 0 2
0.000000
20.000000
40.000000
60.000000
80.000000
1 2 3 4 5 6
Fr equency
D ensit y o f SO2 2 0 0 2
0.000000
10.000000
20.000000
30.000000
40.000000
50.000000
1 2 3 4 5 6
Fr equency
C r oss P e r i odogr a m of S O2 2 0 0 2
-20.000000-10.0000000.00000010.00000020.00000030.00000040.00000050.000000
1 2 3 4 5 6
Fr equency
C r oss D e nsi t y o f S O2 2 0 0 2
-10.000000
0.000000
10.000000
20.000000
30.000000
1 2 3 4 5 6
Fr e que nc y
Cr oss Ampl i t ude of S O2 2 0 0 2
0.0000005.00000010.00000015.00000020.00000025.00000030.000000
1 2 3 4 5 6
Fr equency
S qua r e d Cohe r nc y os S O2 2 0 0 2
0.000000
0.2000000.400000
0.6000000.800000
1.0000001.200000
1 2 3 4 5 6
Fr equency
Phase Sp ect rum o f SO2 2 0 0 2
-4.000000-3.000000-2.000000-1.0000000.0000001.0000002.0000003.0000004.000000
1 2 3 4 5 6
Fr equency
Gai n of SO2 2002
0.000000
1.000000
2.000000
3.000000
4.000000
1 2 3 4 5 6
Fr e que nc y
Monthly Fluctuation of SO2 2002
0
5
10
15
20
Apr May June July Aug Sep Oct Nov Dec Jan Feb Mar
M o nt h
147
Per i odogr am of NO2 2002
0.000000
2000.000000
4000.000000
6000.000000
1 2 3 4 5 6
Fr equency
De nsi t y of NO2 2 0 0 2
0.000000
1000.000000
2000.000000
3000.000000
4000.000000
1 2 3 4 5 6
Fr equency
C r o ss P e r i o d o g r a m o f N O 2 2 0 0 2
0.000000500.0000001000.0000001500.0000002000.0000002500.0000003000.0000003500.0000004000.000000
1 2 3 4 5 6
F r e q u e n c y
C r o ss D e n si t y o f N O 2 2 0 0 2
0.000000500.0000001000.0000001500.000000
2000.0000002500.0000003000.0000003500.000000
1 2 3 4 5 6
F r e q u e n c y
Cross Amplitude of NO2 2002
0.000000
1000.000000
2000.000000
3000.000000
4000.000000
1 2 3 4 5 6
Fr equency
S q u a r e d C o h e r n c y o f N O 2 2 0 0 2
0.000000
0.200000
0.400000
0.600000
0.800000
1.000000
1.200000
1 2 3 4 5 6
F r e q u e n c y
P ha se S pe c t r um of NO2 2 0 0 2
-1.000000
-0.500000
0.000000
0.500000
1.000000
1.500000
1 2 3 4 5 6
Fr equency
Ga i n of NO2 2 0 0 2
0.000000
1.000000
2.000000
3.000000
4.000000
1 2 3 4 5 6
F r e q u e n c y
Monthly Fluctuation of NO2 2002
0
50
100
150
200
Apr May June July Aug Sep Oct Nov Dec Jan Feb Mar
M ont h
148
P er i odogr am of SP M 2003
0.000000
10000.000000
20000.000000
30000.000000
40000.000000
1 2 3 4 5 6
F r e q u e n c y
Densi t y of SP M 2003
0.000000
10000.000000
20000.000000
30000.000000
1 2 3 4 5 6
F r e q u e n c y
Cr oss P er i odogr am of SP M 2003
0.000000
10000.000000
20000.000000
30000.000000
40000.000000
1 2 3 4 5 6
F r e q u e n c y
Cr oss De nsi t y of S P M 2 0 0 3
0.0000005000.00000010000.00000015000.00000020000.00000025000.000000
1 2 3 4 5 6
Fr equency
C r o ss A m p l i t u d e o f S P M 2 0 0 3
0 . 000000
5000 . 000000
10000 . 000000
15000 . 000000
20000 . 000000
25000 . 000000
1 2 3 4 5 6
F r e q u e n c y
Coherncy of SPM 2003
0.000000
0.500000
1.000000
1.500000
1 2 3 4 5 6
Fr equency
Phase of SPM 2003
-0.500000
0.000000
0.500000
1.000000
1 2 3 4 5 6
Fr e que nc y
G a i n o f SP M 2 0 0 3
0. 000000
0. 500000
1. 000000
1. 500000
2. 000000
1 2 3 4 5 6
F r e q u e n c y
Monthly Fluctuation of SPM 2003
0
100
200
300
400
500
Apr May June July Aug Sep Oct Nov Dec Jan Feb Mar
M ont h
149
Periodogram of RPM 2003
0.000000
5000.000000
10000.000000
15000.000000
20000.000000
1 2 3 4 5 6
Fr equency
Density of RPM 2003
0.000000
5000.000000
10000.000000
15000.000000
1 2 3 4 5 6
Frequency
Cr oss Per iodogr am of RPM 2003
0.000000
5000.000000
10000.000000
15000.000000
1 2 3 4 5 6
Fr equency
C r o s s D e ns i t y o f R P M 2 0 0 3
0.000000
2000.000000
4000.000000
6000.000000
8000.000000
10000.000000
1 2 3 4 5 6
Fr e que nc y
C r o ss A m p l i t u d e o f R P M 2 0 0 3
0.000000
2000.0000004000.000000
6000.0000008000.000000
10000.000000
1 2 3 4 5 6
Fr e que nc y
Square d Cohe rncy of RPM 2003
0.0000000.2000000.4000000.6000000.8000001.0000001.200000
1 2 3 4 5 6
F r e q u e n c y
Pgase Spect rum of RPM 2003
-0.500000
0.000000
0.500000
1.000000
1 2 3 4 5 6
Fr equency
G a i n o f R P M 2 0 0 3
0.000000
0.500000
1.000000
1.500000
1 2 3 4 5 6
F r e q u e n c y
Monthly Fluctuation of RPM 2003
050
100150
200250
300
Apr May June July Aug Sep Oct Nov Dec Jan Feb Mar
M ont h
150
P e r i odogr a m of S O2 2 0 0 3
0.000000
20.000000
40.000000
60.000000
80.000000
1 2 3 4 5 6
Fr equency
D ensit y o f SO2 2 0 0 3
0.000000
10.000000
20.000000
30.000000
40.000000
50.000000
1 2 3 4 5 6
Fr equency
Cr oss P e r i odogr a m of S O2 2 0 0 3
-10.000000
0.00000010.000000
20.00000030.000000
40.00000050.000000
60.000000
1 2 3 4 5 6
Fr equency
Cross Density of SO2 2 0 0 3
0.000000
10.000000
20.000000
30.000000
40.000000
1 2 3 4 5 6
Fr equency
Cr oss Ampl i t ude of SO2 2003
0.000000
20.000000
40.000000
60.000000
1 2 3 4 5 6
F r e q u e n c y
S qua r e d Cohe r nc y of S O2 2 0 0 3
0.0000000.2000000.4000000.6000000.8000001.0000001.200000
1 2 3 4 5 6
Fr equency
Phase Spect rum of SO2 2003
-1.500000
-1.000000
-0.500000
0.000000
0.500000
1.000000
1.500000
1 2 3 4 5 6
Fr equency
Ga i n of S O2 2 0 0 3
0.000000
1.000000
2.000000
3.000000
4.000000
1 2 3 4 5 6
Fr equency
Monthly Fluctuation of SO2 2003
0
5
10
15
20
Apr May June July Aug Sep Oct Nov Dec Jan Feb Mar
Fr e que nc y
151
P e r i odogr a m of NO2 2 0 0 3
0.000000
500.000000
1000.000000
1500.000000
2000.000000
1 2 3 4 5 6
Fr equency
De nsi t y of NO2 2 0 0 3
0.000000
500.000000
1000.000000
1500.000000
1 2 3 4 5 6
Fr equency
Cr oss P e r i odogr a m of NO2 2 0 0 3
-200.000000
0.000000
200.000000
400.000000
600.000000
800.000000
1 2 3 4 5 6
Fr equency
C r o ss D e n si t y o f N O2 2 0 0 3
0.000000
100.000000
200.000000
300.000000
400.000000
1 2 3 4 5 6
Fr e que nc y
Cross Amplitude of NO2 2003
0.000000100.000000200.000000300.000000400.000000500.000000
1 2 3 4 5 6
Fr equency
Squared Coherncy of NO2 2003
0.0000000.200000
0.4000000.6000000.800000
1.0000001.200000
1 2 3 4 5 6
Fr equency
Phase Spectrum of NO2 2003
-1.500000
-1.000000
-0.500000
0.000000
0.500000
1.000000
1.500000
1 2 3 4 5 6
Fr equency
Gai n of No2 2003
0.000000
0.200000
0.400000
0.600000
0.800000
1.000000
1 2 3 4 5 6
F r e q u e n c y
Monthly Fluctuation of NO2 2003
0
50
100
150
Apr May June July Aug Sep Oct Nov Dec Jan Feb Mar
M ont h
152
Per i odogr am of SPM 2004
0.000000
20000.000000
40000.000000
60000.000000
1 2 3 4 5 6
Fr e que nc y
De si t y of S P M 2 0 0 4
0.000000
10000.000000
20000.000000
30000.000000
40000.000000
1 2 3 4 5 6
Fr equency
Cr oss P er i odogr am of SP M 2004
-20000.000000
0.000000
20000.000000
40000.000000
60000.000000
1 2 3 4 5 6
F r e q u e n c y
C r o ss D e n si t y o f S P M 2 0 0 4
-10000.000000
0.000000
10000.000000
20000.000000
30000.000000
40000.000000
1 2 3 4 5 6
F r e q u e n c y
C r o s s A m p l i t u d e o f S P M 2 0 0 4
0.000000
10000.000000
20000.000000
30000.000000
40000.000000
1 2 3 4 5 6
F r e q u e n c y
C o h e r n c y o f S P M 2 0 0 4
0.000000
0.500000
1.000000
1.500000
1 2 3 4 5 6
Fr e que nc y
P ha se of S P M 2 0 0 4
-1.000000
-0.500000
0.000000
0.500000
1.000000
1 2 3 4 5 6
F r e q u e n c y
Ga i n of SP M 2 0 0 4
0. 000000
0. 500000
1. 000000
1. 500000
2. 000000
1 2 3 4 5 6
F r e q u e n c y
M ont hl y Fl uc t ua t i on of S P M 2 0 0 4
0100200300400500
Apr May June July Aug Sep Oct Nov Dec Jan Feb Mar
M ont h
153
P e r i odogr a m of R P M 2 0 0 4
0. 000000
5000. 000000
10000. 000000
15000. 000000
1 2 3 4 5 6
F r e q u e n c y
Densi t y of RP M 2004
0.000000
5000.000000
10000.000000
15000.000000
1 2 3 4 5 6
F r e q u e n c y
Cross Per iodogram of RPM 2004
-5000.000000
0.000000
5000.000000
10000.000000
15000.000000
1 2 3 4 5 6
Fr equency
Cr oss Densi ty of RPM 2004
-5000.000000
0.000000
5000.000000
10000.000000
1 2 3 4 5 6
Fr equency
C r os s A mpl i t ude of R P M 2 0 0 4
0. 000000
5000. 000000
10000. 000000
1 2 3 4 5 6
F r e q u e n c y
C o h e r n c y o f R P M 2 0 0 4
0 . 000000
0 . 500000
1 . 000000
1 . 500000
1 2 3 4 5 6
F r e q u e n c y
P h a se o f R P M 2 0 0 4
-0.500000
0.000000
0.500000
1.000000
1.500000
1 2 3 4 5 6
F r e q u e n c y
Ga i n of R P M 2 0 0 4
0.000000
0.5000001.000000
1.500000
2.000000
1 2 3 4 5 6
F r e q u e n c y
Monthly Fluctuation of RPM 2004
0
100
200
300
Apr May June July Aug Sep Oct Nov Dec Jan Feb Mar
M ont h
154
P er i odogr am of SO2 2004
0.000000
50.000000
100.000000
150.000000
200.000000
1 2 3 4 5 6
F r e q u e n c y
D e n si t y o f S O 2 2 0 0 4
0.000000
50.000000
100.000000
150.000000
1 2 3 4 5 6
Fr e que nc y
Cross Periodogram of SO2 2004
-100.000000
0.000000
100.000000
200.000000
1 2 3 4 5 6
Fr equency
Cr oss De nsi t y of S O2 2 0 0 4
0.00000020.00000040.00000060.000000
80.000000100.000000120.000000
1 2 3 4 5 6Fr equency
C r o s s A mp l i t u d e o f S O 2 2 0 0 4
0 . 0 0 0 0 0 0
5 0 . 0 0 0 0 0 0
1 0 0 . 0 0 0 0 0 0
1 5 0 . 0 0 0 0 0 0
1 2 3 4 5 6
F r e q u e n c y
Cohe r nc y of S O2 2 0 0 4
0.000000
0.500000
1.000000
1.500000
1 2 3 4 5 6
Fr equency
Phase Of SO2 2004
-1.500000-1.000000-0.5000000.0000000.5000001.0000001.5000002.000000
1 2 3 4 5 6
Fr equency
G a i n o f SO 2 2 0 0 4
0 . 000000
1 . 000000
2 . 000000
3 . 000000
1 2 3 4 5 6
F r e q u e n c y
M o nt hly F luct uat io n o f SO2 2 0 0 4
05101520253035
Apr May June July Aug Sep Oct Nov Dec Jan Feb Mar
M ont h
155
P er i odogr am of NO2 2004
0.0000001000.0000002000.000000
3000.0000004000.000000
1 2 3 4 5 6
F r e q u e n c y
Densi t y of NO2 2004
0.000000
1000.000000
2000.000000
3000.000000
1 2 3 4 5 6
F r e q u e n c y
Cr oss Per i odogr am of NO2 2004
-1000.000000
0.000000
1000.000000
2000.000000
3000.000000
4000.000000
1 2 3 4 5 6
Fr e que nc y
Cr oss Densi t y of NO2 2004
-5000.000000
0.000000
5000.000000
10000.000000
1 2 3 4 5 6
F r e q u e n c y
C r o s s A mp l i t u d e o f N O 2 2 0 0 4
0 . 000000
500 . 000000
1000 . 000000
1500 . 000000
2000 . 000000
1 2 3 4 5 6
F r e q u e n c y
C o h e r n c y o f N O 2 2 0 0 4
0.000000
0.500000
1.000000
1.500000
1 2 3 4 5 6
Fr e que nc y
P h a s e o f N O 2 2 0 0 4
-0 . 500000
0 . 000000
0 . 500000
1 . 000000
1 . 500000
1 2 3 4 5 6
F r e q u e n c y
Gai n of R P M 2004
0. 0000001. 000000
2. 000000
1 2 3 4 5 6
Fr equen cy
M ont hl y Fl uc t ua t i on of NO2 2 0 0 4
0
50
100
150
Apr May June July Aug Sep Oct Nov Dec Jan Feb Mar
M ont h
156
P er i odogr am of SP M 2005
0.000000
50000.000000
100000.000000
150000.000000
1 2 3 4 5 6
F r e q u e n c y
Densi t y of SP M 2005
0.000000
50000.000000
100000.000000
1 2 3 4 5 6
F r e q u e n c y
Cr oss P er i odogr am of SP M 2005
-50000.000000
0.000000
50000.000000
100000.000000
1 2 3 4 5 6
F r e q u e n c y
Cr oss Densi t y of SP M 2005
-10000.000000
0.000000
10000.000000
20000.000000
30000.000000
1 2 3 4 5 6
F r e q u e n c y
C r o ss A m p l i t u d e o f S P M 2 0 0 5
0.000000
10000.000000
20000.000000
30000.000000
1 2 3 4 5 6
F r e q u e n c y
Coher ncy of SP M 2005
0.000000
0.500000
1.000000
1.500000
1 2 3 4 5 6
F r e q u e n c y
Phase of SPM 2005
-1.000000
-0.500000
0.000000
0.500000
1.000000
1 2 3 4 5 6
Fr equency
Gai n of SP M 2005
0.0000000.5000001.0000001.500000
1 2 3 4 5 6
Fr equency
M ont hl y Fl uc t ua t i on of S P M 2 0 0 5
0.00200.00400.00600.00800.001000.00
Apr May June July Aug Sep Oct Nov Dec Jan Feb Mar
M ont h
157
Per i odogr am of RPM 2005
0.000000
5000.000000
10000.000000
15000.000000
1 2 3 4 5 6
Fr e que nc y
Densi t y of RP M 2005
0.000000
2000.000000
4000.000000
6000.000000
8000.000000
10000.000000
1 2 3 4 5 6
F r e q u e n c y
Cr oss P er i odogr am of RP M 2005
-5000.000000
0.000000
5000.000000
10000.000000
15000.000000
1 2 3 4 5 6
F r e q u e n c y
C r o ss D e n si t y o f R P M 2 0 0 5
0 . 000000
2000 . 000000
4000 . 000000
6000 . 000000
8000 . 000000
1 2 3 4 5 6
F r e q u e n c y
Cr oss Ampl i t ude of RP M 2 0 0 5
0.000000
2000.000000
4000.000000
6000.000000
8000.000000
1 2 3 4 5 6
Fr equency
Coher ncy of RPM 2005
0.000000
0.5000001.000000
1.500000
1 2 3 4 5 6
Fr equency
Phase of RPM 2005
-0.600000
-0.400000
-0.200000
0.000000
0.200000
0.400000
0.600000
1 2 3 4 5 6
Fr equency
Gain of RPM 2005
0.000000
0.500000
1.000000
1.500000
1 2 3 4 5 6
Fr equency
M ont hl y Fl uc t ua t i on of RP M 2 0 0 5
0
100
200
300
Apr May June July Aug Sep Oct Nov Dec Jan Feb Mar
M ont h
158
Per i odogr am of SO2 2005
0.00000
50.00000
100.00000
150.00000
1 2 3 4 5 6
Fr e que nc y
Per iodogr am of SO2 2005
0.000000
50.000000
100.000000
150.000000
1 2 3 4 5 6
Fr equency
Cr oss P er i odogr am of SO2 2005
-50.000000
0.000000
50.000000
100.000000
150.000000
1 2 3 4 5 6
F r e q u e n c y
C r o ss D e n si t y o f S O 2 2 0 0 5
0.00000020.000000
40.00000060.000000
80.000000100.000000
1 2 3 4 5 6
F r e q u e n c y
Cr oss Ampl i tude of SO2 2005
0.000000
50.000000
100.000000
1 2 3 4 5 6
Fr equen cy
C o h e r n c y o f S O 2 2 0 0 5
0 . 0 0 0 0 0 0
0 . 2 0 0 0 0 0
0 . 4 0 0 0 0 0
0 . 6 0 0 0 0 0
0 . 8 0 0 0 0 0
1 . 0 0 0 0 0 0
1 . 2 0 0 0 0 0
1 2 3 4 5 6
F r e q u e n c y
P hase of SO2 2005
-1.500000
-1.000000
-0.500000
0.000000
0.500000
1.000000
1 2 3 4 5 6
F r e q u e n c y
Ga i n of SO2 2 0 0 5
0. 000000
1. 000000
2. 000000
3. 000000
1 2 3 4 5 6
Fr equen cy
M ont hl y Fl uct uat i on of SO2 2005
0
10
20
30
40
Apr May June Jul y Aug Sep Oct Nov Dec Jan Feb Mar
M o n t h
159
Per i odogr am of NO2 2005
0.000000
500000.000000
1000000.000000
1500000.000000
2000000.000000
1 2 3 4 5 6
Fr e que nc y
De nsi t y of NO2 2 0 0 5
0.000000
500000.000000
1000000.000000
1500000.000000
2000000.000000
1 2 3 4 5 6 7
Fr equency
C r os s P e r i odogr a m of N O2 2 0 0 5
-60000.000000
-40000.000000
-20000.000000
0.000000
20000.000000
40000.000000
1 2 3 4 5 6
F r e q u e n c y
C r o ss D e n si t y o f N O 2 2 0 0 5
-25000.000000
-20000.000000-15000.000000
-10000.000000
-5000.000000
0.0000005000.000000
10000.000000
15000.000000
1 2 3 4 5 6
F r e q u e n c y
Cr oss Ampl i t ude of NO2 2 0 0 5
0.0000005000.00000010000.00000015000.00000020000.00000025000.00000030000.000000
1 2 3 4 5 6
Fr equency
Cohe r nc y of NO2 2 0 0 5
0.0000000.2000000.4000000.6000000.8000001.0000001.200000
1 2 3 4 5 6
Fr equency
P ha se of NO2 2 0 0 5
-3.000000-2.000000-1.0000000.0000001.0000002.0000003.0000004.000000
1 2 3 4 5 6
Fr equency
Gai n of NO2 2005
0.000000
1.000000
2.000000
1 2 3 4 5 6
Fr equen cy
M ont hl y Fl uct uat i on of NO2 2005
0
50
100
150
Apr May June Jul y Aug Sep Oct Nov Dec Jan Feb Mar
M o n t h
160
P er i odogr am of SP M 2006
0.00000010000.000000
20000.00000030000.000000
40000.00000050000.000000
60000.000000
1 2 3 4 5 6
F r e q u e n c y
Densi t y of SP M 2006
0.000000
10000.000000
20000.000000
30000.000000
40000.000000
1 2 3 4 5 6
F r e q u e n c y
Cr oss P er i odogr am of SP M 2006
-10000.000000
0.000000
10000.000000
20000.000000
30000.000000
1 2 3 4 5 6
F r e q u e n c y
Cr oss De nsi t y of S P M 2 0 0 6
-5000.000000
0.000000
5000.000000
10000.000000
15000.000000
20000.000000
1 2 3 4 5 6
F r e q u e n c y
C r o s s A mp l i t u d e o f SP M 2 0 0 6
0 . 0 0 0 0 0 0
10 0 0 0 . 0 0 0 0 0 0
2 0 0 0 0 . 0 0 0 0 0 0
3 0 0 0 0 . 0 0 0 0 0 0
1 2 3 4 5 6
F r e q u e n c y
C o h e r n c y o f S P M 2 0 0 6
0 . 0 0 0 0 0 0
0 . 5 0 0 0 0 0
1. 0 0 0 0 0 0
1. 5 0 0 0 0 0
1 2 3 4 5 6
F r e q u e n c y
P hase of SP M 2006
-0.500000
0.000000
0.500000
1.000000
1 2 3 4 5 6
F r e q u e n c y
G a i n o f SP M 2 0 0 6
0.0000000.5000001.0000001.500000
1 2 3 4 5 6
F r e u e n c y
M ont hl y Fl uc t ua t i on of S P M 2 0 0 6
0100200300400500600
Apr May June July Aug Sep Oct Nov Dec Jan Feb Mar
M ont h
161
P e r i odogr a m of RP M 2 0 0 6
0.000000
5000.000000
10000.000000
15000.000000
20000.000000
1 2 3 4 5 6
Fr equency
De nsi t y of RP M 2 0 0 6
0.0000002000.000000
4000.0000006000.000000
8000.00000010000.000000
12000.000000
1 2 3 4 5 6
F r e q u e n c y
C r o s s P e r i o d o g r a m o f R P M 2 0 0 6
- 5 0 00 . 0 00000
0 . 0 00000
5000 . 0 00000
10000 . 0 00000
15000 . 0 00000
1 2 3 4 5 6
Fr e que n c y
C r o s s D e n s i t y o f R P M 2 0 0 6
0 . 0 0 0 0 00
2000 . 0 0 0 0 00
4000 . 0 0 0 0 00
6000 . 0 0 0 0 00
8000 . 0 0 0 0 00
10000 . 0 0 0 0 00
1 2 3 4 5 6
F r e q u e n c y
Cr oss Ampl i tude of RPM 2006
0.000000
5000.000000
10000.000000
1 2 3 4 5 6
f r e que nc y
C o h e r n c y o f R P M 2 0 0 6
0.000000
0.200000
0.400000
0.600000
0.800000
1.000000
1.200000
1 2 3 4 5 6
Fr e que nc y
P hase of RP M 2006
-1.000000
-0.500000
0.000000
0.500000
1.000000
1 2 3 4 5 6
F r e q u e n c y
Gai n of RP M 2006
0.000000
0.500000
1.000000
1.500000
2.000000
1 2 3 4 5 6
F r e q u e n c y
M ont hl y Fl uct uat i on of RP M 2006
0
100
200
300
Apr May June Jul y Aug Sep Oct Nov Dec Jan Feb Mar
M o n t h
162
P er i odogr am of SO2 2006
0.00000020.00000040.00000060.00000080.000000
1 2 3 4 5 6
F r e q u e n c y
D e n si t y o f S O 2 2 0 0 6
0.000000
20.000000
40.000000
60.000000
80.000000
1 2 3 4 5 6
F r e q u e n c y
Cr oss P er i odogr am of SO2 2006
-50.000000
0.000000
50.000000
100.000000
1 2 3 4 5 6
F r e q u e n c y
Cr oss Densi t y of SO2 2006
0.000000
50.000000
100.000000
1 2 3 4 5 6
F r e q u e n c y
Cr oss Ampl i t ude of SO2 2006
0.000000
20.000000
40.000000
60.000000
80.000000
1 2 3 4 5 6
F r e q u e n c y
C ohe r nc y of SO2 2 0 0 6
0. 000000
0. 500000
1. 000000
1. 500000
1 2 3 4 5 6
F r e q u e n c y
P hase of SO2 2006
-0.200000
0.000000
0.200000
0.400000
0.600000
1 2 3 4 5 6
F r e q u e n c y
Gai n of RP M 2006
0.000000
1.000000
2.000000
3.000000
4.000000
1 2 3 4 5 6
F r e q u e n c y
M ont hl y Fl uc t ua t i on of S O2 2 0 0 6
0
10
20
30
Apr May June July Aug Sep Oct Nov Dec Jan Feb Mar
M ont h
163
P er i odogr am of NO2 2006
0.000000
1000.0000002000.000000
3000.000000
1 2 3 4 5 6
F r e q u e n c y
D e n s i t y o f N O 2 2 0 0 6
0 . 000000
500 . 000000
1000 . 000000
1500 . 000000
1 2 3 4 5 6
F r e q u e n c y
Cr oss P er i odogr am of NO2 2006
-500.000000
0.000000
500.000000
1000.000000
1500.000000
1 2 3 4 5 6
F r e q u e n c y
Cr oss Densi t y of NO2 2006
-200.000000
0.000000
200.000000
400.000000
600.000000
800.000000
1 2 3 4 5 6
F r e q u e n c e
C r o s s A mp l i t u d e o f N O 2 2 0 0 6
0 . 0 0 0 0 0 0
5 0 0 . 0 0 0 0 0 0
10 0 0 . 0 0 0 0 0 0
1 2 3 4 5 6
F r e q u e n c y
Coher ncy of NO2 2006
0.000000
0.500000
1.000000
1.500000
1 2 3 4 5 6
F r e q u e n c y
P hase of NO2 2006
-2.000000
-1.000000
0.000000
1.000000
1 2 3 4 5 6
F r e q u e n c y
Gai n of NO2 2006
0.000000
1.000000
2.000000
3.000000
1 2 3 4 5 6
F r e q u e n c y
M o n t h l y F l u c t u a t i o n o f N O 2 2 0 0 6
0
2 0
4 0
6 0
8 0
1 0 0
1 2 0
Ap
r
Ma
y
Ju
ne
Ju
l y
Au
g
Se
p
Oc
t
No
v
De
c
Ja
n
Fe
b
Ma
r
M o n t h
Po
llu
tio
n
164
P er i odogr am of SP M 2007
0.00
20000.00
40000.00
60000.00
1 2 3 4 5 6
F r e q u e n c y
Densi t y of SP M 2007
0.000000
20000.000000
40000.000000
1 2 3 4 5 6
Fr equen cy
Cr oss P er i odogr am of SP M 2007
-20000.000000
0.000000
20000.000000
40000.000000
1 2 3 4 5 6
Fr equen cy
Cr oss Densi t y of SP M 2007
0.000000
10000.000000
20000.000000
30000.000000
1 2 3 4 5 6
F r e q u e n c y
C r os s A mpl i t ude of SP M 2 0 0 7
0.00000010000.00000020000.00000030000.000000
1 2 3 4 5 6
Fr e que nc y
Coher ncy of SP M 2007
0.000000
0.500000
1.000000
1.500000
1 2 3 4 5 6
F r e q u e n c y
P hase of SP M 2007
-1.000000
-0.500000
0.000000
0.500000
1.000000
1 2 3 4 5 6
F r e q u e n c y
Gai n of SP M 2007
0.000000
0.500000
1.000000
1.500000
2.000000
1 2 3 4 5 6
F r e q u e n c y
M ont hl y Fl uct uat i on of SP M 2007
0
200
400
600
Apr May June Jul y Aug Sep Oct Nov Dec Jan Feb Mar
M o n t h
165
P er i odogr am of RP M 2007
0.000000
5000.000000
10000.000000
15000.000000
20000.000000
1 2 3 4 5 6
F r e q u e n c y
Densi t y of RP M 2007
0.000000
5000.000000
10000.000000
15000.000000
1 2 3 4 5 6
F r e q u e n c y
Cr oss Per i odogr am of RPM 2007
-5000.000000
0.000000
5000.000000
10000.000000
15000.000000
1 2 3 4 5 6
Fr equency
Cr oss Densi t y of RP M 2007
0.000000
2000.000000
4000.000000
6000.000000
8000.000000
10000.000000
1 2 3 4 5 6
F r e q u e n c y
C r os s A mpl i t ude of R P M 2 0 0 7
0. 000000
5000. 000000
10000. 000000
1 2 3 4 5 6
Fr equen cy
Coher ncy of RP M 2007
0.000000
0.500000
1.000000
1.500000
1 2 3 4 5 6
F r e q u e n c y
P hase of RP M 2007
-0.500000
0.000000
0.500000
1.000000
1 2 3 4 5 6
F r e q u e n c y
Ga i n of R P M 2 0 0 7
0. 000000
1. 000000
2. 000000
1 2 3 4 5 6
Fr equen cy
M ont hl y Fl uc t ua t i on of RP M 2 0 0 7
0
100
200
300
Apr May June July Aug Sep Oct Nov Dec Jan Feb Mar
M ont h
166
P er i odogr am of SO2 2007
0.000000
5.000000
10.000000
15.000000
1 2 3 4 5 6
F r e q u e n c y
Densi t y of SO2 2007
0.000000
5.000000
10.000000
1 2 3 4 5 6
F r e q u e n c y
Cr oss P e r i odogr a m of S O2 2 0 0 7
-2.000000
0.000000
2.000000
4.000000
6.000000
8.000000
10.000000
1 2 3 4 5 6
F r e q u e n c y
Cr oss De nsi t y of S O2 2 0 0 7
0.000000
2.000000
4.000000
6.000000
8.000000
1 2 3 4 5 6
Fr equency
Coher ncy of SO2 2007
0.000000
0.500000
1.000000
1.500000
1 2 3 4 5 6
Fr equency
C r o s s A mp l i t u d e o f SO 2 2 0 0 7
0 . 0 0 0 0 0 0
5 . 0 0 0 0 0 0
10 . 0 0 0 0 0 0
1 2 3 4 5 6
F r e q u e n c y
P h a s e o f S O 2 2 0 0 7
- 1 . 5 0 0 0 0 0
- 1 . 0 0 0 0 0 0
- 0 . 5 0 0 0 0 0
0 . 0 0 0 0 0 0
0 . 5 0 0 0 0 0
1 . 0 0 0 0 0 0
1 . 5 0 0 0 0 0
1 2 3 4 5 6
F r e q u e n c y
G a i n o f SO 2 2 0 0 7
0 . 0 0 0 0 0 0
2 . 0 0 0 0 0 0
4 . 0 0 0 0 0 0
6 . 0 0 0 0 0 0
1 2 3 4 5 6
F r e q u e c y
M ont hl y Fl uct uat i on of SO2 2007
0
5
10
15
Apr May June Jul y Aug Sep Oct Nov Dec Jan Feb Mar
M o n t h
167
P er i odogr am of NO2 2007
0.000000
500.000000
1000.000000
1500.000000
2000.000000
1 2 3 4 5 6
F r e q u e n c y
Densi t y of NO2 2007
0.000000
500.000000
1000.000000
1500.000000
1 2 3 4 5 6
F r e q u e n c y
Cr oss P er i odogr am of NO2 2007
-500.000000
0.000000
500.000000
1000.000000
1 2 3 4 5 6
F r e q u e n c y
Cr oss Densi t y of NO2 2007
-200.000000
0.000000
200.000000
400.000000
600.000000
1 2 3 4 5 6
F r e q u e n c y
C r o s s A mp l i t u d e o f N O 2 2 0 0 7
0 . 0 0 0 0 0 0
5 0 0 . 0 0 0 0 0 0
10 0 0 . 0 0 0 0 0 0
1 2 3 4 5 6
F r e q u e n c y
C o h e r n c y o f N O 2 2 0 0 7
0 . 000000
0 . 200000
0 . 4000000 . 600000
0 . 800000
1 . 000000
1 . 200000
1 2 3 4 5 6
F r e q u e n c y
P hase of NO2 2007
-1.000000
-0.500000
0.000000
0.500000
1.000000
1 2 3 4 5 6
F r e q u e n c y
G a i n o f N O 2 2 0 0 7
0.0000001.0000002.0000003.000000
1 2 3 4 5 6
Fr e que nc y
M ont hl y Fl uc t ua t i on of NO2 2 0 0 7
0
50
100
150
Apr May June July Aug Sep Oct Nov Dec Jan Feb Mar
M ont h