ArticlePDF Available

Remote sensing-based geostatistical hot spot analysis of Urban Heat Islands in Dhaka, Bangladesh

August 2023
Singapore Journal of Tropical Geography 44(3)

DOI:10.1111/sjtg.12507

License
CC BY-NC 4.0

Authors:

Nur Hussain

McMaster University

S.M. Shahriar Ahmed

Center for Environmental and Geographic Information Services

Amena Muzaffar Shumi

Urban Heat Island (UHI) refers to a phenomenon whereby urban areas experience higher temperatures compared to the surrounding areas. Remote sensing-based Land Surface Temperature (LST) measurements can be utilized to measure UHI. This study emphasized on geostatistical remote sensing-based hot spot analysis (G Ã i) of UHI in Dhaka, Bangladesh as a way of examining the influences of Land Use Land Cover (LULC) on UHI from 1991 to 2015. Landsat 5 and 7 satellite based remote sensing indices were used to explore LULC, UHI and environmental footprints during the study period. The Urban Compactness Ratio (C oR) was used to calculate the urban form and augmented characteristics. The Surface Urban Heat Island (SUHI) intensity (ΔT) was also used to explore the effects of UHI on the surrounding marginal area. Based on our investigations into LULC, we discovered that around 71.34 per cent of water bodies and 71.82 percent of vegetation cover decreased from 1991 to 2015 in Dhaka city. Contrastingly, according to C oR readings, 174.13 km 2 of urban areas expanded by 249.77 per cent. Our hot spot analysis also revealed that there was a 93.73 per cent increase in hot concentration zones. Furthermore, the average temperature of the study area had increased by 3.26 C. We hope that the methods and results of this study can contribute to further research on urban climate.

Content uploaded by S.M. Shahriar Ahmed

Content may be subject to copyright.

Remote sensing-based geostatistical hot

spot analysis of Urban Heat Islands in

Dhaka, Bangladesh

Nur Hussain,

S.M. Shahriar Ahmed

and Amena Muzaffar Shumi

3†

School of Earth, Environment & Society, McMaster University, Hamilton, Canada

Department of Geography and Environment, Jahangirnagar University, Dhaka, Bangladesh

Faculty of Agricultural Sciences, University of Hohenheim, Stuttgart, Germany

Correspondence: Nur Hussain (email: nurhussain55@gmail.com)

Urban Heat Island (UHI) refers to a phenomenon whereby urban areas experience higher temper-

atures compared to the surrounding areas. Remote sensing-based Land Surface Temperature

(LST) measurements can be utilized to measure UHI. This study emphasized on geostatistical

remote sensing-based hot spot analysis (G

i) of UHI in Dhaka, Bangladesh as a way of examining

the inﬂuences of Land Use Land Cover (LULC) on UHI from 1991 to 2015. Landsat 5 and 7 satel-

lite-based remote sensing indices were used to explore LULC, UHI and environmental footprints

during the study period. The Urban Compactness Ratio (C

) was used to calculate the urban form

and augmented characteristics. The Surface Urban Heat Island (SUHI) intensity (ΔT) was also used

to explore the effects of UHI on the surrounding marginal area. Based on our investigations into

LULC, we discovered that around 71.34 per cent of water bodies and 71.82 percent of vegetation

cover decreased from 1991 to 2015 in Dhaka city. Contrastingly, according to C

readings,

174.13 km

of urban areas expanded by 249.77 per cent. Our hot spot analysis also revealed that

there was a 93.73 per cent increase in hot concentration zones. Furthermore, the average temper-

ature of the study area had increased by 3.26C. We hope that the methods and results of this

study can contribute to further research on urban climate.

Keywords: urban expansion, urban heat island, hot spot analysis, ecological imbalance, urban

metabolism

Accepted: 9 February 2023

Introduction

A key aspect of urbanization involves people migrating from rural to urban regions and

cities (Elmqvist et al., 2013; Vinayak et al., 2022). In recent decades, urbanization has

risen dramatically. Currently, over 56 percent of the world’s population resides in

urban areas, and this percentage is projected to rise to 68 percent by 2050

(UN-Habitat, 2020). Urbanization is associated with extensive land use land cover

(LULC) changes (Deng & Srinivasan, 2016). Remote sensing-based indices help to mea-

sure changes in LULC and environmental footprints. The Normalized Difference Vege-

tation Index (NDVI) has become the most widely used index in vegetation, forestry, and

agricultural research (Gao, 1996;Carlson&Arthur,2000). Notably, the Normalized Dif-

ference Water Index (NDWI), Normalized Difference Built-up Index (NDBI), and

Normalized Difference Bareness Index (NDBaI) were developed based on the NDVI

model. The NDVI, NDWI, NDBI, and NDBaI describe the physical footprints or

†

Current afﬁliation: Data Science, Don Valley Advanced Solutions (DVAS), Toronto, Canada.

doi:10.1111/sjtg.12507

Singapore Journal of Tropical Geography (2023)

University of Singapore and John Wiley & Sons Australia, Ltd.

This is an open access article under the terms of the Creative Commons Attribution-NonCommercial License, which

permits use, distribution and reproduction in any medium, provided the original work is properly cited and is not used

for commercial purposes.

critical indicators of land cover distribution and changes in LULC (Cao et al., 2002;

Deng & Wu, 2012; Zhao & Chen, 2005). LULC measurements indicate changes in the

urban physical landscape and growth rate of urban built-up areas compared to other

physical footprints. Urban compactness is one of the leading indicators of urban growth

rate (Chen, 2011;Duet al., 2016). The consequence of urban compactness expansion is

explained in urban metabolism (Wolman, 1965). Urban LULC inﬂuences the urban

area’s Land Surface Temperature (LST) (Shahfahad et al., 2022a).

The urban LST depends on changes in urban LULC, environmental circumstances

and ecological conditions (Qian et al., 2006). The relationship between LST and LULC

ratiﬁes NDVI, NDWI, NDBI, and NDBal indices (Gillies & Carlson, 1995; Chen

et al., 2006a). Unplanned LULC and/or haphazard urbanization is a leading cause in

the formation of LST-based heat islands in many city areas. Urban Heat Islands (UHIs)

are the most prominent example of local weather changes associated with urbaniza-

tion, in which urban areas experience greater temperatures than the surrounding rural

regions (Howard, 1833; Meng et al., 2018). According to Souch and Grimmond (2006),

the UHI can occur at any time of year, depending on local weather conditions. Gener-

ally, higher temperatures in the urban core commercial areas and lower temperatures

in rural areas are considered as UHIs (Souch & Grimmond, 2006). UHIs occur when

built structures cause the destruction of green cover and wetlands inside the urban area

and its surrounding periphery (Jones et al., 1990). The urban structural land surface

absorbs maximum heat in the daytime and increases low atmospheric air temperature

at night by surface absorbed heat release (Sobstyl et al., 2018), which can affect local

weather and climate (Liu & Zhang, 2011; Rosenzweig et al., 2008; Rayk et al., 2016;

Streutker, 2002). Luke Howard (1833)ﬁrst proposed the concept of UHI in 1833. How-

ever for many years there was less work on UHIs in tropical cities than in temperate

zones. UHI research was conducted in 1964 by Nieuwolt (1966) in Singapore’s south-

ern urban area (Chen et al., 2006b; Roth et al., 2012). In the early 1980s, climatologists

used the UHI study’s energy and water balance processes to calculate urban heat stor-

age (Oke & Cleugh, 1987). When an UHI exists within a city, the temperature differ-

ence is generally more signiﬁcant at night in the winter and when the winds are

weaker at night than daytime in summer (Streutker, 2002). However, different scenar-

ios can exist depending on the spatio-temporal context of analysis.

Thermal remote sensing-based UHI was developed in the early 1990s and has

continuously improved. Several studies have employed remote sensing-based UHI in

their investigations on various cities, i.e. Singapore (Goh & Chang, 1999; Shahfahad

et al., 2022b), Texas (Streutker, 2002), Hong Kong (Nichol & Wong, 2005),

Guangzhou, Boluo, Dongguan, Panyu, Foshan, Gaoming, Huadu, Huizhou, Nanhai

and Sanshui in Guangdong Province (Zhang & Wang, 2008), Dhaka-Bangladesh

(Ahmed et al., 2013), the Salt Lake basin area of Turkey (Orhan et al., 2014), Budapest,

Ljubljana, Modena, Padua, Prague, Stuttgart, Vienna, and Warsaw (Damyanovic

et al., 2016; Mahdavi et al., 2016), North-western Siberian cities (Miles & Esau, 2017),

Beijing (Meng et al., 2018), and Shanghai (Tan et al., 2010). These studies were con-

ducted using satellite-retrieved LST concentration. LST-based UHI has also recently

been applied in multi-disciplinary research areas: i.e. urban expansion and UHI inten-

sity (Li et al., 2018; Hu & Brunsell, 2015), phenology of urban ecosystems, spatial and

temporal pattern distribution of UHI (Santos et al., 2017; Yao et al., 2017), urban atmo-

spheric proﬁle (Hu & Brunsell, 2015), urban weather and climatic zone (Chen

et al., 2014; Bechtel et al., 2015; Huang et al., 2017), vegetative cover (Maimaitiyiming

et al., 2014;Luet al., 2017) and urban land use/urban metabolism (Fu & Weng, 2017;

2Nur Hussain, S.M. Shahriar Ahmed and Amena Muzaffar Shumi

14679493, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/sjtg.12507 by Bangladesh Hinari NPL, Wiley Online Library on [30/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

Tran et al., 2017). Ord and Getis (1995) proposed a model of local spatial autocorrela-

tion statistics. Urban metabolism emphasizes city transformation, city expansion, and

environmental degradation within input-throughput-output appearance

(Newman, 1999; Decker et al., 2000). The UHI distribution is responsible for urban eco-

logical degradation, local climate change, micro-level biodiversity, and ecological imbal-

ance. The physical features of LULC are one of the central derivations of urban

ecological footprints (Wackernagel et al., 2002; Holden, 2004; Chen et al., 2006b) asso-

ciated with urban metabolism and urban sustainability. Hence, UHI assessment is

imperative for ecological conservation and sustainability in urban expansion.

A hot spot can be deﬁned as an area with a higher temperature concentration than

the average temperature zone calculated from a random distribution of the study site

(Hussain & Islam, 2020; Chakravorty, 1995). The hot spot method is an autocorrelation

analysis conducted to explore the interactions and relations among the distribution of

variability in a geospatial location (Baddeley, 2010). Geographic Information System

(GIS) applications have dramatically expanded the use of hot spots (G

i) in recent

decades, especially for geostatistical analysis. The hot spot (G

i) analysis comes from

geostatistical and geospatial autocorrelation models. This local spatial autocorrelation

model stands on G

i(d) and G (d) statistics as ﬁrst introduced by Getis and Ord (1992).

In 2005, Arc User (ESRI) extended spatial analysis tools in ArcGIS spatial statistical

analysis segments (Scott & Warmerdam, 2005) according to the local spatial autocorre-

lation G

i(d) and G(d) statistics. These spatial analysis tools include hot spot (G

i) analy-

sis allowing the spatial autocorrelation method to delineate spatial clusters of nearest

neighbouring objects (Hussain & Islam, 2020). In this study, geostatistical hot spot anal-

ysis was used to detect UHI effects over 25 years, helping policymakers to conduct bet-

ter prediction and urban planning strategies.

In this study, we analyse the trends of UHI in Dhaka city using hot spot G



analy-

sis. Dhaka is rapidly growing and one of the most densely populated megacities globally

(Taleb, 2012; Hossain, 2008; Ahmed et al., 2013). To the authors’knowledge, there

appears to be only one UHI-related research work relating to the Dhaka megacity, con-

ducted by Ahmed et al.(

2013). Ahmed et al.(2013) provided UHI and land-use status

change based on yearly one-day remote sensing data for 10-year intervals from 1989

to 2009. In this research, we calculated UHI effects based on 25 years of data (1991 to

2015) with seasonal variation of the spatial and temporal distribution of heat islands

and compared UHI and temperature variations within the rural-urban fringe area. The

Surface Urban Heat Island (SUHI) intensity (ΔT) indicates the effect of UHI on the con-

centration of the surrounding marginal area (Peng et al., 2012). Our research focuses

on urban expansion, land-use change, UHI distribution, and SUHI intensity. These cir-

cumstances are signiﬁcant for environmental conservation and socio-ecological sustain-

ability. Accordingly, the main objectives of this study are to: 1) explore changes in

LULC, urban expansion and compactness from 19912015 based on remote sensing

data; 2) explore the relationship between land cover changes and variations in LST; 3)

determine UHI distribution using hot spot (G

i) and 4) calculate seasonal SUHI intensity

(ΔT) during the last 25 years from 19912015.

Data and methodology

Study area

This study focuses on the anthropogenetic and environmental context of the Dhaka

megacity. Dhaka, the capital city of Bangladesh, is one of the most densely populated

Hot spot analysis of UHI in Dhaka, Bangladesh 3

and rapidly growing megacities in the world (Ahmed et al., 2013). According to the

Bangladesh Bureau of Statistics, Dhaka megacity’s total population in 1991, 2001, and

2011 was 6 620 697, 10 284 947 and 14 730 537 respectively. It reached 17 million in

2017, and has continued to grow since. The study area is located in the tropical humid

climatic zone between 2340’north and 2354’north latitude and 9020’east and

9030’east longitude (Figure 1), bounded by the Buriganga River, Turag River,

Dhaleshwari River, and Shitalakshya River (Hussain, 2018). During the study period

(19912015), the average temperatures recorded in the winter, summer, and autumn

Figure 1. Map depicting a) Bangladesh, b) Dhaka c) Dhaka’s urban area, and d) Landsat-5, 30m spatial

resolution image with RGB 543 band spectrum of Dhaka’s urban area.

4Nur Hussain, S.M. Shahriar Ahmed and Amena Muzaffar Shumi

seasons were 23.8C, 28.7C and 24.2C respectively. According to Shuttle Radar

Topography Mission (SRTM) data, the mean elevation of the study area is 11 meters

above the mean sea level (Figure 4b).

Data

In this study, we investigated urban expansion, urban compactness ratio CoR

ðÞ, land

surface temperature (LST) and land use land cover (LULC) change using Landsat 5 and

Landsat 7 data. Urban physical footprints were detected using the same remote sensing

data with suppositional vital indicators of physical footprints, such as NDVI, NDWI,

NDBI, and NDBaI to explore the relationship between UHI characteristics and these

indices. While the Digital Elevation Model (DEM) of the study area was extracted using

the SRTM data, SUHI intensity (ΔT) was calculated using the retrieved LST. Hot spot

analysis (G

i) was also explored using LST within the geostatistical analysis of Arc GIS

10.4 (Esri). Table 1represents Landsat 5, Landsat 7, and SRTM data references with

acquisition dates. Landsat 5 satellite images were used for the years 1991, 1995, 1999,

2007, and 2011 while Landsat 7 satellite images were used for the years 2003 and

2015 as Landsat 5 data were not available for 2003 and 2015 (USGS, 2011). All data

types were analysed at four-year intervals to understand the urban phenomenon’s

short and long temporal changing characteristics (Table 1). Unprocessed raw data were

download from the United States Geological Survey (USGS) website; URL (https://

earthexplorer.usgs.gov/).

Data analysis

The research process is summarized in Figure 2. Satellite data were acquired from Landsat

5, Landsat 7, and SRTM in four-year intervals from 1991 to 2015. For each year, three sets

Table 1. Data reference.

Data Type Spatial resolution Path/Row Acquisition Date

Landsat 5 30m 137/044 Jan 26, 1991

60m for thermal band 137/044 Apr 16, 1991

137/044 Aug 22, 1991

137/044 Jan 05, 1995

137/044 Apr 11, 1995

137/044 Sep 02, 1995

137/044 Feb 01, 1999

137/044 Apr 22, 1999

137/044 Oct 15, 1999

137/044 Jan 22, 2007

137/044 Apr 28, 2007

137/044 Sep 19, 2007

137/044 Jan 17, 2011

137/044 May 09, 2011

137/044 Sep 30, 2011

Landsat 7 30m 137/044 Jan 19, 2003

60m for thermal band 137/044 May 11, 2003

137/044 Aug 15, 2003

137/044 Feb 05, 2015

137/044 May 28, 2015

137/044 Sep 17, 2015

SRTM 30m Sep 23, 2014

Source: Table produced by author based on U.S. Geological Survey [https://earthexplorer.usgs.gov/].

Hot spot analysis of UHI in Dhaka, Bangladesh 5

of data were obtained for summer, autumn, and winter. The satellite data went through

pre-processing steps consisting of atmospheric and radiometric corrections. NDVI, NDWI,

NDBI, and NDBal were extracted using processed datasets. LST was also retrieved from

processed thermal bands of Landsat images. LST-based temperature anomalies were used

for hot spot (G

i) analysis to ﬁnd the signiﬁcant and non-signiﬁcant areas. LST and NDVI

indices were used for exploring Surface Urban Heat Island (SUHI).

Land use land cover (LULC) change. Land cover change was extracted using Landsat

5 and Landsat 7 data. Both datasets were analysed in ArcMap 10.7 (Esri) and land

cover was detected using supervised classiﬁcation. NDVI, NDWI, NDBI, and NDBaI

were calculated using the same series of data. NDVI was calculated using visible

red bands (0.63 μm - 0.69 μm) and near-infrared bands (0.76 μm - 0.90 μm) of

Landsat 5 and 7 data according to Equation 1(Kogan, 1995). NDWI was calcu-

lated using near-infrared band (0.76 μm-0.90μm) and short-wave infrared

band-1 (1.55 μm-1.75μm) conforming to Equation 2(McFeeters, 1996). NDBI was

calculated using near-infrared band (0.76 μm-0.90μm) and short-wave infrared band-1

(1.55 μm-1.75μm) following Equation 3(Zha et al., 2003). NDBal was calculated using

short wave infrared band-1 (1.55 μm-1.75μm) and thermal band (10.40 μm-12.50

μm) calibrating to Equation 4(Zhao & Chen, 2005).

Figure 2. Research methods.

6Nur Hussain, S.M. Shahriar Ahmed and Amena Muzaffar Shumi

NDVI ¼b4b3

b4þb3ð1Þ

NDWI ¼b4b5

b4þb5ð2Þ

NDBI ¼b5b4

b5þb4ð3Þ

NDBaI ¼b5b6

b5þb6ð4Þ

where b3are visible red bands (0.63 μm - 0.69 μm), b4is the near-infrared band (0.76

μm - 0.90 μm), b5is the short wave infrared band-1 (1.55 μm - 1.75 μm) and b6is the

thermal band (10.40 μm - 12.50 μm).

Urban expansion and compactness. The compactness ratio is a proportion used to measure

the spatial form and density of urban areas calculated by dividing the total area and

built-up area. In this study, we analysed compactness (i) to determine the direc-

tion of urban expansion and extension of built-up areas during the 4-year interval

and also (ii) to explore the actual changes in the urban centre. The urban expan-

sion ratio was analysed using Landsat 5 and Landsat 7 data. The elevation of the

expansion area was detected using DEM. The DEM works in tandem with the Tri-

angular Irregular Networks (TIN) method. This formula of compactness ratio CoR

ðÞ

put forward by Richardson in 1961 (see Haggett et al., 1977) was applied to calculate

urban form and augmented characteristics (Chen, 2011). The urban compactness ratio

CoR

ðÞcan be calculated using the following equation:

CoR ¼2ﬃﬃﬃﬃﬃﬃ

πA

Pð5Þ

where CoR is the compactness ratio, Ais Area and Pis the perimeter of the urban area.

Retrieval of LST. Chen et al.(

2002) proposed a method to simulate brightness tempera-

ture for the retrieval of LST using Landsat 5 data. Firstly, the digital number (DN) of

the thermal band (10.40 μm - 12.50 μm) is used to convert radiation to luminance

using the following formula:

RTM6¼V

255 Rmax Rmin

ðÞ ð6Þ

where Vrepresents the DN of the thermal band, Rmax ¼1:896 mW cm2sr1

ðÞand

Rmin ¼0:1534 mW cm2sr1

ðÞ. Then, the radiation luminance is converted to bright-

ness temperature in Kelvin, T (K), using the following equation:

T¼k1

In K2

RTM6þ1

 ð7Þ

Where Tis the effective at-satellite temperature in Kelvin, K1 =1260.56

(Calibration constant 2) and K2 =607.66 (Calibration constant 1), both of which are

Hot spot analysis of UHI in Dhaka, Bangladesh 7

pre-launch calibration constants under an assumption of unity emissivity. Lastly,

Kelvin is converted to Degree Celsius using the following equation:

Tc¼T273:15 ð8Þ

where Tis the effective at-satellite temperature in Kelvin and Tcis the temperature in

degree Celsius.

The National Aeronautics and Space Administration (NASA) and the United States

Geological Survey (USGS) proposed retrieving LST from Landsat 7 data. This method

converts simulated image pixels into absolute radiance units using 32-bit ﬂoating-point

calculations (USGS, 2011). Subsequently, pixel values of the simulated image are

scaled by their values before media output using the following equations:

Lλ¼Grescale QCAL þBrescale ð9Þ

Lλ¼LMAX LMIN

QCALMAX QCALMIN QCAL QCALMINðÞþLMIN ð10Þ

where Lλis the Spectral Radiance at the sensor’s aperture in watts/ (meter squared *

ster * μm); Grescale and Brescale can be obtained from the header ﬁle of the satellite

image; QCALMIN =1 (minimum quantized calibrated pixel value); QCALMAX =255

(maximum quantized calibrated pixel value); QCAL =the quantized calibrated pixel

value or DN, LMAX =the spectral radiance that is scaled to QCALMAX in watts/(meter

squared * ster * μm); and LMIN =the spectral radiance that is scaled to QCALMIN in

watts/(meter squared * ster * μm). QCALMIN,QCALMAX,QCAL,LMAX, and LMIN

values are also given in the images’header ﬁle.

Subsequently, the spectral radiance of the sensor was converted to brightness tem-

perature in Kelvin, T (K), using the following equation:

T¼k2

In K1

Lλþ1

 ð11Þ

where T =effective at-satellite temperature in Kelvin; K1 =1260.56 (Calibration con-

stant 2); and K2 =607.66 (Calibration constant 1), both of which are pre-launch cali-

bration constants under an assumption of unity emissivity. Lastly, Kelvin is converted

to Degree Celsius using Equation 8.

Hot spot (G

i) analysis. In 2005 ESRI Arc user developed the hot spot analysis (G

i) model

to explore geospatial patterns (Scott & Warmerdam, 2005). The hot spot analysis (G

calculates high or low values of the cluster within the context of neighbouring data.

The algorithm calculates geostatistical hot spots using the nearest neighbours autocor-

relation model. The temperature concentration zone was investigated using hot spot

(G

i) analysis following equations:

G

i¼P

j¼1

wi,jxjXP

j¼1

wi,j

Sﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ

j¼1

i,jPn

j¼1wi,j





n1

ð12Þ

8Nur Hussain, S.M. Shahriar Ahmed and Amena Muzaffar Shumi

X¼P

j¼1

nð13Þ

S¼

ﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ

j¼1

tXðÞ

2ð14Þ

where G

iare the hot spots and cold spots of spatial autocorrelation statistics; xjis the

attribute value for feature j;wi,jis the spatial weight between feature iand j; and

n=total features.

SUHI intensity (ΔT) calculation. SUHI intensity (ΔT) is calculated as the difference in

mean temperature between the Urban Main Built-up Area (UMBA) and the suburban

boundary area (Meng et al., 2018). Peng et al.(

2012) found that the minimum inﬂu-

ence area of the SUHI effect is 150 percent of the urban area (Peng et al., 2012). Fol-

lowing this concentration, SUHI was measured with a 150 per cent buffer area

surrounding Dhaka metropolitan city. Therefore, SUHI intensity (ΔT) was calculated

using the following equation:

ΔT¼TUrban TBoundary ð15Þ

where ΔTis SUHI intensity, TUrban is the mean LST of the UMBA and TBoundary is the

LST of the urban inﬂuencing neighbouring area.

Accuracy assessment of change detection

Generally, classiﬁcation approaches of change detection have focused on the per-pixel

process. The change vector analysis is advantageous for data visualization and attributes

calculation of changes (Kontoes, 2008). However, the difference vector analysis tech-

nique provides a proliﬁc window for geostatistical analysis. This study used change vec-

tor analysis to retrieve change detection from classiﬁed data sets. The traditional

methods of accuracy assessment involve storing, analysing, and presenting spatial data

of maps. The validation of land cover classiﬁcation is required to establish the results.

Validation is realized by comparing the classiﬁed image against the reference ground

truth data. In the present study, we selected a total of 410 ground truth points,

whereby 100 points were delineated for each land classiﬁcation. The confusion matrix

(classiﬁcation accuracy =correct prediction / total prediction) was used to calculate precision,

the user’s and producer’s classiﬁcation accuracy, and overall classiﬁcation accuracy

(Table 2).

Results

Changes in land cover and physical footprints

Figure 3and Table 4represent the land cover changes of the Dhaka Metropolitan

(DMP) area within seven different periods from 1991 to 2015 in four-year intervals.

Land cover has been classiﬁed into four physical features: water bodies, vegetation,

built-up areas, and bare soil. It can be observed from the results that the land cover of

the DMP area changed rapidly. In 1991, seasonal and permanent water bodies

covered 82.78 km

of the area; however, this area decreased to 23.94 km

in 2015.

Hot spot analysis of UHI in Dhaka, Bangladesh 9

Vegetation coverage reduced from 101.91 km

to 28.72 km

in the last 25 years.

Contrastingly, there was an increase in urban built-up and bare soil areas. From 1991

to 2015, urban built areas increased from 69.32 km

to 171.50 km

, while bare soil

areas increased from 48.25 km

to 77.84 km

(Figure 3and Table 3).

Ahmed et al.(

2013) found similar land cover changes in their study of Dhaka city

using a 10-year interval from 1989 to 2009. In order to showcase more speciﬁc chang-

ing trends in land cover, we conducted our study across a four-year interval time

series. According to ﬁeld observations in 2016, most of the wetlands-dominated low-

lands in Dhaka urban area had been converted to bare soil by unauthorized landﬁlling,

during last three decades. Eventually most of these were transformed into built-up

areas. Despite the issuance of stop orders by the High Court, illegal landﬁlling activities

Table 2. Average classiﬁcation accuracy (19912015).

Year User’s Accuracy (%) Producer’s Accuracy (%)

Water

Body

Vegetation Built-up

Area

Bare

Soil

Water

Body

Vegetation Built-up

Area

Bare

Soil

Overall

Accuracy (%)

1991 86.5 77.9 81.5 90.5 94.5 90 91.5 85.6 89.4

1995 86.1 86.2 82.9 91.1 90.5 96.5 90.6 91.2 91.4

1999 83.5 84.3 86.5 90.2 91 82.5 92.4 86.5 89.5

2003 90.2 88.7 89.2 87.5 88.5 79.2 88.6 91.2 86.5

2007 88.2 86.5 93.5 88.6 89.1 77.5 89.5 88.6 90.1

2011 86.5 83.4 91.6 90.5 90.2 81.7 90.2 87.1 88.3

2015 88.95 83.8 85.93 93 91.6 96.3 93.2 91.5 93.1

Figure 3. Land cover changes (19912015).

10 Nur Hussain, S.M. Shahriar Ahmed and Amena Muzaffar Shumi

have continued transforming the city’s ecological and urban structure (Karim, 2003;

UNESCAP 2003;The Daily Star,2012,2019; GoB, 2010). To make matters worse,

developers are particularly interested in wetlands and areas with vegetation project

areas as land prices there tend to be lower as compared to prices of areas in more

elevated land.

Urban expansion and compactness

Urban expansion was traced using geospatial analysis. Urban compactness areas

expanded from 116.27 km

to 290.40 km

in 25 years (Figure 4and Table 4). The

expansion rate was measured to be 16.41 per cent, 16.47 per cent, 21.88 per cent,

16.36 per cent, 14.41 per cent and 13.53 percent respectively every four years in 1995,

1999, 2003, 2007, 2011 and 2015. Furthermore, the land cover of UMBA and the

urban inﬂuencing outer peripheral area also increased due to changes in compactness

Table 3. Land cover change in km

(19912015).

Year Water Body Vegetation Built-up Area Bare Soil

Area Percentage Area Percentage Area Percentage Area Percentage

1991 82.78 27.41 101.91 33.74 69.32 22.95 48.25 15.98

1995 65.32 21.63 80.35 26.61 89.17 29.53 67.37 22.31

1999 63.37 20.98 64.48 21.35 103.92 34.41 70.45 23.33

2003 57.56 19.06 53.86 17.83 113.12 37.46 77.68 25.72

2007 41.39 13.71 45.79 15.16 122.94 40.71 92.10 30.50

2011 34.88 11.55 39.66 13.13 128.83 42.66 98.84 32.73

2015 23.94 7.93 28.72 9.51 171.50 56.79 77.84 25.77

Figure 4. Map depicting: a) urban expansion with changing status of the absolute centre point, and b) Digital

Elevation Model of the study area. The DEM was explored using 30 m spatial resolution SRTM data.

Hot spot analysis of UHI in Dhaka, Bangladesh 11

as evident in the urban compactness ratio (CoRÞmeasurements. The urban compactness

ratio (CoRÞincreased from 0.60 to 0.68 on a scale of 0 to 1. The mean centre of the

urban area also changed due to a change in the compactness ratio (CoRÞ. The mean

centre shifted from the actual mean centre in a north-easterly direction by a subse-

quence of the urban compactness ratio (CoRÞ. The urban area has expanded to high ele-

vated land and also to the low elevated land by land ﬁlling where wetland areas have

been lost (Figure 4).

Table 3presents the quantitative explanation of urban expansion, compactness ratio

(CoRÞ, and shifting direction of the urban mean centre. The mean centre moved from

the actual mean centre by 1.7 km in a north-easterly direction from 1991 to 2015.

Urban expansion increased by 149.77 per cent from 1991 to 2015. The expansion rates

were calculated to be 16.41 per cent, 16.47 per cent, 21.88 per cent, 16.36 per cent,

14.41 per cent and 13.53 per cent every four years, respectively in 1995, 1999, 2003,

2007, 2011 and 2015. Urban compactness areas saw an increase from 1999 to 2003, by

21.88 per cent.

Between 1991 and 2015, there was a signiﬁcant acceleration in structural develop-

ment and the expansion of built-up areas. This growth extended beyond the urban

administrative boundaries into the surrounding rural areas, expanding urban bound-

aries beyond the metropolitan administrative areas. This expansion was mainly driven

by a rapid increase in population density during that period (Streatﬁeld & Karar, 2008;

Ahmed & Bramley, 2015). In 2017, about 17 million people lived in Dhaka City.

According to the Bangladesh Bureau of Statistics (BBS), the urban population

increased at a rate of 113 percent from 1991 to 2011 (MOP-BD, 2013). The rapid pop-

ulation increase has directly inﬂuenced the demographic characteristics in urban

expansion, changing urban compactness. The compactness ratio CoR represents the

shape of the urban area. The elevation proﬁle of the study area was explored using

DEM to ﬁnd out which areas were most impacted by urban growth. The DEM showed

that the urban area had extended from highlands to lowlands mainly because the

wetland-dominated lowlands were more accessible and cheaper to develop.

UHI and physical footprints

UHI was detected to retrieve LST with seasonal variations. Table 5represents the quan-

tities of seasonal LST variations. In 1991 mean temperature was 21.86C, 26.94C and

23.16C, respectively, during the winter, summer, and autumn seasons. 25 years later,

in 2015, the mean temperature was measured at 24.26C, 30.86C, and 26.62C,

respectively, for the winter, summer, and autumn seasons. In the last 25 years,

Table 4. Urban expansion and compactness ratio.

Year Area Expansion % CoR Mean Centre Change (Direction)

1991 116.27 - 0.60 -

1995 135.35 16.41 % 0.63 0.25 Km North-Easterly

1999 157.64 16.47 % 0.57 0.32 Km North-North Easterly

2003 192.13 21.88 % 0.57 0.4 Km South-Easterly

2007 223.57 16.36 % 0.60 0.52 Km North-Easterly

2011 255.80 14.41 % 0.62 0.54 Km East-South-Easterly

2015 290.40 13.53 % 0.68 0.51 Km North

12 Nur Hussain, S.M. Shahriar Ahmed and Amena Muzaffar Shumi

Table 5. Seasonal minimum, mean and maximum land surface temperature (LST) (C) of the study area. Standard deviation and mean values included.

Year Winter Summer Autumn

Minimum Mean Maximum Minimum Mean Maximum Minimum Mean Maximum

1991 15.2 21.86 1.1 26.3 20.06 26.94 1.04 33.59 17.09 23.16 1.66 31.33

1995 12.14 20.95 1.2 25.9 24.58 29.14 0.72 34.54 17.18 23.39 1.4 31.90

1999 13.33 24.98 1.53 28.65 20.96 27.58 1.38 36.83 17.75 23.43 1.35 29.94

2003 13.21 25.96 1.51 29.33 20.04 27.91 1.74 35.76 16.63 23.88 1.56 31.04

2007 14.6 24.09 1.61 29.8 17.01 28.19 1.39 38.77 19.9 23.67 1.65 31.65

2011 14.42 24.48 1.11 30.34 20.63 30.19 1.48 39.28 18.93 24.94 1.67 32.65

2015 10.9 24.26 1.03 32.04 19.04 30.86 1.79 42.22 18.34 26.62 1.85 36.85

Hot spot analysis of UHI in Dhaka, Bangladesh 13

seasonal mean temperatures increased by 2.4C, 3.92C and 3.46C, respectively,

during the winter, summer, and autumn seasons.

An urban area’s physical features (i.e. water bodies, vegetation, temperature, and

structural bodies) are the main factors that contribute to changes in urban ecological

footprints (Holden, 2004). Our research considers how physical footprints yield the

characteristics of LULC change. The NDVI, NDWI, NDBI, and NDBaI indices represent

the scenario of the physical footprints. Changes in LULC can inﬂuence the city’s tem-

perature variation. Table 6demonstrates the relationship among LST, NDVI, NDWI,

NDBI, and NDBaI indices. These four indicators are correlated with LST. While NDVI

and NDWI values have a negative correlation with LST, the NDBI and NDBaI values

have a positive correlation with LST. LST is the dependent value in this relationship,

whereas NDVI, NDWI, NDBI, and NDBaI are independent values.

The LST is a dependent variable, and the other index values are independent values

and negative and positive correlations were among these indicators. When temperature

increases, NDVI and NDWI values decrease, representing a negative correlation. On the

other hand, temperature increases with NDBI and NDBaI values, representing a posi-

tive correlation (Table 6). As land cover changes, physical footprints also change,

resulting in an increase in LST.

Hot spot concentration of UHI

Hot spot analysis (G

i) delineated the geospatial pattern distribution of LST within a per-

centage of hot and cold conﬁdence levels (Gi Bin). While a hot conﬁdence level of

more than 90 per cent was considered as a hot spot zone, a cold conﬁdence level

of more than 90 per cent was considered as a cold zone; ﬁnally, non-signiﬁcant clusters

were assumed to be stable zones. Figure 4represents the spatial distribution of UHI

using the nearest neighbour value with G

iconsideration of UHI’s changing status from

1991 to 2015. For example, from 1991 to 2015, cold spot areas increased

from 19.40 km

to 50.36 km

. Within the same period, non-signiﬁcant zones decreased

from 236.90 km

to 162.14 km

while hot spot zones rapidly increased from

46.74 km

to 90.55 km

Figure 5represents three geostatistical spatial categories of zones: i.e. (a) hot spot,

(b) non-signiﬁcant, and (c) cold spot. Each zone was demarcated based on data sourced

within seven different periods from 1991 to 2015 in four-year intervals. During the last

25 years, cold spots increased by 10.21 percent; non-signiﬁcant zones reduced by 24.67

percent; and hot spot areas increased by 14.46 percent (Figure 6). This demonstrates

that temperature has increased, leading to the creation of hot zones. Conversely,

although cold spots have also increased, the rate of increase is less than that of the hot

spot areas. The decrease in non-signiﬁcant areas represents the UHI’s reduced stability,

and temperature ﬂuctuations.

Our hot spot analysis (G

i) generated the geospatial distribution of UHI in Dhaka

megacity. This analysis produces a zone of categories that considers whole input pixels

statistics and provides hot spot, non-signiﬁcant, and cold spot categories within geo-

graphical coordinates. Hot spot analysis (G

i) is also advantageous because it produces

each pixel’s spatial distribution and geographic references with attribute data. Hot spot

analysis (G

i) in the research was used to identify areas with higher temperatures com-

pared to other regions within the study area. The analysis utilizes the nearest neigh-

bour algorithm from geo-statistics to determine the spatial arrangement of these

temperature hot spots. By applying this method, the research aims to allocate changing

14 Nur Hussain, S.M. Shahriar Ahmed and Amena Muzaffar Shumi

Table 6. Correlation between LST and physical footprint indices.

LST NDVI NDWI NDBI NDBaI

LST 1

NDVI -0.73 1

NDWI -0.77 0.71 1

NDBI 0.88 -0.81 -0.92 1

NDBaI 0.83 -0.78 -0.81 0.69 1

*The p-value is less than 0.05.

Figure 5. Hot spot map of UHI (19912015).

Figure 6. Hot spot bar graph of UHI (19912015).

Hot spot analysis of UHI in Dhaka, Bangladesh 15

temporal records to their corresponding spatial locations during a speciﬁc time interval.

The information obtained from hot spot analysis helps understand the spatial patterns

of temperature variations, which can inform decisions related to urban planning, envi-

ronmental management, and climate adaptation strategies.

Seasonal SUHI intensity

SUHI intensity (Δ,T) was explored by calculating the difference between the mean LST

of the UMBA and the mean LST of surrounding areas. The Δ,Twas measured using

150 per cent of the controlling zone of UMBA area (i.e. 16.8 km distance between the

mean urban centre and outer inﬂuencing boundary). We observed the intensity of UHI

to be higher in the summer season as compared to autumn and winter (Table 7).

SUHI intensity increased by 2.95, 4.21, and 3.00, respectively in the winter, sum-

mer, and autumn seasons from 1991 to 2015 (Table 7). This reveals that UHI was not

only affected in the UMBA but also in outer rural-urban marginal areas. The increasing

trend of SUHI intensity impacts the local climate, ecosystem, and biodiversity of the

UMBA as well as surrounding urban areas.

Discussion

Urban compactness in Dhaka has expanded by more than double in 25 years, urging

the necessity for urban expansion to respond to the population’s need for employ-

ment/jobs, and overall better quality of life. In addition, population density has

increased due to migration into the city, which has led to urban expansion outside the

metropolitan administrative zone. This unplanned urban growth was detected as

the Urban compactness ratio (C

) which increased by 0.08 from 1991 to 2015, starting

at 0.60 to 0.68 on the one scale. Although there was an increasing trend in C

that

started from 16.41 percent in 1991 and continued until 2003 at 21.88 percent, this

expansion rate cooled down to 13.53 per cent in 2015.

In the 25 years of study, water bodies have decreased by 19.48 per cent, vegetation

cover has reduced by 24.23 per cent, while built-up areas and bare soil have experi-

enced an increase of 33.84 per cent and 9.79 per cent, respectively—largely inﬂuenced

by urban expansion and increasing population density. As temperature increased,

NDVI and NDWI decreased, indicating a negative correlation. On the other hand, NDBI

and NDBaI increased with rising temperature, indicating a positive correlation. With

increased population density, built-up and bare soil areas have increased, resulting in

increased movements, overuse of water and other resources, thereby negatively

impacting vegetation growth due to less and less arable land area every year.

Table 7. SUHI intensity of study area.

Year Δ,T (Winter) Δ,T (Summer) Δ,T (Autumn)

1991 1.20 0.87 1.10

1995 1.27 0.22 1.01

1999 1.34 3.14 1.14

2003 1.05 3.03 1.83

2007 1.30 3.97 1.58

2011 4.01 2.22 2.24

2015 4.15 5.08 4.10

jΔ,T 1991 to Δ, T 2015j2.95 4.21 3.00

16 Nur Hussain, S.M. Shahriar Ahmed and Amena Muzaffar Shumi

In 1991, cold spot areas covered 19.40 km

, however, by 2015, coverage had grown

to 50.36 km

. Between 1991 and 2015, non-signiﬁcance zones decreased by

236.90 km

to 162.14 km

, while hot spot zones increased by 46.74 km

to 90.55 km

In the winter, summer, and autumn seasons of the relevant period, SUHI intensity

increased by 2.95, 4.21, and 3.00. As a result, UHI has impacted the UMBA and the

surrounding rural-urban fringe areas.

Our research also detected seasonal variations in LST. LST levels were found to have

increased during the winter, summer, and autumn seasons. In the last 25 years, sea-

sonal mean temperature rose by 2.4C, 3.92C, and 3.46C for the winter, summer,

and autumn seasons. As population encroachment increases, green areas decrease,

making it considerably challenging for the environment to cool down. This adds to the

residual temperature effect every subsequent year, which—based on our ﬁndings—

caused temperatures to rise within the relevant study period.

While urban green cover and water bodies have decreased over the last 25 years,

built-up areas and bare soils have increased. Dhaka city has experienced 71.07 and

71.83 percent of water body and green coverage loss during the last 25 years, respec-

tively. Simultaneously, built-up areas have increased by 147.74 percent. Due to a

decrease in canals and wetlands, ﬂooding was found to have occurred more frequently.

We also observed a 3.26C yearly average LST increase during the last 25 years. The

methods and parameters used in detecting and analysing UHI in this study have con-

structed a clear image of change occurring over 25 years.

Conclusion

This study developed a method of deriving the future UHI of urban areas based on hot

spot analysis as a means of ﬁnding out the status of land cover and temperature

changes in Dhaka city during the period from 19912015. This study derived the UHI

of Dhaka megacity based on hot spot analysis and represented signiﬁcant and non-

signiﬁcant areas using long-term remote sensing data. Landsat 5 and 7 data were uti-

lized to extract land cover, LST, UHI, and UMBA status. A series of data spanning

25 years was used to produce LST which was then translated into spatial and temporal

UHI trends. Urban expansion and compactness were explored using the same dataset

(i.e. Landsat 5 & 7 images). The urban compactness ratio (CoR Þrepresented the chang-

ing status of the mean centre. SRTM images were used to study the physical environ-

ment of the enlarged area. A series of Landsat 5 and 7 data spanning 25 years was also

used to generate NDVI, NDWI, NDBI, and NDBaI to explore changes in the urban

physical footprints arising from land cover change. NDVI, NDWI, NDBI, and NDBaI

were used to examine the relationship between land cover change and LST trends.

SUHI intensity (ΔT) is the status of UHI within inﬂuencing rural-urban fringe areas. As

TUrban and TBoundary were used to calculate SUHI intensity, SUHI intensity (ΔT) served

as urban intensity indicators but also indicated the effect of UHI in surrounding rural

areas. We employed LST for hot spot analysis (G

i) of UHI which demarcated UHI distri-

bution into cooler areas, non-signiﬁcant and hot spot zones. Our methodological

framework may be applied to other cities and sites of urbanization.

Acknowledgements

We would like to thank the Community and Discussion for Environmental Research (CDER)

research group for collecting ﬁeld data for accuracy assessment and Professor Dr. Md. Mizanur

Hot spot analysis of UHI in Dhaka, Bangladesh 17

Rahman, Department of Geography and Environment, Jahangirnagar University, Dhaka,

Bangladesh, for his valuable suggestions. We are sincerely grateful to the manuscript reviewers for

their signiﬁcant contributions and expertise, which greatly improved our work. We also extend

our heartfelt appreciation to the journal editor panel for granting us the opportunity to publish

our research.

References

Ahmed B, Kamruzzaman M, Zhu X, Rahman M, Choi K (2013) Simulating land cover changes

and their impacts on land surface temperature in Dhaka, Bangladesh. Remote Sensing 5(11),

5969–98.

Ahmed S, Bramley G (2015) How will Dhaka grow spatially in future? Modelling its urban growth

with a near-future planning scenario perspective. International Journal of Sustainable Built Envi-

ronment 4(2), 359–77.

Baddeley A, Berman M, Fisher NI et al. (2010) Spatial logistic regression and change-of-support in

poisson point processes. Electronic Journal of Statistics 4, 1151–1201.

Bechtel B, Alexander P, Böhner J et al. (2015) Mapping local climate zones for a worldwide data-

base of the form and function of cities. ISPRS International Journal of Geo-Information 4(1),

199219.

Beverley H (1872) Census of Bengal 1872. The Bengal Secretariat Press, Calcutta.

Cao L, Li P, Zhang L, Chen T (2002) Remote sensing image-based analysis of the relationship

between urban heat island and vegetation fraction. The International Archives of the Photogramme-

try, Remote Sensing and Spatial Information Sciences Vol. XXXVI (B7), 1379–84.

Carlson TN, Traci Arthur S (2000) The impact of land use - land cover changes due to urbaniza-

tion on surface microclimate and hydrology: a satellite perspective. Global and Planetary Change

25 (1–2), 49–65.

Chakravorty S (1995) Identifying crime clusters: the spatial principals. Middle States Geographer 28,

53–58.

Chen F, Yang X, Zhu W (2014) WRF simulations of Urban Heat Island under hot-weather synop-

tic conditions: the case study of Hangzhou City, China. Atmospheric Research 138 (1), 36477.

Chen YH, Wang J, Li XB (2002) A study on urban thermal ﬁeld in summer based on satellite

remote sensing. Remote Sensing for Natural Resources 4,5559.

Chen XL, Zhao HM, Li PX, Yin ZY (2006a) Remote sensing image-based analysis of the relation-

ship between Urban Heat Island and land use/cover changes. Remote Sensing of Environment 104

(2), 133–46.

Chen M, Wang R, Zhang L, Xu C (2006b) Temporal and spatial assessment of natural resource use

in China using ecological footprint method. International Journal of Sustainable Development and

World Ecology 13 (4), 255–68.

Chen Y (2011) Computers, environment and urban systems derivation of the functional relations

between fractal dimension of and shape indices of urban form. Computers, Environment and

Urban Systems 35 (6), 442–51.

Damyanovic D, Reinwald F, Brandenburg C et al. (2016) Pilot Action City of Vienna –UHI-STRAT

Vienna. In Musco F (ed) Counteracting Urban Heat Island Effects in a Global Climate, 257–81.

Springer Nature, Cham.

Decker EH, Elliott S, Smith FA, Blake DR, Rowland FS (2000) Energy and material ﬂow through

the urban ecosystem. Annual Review of Ecology and Systematics 26, 685–740. Available at: https://

doi.org/10.1146/annurev.energy.25.1.685.

Deng C, Wu C (2012) BCI: a Biophysical Composition Index for remote sensing of urban environ-

ments. Remote Sensing of Environment 127, 247–59.

Deng Y, Srinivasan S (2016) Urban land use change and regional access: a case study in Beijing,

China. Habitat International 51, 10313.

Dos Santos AR, De Oliveira FS, Da Silva AG, Gleriani JM et al. (2017) Spatial and temporal

distribution of Urban Heat Islands. Science of the Total Environment 605–606, 94656.

18 Nur Hussain, S.M. Shahriar Ahmed and Amena Muzaffar Shumi

Du H, Wang D, Wang Y et al. (2016) Inﬂuences of land cover types, meteorological conditions,

anthropogenic heat and urban area on surface Urban Heat Island in the Yangtze River Delta

urban agglomeration. Science of the Total Environment 571, 461–70.

Elmqvist T, Fragkias M, Goodness J et al. (eds) (2013) Urbanization, Biodiversity and Ecosystem Ser-

vices: Challenges and Opportunities. Springer, Dordrecht.

Fu P, Weng Q (2017) Responses of Urban Heat Island in Atlanta to different land-use scenarios.

Theoretical and Applied Climatology 133, 123–35.

Gao B (1996) NDWI A Normalized Difference Water Index for remote sensing of vegetation liquid

water from space. Remote Sensing of Environment 58 (April 1995), 257–66.

Getis A, Ord JK (1992) The analysis of spatial association by use of distance statistics. Geographical

Analysis 24 (3), 189206.

Gillies RR, Carlson TN (1995) Thermal remote sensing of surface soil water content with partial

vegetation cover for incorporation into climate models. Journal of Applied Meteorology 34 (4),

74556.

Goh KC, Chang HC (1999) The relationship between height to width ratios and the heat island

intensity at 22: 00 h. International Journal of Climatology 19 (9), 1011–23.

Government of Bangladesh (GoB) (2010) The Bangladesh Environment Conservation (Amendment)

Act, 2010. Available at: https://bangladeshbiosafety.org/bangladesh-doc/bangladesh-environment-

conservation-act-2010-amendment/.

Haggett P, Cliff AD, Frey A (1977) Locational Analysis in Human Geography, 2nd Edn. Edward

Arnold Ltd, London.

Holden E (2004) Ecological footprints and sustainable urban form. Journal of Housing and the Built

Environment 19 (1), 91–109.

Hossain S (2008) Rapid urban growth and poverty in Dhaka city. Urban Studies 5(1), 1–24.

Howard L (1833) The climate of London. London Harvey Dorton 2, 1818–20.

Hu L, Brunsell NA (2015) A new perspective to assess the Urban Heat Island through remotely

sensed atmospheric proﬁles. Remote Sensing of Environment 158, 393406.

Huang W, Li J, Guo Q et al. (2017) A satellite-derived climatological analysis of Urban Heat Island

over Shanghai during 2000-2013. Remote Sensing 9(7), 641. Available at: https://doi.org/10.

3390/rs9070641.

Hussain N, Islam MN (2020) Hot spot (Gi*) model for forest vulnerability assessment: a remote

sensing-based geostatistical investigation of the Sundarbans mangrove forest, Bangladesh.

Modeling Earth Systems and Environment 6(4), 2141–51.

Hussain N (2018) Water quality and status aquatic fauna of Dhaka Mega City, Bangladesh.

Sriwijaya Journal of Environment 3(2), 68–73.

Jones PD, Groisman PY, Coughlan M, Plummer N, Wang W, Karl TR (1990) Assessment of urban-

isation effects in time series of surface air temperature over land, 1990. Nature 347, 169–72.

Kabir A, Parolin B (2012) Planning and Development Of Dhaka –A Story Of 400 Years. Paper pres-

ented at the 15th International Planning History Society Conference, 15-18 Jul, Sao Paolo, 1–20.

Kogan FN (1995) Droughts of the late 1980s in the United States as derived from NOAA polar-

orbiting satellite data. Bulletin - American Meteorological Society 76 (5), 655–68.

Kontoes CC (2008) Operational land cover change detection using change vector analysis. Interna-

tional Journal of Remote Sensing 29 (16), 4757–79.

Li H, Zhou Y, Li X et al. (2018) A new method to quantify surface Urban Heat Island Intensity. Sci-

ence of the Total Environment 624, 26272.

Liu L, Zhang Y (2011) Urban Heat Island analysis using the Landsat TM data and ASTER data: a

case study in Hong Kong. Remote Sensing 3(7), 1535–52.

Lu Y, Coops NC, Hermosilla T (2017) Estimating urban vegetation fraction across 25 cities in Pan-Paciﬁc

using Landsat time series data. ISPRS Journal of Photogrammetry and Remote Sensing 126,1123.

Mahdavi A, Kiesel K, Vuckovic M (2016) Counteracting Urban Heat Island effects in a global cli-

mate change scenario. In Musco F (ed) Methodologies for UHI Analysis,71–91. Springer Interna-

tional Publishing, Cham.

Hot spot analysis of UHI in Dhaka, Bangladesh 19

Maimaitiyiming M, Ghulam A, Tiyip et al. (2014) Effects of green space spatial pattern on land

surface temperature: implications for sustainable urban planning and climate change adapta-

tion. ISPRS Journal of Photogrammetry and Remote Sensing 89,5966.

McFeeters SK (1996) The use of the Normalized Difference Water Index (NDWI) in the delinea-

tion of open water features. International Journal of Remote Sensing 17 (7), 1425–32.

Meng Q, Zhang L, Sun Z, Meng F, Wang L, Sun Yunxiao (2018) Characterizing spatial and tempo-

ral trends of surface Urban Heat Island effect in an urban main built-up area: a 12-year case

study in Beijing, China. Remote Sensing of Environment 204 (November 2016), 826–37.

Miles V, Esau I (2017) Seasonal and spatial characteristics of Urban Heat Islands (UHIs) in north-

ern west Siberian cities. Remote Sensing 9(10), 989. Available at: https://doi.org/10.3390/

rs9100989.

Karim MF, Haider MJ (2003) Engineering geology and geomorphology for ground improvement

in the Dhaka City - Tongi Area. Atlas of Urban Geology 14, 187–94.

MOP-BD (2013) District Statistics, Ministry of Planning. Government of Bangladesh, Dhaka.

Newman PWG (1999) Sustainability and cities: extending the metabolism model. Landscape and

Urban Planning 44 (4), 219–26.

Nichol J, Wong MS (2005) Modeling urban environmental quality in a tropical city. Landscape and

Urban Planning 73 (April 2004), 49–58.

Oke TR, Cleugh HA (1987) Urban heat storage derived as energy balance residuals. Boundary-Layer

Meteorology 39 (3), 233–45.

Ord JK, Getis A (1995) Local spatial autocorrelation statistics: distributional issues and an applica-

tion. Geographical Analysis 27 (4), 286306.

Orhan O, Ekercin S, Celik F (2014) Use of Landsat land surface temperature and vegetation indi-

ces for monitoring drought in the Salt Lake Basin Area, Turkey. The Scientiﬁc World Journal

2014 (Vci). Available at: https://doi.org/10.1155/2014/142939.

Peng S, Piao S, Ciais P et al. (2012) Surface Urban Heat Island across 419 global big cities. Remote

Sensing of Environment 46 (12), 3175–86.

Qian LX, Cui HS, Chang J (2006) Impacts of land use and cover change on land surface tempera-

ture in the Zhujiang Delta. Pedosphere 16 (200523), 681–89.

Rabeya R (2017) Re-Deﬁning Courtyard to Re-Vitalize Urban Community. In Proceedings of UIA

2017 Seoul World Architects Congress; 310 Sept 2017, Seoul, Korea. UIA, Seoul.

Rahman HZ (2014) Urbanization in Bangladesh: challenges and priorities. Paper presented at

Bangladesh Economists’Forum, Jun 2122, Dhaka, Bangladesh.

Rayk R, Kapp R, Reuter U et al. (2016) Pilot actions in European cities –Stuttgart. In Musco F

(ed) Counteracting Urban Heat Island Effects in a Global Climate Change Scenario, 281–305. Springer

International Publishing, Cham.

Rodriguez Lopez JM, Heider K, Scheffran J (2017) Frontiers of urbanization: identifying and

explaining urbanization hot spots in the south of Mexico City using human and remote sens-

ing. Applied Geography 79,110.

Roth M, Chow WT (2012) A historical review and assessment of urban heat island research in

Singapore. Singapore Journal of Tropical Geography 33 (3), 38197.

Rosenzweig C, Karoly D, Vicarelli M et al. (2008) Attributing physical and biological impacts to

anthropogenic climate change. Nature 453 (7193), 353–57.

Scott L, Warmerdam N (2005) Extend crime analysis with ArcGIS spatial statistics tools. ArcUser

Online April-June 8(2). Available at: https://www.esri.com/news/arcuser/0405/ss_crimestats2of2.

html.

Shahfahad NMW, Towﬁqul Islam ARM, Mallick J, Rahman A (2022a) Land use/land cover

change and its impact on surface urban heat island and urban thermal comfort in a metropoli-

tan city. Urban Climate 41, 101052. Available at: https://doi.org/10.1016/j.uclim.2021.101052.

Shahfahad, Talukdar S, Rihan M, Hang HT, Bhaskaran S, Rahman A (2022b) Modelling urban

heat island (UHI) and thermal ﬁeld variation and their relationship with land use indices over

Delhi and Mumbai metro cities. Environment, Development and Sustainability 24 (3), 3762–90.

20 Nur Hussain, S.M. Shahriar Ahmed and Amena Muzaffar Shumi

Sobstyl JM, Emig, T. Qomi, MA, Ulm FJ, Pellenq RM (2018) Role of city texture in urban heat

islands at nighttime. Physical Review Letters 120 (10), 108701. Available at: https://doi.org/10.

1103/PhysRevLett.120.108701.

Souch C, Grimmond S (2006) Applied climatology: urban climate. Progress in Physical Geography 30

(2), 270–79.

Streatﬁeld PK, Karar ZA (2008) Population challenges for Bangladesh in the coming decades. Jour-

nal of Health, Population and Nutrition 26 (3), 261–72.

Streutker DR (2002) A remote sensing study of the Urban Heat Island of Houston, Texas. Interna-

tional Journal of Remote Sensing 23 (13), 2595–608.

Taleb MA (2012) Comparative study of urban area extension and ﬂood risk in Dhaka City of

Bangladesh. Global Journal of Human Social Science Geography & Environmental Geosciences 12 (11),

3740.

Tan J, Zheng Y, Tang X et al. (2010) The urban heat island and its impact on heat waves and

human health in Shanghai. International Journal of Biometeorology 54,7584.

The Daily Star (2012) High Court ruling on illegal landﬁll welcome. Editorial, 1 August. Available

at: https://www.thedailystar.net/news-detail-244281.

The Daily Star (2019) Stop earth-ﬁlling by 24 companies. Staff Correspondent, 28 January. Avail-

able at: https://www.thedailystar.net/country/ensure-status-quo-earth-ﬁlling-in-purbachal-

high-court-1693570.

Tran DX, Pla F, Latorre-Carmona P, Myint SW, Caetano M, Kieu HV (2017) Characterizing the

relationship between land use land cover change and land surface temperature. ISPRS Journal

of Photogrammetry and Remote Sensing 124, 11932.

United Nations Economic and Social Commission for Asia and the Paciﬁc (UNESCAP) (2003)

Review of Developments in Transport in the ESCAP Region 2003. United Nations, New York.

United Nations Human Settlement Programme (UN-Habitat) (2020) Unpacking the value of sus-

tainable urbanization. In World Cities Report 2020. UN-Habitat, Nairobi.

USGS (2011) Landsat 7 Science Data Users Handbook. The United States Geological Survey, Reston.

Available at: https://www.usgs.gov/landsat-missions/landsat-7-data-users-handbook

Vinayak B, Lee HS, Gedam S, Latha R (2022) Impacts of future urbanization on urban microcli-

mate and thermal comfort over the Mumbai metropolitan region, India. Sustainable Cities and

Society 79, 103703. Available at: https://doi.org/10.1016/j.scs.2022.103703.

Wackernagel M, Schulz NB, Deumling D et al. (2002) Tracking the ecological overshoot of the

human economy. In Proceedings of the National Academy of Sciences 99, 9266–71.

Walawender JP, Szymanowski M, Hajto MJ, Bokwa A (2014) Land surface temperature patterns

in the urban agglomeration of Krakow (Poland) derived from Landsat-7/ETM+data. Pure and

Applied Geophysics 171 (6), 913–40.

Wolman A (1965) The Metabolism of Cities. Scientiﬁc American 213 (3), 156–74.

Yao R, Wang L, Huang X, Niu Z, Liu F, Wang Q (2017) Temporal trends of surface Urban Heat

Islands and associated determinants in major Chinese cities. Science of the Total Environment,

609, 742–54.

Zha Y, Gao J, Ni S (2003) Use of normalized difference built-up index in automatically mapping

urban areas from TM imagery. International Journal of Remote Sensing 24 (3), 583–94.

Zhang J, Wang Y (2008) Study of the relationships between the spatial extent of surface Urban

Heat Islands and urban characteristic factors based on Landsat ETM +Data. Sensors 8(11),

7453–68.

Zhao H, Chen X (2005) Use of normalized difference bareness index in quickly mapping bare

areas from TM / ETM +.InProceedings of the 2005 IEEE International Geoscience and Remote Sens-

ing Symposium; 29 Jul, Seoul, Korea. IEEE, Piscataway.

Hot spot analysis of UHI in Dhaka, Bangladesh 21

Impact of planned urban development on urban heat island effect: resilient cities for a sustainable future

Article

Full-text available

Mar 2025
ENVIRON SCI POLLUT R

This comprehensive study delved into the urban dynamics of Bangalore and Hyderabad, focusing on land cover changes, temperature variations, and their implications for sustainable urban development. Analyzing land cover trends from 2001 to 2021, the study revealed substantial urban expansion in both cities, with notable shifts in built-up, vegetation, bare soil, and water bodies. The analysis indicated intensified urbanization, particularly in Hyderabad, raising environmental concerns. The study employed a random forest model to predict contrasting urbanization patterns for 2031, emphasizing the need for resilient and sustainable strategies. The model successfully predicted land surface temperature (LST) and land cover changes, offering insights for informed urban planning. Correlation analyses unveil the influence of factors like proximity to bare soil and built-up areas on LST. The model’s performance is robust, outperforming other models, and is employed for 2021 land cover (LC) and LST predictions. Validation demonstrates the model’s accuracy in predicting LC distribution and LST across diverse land cover types, urban and rural areas. Furthermore, the model is applied to forecast 2031 LC and LST, revealing contrasting urbanization trends. The study incorporates a hot spot analysis, categorizing thermal zones and highlighting the urban heat island (UHI) effect. The model showcases a capacity to project long-term temperature dynamics, making it a valuable tool for urban planning and climate resilience. The findings underscore the need for sustainable land use planning, emphasizing environmental preservation amid rapid urban growth.

Future climate projections and sustainable development: The impact of Urban Heat Islands and land use land cover in Telangana's GHMC area

Article

Full-text available

Feb 2025

High-Resolution Insights into Nighttime Urban Heat Island Detection: a comparative temporal analysis of 1988 and 2015 in Greater Cairo

Article

Full-text available

Jan 2025
GeoJournal

Faten Nahas

Urban climate studies often overlook the study of nighttime urban heat island (UHI) at high spatial resolutions, such as those provided by Landsat. Previous research has primarily focused on daytime and nighttime UHI at coarser spatial resolutions, such as those offered by MODIS. Furthermore, there is a lack of studies that compare two Landsat night images over a period to assess temporal changes, which has not been comprehensively explored. In this study, the Global Artificial Impervious Area dataset was utilized to define urban, while Landsat-5 & 8 daytime and nighttime data (1988, 2015) were employed to classify land cover and extract nighttime land surface temperature (LST). Statistically significant hotspot analysis was applied to quantify the UHI. The findings revealed a 94% increase in urban cover and a 1.8% decrease in vegetation cover between 1988 and 2015. This urban expansion contributed to an average increase of 10 °C in LST across Greater Cairo. The analysis further indicated that every 42.7 km² increase in urban cover corresponds to a 1 °C rise in LST. Streets and industrial areas exhibited the highest LST readings, while compact buildings, especially older structures, recorded higher surface LSTs compared to open buildings with diverse architectural designs, which experienced lower LST values. Hotspot analysis identified an increase in UHI intensity, with a rise from 2.5 °C in 1988 to 2.7 °C in 2015. These findings highlight the critical need for sustainable urban planning and climate-responsive policies to mitigate the adverse effects of urbanization on thermal environments. This study provides a foundation for future research and emphasizes the importance of using high-resolution nighttime thermal imagery for UHI assessments in cities worldwide.

Investigating Sensitivity to Extreme Weather of a Mid-Size Global South City: An Integrated Mcda (Ahp)-Gis Based Modeling Approach

Preprint

Full-text available

Dec 2024

The dynamics of travel patterns are affected by catastrophic weather events though detailed studies on these issues are limited in developing countries. To fill this gap, this study explores the travel patterns of commuters and non-commuters in a city within the global south, focusing on travel mode choice, time, and cost, considering weather variability and spatial heterogeneity. Extreme heat, heavy rain, and waterlogging are defined as extreme weather events in this regard. 1063 data across 12 high- and low-sensitive zones to such weather events were collected by face-to-face, interviewer-administrated survey; and analyzed using inferential statistics, a Multinomial Logistic Regression (MLR), and a Difference-in-Difference (DiD) model. Results show that public transport usage significantly drops by 75% during extreme heat, with females and older adults being less likely to use it. As captive public transport users, low-income individuals exhibit more vulnerability as they have fewer options for alternative modes. Walking decreases across all income groups for both commuting and non-commuting trips, while the use of motorized modes increases by 10-25 times during such events. Additionally, travel time and cost significantly rise under these conditions. The findings highlight significant relevance for policies concerning efficient as well as comfortable public transport mode and active transport to reduce vulnerability, particularly for low-income groups; and to attract females, older adults, and higher-income individuals. Such socio-economic, psychological, and behavioral dynamics should be considered to manage the weather-induced transport demand in the cities of global south or any developing countries in an equitable, reliable, and sustainable manner.

Multi-scenario prediction and attribution analysis of carbon storage of ecological system in the Huaihe River Basin, China

Article

Full-text available

Aug 2024
ENVIRON MONIT ASSESS

Evaluating the impact of large-scale human activities on carbon storage through land use changes is of growing interest in terrestrial ecosystem assessments. The Huaihe River Basin, a vital Chinese grain production area, has undergone marked land use changes amid socio-economic acceleration. Evaluating the impacts of land use change on carbon storage and future carbon sequestration is imperative for regional ecosystem sustainability and Chinese food security, simultaneously, furnishing data support to regional land use planning and decision-making processes. Nonetheless, the mechanisms linking land use changes to carbon storage and the future carbon reservoir responses remain unclear. We utilized a multi-source dataset and representative scenarios, integrating PLUS, InVEST models, and Geodetector to assess land use change impacts on carbon storage in the Huaihe River Basin (2000–2030). The data indicates the following: (1) from 2000 to 2020, cultivated land decreased by 28,344.69 km², construction land increased by 26,914.56 km², and other land types changed little. (2) Land use change resulted a carbon loss of 1.17 × 10⁸ t, primarily due to the expansion of construction land. (3) All four simulation scenarios exhibited diminished carbon storage relative to 2020, with the economic development scenario recording the lowest at 4.98 × 10⁹ t and the ecological protection scenario the highest at 5.06 × 10⁹ t. (4) Elevation predominantly drives carbon storage changes, with its interaction with NPP having the greatest impact. The factors synergistically enhance their explanatory power. The research provides a scientific basis for strategies aimed at augmenting regional carbon sequestration and refining low-carbon land management, safeguarding ecosystem stability.

How Big Data Can Help to Monitor the Environment and to Mitigate Risks due to Climate Change: A review

Article

Jun 2024

Climate change triggers a wide range of hydrometeorological, glaciological, and geophysical processes that span across vast spatiotemporal scales. With the advances in technology and analytics, a multitude of remote sensing (RS), geodetic, and in situ instruments have been developed to effectively monitor and help comprehend Earth’s system, including its climate variability and the recent anomalies associated with global warming. A huge volume of data is generated by recording these observations, resulting in the need for novel methods to handle and interpret such big datasets. Managing this enormous amount of data extends beyond current computer storage considerations; it also encompasses the complexities of processing, modeling, and analyzing. Big datasets present unique characteristics that set them apart from smaller datasets, thereby posing challenges to traditional approaches. Moreover, computational time plays a crucial role, especially in the context of geohazard warning and response systems, which necessitate low latency requirements.

Spatiotemporal Analysis of Urban Heat Islands and Vegetation Cover Using Emerging Hotspot Analysis in a Humid Subtropical Climate

Article

Full-text available

Jan 2024

Research on the temporal and spatial changes of the urban heat island effect can help us better understand how urbanization, climate change, and the environment are interconnected. This study uses a spatiotemporal analysis method that couples the Emerging Hot Spot Analysis (EHSA) technique with the Mann-Kendall technique. The method is applied to determine the intensity of the heat island effect in humid subtropical climates over time and space. The data used in this research include thermal bands, red band (RED) and near-infrared band (NIR), and Landsat 7 and 8 satellites, which were selected from 2000 to 2022 for the city of Sari, an Iranian city on the Caspian Sea. Pre-processed spectral bands from the 'Google Earth Engine' database were used to estimate the land surface temperature. The land surface temperature difference between the urban environment and the outer buffer (1500 m) was modeled and simulated. The results of this paper show the accuracy and novelty of using Emerging Hotspot Analysis to evaluate the effect of vegetation cover on the urban heat island intensity. Based on the Normalized Difference Vegetation Index (NDVI), the city's land surface temperature increased by approximately 0.30 • C between 2011 and 2022 compared to 2001 to 2010. However, the intensity of the urban heat island decreased during the study period, with r = −0.42, so an average −0.031 • C/decade decrease has been experienced. The methodology can be transferred to other cities to evaluate the role of urban green spaces in reducing heat stress and to estimate the heat budget based on historical observations.

Navigating Dhaka's Extreme Weather: Inconveniences and Modal Shifts Among Commuters and Non-Commuters

Preprint

Full-text available

Feb 2025

Modelling urban heat island (UHI) and thermal field variation and their relationship with land use indices over Delhi and Mumbai metro cities

Article

Full-text available

Mar 2022
Environ Dev Sustain

According to the World Urbanization Prospects of United Nations, the global urban population has increased rapidly over past few decades, reaching about 55% in 2018, which is projected to reach 68% by 2050. Due to gradual increase in the urban population and impervious surfaces, the urban heat island (UHI) effect has increased manifold in the cities of developing countries, causing a decline in thermal comfort. Therefore, this study was designed to model the spatio-temporal pattern of UHI and its relationships with the land use indices of Delhi and Mumbai metro cities from 1991 to 2018. Landsat datasets were used to generate the land surface temperature (LST) using mono window algorithm and land use indices, such as normalized difference vegetation index (NDVI), normalized difference built-up index (NDBI), normalized difference bareness index (NDBal), normalized difference moisture index (NDMI), and modified normalized difference water index (MNDWI). Additionally, the urban hotspots (UHS) were identified and then the thermal comfort was modelled using the UTFVI. The results showed that maximum (30.25-38.99 °C in Delhi and 42.10-45.75 °C in Mumbai) and minimum (17.70-23.86 °C in Delhi and 19.06-25.05 °C in Mumbai) LST witnessed steady growth in Delhi and Mumbai from 1991 to 2018. The LST gap decreases and the UHI zones are being established in both cities. Furthermore, the UHS and worst-category UTFVI areas increased in both cities. This research can be useful in designing urban green-space planning strategies for mitigating the UHI effects and thermal comfort in cities of developing countries.

Hot spot (

{G}_{i}^{{*}}

) model for forest vulnerability assessment: a remote sensing-based geo-statistical investigation of the Sundarbans mangrove forest, Bangladesh

Article

Full-text available

Jun 2020

The 5th category super-cyclone Sidr disrupts the world's largest mangrove forest Sundarbans on November 15, 2007. It seriously shatters about 1900 km2 that 31% of the total area of the Sundarbans. That makes a great threat to the mangrove ecosystem and biodiversity, which convey to forest vulnerability monitoring of Sundarbans. This research emphasizes on mangrove forest monitoring with vulnerability assessment using Landsat-5 and Landsat-8 remote sensing data based on geo-statistical hot spot (

{G}_{i}^{{*}}

) model, normalized difference vegetation index (NDVI) and forest discrimination index (FDI). However, the analysis works with statistical algorithm Gi(d) and G(d) in terms of geo-statistical nearest neighborhood spatial autocorrelation analysis. Hot spot (

{G}_{i}^{{*}}

) model used to explore the hot and cold confidence zone, which provided the mangrove vulnerability confidence level. The simulated, ~ 14.1% extreme safe zone is increased from 2001 to 2015 and extremely vulnerable zone also increased 4.1% at the same time, although 4.3% stable zone also decreased in that time. Even, high-density mangrove area was decreased in 2009, and the low-density mangrove area increased due to cyclone Sidr. In addition, FDI denotes the mangrove density, and NDVI provides vegetation health condition and represents the mangrove variability scenario with geospatial location those signify to detect the threatening condition of mangrove population and density. Furthermore, this study’s methods and results will provide the base for further long-term studies on this world’s largest mangrove forest and would have an implication for the mangrove monitoring and disaster risk reduction strategies.

Water Quality and Status Aquatic Fauna of Dhaka Mega City, Bangladesh

Article

Full-text available

Sep 2018

Nur Hussain

Remote sensing images are representation of the earth’s surface as seen from space without any physical contact. It generally detects the surface feature through reflectance of matter and represents various Digital Number (DN) values. In the context of water it also represent DN values according to reflectance of water properties, whatever it can detect from the space. Landsat Operational Land Image (OLI) is more effective to detect the surface features. It has been used to detect water body. The surface water quality is not same all over the surface so the energy absorption and relaxation is not the same. The DN value of surface water is changed due to the change of chemical properties of water. It will be a fundamental study to explore the relationship between DN value and water quality. This research explores a complete integration of DN values of Landsat OLI satellite image with same surface water quality, which are collected from several points. The chemical properties of collected surface water, which are analyzed through lab and is integrated with DN value of the Landsat OLI satellite images. The variations among the chemical properties of collected surface water and the DN values of the exact Landsat OLI images are categorized within an index. The correlations between surface water quality and DN values of Landsat OLI are investigated in this research.

Role of City Texture in Urban Heat Islands at Nighttime

Article

Full-text available

Mar 2018

An urban heat island (UHI) is a climate phenomenon that results in an increased air temperature in cities when compared to their rural surroundings. In this Letter, the dependence of an UHI on urban geometry is studied. Multiyear urban-rural temperature differences and building footprints data combined with a heat radiation scaling model are used to demonstrate for more than 50 cities worldwide that city texture—measured by a building distribution function and the sky view factor—explains city-to-city variations in nocturnal UHIs. Our results show a strong correlation between nocturnal UHIs and the city texture.

A new method to quantify surface urban heat island intensity

Article

Full-text available

May 2018
SCI TOTAL ENVIRON

Reliable quantification of urban heat island (UHI) can contribute to the effective evaluation of potential heat risk. Traditional methods for the quantification of UHI intensity (UHII) using pairs-measurements are sensitive to the choice of stations or grids. In order to get rid of the limitation of urban/rural divisions, this paper proposes a new approach to quantify surface UHII (SUHII) using the relationship between MODIS land surface temperature (LST) and impervious surface areas (ISA). Given the footprint of LST measurement, the ISA was regionalized to include the information of neighborhood pixels using a Kernel Density Estimation (KDE) method. Considering the footprint improves the LST-ISA relationship. The LST shows highly positive correlation with the KDE regionalized ISA (ISAKDE). The linear functions of LST are well fitted by the ISAKDE in both annual and daily scales for the city of Berlin. The slope of the linear function represents the increase in LST from the natural surface in rural regions to the impervious surface in urban regions, and is defined as SUHII in this study. The calculated SUHII show high values in summer and during the day than in winter and at night. The new method is also verified using finer resolution Landset data, and the results further prove its reliability.

Comparative Study of Urban Area Extension and Flood Risk in Dhaka City of Bangladesh” in the Global Journal of Human Social Sciences

Article

Full-text available

Dec 2012

Abu Taleb

Global Journal Inc. USA, Vol. 12, Issue , Version 1.0, Pp-,

Seasonal and Spatial Characteristics of Urban Heat Islands (UHIs) in Northern West Siberian Cities

Article

Full-text available

Sep 2017

Anthropogenic heat and modified landscapes raise air and surface temperatures in urbanized areas around the globe. This phenomenon is widely known as an urban heat island (UHI). Previous UHI studies, and specifically those based on remote sensing data, have not included cities north of 60 • N. A few in situ studies have indicated that even relatively small cities in high latitudes may exhibit significantly amplified UHIs. The UHI characteristics and factors controlling its intensity in high latitudes remain largely unknown. This study attempts to close this knowledge gap for 28 cities in northern West Siberia (NWS). NWS cities are convenient for urban intercomparison studies as they have relatively similar cold continental climates, and flat, rather homogeneous landscapes. We investigated the UHI in NWS cities using the moderate-resolution imaging spectroradiometer (MODIS) MOD 11A2 land surface temperature (LST) product in 8-day composites. The analysis reveals that all 28 NWS cities exhibit a persistent UHI in summer and winter. The LST analysis found differences in summer and winter regarding the UHI effect, and supports the hypothesis of seasonal differences in the causes of UHI formation. Correlation analysis found the strongest relationships between the UHI and population (log P). Regression models using log P alone could explain 65–67% of the variability of UHIs in the region. Additional explanatory power—at least in summer—is provided by the surrounding background temperatures, which themselves are strongly correlated with latitude. The performed regression analysis thus confirms the important role of the surrounding temperature in explaining spatial–temporal variation of UHI intensity. These findings suggest a climatological basis for these phenomena and, given the importance of climatic warming, an aspect that deserves future study.

Impacts of future urbanization on urban microclimate and thermal comfort over the Mumbai Metropolitan Region, India

Article

Jan 2022

This study examines the microclimatic impacts of future urbanization over the Mumbai Metropolitan Region (MMR) in India using a dynamic downscaling approach. Initially, two numerical experiments (curr_exp and fut_exp) were carried out by integrating 2018 and 2050 land-use/land-cover patterns under the same meteorological conditions. Afterwards, the spatial and temporal changes in surface air temperature, wind speed, and thermal discomfort were assessed by comparing the simulated results between curr_exp and fut_exp. To remediate the observed adverse microclimatic impacts of future urbanization, a mitigation strategy was designed and implemented in the third experiment (miti_exp). Simulation results show that the average maximum (minimum) surface air temperature of the region would increase by 1.41°C (1.27°C) due to future urbanization. However, if we adopt the mitigation strategy, this growth can be restricted to 0.69°C (0.92°C). In the context of thermal discomfort, an additional 20% of the total area can undergo hyperthermal treatment by 2050. This study reveals that although the implemented mitigation strategy can restrict the increasing temperature, it is ineffectual in minimizing the thermal discomfort level. The results of this study will be helpful in designing area-specific climate action plans through which goals of sustainable urban development can be accomplished.

Land use/land cover change and its impact on surface urban heat island and urban thermal comfort in a metropolitan city

Article

Jan 2022

The increasing urban heat island intensity (UHII)is a matter of concern for the sustainable urban planning and maintaining urban thermal comfort in the metropolitan cities of developing countries like India. Therefore, in the current research, the land use/land cover (LU/LC) change and its impact on the surface UHI intensity (SUHII) and urban thermal comfort has been analyzed using Landsat datasets and geographically weighted regression (GWR) in Delhi metropolitan city. The result shows that the built-up area has increased from 315.18 to 720.24 sq. km in Delhi during 1991 to 2018 while other LU/LC types have declined. This has resulted in a substantial increase in LST and SUHII. The minimum, maximum and mean SUHII has increased by 1.26 • C, 4.6 • C and 1.18 • C during 1991 to 2018 and hence, the thermal comfort has declined in the city. The GWR analysis showed that the performance of GWR model was very good in showing the association between LU/LC and SUHII and the LU/LC pattern has significant impact on SUHII. The outcome of this research can be utilized for the formulation of SUHI mitigation strategies and maintaining urban thermal comfort in Delhi and cities with similar geographical conditions.

Characterizing spatial and temporal trends of surface urban heat island effect in an urban main built-up area: A 12-year case study in Beijing, China

Article

Sep 2017
REMOTE SENS ENVIRON

Accurately characterizing spatiotemporal changes in surface urban heat islands (SUHIs) is a prerequisite for sustainable urban development. Although urban administrative boundaries have typically been used for SUHI modeling, they are inaccurate, because urban main built-up areas (UMBAs) are not well characterized. In this study, we developed an UMBA extraction method based on impervious surface distribution density (ISDD), to better differentiate suburban boundaries and ensure the integrity of land cover types. Additionally, we propose a new intensity classification method to analyze SUHI spatial distribution and variation. The UMBA was extracted using LANDSAT-8 data, and the temporal dynamics of SUHI intensity (i.e., daily, monthly, seasonal, and yearly changes) were extracted from MODIS data. A case study for Beijing showed that the mean daytime and nocturnal SUHI intensities vary at multiple time scales. In the daytime, SUHI intensities in Beijing UMBA were mainly level-2 and level-3, with central-south Beijing, a high incidence area, at level-3 in spring and summer. At night, with the rise of SUHI intensity levels, the frequency of SUHI intensity levels increased from the periphery to the center within the same season. ISDD had a marked influence on the frequency of SUHI intensity levels during the daytime, and the frequencies of level-1 to level-4 intensities increased with ISDD. This influence tended to weaken when ISDD exceeded 50%.

Remote sensing-based geostatistical hot spot analysis of Urban Heat Islands in Dhaka, Bangladesh

Abstract

Recommended publications

McMaster Engineering: Canada Excellence Research Chairs Faculty Position Available...

Black Academic Excellence Recruitment Opportunities at McMaster Engineering

McMaster Engineering is hiring widely for faculty positions

Empowering older adults to live and age on

Remote sensing-based approach to identify the influence of land use/land cover change on the urban t...

A Remote Sensing Approach for Identifying and Mapping the Coastal Urban Heat Island in Bangladesh th...

Estimation of Urban Heat Islands Effect and Its Impact on Climate Change: A Remote Sensing and GIS-B...

Quantification of Surface Urban Heat Island Intensity Using MODIS Satellite Imagery in Different Ind...

Simulating Land Cover Changes and Their Impacts on Land Surface Temperature in Dhaka, Bangladesh