Bachelor Thesis No: B-018-006

Assessing The Relationship of Air Quality Parameters

on Urban Heat Island Effect in Kathmandu Valley

Palistha Tuladhar,

2018 - 01- 41- 0006

August 2023
Thesis for the Degree of Bachelor of Science in Environment
Management

Assessing The Relationship of Air Quality Parameters

on Urban Heat Island Effect in Kathmandu Valley

Palistha Tuladhar,

2018 - 01- 41- 0006

School of Environmental Science and Management,

Faculty of Science

Pokhara University, Nepal

August 2023

2
Thesis for the Degree of Bachelor of Science in Environment
Management

Assessing The Relationship of Air Quality Parameters

on Urban Heat Island Effect in Kathmandu Valley

Supervised by Prof. Dr. Sanjay Nath Khanal

A thesis submitted in partial fulfilment of the requirements for the degree

of Bachelor of Science in Environment Management.

Palistha Tuladhar,

2018 - 01- 41- 0006

School of Environmental Science and Management,

Faculty of Science

Pokhara University, Nepal

August 2023

3
 DECLARATION

I hereby declare that this study entitled ‘Assessing the Relationship of Air Quality
Parameters on Urban Heat Island Effect in Kathmandu Valley’ is based on my
original research work. Related works on the topic by other researchers have been duly
acknowledged. I owe all the liabilities relating to the accuracy and authenticity of the data
and any other information included hereunder.

Signature
Palistha Tuladhar
P.U. Registration Number: 2018 - 01- 41- 0006
August 2023

ii
 RECOMMENDATION

This is to certify that this thesis entitled ‘Assessing the Relationship of Air Quality
Parameters on Urban Heat Island Effect in Kathmandu Valley’ prepared and
submitted by Palistha Tuladhar in partial fulfillment of the requirements of the degree of
Bachelor of Science (B.Sc.) in Environment Management awarded by Pokhara University,
has been completed under our supervision. We recommend the same for acceptance by
Pokhara University.

Signature
Prof. Dr. Sanjay Nath Khanal
School of Environmental Science and Management (SchEMS)
August 2023

CERTIFICATION

This is to certify that the thesis entitled ‘Assessing the Relationship of Air Quality
Parameters on Urban Heat Island Effect in Kathmandu Valley’ by Palistha Tuladhar
towards partial fulfillment of the degree of Bachelor of Science in Environmental

iii
Management is based on the original research and study under the guidance of Prof. Dr.
Sanjay Nath Khanal. This thesis is property of School of Environmental Science and
Management (SchEMS) and therefore should not be used for the purpose of awarding any
acedimic degree in any other institutions.

Prof. Dr. Sanjay Nath Khanal ……………………………. Date:

(Supervisor)

Mr. Birkha Sunar ……………………………. Date:

(BSc. Program Coordinator)

Associate Prof. Ajay Bhakta Mathema) ………………………… Date:

(Principal)

iv
 ACKNOWLEDGEMENTS

I am sincerely grateful to everyone who contributed to the successful completion of my

thesis. I met a lot of people through my thesis, without their support, guidance and
encouragement, this research would not have been possible. First and foremost, I extend
my heartfelt gratitude to my supervisor, Dr. Sanjay Nath Khanal, for their unwavering
guidance, insights, and support throughout the research. His constructive feedback and
motivation were pivotal in refine my entire thesis. I'd also like to thank him for overseeing
me and connecting me with the correct people for the entirety of my research.

I would like to express my deepest appreciation to Ms. Kripa Khanal without whom the
first idea for the thesis could not have been made. I’d also like to thank Mr. Dhiraj Giri
who was of utmost help to me during the data cleaning and analyzing phase of my
research, without him my data analysis would not have been possible. Similarly, I'd like to
thank Ms. Neevasha Shrestha for encouraging me during this entire journey in completing
my research, she has been there to assist me whenever I encountered a roadblock.

I would also like to thank School of Environmental Science and Management for
providing me this opportunity to carry out the research. A special appreciation to my
academic program coordinator, Mr. Birkha Sunar, who has from Day 1 encouraged my
thesis and work.

My sincere thanks go to the numerous teachers and professionals who shared their
expertise and knowledge which greatly enriched this study. Lastly, I would also like to
acknowledge the support and understanding of family and friends throughout this journey
for their continuous invaluable support, encouragement, and guidance to accomplish my
study successfully.

…………………………….
Palistha Tuladhar
PU Registration No. 2018 - 01- 41- 0006
August 2023

v
 ABSTRACT

Nepal's rapid urbanization, reaching 66.08% in 2021, brings challenges like intensified
energy use, altered land use, expanded built-up areas, and reduced greenery, resulting in
elevated temperatures and heat-related stresses. This study is centered within the
Kathmandu Valley, focusing primarily on two key locations: Ratnapark and Kirtipur
(Tribhuvan University). This study investigates the temperature-air quality (PM2.5, TSP)
relationship across seasons over six years to validate the presence of Urban Heat Island
(UHI), as air quality seems to have a strong link with temperature. The data collection
process utilizes government air quality monitoring stations in Ratnapark and Tribhuvan
University. Additionally, weather data was sourced from the nearest DHM (Department of
Hydrology and Meteorology) weather stations in Tribhuvan International Airport (TIA)
and Khokana. Data from air quality and weather monitoring stations were analyzed using
Google Sheets and SPSS. However, it is important to note that data cleaning was necessary
due to the presence of several data gaps and inconsistencies. Urban-rural temperature
disparities, calculated through Urban Heat Island Intensity (UHII), averaged 0.8°C
annually, with more significant differences in urban centers. Pearson's correlation analysis
revealed varying correlations between air quality and temperature, the relationship was
also examined closely across different seasons. The findings were intriguing, revealing
diverse seasonal correlations and variations indicating that seasonal variations exist in
these relationships. Notably, correlation coefficients between PM2.5, TSP concentrations,
and temperature for Ratnapark were -0.578 and -0.065 at a 0.01 (two-tailed) significance
level with stronger correlations established in Spring and Autumn seasons. Different
associations were explored, and a six-year trendline established a link between air quality
and temperature impacting UHI. The results were enough to prove that air quality has a
relationship with temperature, which affected UHI. This underscores the urgency of
addressing urban challenges caused by rising temperatures. The complex dynamic of air
quality parameters and UHI in KV should be studied so that the city can implement proper
interventions and be more accommodating to its inhabitants.

Keywords: Urbanization, Urban Heat Island, Particulate Pollution, Seasonal Variation

TABLE OF CONTENTS
vi
 Title Page
DECLARATION.....................................................................................................................
RECOMMENDATION......................................................................................................... III
CERTIFICATION.................................................................................................................
ACKNOWLEDGEMENTS....................................................................................................
ABSTRACT............................................................................................................................
TABLE OF CONTENTS......................................................................................................VII
LIST OF TABLES..................................................................................................................
LIST OF FIGURES.................................................................................................................
LIST OF EQUATIONS.........................................................................................................
LIST OF ABBREVIATIONS...............................................................................................XII
LIST OF UNITS AND CONVERSIONS...........................................................................XIV

CHAPTER 1 : INTRODUCTION...........................................................................................

1.1. BACKGROUND.............................................................................................................
1.2. STATEMENT OF THE PROBLEM............................................................................
1.3. RESEARCH QUESTIONS...........................................................................................
1.4. RESEARCH OBJECTIVES.........................................................................................
1.1.1 GENERAL OBJECTIVE...............................................................................................
1.1.2 SPECIFIC OBJECTIVES...............................................................................................
1.5. SIGNIFICANCE OF STUDY.......................................................................................
1.6. SCOPE/ LIMITATIONS OF STUDY..........................................................................

CHAPTER 2 : LITERATURE REVIEW...............................................................................

2.1. OVERVIEW OF UHI: ITS CAUSES AND EFFECTS..............................................

2.2. PREVIOUS STUDIES ON AIR QUALITY PARAMETERS AND UHI...............10
2.2.1. EFFECTS OF TSP ON UHI.............................................................................................10
2.2.2. EFFECTS OF PM 2.5 ON UHI........................................................................................12
2.3. STUDIES ON UHI EFFCET IN KATHMANDU VALLEY...................................14

CHAPTER 3 : MATERIALS AND METHODS................................................................. 18

vii
3.1. STUDY AREA..............................................................................................................18
3.2. RESEARCH DESIGN.................................................................................................20
3.3. METHODS OF DATA COLLECTION.....................................................................21
3.3.1. SECONDARY DATA.......................................................................................................21
3.4. DATA ANALYSIS.......................................................................................................22
3.4.1. URBAN HEAT ISLAND (UHII)......................................................................................22
3.4.2. PEARSON’S CORRELATION COEFFICIENT.....................................................................23
3.4.3. DUAL AXIS TREND CHART..........................................................................................24

CHAPTER 4 : RESULTS AND DISCUSSIONS................................................................. 25

4.1. STATISTICS FREQUENCIES...................................................................................25

4.2. URABAN HEAT ISLAND INTENSITY (UHII).......................................................26
4.3. PEARSON’S CORRELATION COEFFICIENT (R)...............................................29
4.3.1. RELATIONSHIP OF MAXIMUM TEMPERATURE AND AIR QUALITY PARAMETERS FOR
6 YEARS....................................................................................................................................29
4.3.2. SEASONAL ANALYSIS FOR MAXIMUM TEMPERATURE AND AIR QUALITY
PARAMETERS FOR 6 YEARS.......................................................................................................32
4.4. DUAL AXIS TREND CHART...........................................................................................42
4.4.1. DUAL AXIS TREND CHART OF PM 2.5........................................................................43
4.4.2. DUAL AXIS TREND CHART OF TSP.............................................................................44
4.4.3. DUAL AXIS TREND CHART OF TSP AND PM 2.5.........................................................45

CHAPTER 5 : CONLUSIONS AND RECOMMENDATIONS.........................................46

5.1. CONCLUSIONS.............................................................................................................46
5.2. RECOMMENDATIONS.............................................................................................48

REFERENCES.......................................................................................................................50

viii
 LIST OF TABLES

Title
Page

Table 3.1. Interpretation of R-values......................................................................................23

Table 4.1. Total Valid and Missing datasets used during research..........................................25
Table 4.2. Total Valid datasets divided into seasons used for correlation...............................25
Table 4.3. Average Yearly Temperature of TIA (1030) and Khokana (1073) for 6 years......26
Table 4.4. Relationship of Maximum Temperature and Air Quality parameters for 6 years.
..................................................................................................................................................29
Table 4.5. Summary of Seasonal Variation and Air Quality...................................................41

ix
 LIST OF FIGURES

Title
Page

Figure 2.1 Urban Heat Island Effect in a city and its surrounding area; Source: (WMO,
2020)..........................................................................................................................................
Figure 2.2. Causes of UHI, Source (Nuruzzaman, 2015)..........................................................
Figure 2.3. Impact of UHI, Source: (Nuruzzaman, 2015).........................................................
Figure 2.4. Land use/land cover in Kathmandu valley, Source: (Ishtiaque et. al, 2017).........15
Figure 2.5. Land Surface Temperature of Kathmandu Valley from 1995 to 2019, Source:
(Rai, 2017)...............................................................................................................................16
Figure 3.1. Study Area Map.....................................................................................................19
Figure 3.2. Study Design of thesis...........................................................................................20
Figure 4.1. Average Yearly Temperature of TIA (1030) and Khokana (1073).......................27
Figure 4.2. UHII of TIA (1030) and Khokana (1073) seasonally from 2016-2021................28
Figure 4.3. UHII of TIA (1030) and Khokana (1073) of Spring and Winter from 2016-2021
..................................................................................................................................................29
Figure 4.4. Correlation between temperature and PM 2.5 in Ratnapark.................................31
Figure 4.5. Correlation between temperature and TSP in Ratnapark......................................31
Figure 4.6. Correlation between temperature and PM 2.5 in Spring.......................................33
Figure 4.7. Correlation between temperature and TSP in Spring............................................34
Figure 4.8. Correlation between temperature and PM 2.5 in Summer....................................36
Figure 4.9. Correlation between temperature and TSP in Summer.........................................36
Figure 4.10. Correlation between temperature and PM2.5 in Autumn....................................38
Figure 4.11. Correlation between temperature and TSP in Autumn........................................38
Figure 4.12.Correlation between temperature and PM2.5 in Winter.......................................40
Figure 4.13. Correlation between temperature and TSP in Winter..........................................41
Figure 4.14. Dual Axis Line Chart showing trend between temperature and PM 2.5 over 6
years in Ratnapark....................................................................................................................43
Figure 4.15. Dual Axis Line Chart showing trend between temperature and TSP over 6
years in Ratna Park..................................................................................................................44
Figure 4.16. Dual Axis Line Chart showing trend between TSP and PM 2.5 over 6 years in
Ratna Park................................................................................................................................45

x
 LIST OF EQUATIONS

Title Page

Equation 1: Urban Heat Island Intensity..................................................................................22

Equation 2: Pearson’s Correlation Coefficient formula, Source: (Patil, 2021).......................23

xi
 LIST OF ABBREVIATIONS

AQMS : Air Quality Monitoring Station

AUHI : Atmospheric Urban Heat Island
BVOC : Biological Volatile Organic Compounds
CO2 : Carbon Dioxide
COPD : Chronic Obstructive Pulmonary Disease
DoE : Department of Environment
DoHM : Department of Hydrology and Meteorology
EPA : Environment Protection Agency
GHG : Green House Gases
GIS : Geographic Information System
KII : Key Informant Interview
Km : Kilometer
KV : Kathmandu Valley
LST : Land Surface Temperature
LULC : Land Use and Land Cover
Max. : Maximum
MoFE : Ministry of Forest and Environment
NOx : Nitrous Oxides
O3 : Ozone (Ground level)
PBL : Planetary Boundary Layer
PM : Particulate Matter
Ppm : Parts per million
r : Pearson’s Correlation Coefficient
R2 : Coefficient of determination
SOx : Sulfur Oxides
SPSS : Statistical Package for Social Sciences
SUHI : Surface Urban Heat Island
TAP : Transboundary Air Pollution
Temp : Temperature
TIA : Tribhuvan International Airport

xii
TSP : Total Suspended Particles
UHI : Urban Heat Island
UHII : Urban Heat Island Intensity
UN-DESA : United Nations- Department of Economic and Social Affairs
UPI : Urban Pollution Island
VOC : Volatile Organic Compounds

xiii
 LIST OF UNITS AND CONVERSIONS

% : Per Cent
km : Kilometer
M : Meter
o
C : Degree Celsius
µg/m3 : Micrometers/ Meter (cubed)
µm : Micrometers

xiv
CHAPTER 1
INTRODUCTION
1.1. Background
Kathmandu Valley has been experiencing unprecedented levels of urbanization and growth
over the past few decades. According to the World Urbanization prospects 2018, about
20% of the population of Nepal live in cities with an average annual growth rate of
percentage urban in Nepal being almost 1.5% [1]. The valley is inhabited by about 2
million residents and is the country's most populous region [2]. Urbanization has resulted
in a surge in human activity which has increased energy use, altered land use, expanded
built-up areas, increased transportation, and industrial activity, and reduced overall
greenery in cities. This has caused an increase in temperatures and heat related stresses for
both humans as well as the environment. The rising temperature also poses a threat to the
health of urban residents [3]. Urbanization has also led to significant changes in the
environment namely, land-use change, air pollution, urban heat island effect, water
scarcity, waste generation and excessive resource use. In this paper, we explore two major
changes in the environment, particularly air quality and the Urban Heat Island (UHI) effect
in the Kathmandu Valley. We aim to explore and find the relationship between air quality
parameters and UHI effect.

Urban Heat Island (UHI) effect is the increase of temperatures in urban areas compared to
the surrounding rural areas. It is a phenomenon experienced in cities where urban areas
experience higher surface and atmospheric temperatures compared to their surrounding
non-urban areas [4], [5]. Most of the major cities in the world face this problem. Extensive
urbanization leads to change in LULC this includes increase in dense built-up areas,
increment in impervious surfaces; concretization, and high human activity zones which
changes up the energy balance in cities and make the warmer than their nearby surrounding
sub-urban lands [3]. Urbanization also leads to reduction in greenery and concretization
which increase the UHI effect by not moving the heat energy in the environment, favoring
heat trapping, and storing [5]. Thus, growing urbanization has adverse effect on land
surface properties with their thermal capacity [3].

15
Similarly, UHI also has a strong link with human and environment health. There are many
studies linked with the impact of UHI on urban environments, most studies identify the
heat stresses caused on atmospheric conditions human health, environment, and the
ecosystems. The main ramifications of UHI are increased heat stresses in humans and the
environment, higher energy consumption, and decreased air quality. Exacerbated and
excessive urbanization also lead to bad air quality and meteorological conditions that
stagnate air. Consequently, air quality is linked with UHI in two major ways:

i.) When air quality is bad there is a high level of air pollution in urban centers,
this forms a layer of smog called the urban plume, this pollutants in the air trap
solar radiation, heat up the atmosphere and rises temperatures causing UHI. Air
pollution is prevalent in urban regions and city centers as vehicular and
industrial emissions emit pollutants into the environment and trap solar
radiation, which increases the micro-climate impact and raises temperature,
resulting in UHI [6].
ii.) When the air is good on a clear sunny windless day, urban areas heat up directly
due to increased built-up land made from materials such as concrete and
asphalt, which heat up land surface temperatures and cause UHI.

Air quality can primarily effect UHI through modifying the environment's radiative
balance; for example, high air pollution can cause solar radiation to heat up the atmosphere
rather than reaching the ground increasing atmospheric temperatures or on windless days
with good air it can directly heat up the impervious layers of the city increasing surface
land temperatures and causing UHI. Thus, this research assesses the relationship between
air quality parameters (TSP and PM 2.5) and temperature to establish the presence of UHI
in the micro-level in Kathmandu Valley.

In the case of Nepal which is a rapidly developing country with a high urbanization rate,
UHI is of high concern. Kathmandu, the capital city, has been the largest and the fastest
growing city in South Asia [1]. This means that the valley is vulnerable to UHI more so
due to its topography and land use patterns. Numerous studies indicate that cities with
higher levels of anthropogenic heat production and growing urbanization frequently
experience more intense UHI intensities than other areas of the city. Similarly, many
studies have also identified different factors causing the creation in UHI in the valley
[3], [5], [7]
. One of the major reasons is the topography of the valley, the main urban area is in

16
a low-lying valley floor that is encircled by hills, the wind flows over the hills but stills in
the valley, thus, the temperature of the valley is higher than the hills [3]. The low wind
cannot remove the valley's stagnant heat and air pollution, which leads to UHI [7].
Similarly, the land-use has changed drastically with majority of the land converted from
agricultural pastoral lands to dense built-up areas. Rajkarnikar's study on UHI in
Kathmandu Valley identifies five major reasons why the valley is vulnerable to UHI: i)
Increased impermeable and high albedo surfaces, ii) Reduced green covers iii) Increased
urban footprint due to increased population, iv) increased automobile movement and v)
Natural low wind in the valley. These reasons define how UHI is investigated and what
causes UHI [8].

1.2. Statement of the problem

UHI is also a phenomenon that exists in almost every big city [6]. Urbanization increases
the number of concrete roads and high buildings, reduces greenery, increases impervious
surfaces, and halts air movement, all of which adversely affects the valley's land surface
properties with their thermal capacity [3]. The temperatures are rising due to changes in the
micro-climate, and the effect of UHI has resulted in changes of local wind patterns,
formation of fog and clouds, variations in precipitation rates, and humidity changes [7].
This increases the UHI effect of cities making urban areas not just receptors but also
drivers of climate change [8]. Many studies have connected UHI to the onset of
increased storms/precipitation events, increased energy usage and demand, and a
number heat-related mortality [4]. The UHI phenomena is important to investigate because
it has a wide-ranging impact on numerous areas of the economy such as infrastructure,
health, energy consumption, environmental stress, and discomfort, and it also increases to
the cost of developing infrastructure [5]. Similarly, this study will focus on analyzing the
link between air pollution and UHI in the valley by using air quality parameters and
temperature trends in the valley.

As Nepal urbanizes with an unprecedented rate and an urban population of 66.08% [2], it
comes along with its fair share of problems. The capital, Kathmandu, has been the largest
and fastest expanding metropolis in South Asia over the last few decades [1] Numerous
changes have occurred in the city because of the Kathmandu Valley's growing

17
urbanization, the main effect has been high air pollution as a result of the rise in vehicles
and factories. Developing countries including cities like Kathmandu are facing the urgent
need to address the issue of sustainable urban development with the increase challenge of
higher urban temperatures, heat related stresses for people, environment, and infrastructure
as well as poor urban air quality.

Similarly, poor urban air quality is a serious public health issue that has detrimental
impacts on several human body parts and organs and triggers significant changes in the
environment and affects the overall quality of life. High concentration of pollutants like
particulate matter (PM2.5), ground level ozone (O 3), nitrogen dioxide (NO2) and sulfur
dioxide (SO2) are linked with several respiratory and cardiovascular illnesses, neurological
problems, and carcinogenic issues and causes impaired lung development. UHI is linked
with UPI as both are signs of urbanization as well as the thermo-chemical changes that
takes place when conventional built-up components replace natural ones [8].

Urbanization is both a benefit and an affliction. Currently, about 55% of the world lives in
cities, and global projections show that these numbers are expected to rise by 68% by 2050
[1]. As more than half of the world’s population start living in cities, we need urban spaces
to be more comfortable and convenient to its inhabitants. Even in Nepal, according to the
preliminary report of the latest national census (2021), 66.08 percent of Nepal's population
lives in cities, up from 17 percent in 2011 with an expected urbanization rate per year of
6.5% [2]. This stresses the importance of addressing challenges faced in cities. Every city
in the world needs to be more accommodating and sustainable to its inhabitants because
more people will be living in an urban space. A developing country like Nepal where much
urbanization and industrialization are still expected will be at the forefront of climate
catastrophe without any countermeasures in place. This means that UHI & UPI are
interlinked and need to be addressed for growing cities.

1.3. Research Questions

I. What is the relationship between certain air quality parameters (PM2.5 and TSP)
and the UHI effect?
II. How does temperature and air quality parameters correlate in different seasons?

18
 III. How can the relationship between air quality parameters and temperature prove the
existence of UHI in the Valley?

1.4. Research Objectives

1.4.1. General Objective

To determine the existence of urban heat in the Kathmandu valley with relation to
particulate pollution.

1.4.2. Specific Objectives

I. To establish the temperature differences between urban and rural areas

II. To identify and assess the different air quality parameters that affect temperature to
prove UHI exists in the Kathmandu valley.
III. To assess the correlation between temperature and air quality parameters in
different seasons over a 6-year timeframe.
IV. To create a trendline to show the relationship of temperature and air quality
parameters (PM2.5 and TSP).

1.5. Significance/ Rationale of the study

While there have been many studies focusing on the factors that cause UHI, there are still
some gaps that warrant further investigation. Kathmandu Valley has been the focus of few
temperature change studies relating to UHI but none of these studies focused specifically
on UHIs and their interactions with other potential variables that are not related to air
quality [5]. Many studies acknowledge the UHI effect in Kathmandu Valley, but much
research has not been conducted on the relationship between air quality parameters and
UHI effect.

Similarly, not many studies have directly linked air pollution and UHI of different seasons.
A study shows how most studies focus on summer conditions and clear sky conditions,
however, only 40% of those studies developed year-round assessments of UHI [8]. Most

19
studies that have been conducted for UHI has been based on satellite imagery and remote
sensing, while it is good for assessing or evaluating UHI on a larger scale, it rarely gives
the accurate representation at the micro-level. Kathmandu is a valley and temperature
differences are unavoidable with altitude change, satellite experiments may not accurately
portray the UHI effects occurring in the valley's main cities [7].

This research will contribute to develop the understanding of UHI effect in Kathmandu
Valley. The assessment of UHI helps in determining the major components responsible for
accelerating the UHI effect and the suitable mitigation actions. The main purpose of this
thesis is to assess the relationship between air quality parameters (PM 2.5 and TSP) and
UHI in Kathmandu Valley. By investigating their relationship, this research aims to shed
the light on UHI and its interaction with air pollution. A city's urban climate and its level
of air pollution are related in many of ways [9]. The results of this research could improve
our knowledge of the specific pollutants that can significantly aggravate the UHI effect,
these details are necessary if we want to improve urban air quality and reduce the UHI
effect in urban areas. Similarly, it will establish a correlation of air quality parameters and
temperature in different seasons to prove UHI exists in the valley. It will provide key
insights on how different air quality parameters are influenced by temperature changes.
Once, it is established, we can evaluate the data for decision makers and stakeholders who
can develop strategies to mitigate and reduce the UHI effect, target pollution control
measures to reduce UPI in Kathmandu and improve air quality management strategies.
Additionally, understanding the relationship between air quality and UHI is critical for
solving urbanization's environmental and health issues and promoting sustainable urban
development.

20
1.6. Scope/Limitation of the study
I. Data collected from certain locations might not represent the entire Kathmandu
Valley.
II. Currently, Kathmandu and Kirtipur (TU) have similar topographical and urban
characteristics than in the past; thus changes in study area over time could also
influence data sets and values.
III. Air Quality parameters such as O 3, Nox, Sox, VOC, CO2 data are not available in
the AQMS, thus all air quality parameters are not included.
IV. Air Quality data from the government had a lot of null/ empty values, removing all
these values was the only method to get possible results.
V. Other factors such as population density, LULC changes, precipitation and
humidity, human activities, climate change, urban pattern/design, green spaces etc.
will not be analyzed in this research but are factors that affect UHI in cities.
VI. The Department of Environment provided data for only six years, and when
combined with missing data, hindered the Dual-Axis Trend chart from properly
displaying a trend.

21
 CHAPTER 2
LITERATURE REVIEW
2.1. Overview of UHI: Its causes and effects
Urban Heat Island (UHI) effect is the increase of temperatures in urban areas compared to
the surrounding rural areas. It is the heat trapping phenomenon experienced in cities where
urban areas experience higher surface and atmospheric temperatures compared to their
surrounding non-urban areas [3]. This is often due to the absorption of solar radiation by
dark material built-up structures (urban surfaces) as well as the release of heat of stored in
these structures from the day during the night. This change will also affect material flow
and energy flow in urban ecological systems as well as alter their structure and function
[10]. It is commonly observed in many cities and has been studied extensively during the
past several decades due to technological advancement (remote sensing and computational
power) and heightened social and economic concerns (heatwaves, weather extremes, heat
stresses and mortality) [8].

Figure 2.1 Urban Heat Island Effect in a city and its surrounding area; Source: (WMO, 2020)

There are two main types of UHI effect: (i) atmospheric UHI effect (AUHI) and (ii)
surface UHI effect (SUHI). The increase in air temperature from ground level to the tops of
trees and rooftops (canopy layer) is referred to as AUHI, whereas SUHI is the increase in
absorption of solar radiation and heat trapped in the surface, it is dependent on the albedo
and heat capacity of urban surfaces. [7]. AUHIs are more nocturnal; prominent at night or

22
pre-dawn (a few hours after sunset and before sunrise) whereas SUHI is stronger during
the daytime [7].

UHI is a heat accumulation phenomenon within a urban area due to urban activities, it is
now considered as an evident characteristic of the urban climate [10]. Urban growth has
brought about significant changes in the urban environment, particularly in the urban
climate, which has seen a quick rise in temperature beneath the city as opposed to its
surroundings [3]. Much of the urban materials in cities are impermeable, pair it with rapid
urbanization and conversion of natural land into built up space, it becomes the main cause
of UHI effect. Consequently, other causes include dense built-up areas, concrete zones,
high concentration of anthropogenic activities, rapid urbanization, and urban growth,
increasing industrialization and transport activities, reduction of greenery, LULC changes,
GHG emissions in cities, excessive energy use in cities, increase in low albedo surfaces in
cities, increasing urban footprint, population growth, etc. [3], [7] Thus, haphazard
urbanization has a negative effect on land surface properties with their thermal capacity as
it leads to the formation of urban micro-climates, modification of local moisture exchange
and ecosystem services and favors heat trapping and storage to generate UHI [3], [5].
According to Nuruzzaman and Rajkarnikar [6] , the following causes for UHI are listed
below:
 Decrease in green spaces which can intercept heat and keep surfaces from
overheating
 Higher absorption of solar radiation due to low albedo
 Increase in impermeable concrete built-up area
 Increased human activity, which increases air pollution and traps solar radiation
 Urban Canyon effect; high rise buildings in close distances block wind path and
prevent cooling whilst trapping sunlight and heat.

23
 Figure 2.2. Causes of UHI, Source (Nuruzzaman, 2015)

However, UHI also occurs naturally. It is closely linked with local meteorological
conditions. Sandberg (1950) was the first researcher to use a multiple linear regression
method to relate UHI intensity to meteorological variables such cloudiness, wind speed,
temperature, and absolute humidity, he was able to demonstrate that wind speed and
cloudiness parameters are inversely correlated and that the total variance of the regression
model is greater at night than during the day [11]. In the Kathmandu Valley, topology of
the area becomes a large factor for UHI as they valley which is surrounded by natural hills
have more turbulent air than the valley floor, the calm low wind on the valley floor lacks
the ability to clear out stagnant heat and pollutants from the valley contributing to UHI [7].
UHIs become the most prominent on a clear and windless night [11].

Consequently, there are serious impacts of urban heating on different sectors in the
economy. UHI-caused high temperatures have a direct impact on the health and comfort of
city dwellers by causing more intense and prolonged heatwaves, increasing heat stresses
and related mortality such as heat exhaustion, heat cramps, and heat stroke, along with
worsening pre-existing chronic diseases such as respiratory, cerebral, and cardiovascular
diseases [7]. Extreme heat is also linked to lower productivity from outdoor workers [7].
Studies show that extreme heat makes urban air pollution even more prominent by trapping
pollutants at the lower atmosphere in stagnation and increasing levels of ground level O 3

24
[12]. Similarly, it raises energy demand and increases energy cost which in term increases
greenhouse emissions contributing to climate change. The rise in use of energy for cooling
is another effect of UHI effect as a 1oC temperature rise can increase energy demand by 2-
4% [12].

Figure 2.3. Impact of UHI, Source: (Nuruzzaman, 2015)

2.2. Previous studies on air quality parameters and UHI

Different studies have been done to relate different parameters to UHI, among them
Sundborg (1950) was one of the first people to relate meteorological parameters such as
cloudiness, temperature, wind speed and absolute humidity with UHI intensity [11].
Similarly, air pollutants in urban areas released from vehicular and gaseous emissions in
cities are one of the major causes of UHI, temperature rises, and microclimate effects
becomes stronger causing UHI [6]. Thus, different effects of different particulate pollutants
are being studied in this literature review. Air pollutant concentrations of are closely
correlated with several kinds of atmospheric factors [13]. A lot of studies have been done
based on correlation of UHI (both AUHI and SUHI) and different atmospheric conditions.
Fang and Gu mentions how, the geographical distribution of particulate pollution indirectly
influences the UHII by affecting the radiation; when the air quality is bad, the high
concentration of pollutants restricts radiation and leads to increase of AUHI signifying a
positive relationship, however, pollutants can also create a “dust dome” which can affect
infiltration of solar radiation and leads to a temperature drop signifying a negative

25
relationship [14]. Similar studies have been identified where a more positive association
between UHI and pollutants is expected, however, under certain conditions due to the
stimulation of vertical movement and mixing, the relationship between UHI and UPI might
be inversed [8]. Although UHI and UPI formation can account to several other impacts
such as closeness to the city center and/or heat/emission sources, air stagnation, clear
anticyclonic conditions, warm/hot seasons, nocturnal conditions, city size and/or
population, weather conditions like solar radiation, precipitation, temperature, humidity,
wind, etc. still influence UHI and UPI [8].

2.2.1. Effects of TSP on UHI

Total Suspended Particulates (TSP) is the total airborne solid matter in the atmosphere.
The U.S. EPA defines TSP as a type of dust particle which is the mixture of solid and liquid
particles. It is a fraction sampled using high volume samplers with particle diameters more
than 50-100 µm [15]. Smaller particles (PM10 and PM2.5), with an aerodynamic diameter of
less than 10 µm and less than 2.5 µm are called fine particulates whereas, larger particulates
with a size of ≥ 2.5 µm are referred as coarse particulates [16]. It comes from both natural
sources (sea salt, forest fires, pollen, mold) and human activities (burning fuel,
construction sites, industrial emissions, vehicle exhausts). Since, TSP covers a wide range
of particulate matter, it is considered as one of the major causes of smog formation, and air
and environmental pollution. A study about UHI effect with DI was studied in Jakarta, this
study showed a positive correlation that increase in TSP concentration caused higher
discomfort to people [15]. In humans, epidemiological studies have assessed the influence
of TSP in human health with high associations in lung function, respiratory system health,
mortality rates, cardiovascular problems, acute respiratory infections (ARI), asthma,
emphysema, lung cancer, cardiovascular disease, and chronic obstructive lung disease
hypertension and respiratory problems [13], [17].

Meteorological factors such as wind speed, rainfall, temperature, and relative humidity
play a vital role in determining the pollution concentrations of TSP, however, for the sake
of this study only the relationship of TSP with temperature is assessed. TSP can influence
the metrological conditions such as wind patterns, air circulation, and local atmospheric
dynamics, potentially leading to a spatial distribution of higher temperatures and
intensifying the UHI in urban areas. Studies found TSP showing significant seasonal

26
variations with TSP levels higher in winter but lower in the monsoon [13], [15], [18] .
Similarly, TSP and temperatures were significantly correlated with values coefficients (r)
as 0.609137, 0.710154 and 0.327348 in winter, transition, and summer, respectively [13].

The UHI effect can be impacted by high TSP concentrations in the atmosphere through a
variety of mechanisms. In a study, annual TSP data was recorded for 42 years and its
relationship with fog is studied, it was shown how TSP was significantly relevant in
explaining the variability of dense fog hours, it is also shown that as TSP levels continue to
fall, less condensation nuclei are available to produce fog droplets [19]. TSP and UHI;
along with high temperatures, can bring secondary effects such as changes in local wind
pattern, precipitation rates and humidity as well as formation of fogs and clouds on local
climate and weather [20]. Similarly, a study in Turkey found higher TSP concentrations
were highly correlated with colder temperatures, lower wind speeds, higher pressure
systems, weakly reduced precipitation, and higher relative humidity [15]. Many research
suggested that various meteorological parameters such as wind speed, wind direction,
rainfall, and relative humidity influence PM10 and TSP concentrations, a study by Karar
et.al showed an inverse relationship of temperature and particulate concentration at -0.46
(P <0.05 ) [21]. A multiple linear regression analysis was used to determine the impact of
SO2 and TSP on meterological conditions in Turkey, the study discovered that higher TSP
and SO2 concentrations were more related to colder temperatures, slower winds, larger
atmospheric pressure, less rain, and higher humidity [22]. This suggests that TSP might
have an inverse relationship with temperature.
TSP is another regulated pollutant measured in the monitoring stations [18], this is why the
relationship of TSP and UHI can be studied. Scientifically, high levels of TSP should
contribute to an intensification of UHI which implies a positive relationship between TSP
and UHI. However, this is a complex process that varies on specific conditions of the
urban environment such as meteorological conditions, urban morphology, and local
emission sources. A few studies have directly linked temperature and TSP, most studies
examine LST and air pollution through satellite and remote sensing. A researcher found a
strong negative correlation with TSP but a weaker positive correlation with other air
quality parameters, however, he also mentions another paper where weak correlation was
discovered after using AVHRR which increased the pixel size [18].

27
2.2.2. Effects of PM 2.5 on UHI

Particulate Matter (PM) is a mixture of solid particles, aerosols and liquid droplets found in
the air [23]. In this study we are assessing PM2.5, it is a fine inhalable particle of 2.5
microns or less in diameter. It is emitted direct from a point source such as
pollutants emitted from power plants, industries and automobiles, construction sites,
unpaved roads, fields, smokestacks, or fires. These droplets are so miniscule that they can
be directly inhaled into our lungs or bloodstream [23]. These interactions between UHI and
PM2.5 will alter the urban environment in ways that are detrimental to both the economy
and public health [14] as PM2.5 has serious health implications such as acute and chronic
bronchitis, asthma attacks, respiratory and cardiovascular illnesses, COPD, atherosclerosis,
lung cancer, etc. The most standard air pollution detection indicator for urban air pollution
is PM2.5, the greater the PM2.5 concentration, the higher bigger harm it is to human health
[14].

There is a growing understanding about UHI and PM2.5 concentrations. Several studies
have been examined to investigate the relationship between UHI and PM2.5, and the
results indicate that these two factors are not unrelated and greatly influence and enhance
each other in the urban climate environment [14], [24]. This would mean that the presence
of PM2.5 could contribute to the intensification of UHI effect. Studies specify how PM2.5
concentrations worsen urban climate in terms of UHI [14]. The UHI effect is characterized
by higher temperatures and altered atmospheric conditions in urban areas, this can
influence the dispersion and accumulation of pollutants such as PM2.5, similarly, high
PM.5. pollution leads to UHI effect as these particles absorbing and re-emitting solar
radiation which increases temperature in areas with high air pollution [14]. Evidence from
studies show that the particles of PM 2.5 can reduce UHI intensity in the daytime due to
the haze whereas, at night UHI intensity could be exacerbated by the same said haze UHI
[8]. Some studies show a positive relationship between PM2.5 and temperature, this is
because temperature can influence particle output; consequently, high temperatures can
increase photochemical reactions [25], [26].

When comparing SUHI and PM2.5, the correlation coefficient was smaller (-0.6115 with
p< 0.001) than it was for AUHI and PM2.5. This shows that both CUHII and SUHII were
inversely influenced by PM2.5 concentrations [24]. A study by Yang in Beijing states how

28
seasonal variation also had a link with UHI intensity and PM2.5 concentration in urban
areas as PM2.5 pollution reduces UHII throughout the summer and winter nights but
increases it during the winter daytime [27]. This study concludes that UHI tendencies
decrease in summer and winter nights due to the aerosol-radiation interaction in summer
and aerosol-PBL interaction in winter; PBL is the lowest layer of the troposphere where
weather occurs and friction affects wind whereas the aerosol-PBL is the interaction among
aerosol and PBL which influences atmospheric stability, radiative forcing, aerosol-cloud
interactions, etc. It is hypothesized that PM2.5 can reduce energy emissions from urban
land surfaces which is why increased PM2.5 concentrations may lead to reduced UHII
[24]. Another study which uses correlation between uses correlation analysis show that
PM2.5 concentrations have a positive relationship with pressure, and a negative
relationship with other atmospheric parameters such as temperature, RH, rainfall, and wind
speed, with correlation coefficients of -0.512, -0.237, -0.524, and -0.284, respectively [28].

Consequently, when air quality is extremely bad, a phenomenon called the "dust dome" is
formed where high concentration of pollutants limits the radiation from the sun to the
ground during the day and the long-wave radiation from the earth at night, limiting heat
emission and causing the city's temperature to fall [14]. Papers suggest how differing
PM2.5 concentrations in summer led to a slight variation in temperature, and the chemical
composition of the particles dictated whether they merely scattered light or also absorbed
light. There have been similar studies suggesting that PM2.5 concentrations decrease as
temperature rises or vice versa; the reasoning is that when temperature rises, air convection
at the lower surface becomes stronger, allowing specific materials to be transferred higher
[28]. Zhong et al. mentions that cities might have significantly lower the temperature due
to the particles' ideal scattering properties, which is why a decrease in urban temperature is
frequently caused by an increase in particle concentrations [24].

2.3. Studies on UHI effect in Kathmandu Valley

Kathmandu is a bowl-shaped valley located at a mean elevation of 1300m above sea level
with an average temperature range from 30oC in summerand 3oC in winter along with an
average rainfall of 1500 mm, 80% of which fall in the summer monsoon season [3].
During the period 2000-2016, the average annual air surface temperature in Kathmandu
valley was 18.06°C, with a maximum of 24.15°C. In contrast, the average annual land

29
surface temperature in 2000 ranged from 15.84°C to 39.17°C and from 16°C to 33.98°C in
2014 [29]. In the last few decades, Kathmandu has been the largest and the fastest growing
city in South Asia with a population density of 10,00/km 2 [1], [30]. Less than 1% of the
country's total area is covered by it, yet 31% of Nepal's urban population lives there [3].
Numerous changes have occurred in the city because of the Kathmandu Valley's intense
urbanization such as increase in factories and vehicles, bad urban air quality, increased
built-up spaces and loss of greenery, etc. The built-up area in 1989 accounted to 5.1% but
reached to 26.06% in 2016 [3]. According to a study, Kathmandu’s agricultural land has
changed into built-up areas by 412% as shown in Fig 2.4. [31]. Similarly, another study
found that agricultural area decreased from 62 to 42% between 1984 and 2000 and
reported that the Kathmandu valley will have no agricultural land by 2025 [26].

Figure 2.4. Land use/land cover in Kathmandu valley, Source: (Ishtiaque et. al, 2017)

From the data and figures, we can clearly see that the LULC of Kathmandu Valley has
changed due to haphazard and unmanaged urbanization. All these factors have led to the
increase of temperature in the valley which has created the UHI effect. According to
different papers, the annual mean temperature of a city with about 1 million residents can

30
experience 1-3°C warmer temperature compared to its rural surroundings [7]. Many studies
have highlighted the temperature increase in the valley, Chidi states how the heat stress of
the Inner-City Zone of Kathmandu valley is higher than nearby areas [3]. Another article
claims that the core Kathmandu valley is warming, with an annual temperature trend of
0.5-0.8°C between 1976 and 2008 [7]. Studies have shown the temperature of the valley to
be rising at rate of 0.04°C per year with a maximum trend of 0.06°C [7]. A study by
Mishra on UHI in Kathmandu valley depicts a 5°C temperature difference between forest land
and developed land and an increase in temperature Kathmandu by 0°C to 2°C between 2000 and
2018 [5]. Baniya’s study mentions the high temperature in the core city (20.5°C to 21.5°C)
and compares it with rural fringes Nagarkot (0.02°C per year) [30]. Similarly, hotspots in
the city are also rising yearly, satellite images (Landsat imagery) show rising hotspots in
cities from 1995- 2015 [5], [32]. The UHI condition in Kathmandu is not as severe as in
other capital cities, but the escalating trend foreshadows a disastrous future for all its
residents. The rising temperatures will bring negative impacts on the people, environment,
and the economy.

Figure 2.5. Land Surface Temperature of Kathmandu Valley from 1995 to 2019, Source: (Rai, 2017)

31
The urbanization of Kathmandu is one of the main reasons for the increased UHI effect.
Zhong et al. mentions that spatial distribution of SUHII depends on the type of land surface
[24]. However, the topology of the area also becomes a large factor for UHI. Valleys
which are surrounded by natural hills have more turbulent air than the valley floor, the
calm low wind on the valley floor lacks the ability to clear out stagnant heat and pollutants
from the valley contributing to UHI [7]. Thus, the valley becomes more vulnerable to UHI
more due to its topography and land use patterns. Numerous studies indicate that cities
with higher levels of anthropogenic heat production and growing urbanization frequently
experience more intense UHI intensities than other areas of the city. Similarly, many
studies have also identified different factors causing the creation in UHI in the valley
[3], [5], [7]
. One of the major reasons is the topography of the valley, the main urban area is in
a low-lying valley floor that is encircled by hills, the wind flows over the hills but stills in
the valley, thus, the temperature of the valley is higher than the hills [3]. The low wind
cannot remove the valley's stagnant heat and air pollution, which leads to UHI [7].
Similarly, the land-use has changed drastically with majority of the land converted from
agricultural pastoral lands to dense built-up areas.

Rajkarnikar's study on UHI in Kathmandu Valley identifies five major reasons why the
valley is vulnerable to UHI: i) Increased impermeable and high albedo surfaces, ii)
Reduced green covers iii) Increased urban footprint due to increased population, iv)
increased automobile movement and v) Natural low wind in the valley. He writes that over
80% of the 2 million household are made of cement, mortar or concrete and that 90% of
the total roads are black topped, all of which are impervious surfaces with high low albedo
[7]. These factors govern how UHI is examined and what causes it in the valley [5].

32
 CHAPTER 3
MATERIALS AND METHODS
III.1. Study Area
Kathmandu valley is located at a latitude of 27°38’32”-27°45’7” N and longitude of
85°16’5”-85°22’32” E. The bowl-shaped valley is 1,100–2,700 meters (on average,
1,350 meters) above mean sea level [5]. It covers an area of 665 km2. Despite taking up
less than 1% of the country's total space, the Kathmandu Valley is home to 31% of
Nepal's urban residents. In 1989, there were 5.1% of people living in the Kathmandu
Valley in 2016, that number had increased to 26.06%, more than a four-fold increase
[3].

The climate of Kathmandu valley is sub-tropical cool temperate. The average

temperature ranges from 30oC in summer and 3oC in winter with an average rainfall of
1500 mm, 80% of which fall in the summer monsoon season [3]. It has a moderate,
warm, and temperate climate. Summers in the valley are wetter and rainier than winter.
The climate is influenced by tropical monsoon of southeast and receives average
rainfall of 1400 mm during June to August; summer monsoon season [7]. The least
amount of precipitation occurs from November to March, whereas the rainy season
lasts from mid-June to early October. This valley includes the area of three major cities
which are Kathmandu, Bhaktapur and the northern part of Lalitpur District and is
surrounded by four major mountains, namely: Shivapuri, Phulchowki, Nagarjun and
Chandragiri. This research will be carried out to establish the urban heat island in
Kathmandu Valley. The study will be based on two main cities: Kathmandu
(Ratnapark) and Kirtipur (Tribhuvan University).

Since the thesis objective was to establish UHI in the valley, two locations were
identified as a sample to establish the urban and rural cities. The locations were
identified as per the availability of nearby air quality monitoring stations and
temperature stations. The two locations selected for the quality stations were Ratnapark
and Kirtipur. Similarly, the temperature station locations were the Tribhuvan
International Airport and Khokana, which were located approximately 5–6 kilometers

33
 from the respective air quality stations. These areas were good at assessing the
relationship between air quality parameters and temperature due to their urbanization
levels, traffic conditions, and local weather conditions, i.e., temperature and
precipitation.

The first location selected was Ratnapark. It is in the city center and is one of the
largest commercial and transportation hubs in the city, with a high volume of traffic
passing daily. Thus, this area experiences heavy traffic congestion, which increases air
pollution due to the accumulation of air pollutants such as TSP, PM2.5, and NOx. It is
also at a lower elevation than most parts of the city, which means that air pollution
from traffic can create a dome-like effect that can increase UHI, making it an ideal
study area for assessing the relationship between air quality parameters and the UHI
effect. Temperature data from the Tribhuvan International Airport was taken as another
variable to assess the relationship between air quality parameters and temperature.

The second location selected was Kirtipur; although the elevation and temperature are
similar to those of Ratnapark, it is still a fast-developing sub-urban area that is located
on the outskirts of Kathmandu city. Despite being called a neutral area, neither urban
nor rural, it is still heavily urbanized with increased settlements and commercial areas,
high traffic (proximity to the highways), an education hub (Tribhuvan University), etc.
The study area was chosen because the AQMS is on the premises of Tribhuvan
University. The area around the university is lush and green, with traces of agriculture
and open spaces. Similarly, Khokana DHM station was taken as another variable to
assess the relationship between air quality parameters and temperature.

34
 Figure 3.6. Study Area Map

35
 III.2. Research Design

Assessing the relationship of

Air Quality parameters on
Urban Heat Island Effect in
Kathmandu Valley

Objectives

Establish the Assess the correlation Create a trendline to

Assess the correlation
temperature between temperature show the relationship of
between temperature and
differences between and PM2.5 and TSP
PM2.5 and TSP for 6 years temperature and PM2.5
study areas seasonally for 6 years and TSP

Collection of temperature and

air quality ( Secondary
Data Collection Government Data)

Urban Heat Dual Axis Line

Data Analysis Island Intensity Pearson’s Chart
Correlation
Coefficient

Correlation for 6 years= -

UHII average= 0.578 and -0.065 (0.01
6 year trendline
Results significance level). Positive
0.8 o C per year Corrleation in spring; 0.166 established a link
and 0.212 and Negative between air quality and
correalion in Autumn; temperature

Final Findings

Figure 3.7. Study Design of thesis

36
 III.3. Methods of data collection
Both the temperature and air quality data were obtained from government agencies, thus
the data obtained were from stations calibrated and were accurate in their measurements.
However, there were a lot of the missing which had to be manually removed. Thus, data
cleaning was also a big part of data collection and handling.

III.3.1. Secondary Data

I. Measurement of air quality parameters

In Kathmandu Valley, the air quality is measured by the government or inter-government

agencies using Automatic AQMS located at various points throughout the valley. The
device that was used to measure air quality is the Environmental Dust Monitor (Grimm
EDM 180+), it uses laser light-scattering technology to count particles in sample air, which
is measured by the number of particles and the volume flow rate [10]. These stations
measure different air quality parameters such as PM2.5, PM10, TSP, etc. To obtain data
from Ratnapark and Kirtipur for different seasons, we took the secondary data from the
DoE for the AQMS stations of Ratnapark and Kirtipur for six years. The analysis was done
for the air quality data to identify if there were differences in air quality parameters
between urban and rural areas.

II. Measurement of temperature

Similarly, we can use the local weather stations of the DHM to collect temperature data.
The government has several stations that provide temperature data from their network of
weather stations that can be accessed by requesting data directly from them. For this
research, the data from the Tribhuvan National Airport (1030), which is an aeronautical
type of station in Kathmandu, and Khokana (1073), which is a climatology type of station
in Lalitpur, were used to obtain the temperature data for 6 years. After collecting the
temperature data, analysis was done to identify if there were temperature differences
between urban and rural areas. The data was cleaned up and used for further statistical
analysis.

37
 III.4. Data Analysis
Data Analysis was done using Google Sheets and SPSS software. Google Sheet is a web-
or cloud-based spreadsheet application that helps store, manage, and analyze spreadsheets
and share the data on a real-time basis. It was used to identify the UHIII and the trend of
temperature and air quality parameters. Google Sheets were preferred over MS Excel due
to its real-time sharing option and user-friendly simplicity. SPSS software is a statistical
software for data management and analysis and was used to find the Pearson’s Correlation
Coefficient (r). SPSS was preferred over MS-Excel as it was a better data analysis tool and
software to handle large volumes of data and perform the Pearson’s Correlation Coefficient
(r).

III.4.1. Urban Heat Island (UHII)

Urban Heat Island Intensity (UHII) is measured by the difference between urban
temperatures and rural temperatures. The formula used to obtain the difference between the
was as follows:

ΔT =t 1−t 2

Equation 1: Urban Heat Island Intensity

where, ΔT is the temperature difference, t 1 is the average temperature of urban area (Ratna
Park), and t2 is the average temperature of rural area (Kirtipur). UHII is a common
technique to determine the impact of urbanization which takes into account the variation in
temperature between representative urban and rural sites [11]. UHII is a well-known
measure of UHI which measures the average differences in surface or air temperature
between an urban area and its rural surroundings [33], [34] . This method has been
extensively used and adopted for UHI investigations. The daily maximum and minimum
temperature of 6 years from two stations TIA (1030) and Khokana (1073) were averaged
into monthly and yearly average temperatures. The average yearly temperatures from 2017
to 2021 were tabulated for both station and the formula were used to obtain the UHII.

38
 III.4.2. Pearson’s Correlation Coefficient

The collected temperature data and air quality parameters will be further analyzed in MS-
EXCEL to analyze the response between TSP and PM 2.5 concentrations and the local
temperature using Pearson’s correlation. In a paper which reviewed 72 existing papers
related to UHI, it was found that 68% of papers used Pearson’s Correlation Coefficient or
Ordinary Least Square Regression to analyze data [4].

Correlation is the measurement of association or the relationship between two variables

regardless of positive, negative, or null relationship. If changes in one variable affect or
influence changes in the other variable, they are said to be related. Correlation coefficients,
in other words, quantify the strength of an association or relationship between two
variables.

Using Pearson’s correlation analysis, it is determined if there is a relationship between

temperature and air quality parameters. Pearson’s Correlation Coefficient (r) is used to
determine the strength of the relationship between temperature and air quality parameters
as it is the most common technique for determining the relationships between variables.
The Pearson correlation coefficient assesses the strength of a two-variable linear
relationship. It has a value ranging from -1 to 1, with -1 indicating perfect negative linear
correlation, 0 indicating no connection, and + 1 indicating perfect positive correlation [35].
Similarly, correlation coefficients lower that 0.40 (whether negative or positive 0.40) are
said to be low, between 0.40 and 0.60 are moderate, and above 0.60 are high [35]. The
mathematical formula is as follows:

Equation 2: Pearson’s Correlation Coefficient formula, Source: (Patil, 2021)

Where,
r = Pearson’s Correlation Coefficient
xi = x variable samples
yi = y variable samples
x = mean of values of x variables

39
y = mean of values of y variables
Table 3.1. Interpretation of R-values

Value of r Correlation between variables

If, r = -1 Perfect negative correlation

If, -0.99< r <= -0.5 Moderately negative correlation

If, -0.49< r > 0 Weak negative correlation

If, r = 0 No correlation

If, 0< r < = 0.49 Weak positive correlation

If, 0.5< r > = 0.99 Moderately positive correlation

If, r = +1 Perfect positive correlation

III.4.3. Dual Axis Trend Chart

A Dual Axis Line Chart is a line chart that has multiple axes and shows the connections
between two variables with different magnitudes and measurements. This multi-linear
chart gives us better visual representation of different data sets and helps us input data
using limited space, showing trends and patterns. In this research, we are using the six-year
average temperatures in the left Y-axis and air quality parameters such as PM2.5 and TSP
in the right Y-axis whereas the time (In years) is in the X-axis. This is done to plot
different data sets with separate units on a same chart while making it easier to compare
trends over time.

40
 CHAPTER 4
RESULTS AND DISCUSSIONS
4.1. Statistics Frequencies
Table 4.2. Total Valid and Missing datasets used during research

Year Year

2016 N Valid 111 2019 N Valid 319

Missing 254 Missing 46

2017 N Valid 336 2020 N Valid 229

Missing 29 Missing 136

2018 N Valid 317 2021 N Valid 307

Missing 48 Missing 58

A total of 1619 valid days were recorded and were used for correlation, among which 571
data sets were missing from the government database. Similarly, a total of 373 Spring
days, 419 Summer days, 378 Autumn days and 449 Winter days were valid and was taken
for the Pearson’s correlation analysis to show the results.

Table 4.3. Total Valid datasets divided into seasons used for correlation

YEA Frequen Perce YEA Frequen Perce

R Seasons cy nt R Seasons cy nt
SPRING
Vali Vali (March-
2016 d 2019 d May) 75 23.5
SUMMER SUMMER
(June- (June-
August) 4 3.6 August) 70 21.9
AUTUM AUTUM
N N
(Sepetmbe (Sepetmbe
r- r-
November November
) 77 69.4 ) 86 27
WINTER WINTER
(Decembe (Decembe
r- r-
February) 30 27 February) 88 27.6

41
 Total 111 100 Total 319 100
SPRING SPRING
Vali (March- Vali (March-
2017 d May) 81 24.1 2020 d May) 79 34.5
SUMMER SUMMER
(June- (June-
August) 92 27.4 August) 78 34.1
AUTUM
N
(Sepetmbe WINTER
r- (Decembe
November r-
) 80 23.8 February) 72 31.4
WINTER
(Decembe
r-
February) 83 24.7
Total 336 100 Total 229 100
SPRING SPRING
Vali (March- Vali (March-
2018 d May) 49 15.5 2021 d May) 89 29
SUMMER SUMMER
(June- (June-
August) 87 27.4 August) 88 28.7
AUTUM AUTUM
N N
(Sepetmbe (Sepetmbe
r- r-
November November
) 91 28.7 ) 44 14.3
WINTER WINTER
(Decembe (Decembe
r- r-
February) 90 28.4 February) 86 28
Total 317 100 Total 307 100

4.2. Urban Heat Island Intensity (UHII)

The daily maximum and minimum temperatures for 6 years from TIA (1030) and Khokana
(1073) were plotted on the Google Sheet. The yearly average temperature was calculated
by averaging each day with the monthly average for each year. Each average yearly
temperature was plotted on a graph for both study areas, and their differences were
calculated to obtain their UHII. The difference between two temperature points ranged
from 0.50 to 1.11. The calculated UHIII on average is 0.8 oC per year. There were

42
differences seen between the urban center (TIA) and what was assigned the rural center
(Khokana).

Table 4.4. Average Yearly Temperature of TIA (1030) and Khokana (1073) for 6 years

Yearly TIA Yearly KhK

Year temp avg Temp avg Difference

2016 19.49 18.99 0.50

2017 19.68 18.70 0.79

2018 19.21 18.10 1.11

2019 19.41 18.31 1.10

2020 18.90 18.18 0.72

2021 19.50 18.42 1.08

The outcomes were represented on a graph. As illustrated in Fig. 4.1., the temperature in
TIA is greater than at Khokana. The highest average temperature recorded in TIA was 34.2
o
C in 2016, whereas the highest average temperature recorded in Khokana was also 34 o C
also in 2016. The temperature at TIA was used as a reference point for urban Kathmandu,
whereas the temperature in Khokana was used as a reference point for rural Kathmandu.
These stations are approximately 12 kilometers apart. This shows that urban temperatures
are higher than (assigned) rural temperatures, proving that UHI exists in the valley.

43
 Figure 4.8. Average Yearly Temperature of TIA (1030) and Khokana (1073)

According to different papers, the annual mean temperature of a city with about 1 million
residents can experience 1-3°C warmer temperature compared to its rural surroundings [7].
A paper by Ulpiani noted the average UHII range with an average of 1-2 °C, while the
peak UHII reached 12 °C, with mean values ranging between 3-5 °C [8]. In Baniya’s
study, the average annual temperature of temperature of Kathmandu Airport was 19.50°C,
showing that the urban center has higher temperatures than surrounding semi-urban areas
like Khokhana, Nagarkot and Godavari [30]. The study above showed similar results TIA
having higher temperature than Khokana with an average difference of 0.88 o C. Another
article claims that the core Kathmandu valley is warming, with an annual temperature trend
of 0.5-0.8°C between 1976 and 2008 [7]. Studies have shown the temperature of the valley
to be rising at rate of 0.04°C per year with a maximum trend of 0.06°C [7]. A study by
Mishra on UHI in Kathmandu valley depicts a 5°C temperature difference between forest land
and developed land and an increase in temperature Kathmandu by 0°C to 2°C between 2000 and
2018 [5].

Similarly, UHII in different seasons were assessed to check if seasonal variation for
temperatures were also present. According to a 24-year study conducted in Seoul using a
regression line, UHII was higher in the spring (0.95°C) [11]. UHI effects are more
prominent in summer, however, both areas in the cities have higher temperatures of 24°C
meaning that UHI is prominent in both areas. According to Fig 4.2., clear differences of

44
temperatures are seen in Spring and Winter, UHII for TIA (1030) than Khokana (1073)
which would mean that temperature during Spring and Winter are higher in the inner city
than the surrounding areas. There are studies that state how intensity of UHI is greater in
the spring and summer than in the autumn and winter [4]. Fig 4.3. shows the difference of
temperatures for 6 years for Spring and Winter at TIA (1030) than Khokana (1073).

Figure 4.9. UHII of TIA (1030) and Khokana (1073) seasonally from 2016-2021

45
 Figure 4.10. UHII of TIA (1030) and Khokana (1073) of Spring and Winter from 2016-2021

46
4.3. Pearson’s Correlation (r)

4.3.1. Relationship of Maximum Temperature and Air Quality

parameters for 6 years
Table 4.5. Relationship of Maximum Temperature and Air Quality parameters for 6 years.

Maximum
Temperature

RP PM2.5 (µg/m3) Pearson Correlation -0.578**

Sig. (2-tailed) 0

N 1618

RP TSP (µg/m3) Pearson Correlation -0.065**

Sig. (2-tailed) 0.009

N 1618
** Correlation is significant at the 0.01 level (2-tailed).

The relation between PM2.5 and TSP concentrations and the local temperature was
comprehended using Pearson’s correlation. The values of R, correlation coefficients
between PM2.5 and TSP concentrations, and the temperature for Ratnapark were -0.578
and -0.065 at the 0.01 (2-tailed) significance level. The result showed that the relationship
between PM2.5 concentration and temperature in Ratnapark showed a moderately negative
correlation, and the relationship between TSP concentration and temperature in Ratnapark
showed very weak negative correlation. A correlation coefficient between -0.9 to -0.5
suggests a moderately negative relationship, that means as temperatures increase, PM2.5
levels tend to decrease and vice versa; however, it also means that the relationship between
these variables is not very strong or is ambiguous. Similarly, a correlation coefficient
between -0.49 to 0 suggests a weak negative relationship, meaning that as temperatures
increase, TSP levels tend to decrease and vice versa; however, it also means that the
relationship is weak and negligible and that other factors might have larger influences on
temperatures more than TSP.

Different studies have been found where PM2.5 and TSP shows inverse relationship with
temperature [13], [15], [18] . A study in Turkey found higher TSP concentrations were

47
highly correlated with colder temperatures, lower wind speeds, higher pressure systems,
weakly reduced precipitation, and higher relative humidity [15], another study by Karar
et.al showed an inverse relationship of temperature and particulate concentration at -0.46
(P <0.05) [21]. Consequently, Zhong et al. mentions that cities might have significantly
lower the temperature due to the particles' ideal scattering properties, which is why a
decrease in urban temperature is frequently caused by an increase in particle concentrations
[24]. This helps prove the above results of the study. However, due to the values having
lower values, it also means that the relationship is weak and that other factors might have
larger influences on temperatures more than particulate pollution. Similarly, a lower R2
value indicates that variations in air quality parameters do not have a straight-forward
relationship and can explain only a small portion of the variations in temperature which
means that air quality parameters alone are not very good predictors of temperature
changes in the context of UHI.

A scatter plot of temperature against PM2.5 and TSP concentrations was created to observe
how data points are distributed along the plot. Each year was coded in a different color, the
R2 values were calculated, and a linear equation; (y = mx + b); was passed through the
scatter plot. The coefficient of determination or R 2 gives the strength of a linear
relationship between two variables; how can one variable (dependent) explained by
another variable (independent) and is calculated by squaring the (r) coefficient [36]. The R2
value ranges from 0 to 1, with 0 indicating no relationship and 1 indicating a perfect linear
relationship. In Fig. 4.4., the correlation between PM2.5 and Maximum Temperature is -
0.578, and the R2 value is 0.334 with a linear equation of y = -0.88x + 29.09. Here we can
see that the scatterplot shows a general trend downward due to a moderately negative
correlation with data points scattered around the trendline.

48
 Figure 4.11. Correlation between temperature and PM 2.5 in Ratnapark

In Fig. 4.5., the correlation between TSP and Maximum Temperature is -0.065, and the R2
value is 0.004 with a linear equation of y = -0.002x + 25.8. Here we can see that the
scatterplot shows a slight trend downward due to a weak negative correlation with data
points scattered and clustered around the trendline without a clear trend.

Figure 4.12. Correlation between temperature and TSP in Ratnapark

49
 4.3.2. Seasonal Analysis for Maximum Temperature and Air Quality
parameters for 6 years

The correlation analysis done for the Maximum Temperature and Air Quality parameters
for 6 years did not provide an in-depth analysis for the relationship between maximum
temperatures and different air quality parameters; PM 2.5 and TSP. According to one
study, assessing UHI intensity requires excluding the effects of large-scale changes and
narrowing down local effects [11], which is why seasonal analysis is used to assess the
relationship between temperature and air quality. In a paper which reviewed 72 existing
papers related to UHI, it was found that 33.3% of papers checked for seasonal variation to
analyze UHI [4]. It is highly possible that certain seasons and weather conditions exhibit
stronger correlations than others. Thus, the seasonal analysis of temperature and air quality
such as PM 2.5 and TSP was conducted. The relation between PM2.5 and TSP
concentrations and the local temperature was comprehended using Pearson’s correlation.
Four seasons were divided, with each of the six years containing 3 months each for Spring,
Summer, Autumn, and Winter. The results are given down below:

I. Spring (March-May)

The spring season contained 373 data sets among the expected 540. The values of R,
correlation coefficients between PM2.5 and TSP concentrations, and the temperature for
Ratnapark were 0.166 and 0.212, respectively, at the 0.01 (2-tailed) significance level. The
result showed that the relationship between PM2.5 and TSP concentrations and
temperature in Ratnapark both showed a weak positive correlation, which means that as
temperatures increase, PM2.5 and TSP levels tend to increase and vice versa in Spring
(March–May). However, since the relationship is weak, the changes in temperature might
not show significant effects on PM2.5 and TSP concentration levels, or other factors might
have a more dominant influence.

Some explanations for this positive association have been identified. The Spring season in
Nepal is a dry season, here temperatures are expected to rise, and more human activity is
also expected. According to different reports, Nepal often experiences windstorms,
droughts, and heatwaves with a high number of wildfires, forest fires and crop residue
burning of post wheat harvest in the dry season between March to May [37], [38]. This
coincides with our positive correlation value as pollution levels both on PM2.5 and TSP

50
tend to increase with such activities. Higher temperatures are also linked to increased and
enhanced formation of secondary pollutants which increase formation of PM2.5 which
ultimately increases TSP of an area [14]. Dry seasons are also linked with higher levels of
pollution within an urban space due to higher atmospheric stability which explains the
positive correlation [8].

Similarly, a scatter plot of temperature against PM2.5 and TSP concentrations were created
to observe how data points are distributed along the plot. Each year was coded in a
different color, the R2 values were calculated, and a linear equation; (y = mx + b), was
passed through the scatter plot. In Fig. 4.6., the correlation between PM2.5 and Maximum
Temperature is 0.166 , and the R2 value is 0.045 with a linear equation of y = 0.00291x +
26.42. Here we can see that the scatterplot shows a slight trend upwards due to a weak
positive correlation with data points scattered around the trendline.

Figure 4.13. Correlation between temperature and PM 2.5 in Spring

Similarly in Fig. 4.7., the correlation between TSP and Maximum Temperature is 0.212,
and the R2 value is 0.028 with a linear equation of y = 0.02x + 26.25. Here we can see that
the scatterplot shows a slight trend upward due to a weak positive correlation with data
points scattered around the trendline.

51
 Figure 4.14. Correlation between temperature and TSP in Spring

II. Summer (June-August)

The summer season contained 419 data sets among the expected 540. The values of R,
correlation coefficients between PM2.5 and TSP concentrations, and the temperature
for Ratnapark were -0.038 and 0.142, respectively. The TSP data is significant at 0.01
level (two-tail), however, PM2.5 data did not show a significance level. This means
that the p-value or correlation between PM2.5 and temperature may not represent a true
or association, and that the results may have happened by random chance. Because the
p-value (0.434) is bigger than the significance level, the observed correlation between
these two variables is not statistically significant. The results showed that the
relationship between PM2.5 and temperature in Ratnapark showed a weak negative
correlation, which means that as temperatures increase, PM2.5 levels tend to decrease
and vice versa in Summer (June-August) whereas TSP concentrations show weak
positive correlation which means that as temperatures increase, TSP levels tend to
increase and vice versa.

52
 Some explanations for these associations have been identified. Summer season in
Kathmandu is hot and wet, here temperatures are expected to rise, rain is present
throughout the mid of summer and early Autumn season and more human activity such
as tourism, increased transportation, etc. is also expected. According to different
reports, Nepal often experiences wind and thunderstorms which can impact air quality
in the wet season between June to August [37], [38] . For PM2.5 concentrations,
negative correlation is brought by the washout effect from the rain, the rainfall and
thunderstorms disperse pollution which lowers PM2.5 levels. PM2.5 pollution is also
reduces UHII throughout the summer [27] However, since the significance level is low,
we also need alternate hypothesis, a study measured the impact of rainfall on PM10,
PM2.5 concentrations and found that heavy rain lowered larger particle pollutants by
coagulating them with water but has very minimal effect on small particles (PM 2.5
and PM1) [39]. Consequently, the positive correlation value for TSP and temperature
would mean that as temperature increases so does TSP levels. Higher temperatures are
also linked to increased and enhanced formation of secondary pollutants which
increase formation of PM2.5 which ultimately increases TSP of an area [14]. Summer
heats can lead to cases of a haze due to high increase of photochemical smog or
ground-level O3 due to mixing of nitrous oxides, VOCs and BVOCs which increases
temperatures and increases TSP concentrations which decrease air quality. Summers
with high temperatures can also lead to increase of BVOCs and biological particles as
rainy season promotes vegetation growth which can increase biogenic emissions [40].
Similarly, summers are also associated with higher movement of people, since the
study was done in Ratnapark higher vehicular emissions, urban emissions and
secondary aerosols could have increased TSP levels, thereby explaining positive
correlation.

Similarly, a scatter plot of temperature against PM2.5 and TSP concentrations were
created to observe how data points are distributed along the plot. Each year was coded
in a different color, the R2 values were calculated, and a linear equation; (y = mx + b),
was passed through the scatter plot. In Fig. 4.8., the correlation between PM2.5 and
Maximum Temperature is -0.038, and the R2 value is 0.001 with a linear equation of y
= -0.00858x + 29.27. Here we can see that the scatterplot shows a slight trend
downwards due to a weak negative correlation with data points scattered around the
trendline.

53
 Figure 4.15. Correlation between temperature and PM 2.5 in Summer

Similarly in Fig. 4.9., the correlation between TSP and Maximum Temperature is
0.142, and the R2 value is 0.020 with a linear equation of y = 0.00294x + 28.89. Here
we can see that the scatterplot shows a slight trend upward due to a weak positive
correlation with data points scattered around the trendline.

Figure 4.16. Correlation between temperature and TSP in Summer

54
 III. Autumn (September-November)

The autumn season contained 378 data sets among the expected 540. The values of R,
correlation coefficients between PM2.5 and TSP concentrations, and the temperature for
Ratnapark were -0.677 and -0.332, respectively, at the 0.01 (2-tailed) significance level.
The result showed that the relationship between PM2.5 and TSP concentrations and
temperature in Ratnapark both showed a moderate to weak negative correlation, which
means that as temperatures decrease, PM2.5 and TSP levels tend to increase and vice versa
in Autumn (September-November).

Some explanations for these associations have been identified. Autumn season in
Kathmandu is wet in the beginning of September but turn dry and cooler throughout
November. The temperatures in these months are warm but gradually turn cool. In the
beginning of September rainfall brings the washout effect where the rain clears up
particulate pollution, this means that even if temperature is warm, PM 2.5 and TSP fall.
However, as we move on to the other months, temperatures start decreasing and many
factors drive the relationship of temperature and air quality parameters. The major driver of
negative correlation and bad air quality could be biomass burning as open agricultural
residue burning occurs the most during October and November (post rice harvest) [38].
Additionally, this region also experiences TAP from regional areas such as nearby
provinces or countries (Asian Brown Cloud). Another important contributor to increased
PM 2.5 and TSP particles in Kathmandu Valley is the brick kiln industry which fire up
bricks for the season at this time due to favorable atmospheric conditions which increases
industrial activity and increases industrial emissions. In general, due to cooling
temperatures temperature start stabilizing which can lead to poor dispersion of air
pollutants. Dry seasons are also linked with higher levels of pollution within an urban
space due to higher atmospheric stability which explains the negative correlation [8].

Similarly, a scatter plot of temperature against PM2.5 and TSP concentrations were created
to observe how data points are distributed along the plot. Each year was coded in a
different color, the R2 values were calculated, and a linear equation; (y = mx + b), was
passed through the scatter plot. In Fig. 4.10., the correlation between PM2.5 and Maximum
Temperature is -the correlation between PM2.5 and Maximum Temperature is -0.677, and
the R2 value is 0.458 with a linear equation of y = -0.08x + 29.5. Here we can see that the

55
scatterplot shows a downward trend due to a moderately negative correlation with data
points scattered around the trendline.

Figure 4.17. Correlation between temperature and PM2.5 in Autumn

In Fig. 4.11., the correlation between PM2.5 and Maximum Temperature is -0.332, and the
R2 value is 0.110 with a linear equation of y = -0.008x + 28.03. Here we can see that the
scatterplot shows a downward trend due to a weak negative correlation with data points
scattered and clustered around the trendline without a clear trend.

Figure 4.18. Correlation between temperature and TSP in Autumn

56
 IV. Winter (December-February)

The winter season contained 449 data sets among the expected 540. The values of R,
correlation coefficients between PM2.5 and TSP concentrations, and the temperature
for Ratnapark were -0.087 and 0.107, respectively. The TSP data is significant at 0.05
level (two-tail), however, PM2.5 data did not show a significance level. This means
that the p-value or correlation between PM2.5 and temperature may not represent a true
or association, and that the results may have happened by random chance. Because the
p-value (0.064) is bigger than the significance level, the observed correlation between
these two variables is not statistically significant. The results showed that the
relationship between temperature an in Ratnapark showed a weak negative correlation,
which means that as temperatures increase, PM2.5 levels tend to decrease and vice
versa in Winter (December-February) whereas TSP concentrations show weak positive
correlation which means that as temperatures decrease, TSP levels tend to increase and
vice versa.

Some explanations for these associations have been identified. Winter season in
Kathmandu is dry and cold and, here temperatures are expected to drop sharply from
late December to January. Cold temperatures are associated with increased energy
consumption and increased automobile emissions due to less efficient combustion [41].
This would mean higher concentration of PM2.5 in Ratnapark which is a major
transportation hub in Kathmandu Valley. Consequently, cooler temperatures can also
reduce dilution and dispersion of air, reduced temperatures still particulate pollutants
which get trapped closer to the surface, which leads to poor dispersion and thus, traps
PM 2.5 particles [41]. The last theory is that of temperature inversion. The valley often
has weather conditions that lead to a formation of a warm layer above the surface,
which acts like a lid trapping colder air and temperatures closest to the ground, this is
known as temperature inversion. This not only prevents heat from reaching the ground
surface but also prevents pollution such as PM 2.5 and TSP from dispersing into the
atmosphere. This stagnant air causes lowered temperatures but high rise of PM2.5
levels. However, since the significance level is low, it is important to note that
correlation between PM 2.5 and temperatures might be influenced by other local
factors, emission sources and meteorological conditions.

57
 As for the TSP with a positive correlation, it would mean that falling temperatures
would lead to falling TSP levels. This could happen in a few different ways. Winter in
Kathmandu may experience occasional rain or favorable wind patterns which help
cleanse atmosphere by diluting and dispersing particulate matter from air, thus
removing TSP from the atmosphere. Similar concept can be used for improved
atmospheric dispersion during warmer temperatures after a temperature inversion
event. This leads to reduced TSP levels during a period of higher temperatures in
winter. However, many studies found higher TSP concentrations were highly
correlated with colder temperatures (winter), thus a negative correlation in this study
would have made more sense [13], [15], [18].

A scatter plot of temperature against PM2.5 and TSP concentrations were created to
observe how data points are distributed along the plot. Each year was coded in a
different color, the R2 values were calculated, and a linear equation; (y = mx + b), was
passed through the scatter plot. In Fig. 4.12., the correlation between PM2.5 and
Maximum Temperature is -0.087, and the R2 value is 0.0008 with a linear equation of y
= -0.0097x+ 20.54. Here we can see that the scatterplot shows a slight trend
downwards due to a weak negative correlation with data points scattered and clustered
around the trendline without a clear trend.

Figure 4.19.Correlation between temperature and PM2.5 in Winter

58
Similarly in Fig. 4.13., the correlation between TSP and Maximum Temperature is 0.107
and the R2 value is 0.011 with a linear equation of y = 0.00362x+19.23. Here we can see
that the scatterplot shows a slight trend upward due to a weak positive correlation with data
points scattered around the trendline.

Figure 4.20. Correlation between temperature and TSP in Winter

59
V. Relationship of maximum temperature and air quality with different seasons

Table 4.6. Summary of Seasonal Variation and Air Quality

PM 2.5
Correlatio R^2 TSP R^2
Area Season n value Correlation value
Ratnapark Spring 0.166** 0.045 0.212** 0.028
Summer -0.038 0.001 0.142** 0.02
Autumn -0.677** 0.458 -0.332** 0.11
Winter -0.087 0.0008 0.107* 0.011

Positive Correlation
Negative Correlation
* Correlation is significant at the 0.05 level (2-tailed).
** Correlation is significant at the 0.01 level (2-tailed).

According to the results calculated above, Table 4.5. was created to tabulate the results of
the relationship of maximum temperature and air quality such as PM2.5 and TSP with
different seasons. The final verdict is that PM2.5 shows a mostly negative weak correlation
with temperature whereas TSP shows a weak positive correlation with temperature. The
strongest correlation is shown in the Autumn season whereas the weakest is in summer and
winter. Different explanations have been identified for the results above.

4.4. Dual Axis Trend Chart

For the dual-axis line chart, the average yearly temperature of TIA (1030), average yearly
PM 2.5 concentration, and average yearly TSP of Ratna Park for six years were calculated
using Google Sheets. These were plotted on a graph, and the results were calculated using
the line chart in Google Sheets. The Dual Axis Line chart shows different variables in one
chart to give us a better representation of temperature and PM 2.5. concentration over 6
years. It shows us the patterns and trends of these variables. In this analysis, we inputted
the six-year average temperatures on the left Y-axis and air quality parameters such as
PM2.5 and TSP on the right Y-axis, whereas the time (In years) is on the X-axis. To check
for the data accuracy, results were compared with the Status of Air Quality in Nepal
Annual Report 2016-20 and Status of Air Quality in Nepal Annual Report 2021 [42], [43].

60
 4.4.1. Dual Axis Trend Chart of PM 2.5.

Figure 4.21. Dual Axis Line Chart showing trend between temperature and PM 2.5 over 6 years in
Ratnapark.

According to the results in Fig. 4.14., both temperature and PM2.5 show similar trends and
a moving average trendline. With the average annual temperature rising to 20.19 o C in
2017 and gradually falling for the next couple of years, whereas the PM 2.5 falls in 2017 to
40.31 µg/m3 but rises for the next couple of years until 2020. The year 2020 is subjected to
a sharp fall in PM 2.5 concentrations and is considered an outlier year in this time series
because of the lockdown era from the pandemic COVID-19; however, after this
externality, the PM 2.5 concentrations seem to rise again in 2021 with an average
concentration of 47.43 µg/m3. Both these graphs do not have all 365 days of data, thus
could be subjected to certain inconsistencies. The initial reasoning for the graph was to
show the positive or negative relationship of temperature and PM 2.5, however, we can see
that the relationship of these parameters is more complicated than it seems and can be
accounted to other influences in the environment such as meteorological conditions,
pollution levels, seasonal variation, human activities, etc.

61
 4.4.2. Dual Axis Trend Chart of TSP

Figure 4.22. Dual Axis Line Chart showing trend between temperature and TSP over 6 years in Ratna
Park

According to the results in Fig. 4.15., temperature and TSP show different trends. The
average annual temperature rises to 20.09 o C in 2017 and falls for the next couple of years,
whereas the TSP falls drastically in year 2019 and 2020 to 99.53 µg/m 3 and 100.59 µg/m3.
The year 2020 is subjected to sharper fall in TSP concentrations and is considered an
outlier year in this time series because of the lockdown era from the pandemic COVID-19.
This is when minimum human activity was recorded leading to lowered levels of pollution.
However, after this externality, the TSP concentrations seem to rise again in 2021 with an
average concentration of 159.45 µg/m3. Both these graphs do not have all 365 days of data,
thus could be subjected to certain inconsistencies. The initial reasoning for the graph was
to show the positive or negative relationship of temperature and TSP, however, we can see
that the relationship of these parameters is more complicated than it seems and can be
accounted to other influences in the environment such as meteorological conditions,
pollution levels, seasonal variation, human activities, etc.

62
 4.4.3. Dual Axis Trend Chart of TSP and PM 2.5.

Figure 4.23. Dual Axis Line Chart showing trend between TSP and PM 2.5 over 6 years in Ratna Park

According to the results in Fig. 4.16, TSP and PM2.5 show similar trends. According to
Pearson’s correlation, these data sets are positively correlated at 0.593, meaning that as PM
2.5 rises so does TSP and vice-versa. The results show a distinct rise and fall together in
the past 6 years to support the moderately positive correlation of these variables. PM 2.5
and TSP concentrations rise in Ratna Park due to various reasons. The area has a high
traffic congestion and is in the heart of the city, thus representing an urban area, therefore
subjected high levels of pollution. Similarly, the dip in 2020 represents the country wide
lockdown due to the spread of the COVID-19 pandemic. A study by Dhital mentions how
air quality for PM 2.5 decreased by 38.1% during the dry season segment of the lockdown
[44], thus giving us insights on the dip seen in both PM 2.5 and TSP levels.

63
 CHAPTER 5
CONLUSIONS AND RECOMMENDATIONS
5.1. Conclusions
This study aimed at assessing the relationship between air quality parameters and
temperatures to prove that UHI effect exists in Kathmandu Valley. It addressed four
specific objectives: (i) establishing the temperature differences between study areas, (ii)
identify and assess the different air quality parameters that affect temperature to prove UHI
exists in the Kathmandu valley (iii) assess the correlation between temperature and air
quality parameters in different seasons over a 6-year timeframe and (iv) creating a
trendline to show the relationship of temperature and air quality parameters (PM2.5 and
TSP). Through a comprehensive research design including data collection, analysis, and
interpretation all these objectives were met.

The analysis of temperature variations between urban and rural areas in the Kathmandu
Valley confirmed the presence of the UHI effect. The urban center which was the
International Airport had temperatures higher than that of assigned the rural center
(Khokana). The UHII in different seasons were also assessed to check for the seasonal
variation for temperatures, the study found that clear differences of temperatures are seen
in Spring and Winter, UHII for TIA (1030) than Khokana (1073) which would mean that
temperature during Spring and Winter are higher in the inner city than the surrounding
areas. Temperatures in urban regions were consistently higher than in rural areas,
emphasizing the necessity of addressing this phenomenon to improve urban livability and
prevent potential negative implications on public health and energy consumption.
Similarly, the findings of this study also show the intricate interplay between air pollutants
and temperature within the urban environment of the Kathmandu Valley. The research
revealed that PM2.5 and TSP have negative correlations, however, when examined closely
across different seasons, some relationships seem to have a positive correlation as well,
indicating that seasonal variations exist in these relationships.

The study’s significance is that it assesses the relationship of UHI with a specific
parameter which is air quality, studies have not directly linked air pollution and UHI of

64
different seasons. Similarly, this research can give an accurate representation of UHI at the
micro-level as KVs topography and temperature differences may not accurately portrayed
with satellite experiments. By investigating relationship between air quality parameters and
temperatures to prove that UHI, this research aims to shed the light on UHI and its
interaction with air pollution. The assessment of UHI helps in determining the major
components responsible for accelerating the UHI effect and the suitable mitigation actions.
The discovered correlations between air quality metrics and the UHI impact can help urban
planners and policymakers develop evidence-based policies, effective measures to improve
air quality, lower the UHI effect, and promote sustainable and resilient cities.

However, the study also acknowledges some limitations, including data availability and
data cleaning, representativeness of data and the complexity of the relationship between
UHI and other factors like population density, LULC changes, precipitation and humidity,
human activities, climate change, urban pattern/design, green spaces etc. Future research
could benefit from long-term monitoring and more extensive spatial coverage to gain a
deeper understanding of this event.

In conclusion, this thesis highlights the critical relationship between air quality parameters
and the UHI effect in the Kathmandu Valley. Temperatures are increasing yearly, this
increase will have a negative impact on humans, environment, and the economy, hence, to
mitigate this problem an interdisciplinary approach must be taken to address complex
urban challenges. Concerned authorities must plan and implement efforts to prevent rural
migration [32]. The results of this study opens the way for evidence-based policies,
interventions, and long-term urban planning with the goal of making cities healthier, more
resilient, and habitable. Kathmandu and other cities should work toward creating a
sustainable future to ensure the well-being of urban inhabitants by prioritizing air quality
control and tackling the UHI effect.

65
 V.2. Recommendations

The findings of this thesis provide important insights into the relationship between air
quality measures and the Kathmandu Valley's UHI impact. Some findings were not as
anticipated; however, this study can help to direct future research efforts and enhance the
quality of studies on the complex relationship between air quality indicators and UHI. The
unexpected negative low correlation coefficients underline the need for a more refined and
comprehensive approach in future research. Future researchers should consider the
following recommendations to produce more strong and conclusive results:
I. Collecting data from more diverse locations to provide a wider range and
accurate representation of the entire valley.
II. Incorporating a longer time frame to establish better trends; historical trends
should be integrated into the analysis as well.
III. Expanding the scope of air quality parameters that can affect UHI such as O3,
NOx, SOx, VOC, and CO2 to offer a more holistic understanding of how
various pollutants interact with temperature.
IV. Improving data quality to address the issue of null/empty values by using
interpolation of seasonal/monthly averages to minimize data gaps and improve
data reliability.
V. Using multifactor analysis or a more advanced statistical analysis method like
multi-linear regression analysis to establish causation or deeper analysis of the
relationship of air quality and temperature.
VI. Incorporate some aspects of spatial techniques such as GIS or remote sensing to
account for spatial variability.

The unexpected results for the study should serve as a steppingstone for future research as
the relationship of air quality and temperature is multi-faceted. These recommendations
can help researchers refine their methodologies and develop better results to form better
understanding of UHI dynamics. Similarly, mitigation efforts for UHI have been reviewed
to understand solutions for urban heat.

66
 V.2.1. Mitigation efforts for UHI

Multiple studies and papers were assessed and reviewed, and several significant
recommendations and mitigations strategies have been compiled to address the issues
created by UHI. These guidelines can help policymakers, urban planners, researchers,
and other relevant stakeholders toward creating effective policies that promote
sustainable urban development. The recommendations are as follows:

I. Increasing amount of vegetation: Tree plantation is an effective strategy to reduce

UHI as trees absorb CO2, improve air quality, increase wind flow, cool
surroundings through evapotranspiration and produce a natural cooling effect
[29].

II. Green roof or rooftop gardens: A vegetative layer grown on rooftops to provide
shade, lower temperatures (evapotranspiration), filter air, promote local
biodiversity and bring energy balance in cities [29].

III. Cool Roofs or High Albedo Roofing Materials: These are roofs made of light
colored highly reflective materials that help buildings remain cooler and reduce
their energy use [29].

IV. Cool pavements, or whitewashing roads and sidewalks: Studies have shown UHI
effect can be reduced by replacing dark surfaces in urban areas with high albedo
surfaces which reflect solar radiation [29].

V. Pervious Pavements: Pavements built up permeable materials to allow water

infiltration, thereby bringing a cooling effect and lowering temperatures [6].

VI. Water Body Formation: Increasing surfaces with water bodies will increase the
LST through their evaporative action, thus, reducing temperatures. Similarly,
water also has high heat absorption capacity [6]. Water sensitive city design

67
 polices have been formed to reduce UHI in many cities as waterbodies provide a
cooling effect [4].

VII. Improving urban ventilation: Increasing green areas along with systematic and
efficient building arrangements in urban areas [4].

VIII.Proper Urban Planning: Considering layout of streets and buildings can be an

efficient and holistic approach for city planners. This could mean planning

buildings in a systematic way; at a 45o angle, widening width of streets, varying

heights of buildings, etc. to allow certain wind paths and circulation for a cool
airflow [29].

IX. Green city policy: Implementing green strategies like planning urban forests
(parks), street trees, private green in gardens, and green roofs or facades,
monitoring air quality, interdisciplinary department collaboration to tackle UHI
and knowledge dissemination as a method for policy implementation.

68
 REFERENCES

[1] UN-DESA, “World Urbanization Prospects 2018: The 2018 Revision,” 2018.

[2] National Statistics Office, “National Population and Housing Census 2021,” Mar. 2023.
Accessed: May 11, 2023. [Online]. Available: file:///Users/pal/Downloads/National
%20Report_English.pdf

[3] C. L. Chidi, R. K. S. Magar, and D. S. Magar, “Assessment of urban heat island in

Kathmandu valley (1999-2017),” Geographical Journal of Nepal, vol. 14, pp. 1–20, Mar.
2021, doi: 10.3126/gjn.v14i0.35544.

[4] K. Deilami, M. Kamruzzaman, and Y. Liu, “Urban heat island effect: A systematic review
of spatio-temporal factors, data, methods, and mitigation measures,” International Journal
of Applied Earth Observation and Geoinformation, vol. 67. Elsevier B.V., pp. 30–42, May
01, 2018. doi: 10.1016/j.jag.2017.12.009.

[5] B. Mishra, J. Sandifer, and B. R. Gyawali, “Urban Heat Island in Kathmandu, Nepal:
Evaluating Relationship between NDVI and LST from 2000 to 2018,” International
Journal of Environment, vol. 8, no. 1, pp. 17–29, Jan. 2019, doi: 10.3126/ije.v8i1.22546.

[6] Md. Nuruzzaman, “Urban Heat Island: Causes, Effects and Mitigation Measures - A
Review,” International Journal of Environmental Monitoring and Analysis, vol. 3, no. 2, p.
67, 2015, doi: 10.11648/j.ijema.20150302.15.

[7] A. Rajkarnikar and I. Pradhan-Salike, “Urban Heat Island in Kathmandu Valley: A Case of
Patan,” Month, no. 8, pp. 2350–8906, 2020.

[8] G. Ulpiani, “On the linkage between urban heat island and urban pollution island: Three-
decade literature review towards a conceptual framework,” Science of the Total
Environment, vol. 751. Elsevier B.V., Jan. 10, 2021. doi: 10.1016/j.scitotenv.2020.141727.

69
[9] P. Jonsson, C. Bennet, I. Eliasson, and E. Selin Lindgren, “Suspended particulate matter
and its relations to the urban climate in Dar es Salaam, Tanzania,” Atmos Environ, vol. 38,
no. 25, pp. 4175–4181, Aug. 2004, doi: 10.1016/j.atmosenv.2004.04.021.

[10] L. Yang, F. Qian, D. X. Song, and K. J. Zheng, “Research on Urban Heat-Island Effect,” in
Procedia Engineering, Elsevier Ltd, 2016, pp. 11–18. doi: 10.1016/j.proeng.2016.10.002.

[11] Y.-H. Kim and J.-J. Baik, “Maximum Urban Heat Island Intensity in Seoul,” 2002.

[12] Climate Central, “HOT ZONES: URBAN HEAT ISLANDS,” Jul. 2021.

[13] O. Rimawi, Al-Dahabi Indira, Al-Zu’bi Yasin, and Al-Zu’bi Jarrah, “The relationship
between total suspended particulate matter (TSP) and different climatic factors: The case
of Jordan,” Science and Technology, vol. 8, no. 1, pp. 308–312, Jan. 2010, [Online].
Available: https://www.researchgate.net/publication/266344531

[14] Y. Fang and K. Gu, “Exploring coupling effect between urban heat island effect and
PM2.5 concentrations from the perspective of spatial environment,” Environmental
Engineering Research, vol. 27, no. 2, Apr. 2022, doi: 10.4491/eer.2020.559.

[15] T. Wati and R. I. Nasution, “Evaluation of urban pollution and bio-climate using total
suspended particles and discomfort index in Jakarta City,” IOP Conf Ser Earth Environ
Sci, vol. 203, p. 012003, Dec. 2018, doi: 10.1088/1755-1315/203/1/012003.

[16] T. Ajie Ronowijoyo, M. Arief Budiharjo, and S. Sumiyati, “Analysis of Ambient Air
Quality Conditions of TSP Parameters and Its Source in Temon District,” E3S Web of
Conferences, vol. 202, p. 02009, Nov. 2020, doi: 10.1051/e3sconf/202020202009.

[17] World Health Organization (WHO), “Exposure to air pollution: A major public health
concern,” 2017.

[18] Q. Weng and S. Yang, “Urban Air Pollution Patterns, Land Use, and Thermal Landscape:
An Examination of the Linkage Using GIS,” Environ Monit Assess, vol. 117, no. 1–3, pp.
463–489, Jun. 2006, doi: 10.1007/s10661-006-0888-9.

70
[19] S. LaDochy and M. Witiw, “The Continued Reduction in Dense Fog in the Southern
California Region: Possible Causes,” Pure Appl Geophys, vol. 169, no. 5–6, pp. 1157–
1163, May 2012, doi: 10.1007/s00024-011-0366-3.

[20] Conserve Energy Future, “Causes, Effects and Solutions To Urban Heat Island,” 2023,
2023.

[21] K. Karar, A. K. Gupta, A. Kumar, and A. K. Biswas, “Seasonal Variations of PM10 and
TSP in Residential and Industrial Sites in an Urban Area of Kolkata, India,” Environ Monit
Assess, vol. 118, no. 1–3, pp. 369–381, Jul. 2006, doi: 10.1007/s10661-006-1503-9.

[22] F. Sezer Turalıoğlu, A. Nuhoğlu, and H. Bayraktar, “Impacts of some meteorological

parameters on SO2 and TSP concentrations in Erzurum, Turkey,” Chemosphere, vol. 59,
no. 11, pp. 1633–1642, Jun. 2005, doi: 10.1016/j.chemosphere.2005.02.003.

[23] EPA, “Particulate Matter (PM) Pollution,” Jul. 18, 2022. https://www.epa.gov/pm-
pollution/particulate-matter-pm-basics (accessed May 17, 2023).
[24] C. Zhong, C. Chen, Y. Liu, P. Gao, and H. Li, “A Specific Study on the Impacts of PM2.5
on Urban Heat Islands with Detailed In Situ Data and Satellite Images,” Sustainability
(Switzerland), vol. 11, no. 24, Dec. 2019, doi: 10.3390/su11247075.

[25] J. Wang and S. Ogawa, “Effects of Meteorological Conditions on PM2.5 Concentrations in

Nagasaki, Japan,” Int J Environ Res Public Health, vol. 12, no. 8, pp. 9089–9101, Aug.
2015, doi: 10.3390/ijerph120809089.

[26] C. Zhao, Y. Wang, Y. Wang, H. Zhang, and B. Zhao, “Temporal and spatial distribution of
PM2.5 and PM10 pollution status and the correlation of particulate matters and
meteorological factors during winter and spring in Beijing,” pp. 418–427, Feb. 2014.

[27] G. Yang et al., “Influence on Urban Heat Island (UHI) Effect in Beijing and the Possible
Mechanisms,” Journal of Geophysical Research: Atmospheres, vol. 126, no. 17, Sep.
2021, doi: 10.1029/2021JD035227.

71
[28] X. Li, Y. J. Feng, and H. Y. Liang, “The Impact of Meteorological Factors on PM2.5
Variations in Hong Kong,” IOP Conf Ser Earth Environ Sci, vol. 78, p. 012003, Jul. 2017,
doi: 10.1088/1755-1315/78/1/012003.

[29] D. A. Sachindra, A. W. M. Ng, S. Muthukumaran, and B. J. C. Perera, “Impact of climate

change on urban heat island effect and extreme temperatures: A case-study,” Quarterly
Journal of the Royal Meteorological Society, vol. 142, no. 694, pp. 172–186, Jan. 2016,
doi: 10.1002/qj.2642.

[30] B. Baniya, K.-A. Techato, S. K. Ghimire, and G. Chhipi-Shrestha, “A Review of Green

Roofs to Mitigate Urban Heat Island and Kathmandu Valley in Nepal,” Applied Ecology
and Environmental Sciences, vol. 6, no. 4, pp. 137–152, 2018, doi: 10.12691/aees-6-4-5.

[31] A. Ishtiaque, M. Shrestha, and N. Chhetri, “Rapid Urban Growth in the Kathmandu Valley,
Nepal: Monitoring Land Use Land Cover Dynamics of a Himalayan City with Landsat
Imageries,” Environments, vol. 4, no. 4, p. 72, Oct. 2017, doi:
10.3390/environments4040072.

[32] R. Rai, “Assessment of LST variation in Kathmandu, NEPAL,” Oct. 2019. Accessed: Jul.
27, 2023. [Online]. Available:
https://storymaps.arcgis.com/stories/a70d27a801bf4972a005e03cb004e068

[33] R. A. Memon, D. Y. C. Leung, and C.-H. Liu, “An investigation of urban heat island
intensity (UHII) as an indicator of urban heating,” Atmos Res, vol. 94, no. 3, pp. 491–500,
Nov. 2009, doi: 10.1016/j.atmosres.2009.07.006.
[34] A. Tzavali, J. Paravantis, G. Mihalakakou, and A. E. Fotiadi, “Urban heat island intensity:
A literature review,” Fresenius Environmental Bulletin, vol. 24, pp. 4537–4554, Jan. 2015.

[35] E. I. Obilor and E. C. Amadi, “Test for Significance of Pearson’s Correlation Coefficient
( )OBILOR,” International Journal of Innovative Mathematics, Statistics & Energy
Policies, vol. 6(1), pp. 11–23, 2018, [Online]. Available:
https://seahipaj.org/journals-ci/mar-2018/IJIMSEP/full/IJIMSEP-M-2-2018.pdf

72
[36] Stephanie Glen, “Coefficient of Determination (R Squared): Definition, Calculation,”
Statistics How To, 2023.
https://www.statisticshowto.com/probability-and-statistics/coefficient-of-determination-r-
squared/ (accessed Jul. 22, 2023).

[37] B. Rohwerder, “Seasonal vulnerability and risk calendar in Nepal ,” UK: GSDR, Apr.
2016. https://gsdrc.org/publications/seasonal-vulnerability-and-risk-calendar-in-nepal/
#:~:text=The%20dry%20season%20(March%2DJune,avian%20influenza%2C%20and
%20crop%20damage. (accessed Jul. 23, 2023).

[38] ICIMOD, “Investigating air quality in the Kathmandu Valley: The need for data,”
ICIMOD, May 03, 2021. https://www.icimod.org/investigating-air-quality-in-the-
kathmandu-valley-the-need-for-data/ (accessed Jul. 23, 2023).

[39] X. Feng and S. Wang, “Influence of different weather events on concentrations of

particulate matter with different sizes in Lanzhou, China,” Journal of Environmental
Sciences, vol. 24, no. 4, pp. 665–674, Apr. 2012, doi: 10.1016/S1001-0742(11)60807-3.

[40] T. Miyama et al., “Increases in Biogenic Volatile Organic Compound Concentrations

Observed after Rains at Six Forest Sites in Non-Summer Periods,” Atmosphere (Basel),
vol. 11, no. 12, p. 1381, Dec. 2020, doi: 10.3390/atmos11121381.

[41] A. Watson, R. Bates, and D. Kennedy, Air Pollution, the Automobile, and Public Health.
Washington (DC): National Academies Press (US), 1988. Accessed: Jul. 23, 2023.
[Online]. Available: https://www.ncbi.nlm.nih.gov/books/NBK218142/

[42] Government of Nepal Ministry of Forests and Environment Department, “Environment

Status of Air Quality in Nepal Annual Report, 2021,” Kathmandu, 2023. [Online].
Available: www.doenv.gov.np

[43] Government of Nepal, “Ratnapark Air Quality Monitoring Station,” Kathmandu, 2016.

[44] N. B. Dhital, D. R. Bhattarai, R. P. Sapkota, K. Rijal, R. M. Byanju, and H.-H. Yang,

“Comparing the Change in Air Quality during the COVID-19 Lockdown between Dry and

73
Wet Seasons in Nepal,” Aerosol Air Qual Res, vol. 22, no. 8, p. 220201, 2022, doi:
10.4209/aaqr.220201.

74