Using landsat data to monitor the land cover change in vientiane capital, lao pdr

THAI NGUYEN UNIVERSITY UNIVERSITY OF AGRICULTURE AND FORESTRY NAMKHO SONGKHAM USING LANDSAT DATA TO MONITOR THE LAND COVER CHANGE IN VIENTIANE CAPITAL, LAO PDR BACHELOR THESIS Study

THAI NGUYEN UNIVERSITY UNIVERSITY OF AGRICULTURE AND FORESTRY

NAMKHO SONGKHAM

USING LANDSAT DATA TO MONITOR THE LAND COVER CHANGE IN

VIENTIANE CAPITAL, LAO PDR

BACHELOR THESIS

Study mode : Full-Time Major : Environmental Science and

Management Faculty : International Program Office Batch : ESM48 (2017 – 2020)

Thai Nguyen, 8 th September 2021

DOCUMANTATION PAGE AND ABSTRACT

Thai Nguyen University of Agriculture and Forestry

Degree Program Bachelor of Environmental Science and

by using intensive desktop for analysis land cover in particular area by using land satellite image, land satellite image bring a benefit for many proposes especially analysis area without time-travelling Vientiane is located on center

of Lao PDR and considered as capital of Lao PDR this capital is also a development area By using ArcGIS software to classify a land cover types by using supervised classification The result illustrates the decreasing of forest area and increase in built-up scale In term of environmental impacts, extending of settling area is the factor of urban population growth which leads to environment issues This research will gather a result and provide a changing detail on land cover in Vientiane and in provision of development plan to mitigate the issues that probably happen in the future

Keywords Land cover, Land cover change, GIS, Urbanization,

Landsat image, Vientiane Lao, USGS

Number of Pages 56

Submission date

ACKNOWLEDGEMENT

I would like to express the deepest appreciation to my supervisor, Mr Nguyen Van Hieu, who has the attitude and substance of a genius For this patience, motivation, enthusiasm, and immense knowledge His suggestion helped

me in all the time of the research and conducting of this thesis Without his advice and persistent help this thesis would not have been possible

Besides my supervisor, I would like to thank the rest of my thesis committee: Mr Bui Minh Tung For his cooperation and suggestion to my thesis about comment and question Moreover, I would like to say thank you to Geoinformatics Research Center (GIRC) for offering a chance to conduct research

Last but not least, I would like to thank my family, my parent Kay Namkho and Sonechan Namkho For giving birth to me at the first place as well as encouragement and support throughout my life

TABLE OF CONTENT

DOCUMANTATION PAGE AND ABSTRACT i

ACKNOWLEDGEMENT iii

TABLE OF CONTENT iv

LIST OF FIGURES vii

LIST OF TABLES viii

LIST OF ABBREVIEATION ix

PART I INTRODUCTION 1

1.1 Research Rationale 1

1.2 Research Questions and Hypothesis 3

1.3 Research Objective 4

1.4 Significance of Study 4

1.5 Scope and Limitations 5

1.6 Definitions 5

PART II LITERATURE REVIEW 7

2.1 Urbanization 9

2.2 Urbanization in Vientiane 11

2.3 Land use and Land cover study 14

PART III METHODOLOGY 18

3.1 Materials and Methods 18

3.2 Study area 19

3.3 Introduction to the preprocessing 20

3.4 Software used in this study 20

3.5 Data collection 20

3.6 Band combination 21

3.7 Image clip 24

3.8 Image analysis and classification 25

3.9 Supervised classification 26

PART IV RESULT 28

4.1 Land Cover Map In 2000 28

4.2 Land cover map in 2013 29

4.3 The land cover map of 2020 30

4.4 Land cover between 2000 and 2013 31

4.5 Land cover between 2013 and 2020 34

4.6 Land cover between 2000 and 2020 37

4.7 Land cover change in Vientiane from 2000 and 2020 39

4.8 Field survey 42

PART V DISCUSSION 44

5.1 How has land use in Vientiane changed from 2000 to 2020 44

5.2 Reason Behind Land Use Changes in Vientiane 45

5.3 Urbanization and Its Impacts on the Environment: the case of Vientiane, Laos 47

5.4 Implications for Urban Planning and Management: 50 PART VI CONCLUSION 53 REFERENCES 55

LIST OF FIGURES

Figure 3.1 Flowchart of the methodology 18

Figure 3.2 Study area Vientiane 19

Figure 3.3 Process of band combination 23

Figure 3.4 Band combination completed 23

Figure 3.5 Clipping image process using Vientiane boundary 24

Figure 3.6 Taking training samples 27

Figure 3.7 Types of samples 27

Figure 4.1 Vientaine’s land cover map in 2000 28

Figure 4.2 Vientaine’s land cover map in 2013 29

Figure 4.3 Vientaine’s land cover map in 2020 30

Figure 4.4 Land cover Vientiane 2000 and 2013 31

Figure 4.5 Graph of land cover change 2000 and 2013 32

Figure 4.6 Land cover Vientiane 2013 and 2020 34

Figure 4.7 Comparison graph of land cover 2013 and 2020 35

Figure 4.8 Land cover of Vientiane 2000 and 2020 37

Figure 4.9 Comparison graph of land cover 2000 and 2020 38

Figure 4.10 Land cover change map of Vientiane 39

Figure 4.11 Graph of Land cover change 41

Figure 4.12 Google Earth image of December 2000 and February 2020 42

LIST OF TABLES

Table 3.1 Data sources 21

Table 3.2 Band combination for natural color (ESRI) 22

Table 4.1 Details of land use of Vientiane in 2000 29

Table 4.2 Details of land use of Vientiane in 2013 30

Table 4.3 Details of land use of Vientiane in 2020 31

Table 4.4 Details of land use of Vientiane 2000 and 2013 32

Table 4.5 Details of land use of Vientiane 2013 and 2020 35

Table 4.6 Details of land use of Vientiane in 2000 and 2020 38

Table 4.7 Details of land cover change in Vientiane 2000 and 2020 40

LIST OF ABBREVIEATION

CAC Center for Clean Air Coalition

JICA Japan International Cooperation Agency

Lao PDR Lao People Democratic Republic

NOAA National Oceanic and Atmospheric Administration

In parallel with the rapid development of all nations across the globe, the

natural resources of such countries are now coping the severe crisis The

conversion from natural resources such as the plantation areas, the forest patches and natural land area to the infrastructure and expansion of agricultural is putting such an enormous pressure on the global biodiversity According to the report of United nation in 2018, people more than 55 percent are dwelling in urban area meaning that more than a half of total population are contributing to the global biodiversity degradation

Lao PDR, known as a country with the abundance and diversity of forests and natural resources, are now being affected by urbanization in the similar way Vientiane capital, the small and relaxed city located on the central of Laos, it is the large city and flattening in geography that situated along the Mekong River (lao-guide-999.com) City has a monsoon climate with the average daytime temperature above 80 ºF (27ºC) and more than 80% of average annual participation

falling in five months from May to September The city is modernized by the infrastructure, building and facilities, however, it still keeps the beauty of traditions and culture since many old temples are located inside the town (Augustyn, 2016) Additionally, with more than 948,000 people are settling down, Vientiane is considered as a largest city where people mostly embody to earn a living As an economic city, most of citizens in Vientiane capital occupied as a worker, officer, and farmer in suburb In view of the fact that most people in Vientiane is farmers, thus, the large paddy field and agricultural area is plenty to be seen along the capital city Interestingly, with the potential of being an attraction of inventors to contribute their business, Vientiane is popularly known as one of economic attraction cities in Laos Therefore, due to this issue, Vientiane have expended the urban to suburb with growing of economic and develop a large area, this resulted

in the transforming from forest area to infrastructure and agricultural land (JICA, 2011) This necessitated to monitor the change in land use of Vientiane capital through time to compare how the biodiversity has been degrading and how the urbanization has undertaken

In order to monitoring the land use, several elements will be produced in ArcGIS software including Land use change and land cover change mapping To specify, Land use is the specification of land such as building, vegetation and agriculture that can be classified due to what can be built on the land and what the land can be used for Category of community, environment or placing that determined can be used on a specific type of land The arrangements, activities, and inputs that people attempt in a certain land cover type are called land use

(DIANA EASTMAN, 2018) In order to identify what a region is covered by forest, instruction, agriculture, and another land cover types and types of water it can be classified by Land cover data Water types include wetlands or open water Land use is an indicator of the propose of using by people in that area use the land field whether for development, conservation, invasion, or mixed uses Land cover can be managed with the different types and use differently (NOAA, 2020)

It is necessary and worthwhile to study how the land use had been changed according to time Therefore, to the best knowledge, this study attempts to perform the land use change and urbanization mapping of Vientiane capital of Laos by employing the modern technology tools to compare the transformation and variation of Land use between the year of 2000 and 2020 with LULC by using Landsat image via ArcGIS software

1.2 Research Questions and Hypothesis

This study will address the following questions:

1 How the land use has been changed during the period of 2000 to 2020 in Vientiane capital, Laos PDR?

2 How the urbanization affects the environment?

3 What could be the suitable plan urbanization management in the area ? what could be a practical practice?

Null hypothesis:

1 The land use/land cover change in Vientiane capital, Lao PDR have severe changes from 2000 to 2020

2 The land use/land cover change in Vientiane capital, Lao PDR impacts on environment since 2000 to 2020

This research aims to accomplish the following to:

1 Assess the land use change of Vientiane capital of Laos during the period

1 Student: the study about land use and land cover change in capital city of

Laos is required more studies conduct Moreover, the information that will be generated from this study can be the secondary data that other students can use for

further research

2 Society: the result of this study could be applied to the urbanization

management strategies in the area Moreover, this finding will also contribute to the better environmental management as well as raise people’s awareness towards

the impact of urbanization to biodiversity

1.5 Scope and Limitations

The scope of this study will cover all the area of municipality of Vientiane Required data information regarding to this study and some advice from the GIS specialist will be collected to escalate the quality of this study

However, there might be some factors limiting the study, firstly the potential limitation such as data collection, survey, information downloading with some sources are not available and need to be paid, GIS license Moreover, there are some difficulties of the research such as the resolution of the image after classified quite difficult to recognized based on the inaccuracy of the data and the source such as land satellite image are rarely available for collecting Especially, the time for gathering the information is quite short

1.6 Definitions

Nowadays, human activities have been the main cause of forest areas transformation Conversion to plantation and agriculture represents a significant contributor to loss of native forest that has been rapidly increasing in tropical regions caused by these activities include agriculture spreading, plantation establishment, infrastructure development, hydropower or energy development and mining (Thapa et al 2013) In order to improve the understanding of human and environmental interactions with vegetation

condition and structure by using LULC method (Coops et al 2009) Biodiversity loss based on land cover change is one of the center management challenges

at both global and regional scale (Ghazoul 2013)

Personally, as being born and grown up in Vientiane, I have seen the continuingly gradual changes in term of the density of the population and utilization of land Because of that, it makes me interested in conducting research

on Vientiane I decided to conduct this research by applying GIS skill that I have learned during my academic, together with doing a literature review By doing so,

I wish to understand the current state of urbanization of my hometown and legal framework related to it

PART II LITERATURE REVIEW

At local, regional, and global scales can be affected by Land use and land cover changes with significant environmental consequences Land cover change is the factor of biodiversity loss, disturb in hydrological cycles, increase in soil erosion, and sediment loads of the regional and global scales (Lambin et al 2006) With Sustainable assessment of LULC changes provide environmental, social, and economic dimensions For land management, monitoring, and degradation these can be observed by using earth observation Landsat data Via GIS techniques and remote sensing to assess LULC changes, especially in mountainous area where accessibility is limited it can carry out many studies around the globe (Mottet et al 2006) At the local level, such as processes of degradation of land and landscape-level biodiversity, watershed runoff, microclimatic resources, soil erosion, and sediment loads are the consequences of changing in the use of land and cover of land (Sultan, 2016) Commonly, study of LULC change is based on a comparison

of present and historical data that are available through ground, airborne, and satellite means The gathered techniques of remote sensing and GIS provide the important information with accuracy over time, many sectors are able to know the changes and their expansion of area, and the pattern of changes (Islam et al 2018) The LULC pattern of a region is a benefit of natural and socio-economic factors, and their utilization are considered a central component for protecting natural resources and monitoring environmental changes (Su et al 2011)

Furthermore,the term ‘LULC change’ known as the human modification of the terrestrial surface of the earth as well as the study of land surface change (Ellis

et al 2010) The LUCC drivers analyzed in this study include biophysical changes for example, climate change and human activities such as, population growth, poverty, land policies, and GDP growth (Palamuleni, 2009), (Wiyo et al 2015) For decision-makers, LUCC analysis can provide important information for global environment change and sustainable development studies (Hu et al 2019) LULC change can lead to effect of human settlements and livelihoods, as well as the natural resources that are depended upon by the inhabiting humanity, wildlife, fisheries, and other valued wild and domestic life forms (MRC, 2010) Even though, Land cover map is provided a useful understanding of the factor By the way, there are imperfect and have biases of all maps of LULC and LULC change for example classification and positional errors and uncertainties about the standard of such products (Verberg et al., 2011) The biases like errors, how it is processed, minimum mapping unit, classification scheme, characteristics of the study area, and other factors can occur rely on kind and quality of remotely sensed data from given land cover change map data product (Congalton, 2015) The limitation and pose challenges on how digital LULC maps should be used, looking

on the appliance are affected fromthose imperfections Accuracy assessments of Land use and Land cover change map can provide information on the standard of the map data set and the way to effectively use these products

The useful of present study about spatial geo-integrated analysis techniques, which offer information about the pronounced changes within the epicentral and surrounding areas that are related to the earthquake at an everyday quantity (Nath

et al 2018) Moreover, Forestland and shrubland are declining, while tillage land

and artificial surfaces have increased within the area, and deforestation appears and become serious problem by LULCC include occurring of rapid growth and GDP, macroeconomic activities, especially within the western part of the river such as manufacturing industries, and insufficient national policies that have failed

to effectively enforce ban of uncontrolled harvesting of forest resources In step with Xiao and Weng (2007), LULC were largely attributed to density of population

on the land, a rapidly growing of economic, poor land use planning, and therefore the inconsistency within the govern-mental policies and also Changes in LULC were in the middle of changes in LST cause by explanation for urban/built-up leaded temperature differences between the and the surrounding rural areas significantly widened This might result in an intensified urban heat island effect within the urban areas The abundance of forest was crucial factor influencing LST This study clearly indicates the many impacts from development activities

by demanding of its citizen on LU/LC change (Praveen et al 2013) Chang and Hou (2013) states that, modern remote sensing technology has been widely used, but there are some problems for instance the environment factor because of the limitation of data quality and there is no unified theoretical system for reference, that the methods and models employed by researchers have obvious regional limitations

2.1 Urbanization

At the level of increasing world urbanization nowadays and also the amount and unprecedented size of the world’s largest cities At the 20th century, just 16 cities within the planet the overwhelming particularly in advanced industrial countries contained more than a million people Nowadays, there are

more than 300 cities comprise a million people or more, and nearly seventy percent

of citizen are found in the developing world In the year 2007, for the primary time

in human history, there were low amount of people living in rural area, more people

in the world will be living in cities by 2017 within the developing world is probably to became more urban in outstanding than rural (United Nation, 2004) Causes of urbanization and city growth are a variety of various factors including migration of rural–urban, increasing of natural population, and annexation In rural areas the rate of natural is higher than the urban area, that generally decrease every day, rural–urban migration are therising levels of urbanization principal reasons,supplementing is the factor of geographic expansion of urban areas, and the transformation and reclassification of rural villages became small urban settlements The metropolitan periphery is expanded might be caused both, firstly the arrival of recent migrants and second the sub-urbanization of the center class out of the central city The associated with importance of each of these different causes of urbanization and suburbanization varies both amongst and between regions and countries (Cohen, 2006) Cities are also centering of contemporary living, where female working class is greatest participated and where indicators of general health and well-being, literacy, women’s status, and social mobility are typically highest Finally, cities are also important social and the integrating

of cultural centers for instance, house museums, art galleries, film industries, theaters, fashion houses, and other important cultural centers (Cohen, 2006)

High population density in the city may also be good for reducing the effect of human on local ecosystems High density of population typically assumes lower per capita cost of providing infrastructure and basic services And even though, the urban poverty with high rates is found in many cities, urban

residents, on average, however, there are a better access to education and health care, similarly other convenient basic public services for instance electricity, water, and sanitation than people in rural areas (Cohen, 2006) There is an increasing rapidly of the amount and proportion of individual living in urban areas between

1970 and 2000, there was a rising of urban area to 5.8 million ha globally (Seto et al., 2011) China was urbanized alone around 2.8 million ha in Between 2000 and

2010 (World Bank Group, 2015) Habitat loss and degradation globally caused by expansion of urbanization It has several impacts on the environment for example, changing onto land use and impact on hydrological system (Grimm et al., 2008) Moreover, in urban areas consist of Parks, yards/gardens, medians on highways, and golf-courses included Another method of measuring urbanization includes built-up areas or place dominated by the built environment or impervious surfaces (Liu et al., 2014) Admire urbanization is that the increase in human populations near or in protected areas

2.2 Urbanization in Vientiane

Vientiane, as the capital of a rustic that passes three main eras which are a kingdom empire, colonial and independent eras(Vongpaseuth et al 2020) It became a capital city of the country since 1560 Ever since it has always been a crucial city of the country in term of social, economic, and political status (Viravong, 1964 as cited in Rafiqui & Gentile, 2009) From 1893 to 1953, Lao was under French colonization (Vongvilay et al., 2016) During that time, it did not gain interest for infrastructure construction such as traffic from the French compared to other Indochina countries like Vietnam and Cambodia due to its lack of access to the sea (Okamoto et al., 2014) According to

Okamoto et al (2014), thank to the fear of the influence of communist, the United States of America financed plenty in Laos, particularly in Vientiane, leaving other provinces of the country remain in rural situation According

to Rafiqui & Gentile (2019), since 1975, the evacuation of US (Okamoto et al., 2014) (or also known as a communist takeover in Rafiqui & Gentile, 2019), the urbanization in Laos came to rest until the introduction of “marketizing economy” in 1986 Since then, Laos began to share similarity urbanization characteristics to those in other regions in Monsoon Asia.(Sharifiet al., 2014; Okamoto et al., 2014).In 1995, Vientiane had experience expansion within the built-

up area(Okamoto et al., 2014) For the past 15-20 year, the population of Vientiane has doubled (Rafiqui & Gentile, 2009) The GDP of Laos has increased rapidly, from 473 USD in 2005 to 1077 USD in 2010 and 1850 USD in 2015 (UNSD 2017) Additionally, the population of Vientiane is estimated to grow dramatically from 761,000 in 2010 to 1540,000 in 2025 (United Nations Population Division 2015) Its population increased from 200.000 in 1975 to half of million within the 1990s because of the migration caused by the rise in demand for monetary income

In 2045, the population of Vientiane is expected to reach 1.4 million (Centre for Development and Environment, 2018)

According to Lao Statistics Bureau (2016), the population density of Vientiane is eight times beyond the national average in which 4 out of 10 people in the city were lifetime migrants mostly from the rural areas of the country(Lao Statistics Bureau,2016) Currently, despite being the foremost populated city of the country, Vientiane is still a small capital compared to other

capital cities in its region, such as Bangkok, Hanoi, and Phnom Penh with the common of only 1744 persons per square kilometer in the four metropolitan districts (Centre for Development and Environment, 2018) Yet, 78% of the urbanized area is still situated in Vientiane (Lao Statistics Bureau,2016) Currently, over half of the urban population that concentrates in Vientiane is creating challenges for authorities to plan and provide services, especially

in term of infrastructure provision and efficient use of land (Centre for Development and Environment, 2018) At present, most of the population seems to live near the business zones or concentrate long the transportation route, meaning that there are more built-up areas along Mekong River and national roads (JICA, 2011) This case could result in the fragmentation of land use in Vientiane (Cao etal.,2019) One uniqueness of Vientiane is that even though it has high urban growth, but there is no shantytown The urbanization

of Vientiane is seen as the expansion toward the surrounding suburbs along the national roads (JICA, 2011)

This expansion of the geographic area to the inadequate infrastructure suburbs may lead to the worsening living condition of its population thank to inadequate infrastructure and social services (Centre for Development ad Environment, 2018) In 2009 The population of Vientiane Capital was 795,000 The annual growth rate of Vientiane Capital was 3.1% between 1985 and 2005 The GDP rate was 7.6% in 2008 For the economic output, manufacturing, mining quarrying and repair sector are major contributors to GDP growth (World Bank, Lao office, 2010) Focusing on land use of Vientiane Capital in 2005, Vientiane was covered forest area, which occupied about 68% of total area, and

17% of paddy area respectively Rapid economic development and increment may have largely affected the land use pattern The built-up area doubled from approximately 3% of total area in 1995 to six percent in 2005 ( JICA, 2011) Personally, Vientiane is the Centre of Lao as known as the capital This area is that the Centre of the office and organization According to Lao Statistics Bureau (2016), About one million people added to the country’s population every decade

At its current population, the country remains one amongst the tiniest in geographical region slightly bigger than Singapore, but less than half of neighboring Cambodia

2.3 Land use and Land cover study

Hasan et al (2019) stated that land-use changes had been researched to have adverse environmental impacts such as releasing greenhouse gases, depleting ozone layers Their impacts are affecting various sectors such as agriculture, biodiversity in both local and global levels (Hasan et al., 2019) Therefore, it is essential to study land-use and land cover change.Landsat program was first launched in the 1970s, followed by the developed of many classification techniques which were the result from it (for example, random forest and maximum likelihood classifiers) (Ha et al., 2018) Provide better understanding

of socio-economic and physical drivers of forest cover change at a local level, which is useful for policy makers to ensure the effective management of land use and forest resources However, expanding the social and environmental factors (e.g., income and timber species) included could improve the analysis and provide deeper understanding of factors driving forest clearance, but is dependent on

availability of suitable data (Phomphila et al., 2017) The research on urban land use change has not only revealed spatial and temporal characteristics of urban expansion

or land use dynamics but has also provided support and evidence to address serious environmental issues threatening urban sustainable development (Su et al., 2012)

Using land use maps, in order to avoid negative impact associated with urban expansion, using land use maps is great method for planners and policy makers Although GMS covers a large extent, various climate zones and eco-regions, and offers an extremely interesting area from all imaginable application fields related to Earth observation, however not many LULC studies have been published (Kuenzer, 2014) The technologies used for identifying different land cover types have evolved from simple unsupervised approaches to various complex supervised classifications (e.g., maximum likelihood classification, support vector machine, decision tree, wavelet transform classification, artificial neural networks classification, and so on), and from pixel-based methods to subpixel (e.g., linear spectral unmixing, support vector regression, and decision tree regression) and object-oriented methods Moreover, Studies are continuously improving LCLUC detection methods, which are evolving along with newly developed classification methods for the scales of studies range from regional

to continental and global (Lu et al.,2004) According to Paul & Rashid (2016), Land use change is a process by which human activities transform the natural landscape, referring to how land has been used, usually emphasizing the functional role of land for economic activities Land use changes are often not straight line and

might trigger feedbacks to the system, stress living conditions, and threaten people with vulnerability

According to Pheng Souvanthong (1995), Rural land use in Lao PDR covers

a spectrum from extensive, pioneer shifting cultivation at high altitudes, through a range of forest types and uses, to intensive, permanent, irrigated lowland agriculture Cultivation with shifting might covers a range of farming practices that occur within this spectrum In this report, farming practices are grouped according

to the social categories generally in use, i.e., Lao Lum, Lao Theung, Lao Sung However, it may be more appropriate to categorize land use systems according to region: north, central, and southern Lao PDR show quite specific geographical characteristics, and it is probably that future forestry planning will be based on these three categories Types of shifting cultivation are described below, following

a background description of forestry and agricultural activities Most of the changes in land cover type came from the changing of forest and dry forest to agricultural areas, and this might be the main effect for the changes in the potential provisions of ecosystem over two decades Most of the potential provisions of ecosystem decreased due to these changes (Thipphachanh et al., 2017) Thanks to the development of remote sensing technologies, the research on the field of land-use and land cover changes are now benefited from that (Mewenda, Watanabe, & Avtar, 2020) With remote sensing and GIS data, the researcher can now study the changes in land-use accurately and timely (Reis, 2018) Land-use managers and planners are using remote sensing data to gain fundamental data in a non-expensive cost, especially in developing countries where there is not much information

available (Mewenda, Watanabe, & Avtar, 2020) Laos, which is experiencing

the limited access to information, is also benefiting from the remote sensing data

Especially, land use and land cover (LULC) change are playing an important role

in global change, Earth systems, and sustainable development with the

development Earth observation techniques (Lambin et al., 2006)

PART III METHODOLOGY 3.1 Materials and Methods

The detailed methodology adopted in this thesis to achieve the above objective is described in this part The land satellite images were obtained from United State Geological Survey (USGS) www.earthexplorer.usgs.gov and Vientiane boundary was obtained from www.diva-gis.org

Flowchart of the broad steps followed in this work for deriving statistics of land use pattern of the area is shown in Figure 3.1

Figure 3.1 Flowchart of the methodology

Data collection ( Satellite image, boundary of study

3.2 Study area

Vientiane is the capital city of Laos It is located in the Centre of the country,

on the left of Mekong riverbank to the north of Nongkhai province(a province in Thailand)with the estimated total area of 3583 km2 (Sharifi et al 2014) There are nine districts in Vientiane which are: Chanthabuly, Sikhottabong, Xaysettha, Sisattanak, Naxaithong Xaythany, Hadxayfond, Sangthong, and Pakngum in which the first four districts are considered as the metropolitan area

As being the capital city of one of the poorest countries in the world, it is the smallest capital city in the region but with high potential on economic and demographic growth (Rafiqui & Gentile, 2009) At present, it is the location

of centralized governmental institutions such as the Presidential palace, government offices, ministries, and governmental institutions

Figure 3.2 Study area Vientiane

3.3 Introduction to the preprocessing

In this study the satellite image data contains very large area, but the study area in this thesis is small, and different images have different spatial resolutions One advantage of the remote sensing data is the ability to perform image processing for modifying or improving the image However, the data

of the study area is rarely to collect and the image that obtained from the website were took by different time for example, one took at nighttime and another one took at daytime so, there are some reflection and refraction that affects to the image quality Moreover, there will be some noises in data, and it will be difficult to identify the distinctions between features on the image when the edges of them are not clear

3.4 Software used in this study

Three following software were used for this project

1 ArcMap 10.3 – used for operating those process such as combining

bands and classification the image

2 Google Earth Pro 7.3 – used for comparing the image, in order to ensure the quality of image analyzing

3 Microsoft Excel 2016 – used for calculating the area after classification by importing the data from ArcMap to summarize the area

3.5 Data collection

For this thesis, land satellite image and land satellite data were acquired for three years namely, 2000, 2013 and 2020 Firstly, the land satellite image of 2000

there was only Landsat7 The imagery belonged to month of October and November, with less than 10% land cloud cover Next, the land satellite image of

2013 is Landsat8, and the imagery belonged to month of February, July, and September with land cloud cover less than 10% Lastly, land satellite image of

2020 is landsat8 belonged to month of January by less than 10% cloud cover Moreover, the resolution of all imageries is 30m Furthermore, the obtained imageries are very large area the study area is small thus, the boundary of study area is necessary for narrowing the image therefore, the boundary was collected from diva-gis.org In addition, those processing operated with WGS 1984 UTM Zone 48N

No Data type Data Production Resolution (m) Source(s)

1 Landsat7 2000

October and November

Combining band is use for different purposes for instance, band 6,5,2 used for agriculture analyzing and band 5,6,4 with color land/water (Kevin Butler, 2013)

In this study, three Landsat images are needed for contrast between year 2000,

2013 and 2020 with natural color red, green and blue For combining band in this study band1 to band7 were combined for both images in order to get a good result and classify easily For Landsat image of 2000 there was only Landsat7 and Landsat image of 2013 and 2020 is Landsat8 Based on the table 3.2

Band Name Bandwidth (µm) Resolution (m) Band Name Bandwidth (µm) Resolution (m)

Band 1 Blue 0.45 – 0.52 30 Band 2 Blue 0.45 – 0.51 30 Band 2 Green 0.52 – 0.60 30 Band 3 Green 0.53 – 0.59 30

Band 3 Red 0.63 – 0.69 30 Band 4 Red 0.64 – 0.67 30

Table 3.2 Band combination for natural color (ESRI)

By using image analysis tools in ArcMap, the satellite image included band7 were added to the program, to perform the composite band Furthermore, after adding 7 bands, open image analysis by selecting those band in the windows

band1-of image analysis However, this process must be performed for each image respectively

Figure 3.3 Process of band combination

Figure 3.4 Band combination completed

3.7 Image clip

As mentioned before, the satellite image contains a very wide area But

in practical application, most of the areas in the satellite image are not necessary for the research To decrease the time consuming and storage space, the images should be clipped to extract the study areas in this study This step operated

on ArcMap software and the study area is Vientiane capital of Laos For clipping the image, clip function is used by using polygon, Polygon subset is performed

by setting up a polygon to clip the image For instance, at first click on the effect icon on processing the window of function will appear, and choose the input means Landsat image from mosaic function, for Clipping geometry means study area Vientiane so, use Vientiane boundary and tick on Use Input Feature for Clipping Geometry and Landsat image of Vientiane will be appeared

Figure 3.5 Clipping image process using Vientiane boundary

3.8 Image analysis and classification

Nowadays government and private agencies apply remote sensing imagery for a wide range of applications from military applications to farm development and analysis for a large project The images may be a panchromatic, multispectral, hyperspectral, or even ultra-spectral of terra bytes Remote sensing image classification is one amongst the most significant application worlds for remote sensing (K Perumal, 2010) Therefore, in this study focus on five types of the surface of the land namely, Water (surface water), Forest (plant and forest), Agriculture(including farm and rice field), Built-up (including building and residential) and other land field (open land, industrial area, and land field), in this study applied supervised classification in order to classify the land cover of Vientiane

A wide variety of techniques for classifying the different ground features from RS data have been developed The objective of the processing of image classification is to classify all the pixels in an image into different categories by users (Lillesand, 2004) Supervised classification associated with the use of training area data that are meditated representative of each rock type or surficial unit to be classified (Ford et al., 2008a,b) Unsupervised classification is

a computer-automated process which can be called “clustering” due to it is operated based on the natural classifying of pixels of the remote sensing data (Bakr, et al.,2010)

3.9 Supervised classification

Supervised classification is the technique frequently used for the quantitative analysis of remote sensing image data At its core is the feature of separating the spectral domain into regions that can be associated with the ground cover types of interest to a particular application However, in practice those regions may sometimes overlap A variety of algorithms is available for the task, and it is the purpose of the research to cover those most commonly encountered Essentially, the different methods vary in the way to identify and describe the regions in spectral space (Richards J.A 2013)

Supervised classification is more flexible than the unsupervised classification due to the selection of pixels which can be identified directly

or based on other sources like Google Earth which is available for user to search the same position and observe the type of land use and land cover

of this position is up to users In additional, this research performed supervised classification, with 5 types of land cover for example, Water, Forest, Agriculture, and another field of land cover In order to perform this process, all samples of those fields must be taken with accurately Moreover, The Interactive Supervised Classification tool accelerates the maximum likelihood classification process due to the figure It works the same as the Maximum Likelihood Classification tool with default parameters All the bands from the selected image layer are used by this tool in the classification The classified image is added to ArcMap as a raster layer

Figure 3.6 Taking training samples

Figure 3.7 Types of samples