*2.1.2.2.1. High spatial resolution remote sensing satellite*

High spatial resolution remote sensing data have been a fundamental and strategic national resource, serving to provide accurate mapping, urban planning, land resource management, environmental monitoring, ground mapping, military mapping, and intelligence gathering. Because of its huge economic and military benefits, high spatial resolution remote sensing satellites are quickly developing all over the world. The IKONOS satellite, which was launched in 1999, marked the beginning of the commercial high‐resolution satellite era; in 2001, the QuickBird satellite was developed by Digital Globe Company. **Table 4** shows the high spatial resolution remote sensing satellites that have a resolution of no more than 1 m [55].


**Table 4.** Bands included in ETM+ and OLI.

#### *2.1.2.2.2. Hyperspectral remote sensing satellite*

Hyperspectral resolution remote sensing is performed to obtain many narrow‐ and continu‐ ous‐spectrum remote sensing images in the visible light, near infrared, intermediate infrared, and thermal infrared of the electromagnetic spectrum [56]. Hyperspectral resolution remote sensing technology contains the special properties of a fine structure of spectra and abundant date information; moreover, it has incomparable application advantages with respect to the identification, assortment, and information extraction of ground objects, which give it great potential for application to ecological environment monitoring [57].

On August 23, 1997, the first hyperspectral remote sensing satellite (LEWIS) was launched in the United States. After years of development, many hyperspectral remote sensing satellites


have been developed and successfully operated. **Table 5** shows some of the hyperspectral resolution remote sensing satellites [57].

**Table 5.** Common hyperspectral resolution remote sensing satellites.

*2.1.3. Extracting urban land areas using remote sensing in the Indochinese Peninsula*

## *2.1.3.1. Remote sensing data sources*

In this chapter, for the study of urban expansion of the primary cities in the Indochinese Peninsula, Landsat TM/ETM+ and Landsat‐8 images from 2000 to 2015 were primarily selected for urban area identification. Dynamic changes were analyzed using results from multiple years. In the process of interpreting the remote sensing data, Google Maps is an important reference for this region. **Table 6** shows the various types of high‐quality satellite remote sensing data used in this study. To study urban expansion at the national level, spatial data on built‐up areas in the East Asia region for the period from 2001 to 2010 were also used in this chapter.

#### *2.1.3.2. Image processing of remote sensing*

To study urban development in the Indochinese Peninsula, eight primary cities with popula‐ tions of more than 500,000 were selected in this chapter: Naypyidaw and Yangon (Myanmar), Hanoi and Bien Hoa (Vietnam), and Bangkok and Chon Bury (Thailand) for the period from 2000 to 2015 and Vientiane (Laos) and Phnom Penh (Cambodia) for the period from 2000 to 2010 because the remote sensing data for Vientiane and Phnom Penh were deficient in 2015 (**Figure 1**). In this study, we extracted the urban land information from the remote sensing


**Table 6.** Types of high‐quality satellite remote sensing data used in this study.

have been developed and successfully operated. **Table 5** shows some of the hyperspectral

**bands**

1600–2430 6 30 8125–11,650 5 90

400–1050 18–62 17

900–2500 160

900–2500 160

**Spatial resolution (m)**

resolution remote sensing satellites [57].

14 Sustainable Urbanization

CHRIS 2001 European Space

*2.1.3.1. Remote sensing data sources*

*2.1.3.2. Image processing of remote sensing*

this chapter.

**Satellites Launch time State Band (nm) Number of**

ASTER 1999 USA–Japanese 520–860 3 15

MightySat 2000 USA 500–1050 256 or 512 30 HYPERION 2000 USA 400–1100 60 30

ARIES 2004 Australian 400–1100 60 30

HSI 2008 Chian 450–950 115 100

*2.1.3. Extracting urban land areas using remote sensing in the Indochinese Peninsula*

In this chapter, for the study of urban expansion of the primary cities in the Indochinese Peninsula, Landsat TM/ETM+ and Landsat‐8 images from 2000 to 2015 were primarily selected for urban area identification. Dynamic changes were analyzed using results from multiple years. In the process of interpreting the remote sensing data, Google Maps is an important reference for this region. **Table 6** shows the various types of high‐quality satellite remote sensing data used in this study. To study urban expansion at the national level, spatial data on built‐up areas in the East Asia region for the period from 2001 to 2010 were also used in

To study urban development in the Indochinese Peninsula, eight primary cities with popula‐ tions of more than 500,000 were selected in this chapter: Naypyidaw and Yangon (Myanmar), Hanoi and Bien Hoa (Vietnam), and Bangkok and Chon Bury (Thailand) for the period from 2000 to 2015 and Vientiane (Laos) and Phnom Penh (Cambodia) for the period from 2000 to 2010 because the remote sensing data for Vientiane and Phnom Penh were deficient in 2015 (**Figure 1**). In this study, we extracted the urban land information from the remote sensing

Agency (ESA)

**Table 5.** Common hyperspectral resolution remote sensing satellites.

MODIS 1999 USA 400–1400 36 250, 500, 1000

image by visual interpretation. This approach was used because of its advantages of simplicity and accuracy, although it is also time‐consuming and costly. According to the shape and image features of the ground objects, most of the study area can be identified using this approach. For example, farmlands, water bodies, residential blocks, etc., can be easily recognized. Remote sensing TM7 is a medium‐infrared waveband in which the rock shows a strong reflection, TM4 is a near‐infrared waveband in which vegetation can be strongly reflected, and TM3 is the red waveband that shows the primary absorption of vegetation chlorophyll. Thus, we selected band combinations 7, 4, and 3, which can be used to identify the urban area with the charac‐ teristic of the built‐up areas on less vegetation biota, whereas the suburban area shows abundant vegetation biota. **Figure 2** shows the technical route for the built‐up area extraction, and **Figure 3a–d** shows the distribution of built‐up areas in the representative cities of Yangon, Chon Bury, Bangkok, and Hanoi during various periods.

**Figure 2.** Map showing the technical route for the built‐up area extraction.

**Figure 3.** Maps showing the distribution of built‐up areas in the cities of Yangon **(a)**, Chon Bury **(b)**, Hanoi **(c)**, and Bangkok **(d)** in the Indochinese Peninsula in different periods.

#### *2.1.4. Urban expansion rate*

In some of those previous studies [58, 59], the built‐up area is considered an indicator for urban sprawl monitoring, and these areas always represent the status of a city's construction and development from the perspective of space in urban geography. Thus, in this study, the urban expansion rate was adopted to evaluate the spatial distribution and rate of urban sprawl in the Indochinese Peninsula for the period 2000 to 2015. The urban expansion rate that can be defined in Equation (2) shows changes in the quantity of the urban area per unit time and is a key parameter for evaluating spatial changes in urban sprawl [59, 60].

$$R\_{UL} = \frac{UL\_{u \times i} - UL\_i}{UL\_i} \times \frac{1}{n} \times 100\% \tag{2}$$

where *RUL* stands for the expansion rate of urban land; *ULn+i* and *ULn* stand for the built‐up area in the target unit at times *n*+*i* and *i*, respectively; and *n* is the interval of the calculation period (in years).

#### **2.2. Methodology for driving force analysis for urban expansion**

The dynamic changes of urban areas meet socioeconomic development, along with land use in the urban fringe area and the interior region, after continuous adjustment and configuration result in a transformation into urban land. With increased population, an increasing amount of the rural population is changed into an UP [61]. The dynamic changes of urban areas express urbanization in space and are an inevitable consequence of urbanization. Pattern‐process‐ mechanism always guides the geographical study, and pattern is the distribution of the geographical objects and phenomena; process stands for the analysis of changes in the geographical objects and phenomena in time and space; and mechanism finds the reasons for these changes. Thus, driving force analysis for urban expansion can enable a better under‐ standing of urban development and policy decisions [62]. This chapter presents a multiple‐ factor model (geographic position, regional economic development, population, infrastructure, and foreign economic and trade relations) to explore the driving forces of urban expansion in countries of the Indochinese Peninsula using multiple principles and multiple‐ level data.

**Figure 4.** Map showing GDP in the countries of the Indochinese Peninsula.

#### *2.2.1. Data sources for driving force analysis*

#### *2.2.1.1. Regional economic development*

**Figure 3.** Maps showing the distribution of built‐up areas in the cities of Yangon **(a)**, Chon Bury **(b)**, Hanoi **(c)**, and

In some of those previous studies [58, 59], the built‐up area is considered an indicator for urban sprawl monitoring, and these areas always represent the status of a city's construction and development from the perspective of space in urban geography. Thus, in this study, the urban expansion rate was adopted to evaluate the spatial distribution and rate of urban sprawl in the Indochinese Peninsula for the period 2000 to 2015. The urban expansion rate that can be defined in Equation (2) shows changes in the quantity of the urban area per unit time and is a

<sup>1</sup> <sup>+</sup> 100% - = ´´ *ni i*

where *RUL* stands for the expansion rate of urban land; *ULn+i* and *ULn* stand for the built‐up area in the target unit at times *n*+*i* and *i*, respectively; and *n* is the interval of the calculation

The dynamic changes of urban areas meet socioeconomic development, along with land use in the urban fringe area and the interior region, after continuous adjustment and configuration

*UL n* (2)

*i*

key parameter for evaluating spatial changes in urban sprawl [59, 60].

*UL UL <sup>R</sup>*

*UL*

**2.2. Methodology for driving force analysis for urban expansion**

Bangkok **(d)** in the Indochinese Peninsula in different periods.

*2.1.4. Urban expansion rate*

16 Sustainable Urbanization

period (in years).

Urban development primarily rests on financial strength, and economic development accel‐ erates city changes and urban expansion. To an extent, urban land‐use can be viewed as an economic issue, which is also noted in prior studies [63, 64]. Thus, Gross Domestic Product (GDP) can be regarded both as an integrated index reflecting regional economic development and as a predictive factor for urban development. **Figure 4** shows GDP in the countries of the Indochinese Peninsula. These data were obtained from the GMS Statistics data set on the ADB Website, and the data set provided the latest state and trend of key GMS economic data according to the International Monetary Fund (IMF) World Economic Outlook [65, 66]. From 2000 to 2014, the economies of the countries of the Indochinese Peninsula have experienced rapid growth. Except for Thailand, the peninsula's economy has grown almost 7‐fold over the last 15 years in Myanmar, Laos, Cambodia, and Vietnam.

## *2.2.1.2. Population data*

To analyze the driving force of urban expansion, data on the UP (% of total) were used in this study. The number of people living in the urban land area is generally defined as UP, and the ratio of UP to total population relates to the percentage of the total population living in cities; UP (% of total) is usually regarded as an indicator of urbanization additional to built‐up areas [67, 68]. The UP (% of total) data set for countries in the Indochinese Peninsula for the period from 2001 to 2014 used in this chapter were obtained from the WB's World Development Indicators (WDI), and these data show the numbers of urban residents per 100 total population [69]. The UN Department of Economic and Social Affairs, using the cohort component method, has developed population estimates for developing countries that lack census data. The Department calculated this data set and provided information that is convenient for popula‐ tion studies. These data are considered a valuable scientific reference for population studies, although there is some uncertainty caused by data limitations. **Figure 5** shows the UP (% of total) for countries in the Indochinese Peninsula for the period from 2001 to 2014. According to that figure, the UP proportion increased during the 14 years and the region appeared to experience rapid urbanization, especially in Thailand.

**Figure 5.** Map showing the UP (% of total) for countries in the Indochinese Peninsula for the period from 2001 to 2014.

To explore the relationship between urban expansion and UP, the UN's population statistics for urban agglomerations with 300,000 inhabitants or more in 2014 by country were used in this chapter to perform an urban expansion analysis of the eight cities in the Indochinese Peninsula [1]. Based on national statistics data (population censuses are the most commonly used sources), the UN developed the UP estimation to respond to the sustainable development challenges of urbanization. **Figure 6** shows the UP in Naypyidaw, Yangon, Hanoi, Bien Hoa, Bangkok, Chon Bury, Vientiane, and Phnom Penh in 2000, 2010, and 2015.

2000 to 2014, the economies of the countries of the Indochinese Peninsula have experienced rapid growth. Except for Thailand, the peninsula's economy has grown almost 7‐fold over the

To analyze the driving force of urban expansion, data on the UP (% of total) were used in this study. The number of people living in the urban land area is generally defined as UP, and the ratio of UP to total population relates to the percentage of the total population living in cities; UP (% of total) is usually regarded as an indicator of urbanization additional to built‐up areas [67, 68]. The UP (% of total) data set for countries in the Indochinese Peninsula for the period from 2001 to 2014 used in this chapter were obtained from the WB's World Development Indicators (WDI), and these data show the numbers of urban residents per 100 total population [69]. The UN Department of Economic and Social Affairs, using the cohort component method, has developed population estimates for developing countries that lack census data. The Department calculated this data set and provided information that is convenient for popula‐ tion studies. These data are considered a valuable scientific reference for population studies, although there is some uncertainty caused by data limitations. **Figure 5** shows the UP (% of total) for countries in the Indochinese Peninsula for the period from 2001 to 2014. According to that figure, the UP proportion increased during the 14 years and the region appeared to

**Figure 5.** Map showing the UP (% of total) for countries in the Indochinese Peninsula for the period from 2001 to 2014.

To explore the relationship between urban expansion and UP, the UN's population statistics for urban agglomerations with 300,000 inhabitants or more in 2014 by country were used in this chapter to perform an urban expansion analysis of the eight cities in the Indochinese Peninsula [1]. Based on national statistics data (population censuses are the most commonly used sources), the UN developed the UP estimation to respond to the sustainable development

last 15 years in Myanmar, Laos, Cambodia, and Vietnam.

experience rapid urbanization, especially in Thailand.

*2.2.1.2. Population data*

18 Sustainable Urbanization

**Figure 6.** Map showing the UP in Naypyidaw, Yangon, Hanoi, Bien Hoa, Bangkok, Chon Bury, Vientiane, and Phnom Penh in 2000, 2010, and 2015.

**Figure 7.** Map showing the existing, under construction, and planned/potential railways in GMS countries in 2012.
