See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/334193571

LANDUSE AND LAND COVER CHANGE IN DHAKA METROPOLITAN AREA

DURING 1991-2022

Article · January 2019

CITATION READS

1 389

2 authors, including:

Md. Sofi Ullah

University of Dhaka
11 PUBLICATIONS   43 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Detecting Agricultural Landuse Change and Prediction in the Tarakanda Upazila of Mymensingh District During 1989 - 2026: A Geospatial Analysis using GIS and Remote
Sensing View project

Geospatial Modeling of Landslide Vulnerability in Bandarban District of CHT View project

All content following this page was uploaded by Md. Sofi Ullah on 03 July 2019.

The user has requested enhancement of the downloaded file.

LANDUSE AND LAND COVER CHANGE IN DHAKA METROPOLITAN AREA 1

Oriental Geographer
Vol. 60, No. 1&2, 2016
Printed in March 2019

LANDUSE AND LAND COVER CHANGE

IN DHAKA METROPOLITAN AREA DURING
1991-2022
MD. SOFI ULLAH1
MUHAMMAD ESMAT ENAN2

Abstract: The study aims at finding the changing pattern of landuse/cover of Dhaka
Metropolitan Area (DMA). Dhaka is one of the most populated megacities in the world.
The city built-up area is increasing day by day. As a result, the natural covers like water
body, green coverage and cultivable land is decreasing. To evaluate this change, different
geospatial indices like MNDWI, NDVI, and NDBI were used. Further, the Markov model
was used to predict the future state of landuse change in DMA. Among the indices
MNDWI was used to identify water body from satellite images, NDVI helps identify the
vegetation/green coverage and NDBI was used to find out the built-up area in DMA. The
result shows that, in 1991, the extension of built-up area was only 100 km2 and later, after
26 years in 2016, it was 220 km2 and will be 260 km2 in 2022. But water body is
continuously decreasing which was 123 km2 in 1991 and later in 2016, it was 41.8 km2
and after six years it will be 20 km2. The similar trend shows in green coverage which is
decreasing alarmingly. The trend shows that in 1991 there was 58.6 km2 green coverage,
later in 2016 it stood 38 km2 and it will be 23 km2 in 2022. The natural covers like water
bodies and greeneries should be protected for the sake of having livable environment for
Dhaka city dwellers.
Keywords: Geospatial techniques, GIS, Remote sensing, Markov probability matrix,
Indices, NDVI, MNDWI, and NDBI.

INTRODUCTION
Landuse/cover change is one of the most significant indicators that explain the impact of
human activities on environment. Both natural and manmade factors are responsible for
changing landuse/cover in local and regional levels. Such changes through human
intervention are common, that has seriously affected the earth’s ecological systems
(Lambin et al., 2001). Landuse change (LUC) is the modification of land surface. This
change is based on the purposes of need, which is not necessarily only making the change
in land cover but also change in intensity and management (Verburg et al., 2000).

1
Md. Sofi Ullah is Associate Professor, Department of Geography and Environment, University of Dhaka, Bangladesh
2
Muhammad Esmat Enan is an MS Research Student, Department of Geography and Environment, University of Dhaka,
Bangladesh
2 ORIENTAL GEOGRAPHER

The International Geosphere-Biosphere Program (IGBP) and the International Human

Dimension Program (IHDP) initiated a joint program to study the landuse/cover change
(LULC) considering the enormous impacts generating from such changes of landuse/land
cover on human life and environment. It emphasized the necessity of understanding,
modeling and projections of landuse dynamics from global to regional scale with focus
particularly on the spatial explicitness of processes and outcomes (Geoghegan et al.,
2001).

Vegetation cover changes due to human intervention in the landscape have resulted a
variety of negative environmental consequences. Deforestation, for example, can affect
the vegetation composition and water balance and can increase soil erosion (Glade, 2003;
Ghimire et al., 2013). This leads to increased environmental risks, such as landslide and
can have strong impacts on the human wellbeing on a larger scale (Tasser et al., 2003;
Korner et al., 2005; Papathoma-Köhle and Glade, 2013). Besides, forest cover change
produces a net carbon flux to the atmosphere (Houghton and Skole, 1990; Alves and
Skole, 1996), which in turn contributes to global climatic change. Another big concern is
urban area expansion by converting rural lands to urban lands. Globally, a large number
of the people live in the urban areas and since 2007 the global urban population has
exceeded the global rural population (UN, 2014). This process has impacts on energy
flow, biogeochemical cycles, biodiversity and climatic conditions at local, regional and
even global levels (McDonnell et al., 1997; Baker et al., 2001; Green and Baker 2003).
These changes are significant in understanding the global change of environment
(Nagendra et al., 2004).

Remote sensing and GIS are important tools in detecting and identifying the
landuse/cover changes and their geographical dynamics and its association with human
activities (Jensen and Cowen, 1999). These tools however are often inadequate to derive
detailed landscape properties such as its composition and configuration (Narumalani et
al., 2004). Various landscape metrics are available in ArcGIS to calculate and quantify
spatial patterns of land cover (McGarigal and Marks, 1995).

During last few decades, a number of researchers have identified, detected and monitored
urban landuse/cover from remotely sensed data (Ridd, 1995; Yang and Lo, 2002; Haack
and Rafter, 2006; Yuan, 2010). These data from various satellites are very expensive
depending on their resolutions (Blodget et al., 1991), provides an excellent opportunity to
know the historical landuse/cover changes. Such changes can be related to other
environmental and human factors as well. Therefore, a large number of researches have
utilized satellite-derived landuse/cover data with GIS landscape metrics to detect the
impacts on landscape structure (Lausch and Herzog, 2002; Herold et al., 2002;
Southworth et al., 2002; Li et al., 2004; Seto and Fragkias, 2005; Yu and Ng, 2006;
Kamusoko and Aniya, 2007; Jat et al., 2008; Hargis et al., 1998; O’Nell et al., 1999;
Dramstad et al., 2001; Herold et al., 2002; Civco et al., 2002; Luck and Wu, 2002;
Kamusoko and Aniya, 2007; Tzanopoulos and Vogiatzakis, 2010). These studies further
open the door of comprehensive knowledge of landuse change and its impact on
environment ((Leitao and Ahern, 2002). Such knowledge can help to restore the
LANDUSE AND LAND COVER CHANGE IN DHAKA METROPOLITAN AREA 3

ecological degradation (Egan and Howell 2001) and can help to formulate appropriate
landuse planning (Berger, 1987).
Bangladesh has already experienced a rapid increase in urban population in the few
decades. The total urban population was 13.2 million in 1981, 20.8 million in 1991, 29.2
million in 2001 and 33.5 million in 2011 (BBS, 2014) (Figure 1). The urban livelihood
thus converts rural land to the built-up areas to meet their necessary demand. It is
estimated that every year more than 809 km2 of arable land is being converted to cities’
roads and infrastructures in Bangladesh (BBS, 1996).

Figure 1: Population Growth Trends of Dhaka City

Data Source: BBS, 2011; UN, 2014.

Dhaka is one of the fastest growing megacities in the world. In 1991, Dhaka SMA
(Statistical Metropolitan Area) has been raised to the status of Dhaka Megacity with a
population of 6.8 million (BBS 1998). Between 1990 and 2005, the population of Dhaka
City doubled in number to 12 million (Khondker, 2010). In 2010, Dhaka was a city of 15
million populations (World Bank, 2013). Dhaka will become the second most densely
populated city in the world by 2020 (World Bank, 2007). By 2025, the UN predicts
Dhaka will be the home of more than 20 million people, larger than Mexico City, Beijing
or Shanghai (UN, 2014). The landscape of Dhaka City is undergoing a continuous
changes and modifications due to rapid urbanization (WASA, 1991). As the fastest
growing megacity, Dhaka has been facing multifarious challenges and numerous
difficulties (Islam and Ahmed, 2011).
The environmental consequences associated with fast and unplanned growth of Dhaka
have considerable impacts on natural habitats and on land (Amin et al., 2008; Dewan and
Yamaguchi, 2009). Such stresses induce high rate of poverty, unplanned urbanization, the
growth of urban slums and squatters, traffic jams, environmental pollution and other socio-
economic problems on the one hand and on the other hand, the landuse of DMA is
changing rapidly.
4 ORIENTAL GEOGRAPHER

Applications of geospatial technology enable us to explore the existing landuse condition of

a particular area. However, there is little attempt that has been made to employ geospatial
techniques to determine and detect landuse/cover changes in a fastest growing
metropolis, Dhaka. This paper aims at quantify land cover patterns of DMA for a specific
time period and predict the future scenario of land cover using remote sensing, GIS, and
metrics.

Considering these, the aims of this paper are to detect and identify the changes in
landuse/cover from 1991 to 2016 and predict it up to 2022. The specific objectives are:
 To find out the landuse pattern of DMA from 1991 to 2016.
 To analyze inter-sectoral changing pattern of landuse in DMA from 1991 to 2016
using Markov Matrix, and
 To simulate landuse change using CA Markov Matrix of probability for the year
of 2022

MATERIALS AND METHOD

Study Area
Dhaka Metropolitan Area (DMA) has been chosen for the study. It lies between 23.55
and 24.18N latitudes and 90.18 and 90.57E longitude (Figure 2). Dhaka is surrounded
by six rivers, like a garland. Balu and Sitalakhya Rivers on the eastern side, Turag and
Buriganga on the western side, Tongi Khal in the north and Dhaleshwari in the south
(Figure 2). Tongi Khal connects Turag and Balu rivers. Dhaleshwari and Sitalakhya join
in the south of Narayanganj and flow into the Meghna River.

There are different administrative bodies in Dhaka city. According to BBS report 2001,
the area under DCC (both North and South City Corporation) was 276 km2. There are
other areas, about 120 km2 outside DCCs but within DMA. These areas are administered
mainly by Union Parishads (Rural Local Governments).

Topographically, the area is flat partly flood plain and partly alluvial terrace, popularly
known as the Modhupur terrace of the Pleistocene period (Miah and Bazlee, 1968). The
elevation of the surface of the area is between 1 and 14 m and most of the built-up areas
are located at the elevations of 6 to 8 m (FAP 8A, 1991).

Data Sources
Both primary and secondary data were considered in this study. Primary data ) generated
from toposheets (with a scale of 1:50,000), collected from the cartographic lab,
Department of Geography and Environment, University of Dhaka and satellite data
comprised with Landsat 4-5 TM ( for the year 1991, 2001and 2011) and Landsat
OLI_TIRS for 2016 were considered and collected from USGS GloVis (Table 1).
LANDUSE AND LAND COVER CHANGE IN DHAKA METROPOLITAN AREA 5

Figure 2: Study Area: Dhaka Metropolitan Area (DMA)

Table 1: Specification of Satellite Data

Satellite Sensor Path/Row Acquisition Date Resolution
16/01/1991
Landsat 4-5 TM 137/44 21/01/2001
30/01/2011 30M
Landsat 8 OLI_TIRS 137/44 17/01/2016
Source: USGS GloVis

All the images were collected in the month of January considering sky condition. Spatial
resolution of these images was 30 meters.

Method
A lot of indices are available to calculate and quantify land cover change and to
understand their geographical pattern. For example, Normalized Difference Vegetation
6 ORIENTAL GEOGRAPHER

Index (NDVI), Normalized Difference Water Index (NDWI), Modified Normalized

Difference Water Index (MNDWI), Normalized Difference Built-up Index (NDBI),
Green Normalized Difference Vegetation Index (GNDVI), Normalized Difference Red
Edge Index (NDRE), etc.

Erdas Imagine and ArcGIS software were used for extracting the landuse/cover data of
different periods. All images were rectified using geometric correction and radiometric
correction. To find out the green coverage, Normalized Difference Vegetation Index (NDVI)
has been used. NDVI is a widely used vegetation index and is basically the difference of
vegetation reflectance in Near Infrared and Red bands (Rouse et al., 1974). The NDVI is
expressed as follows, NDVI = (NIR - RED) / (NIR + RED) , where NIR is a near infrared
band (0.76-0.90m) such as TM band 4 and RED is RED band such as TM band 3 (0.63-
0.69m). NDVI values range from +1.0 to -1.0. Areas of barren rock, sand, or snow
usually show very low NDVI values (for example, 0.1 or less). Sparse vegetation such as
shrubs and grasslands or senescing crops may result in moderate NDVI values
(approximately 0.2 to 0.5). High NDVI values (approximately 0.6 to 0.9) correspond to
dense vegetation such as found in temperate and tropical forests or crops at their peak
growth stage.

The second Index that has been used in this study is Modified Normalized Difference Water
Index (MNDWI). MNDWI helps finding out surface water by suppressing errors
produced by built-up area. Since water features extracted using the NDWI include false
positives from built-up land, a modified NDWI (MNDWI) was developed in which the
middle infrared (MIR) band was replaced with the near-infrared (NIR) band. The MNDWI
extracts surface water while suppressing errors from built-up land as well as vegetation and
soil (Xu, 2006). Modified NDWI can be expressed as follows, MNDWI = (Green –MIR) /
(Green + MIR), where MIR is a middle infrared band, which is Shortwave Infrared
(SWIR-1) and Green is the GREEN band. The computation of the MNDWI will produce
three results: (1) water will have greater positive values than in the NDWI as it absorbs
more MIR light than NIR light; (2) built-up land will have negative values as mentioned
above; and (3) soil and vegetation will still have negative values as soil reflects MIR light
more than NIR light (Jensen 2004) and the vegetation reflects MIR light still more than
green light.

The third index, Normalized Difference Built-up Index (NDBI), has been used in this
study to calculate the built-up area. NDBI is a new method to automate the process of
mapping the built-up areas (Zha et al., 2003). Like previous two indices, this index also
helps us to separate built-up area from the images of different years according to
reflectance value. Normalized Difference Built Index can express as follows,
NDBI = (SWIR – NIR) / (SWIR + NIR). Finally, the outputs of all indices have merged
and found out the fourth type of landuse, the bare soil of Dhaka City. A total four
landuse/cover types have been identified in the study area, these are: water body, green
coverage, built-up area and bare soil (Table 2 and Figure 3).
LANDUSE AND LAND COVER CHANGE IN DHAKA METROPOLITAN AREA 7

Table 2: Landuse/Cover Types in Dhaka Metropolitan Area

Landuse/cover types Description
Bare soil Landfills and bare area
Residential, commercial, service, transport, land prepared
Built-up area
for settlement, and communication
Green coverage Crop field, vegetation, natural and manmade forest
Permanent and seasonal wetlands, low lands, marshy land,
Water body
river and khal, etc
Source: Based on image analysis

Markov Model in Landuse/Cover Change

The classified landuse/cover (LU/LC) were used to create Markov Transitional Matrix of
landuse/cover change between 1991 and 2016. Then the Markov Matrix of Probability
was used to simulate landuse/cover change for the period of 2016 to 2022. Markov model
is a convenient tool for simulating LU/LC change when changes and processes in the
landscape are difficult to describe. Markov model describes LU/LC change from one
period to another and used this as the basis to project future changes (Logsdon et al.,
1996). Markov model has provided a simple methodology by which a dynamic system
could be analyzed and examined (Muller and Middleton, 1994; Dongjie et al., 2008;
Huang et al., 2008; Dadhich and Hanaoka, 2010). Researchers have tested the accuracy
of the Markov model (Jianping et al., 2005; Zhang et al., 2011).

Figure 3: Flow Chart of Methodology

8 ORIENTAL GEOGRAPHER

Accuracy Assessment
To assess the accuracy, some 200 sample points were cross verified with the Google
Earth Pro. Finally, the overall accuracy for different year was calculated. According to
accuracy assessment in 1991, there was a 3.85 percent absolute error, which is 118.02
hectares of the total land area at 96.15 percent accuracy level with a Kappa coefficient of
0.92. In 2001, there was a 7.69 percent absolute error, which is 236.04 hectares of the
total land area at 92.31 percent accuracy level with a Kappa coefficient of 0.83. In 2011,
there was a 3.85 percent absolute error, which is 118.02 hectares of the total land area at
96.15 percent accuracy level with a Kappa coefficient of 0.92. Similarly, in 2016, there
was a 3.85 percent absolute error, which is 118.02 hectares of the total land area at 96.15
percent accuracy level with a Kappa coefficient of 0.92 (Table 3).

Table 3: Overall Accuracy Assessment for Land Cover Analysis

Absolute Absolute Errors Overall Accuracy Kappa Coefficient

Year
Errors (percent) (in Hectares) (percent) (T)
1991 3.85 118.02 96.15 0.92
2001 7.69 236.04 92.31 0.83
2011 3.85 118.02 96.15 0.92
2016 3.85 118.02 96.15 0.92
Source: Satellite image and Google Earth Pro

RESULTS AND DISCUSSIONS

Landuse/Cover Change from 1991 to 2016
The spatial pattern of landuse/cover change in DMA for the year 1991, 2001, 2011 and
2016 is shown in Figure 4. It is revealed that in 1991, the city land cover was dominated
by water body. The spatial extent of this water body was covered the eastern part and
north-western part of the city, and it was replaced by green coverage in 2001. Similarly,
in 1991, the northern part of DMA (Uttarkhan and northern Uttara) was covered by green
coverage and the north-eastern portion was converted to the built-up area.

Furthermore, in 2001, the eastern part, which was low lying area, was filled-up with
plants temporarily. Later most of this area was converted into plain land through land
filling and turn to bare land area. The research findings show that, in 2011 the western
part, the vegetative area was converted into the built-up area and in the northern part
water and green coverage were narrowed down and shifted to built-up and bare land.
Finally, in 2016, the northern part of Uttarkhan, the Gazaria area and eastern part of
Matuail area were under green coverage. Similarly, in 2016, the water body has come
down to the lowest level (Figure 4).
LANDUSE AND LAND COVER CHANGE IN DHAKA METROPOLITAN AREA 9

Figure 4: Landuse/Cover Change in Dhaka Metropolitan Area from 1991 to 2016

Source: Extracted result from Landsat image analysis

Table 4 shows the water body was 123.4 km2 in 1991 which stood at 84 km2 in 2001,
50.39 km2 in 2011 and about 42 km2 in 2016. The decreasing rate of the water body is
strongly negatively correlated with the increase of built-up area and the rate of change
was -0.99. With this trend of change it can be speculated that in the near future, the water
body of the city will be nil if this rate continues. Similarly, the green coverage of DMA is
also decreasing. In 1991, City’s green coverage was 58.6 sq km which stood at 72 km2 in
2001, 49 km2 in 2011 and 38 km2 in 2016. The change rate of the green coverage is -0.71,
which is also negatively correlated. Another sector of landuse/cover, the bare soil, has
10 ORIENTAL GEOGRAPHER

also decreased between 1991 and 2016. In 1991, the bare soil was 24.3 km2, which stood
20 km2 in 2001, 14.8 km2 in 2011 and 6.3 km2 in 2016. Generally, the bare soil is rapidly
transformed into the built-up area. In DMA this is an excellent example of land
transformation. Finally, the dominated landuse/cover, the built-up area, which was 100
km2 in 1991, stood at 130.5 km2 in 2001, 192 km2 in 2011 (doubled between the year
1991 and 2011) and 220.5 km2 in 2016. The temporal increasing rate of the built-up area
is strongly positively correlated with urbanized land and the rate is 0.988 (Table 4). The
landuse/cover change trend is represented in Figure 5.

Table 4: Landuse/Cover Statistics of Dhaka Metropolitan Area from 1991 to 2016

Area (sq km)
Landuse Category Correlation (r)
1991 2001 2011 2016
Bare soil 24.34 20.24 14.87 6.32 -0.952
Built up area 100.43 130.57 192.32 220.53 0.988
Green coverage 58.64 72 49.27 38.16 -0.710
Water body 123.43 84.05 50.39 41.83 -0.993
Total 306.84 306.84 306.84 306.84
Source: Result of Landsat image analysis

Figure 5: Landuse/Cover Change Statistics in Dhaka Metropolitan Area from 1991 to 2016
Source: Result from Landsat image analysis

Proportional Area in Each Category of Landuse/Cover from 1991 to 2016

The proportional area in each of the landuse categories in DMA from 1991 to 2016 is
presented in Table 5. In 1991, the highest portion of the land cover was water body
(40%), the second highest was built-up area (33%), green coverage was 19 percent and
the smallest portion was bare soil, which was 8 percent only. In 2001, the built-up area
occupied 43 percent of the land and secured the first place among the land coverage
LANDUSE AND LAND COVER CHANGE IN DHAKA METROPOLITAN AREA 11

categories. In this period, water body was 27 percent, placed at the second position, green
coverage was 23 percent, and bare soil was 7 percent. In 2011, the built-up area
increased at 63 percent, the water body and green covers were the same in this period
(16%). In 2016, the built-up area covered about 72 percent of DMA and water body
covered at 14 percent. Green coverage was 12 percent only (Table 5).

Table 5: Proportional Area in Each Category of Landuse/Cover in Dhaka Metropolitan

Area from 1991 to 2016 (in percentage)
Landuse Category 1991 2001 2011 2016

Bare soil 08 07 05 02

Built up area 33 43 63 72

Green coverage 19 23 16 12

Water body 40 27 16 14

Grand Total 100 100 100 100

Source: Result from Landsat image analysis

Transitional Matrix of Landuse/Cover Change

In control theory, the state-transition matrix is a product with the state vector at an initial
time gives at a later time. The state-transition matrix can be used to obtain the general
solution of linear dynamic systems. A transitional matrix can represent changes from one state
to another state. According to the transition matrix, the LU/LC class will change to another class
in the future, given the present state of the class (Kumar et al, 2014). The matrix equation is as
follows:

On the basis of the given equation, the transition probability of a certain LU/LC type in 1991
was converted into the same LU/LC type in 2001. Similarly, the transitional probability of 2001-
2011 and 2011-2016 were calculated. Table 6 shows the transitional probability of 2011-2016.
Finally, the transitional probability of a certain LU/LC type in 2016 converted into the same
LU/LC type in 2022 by using Markov Matrix. Table 7, shows the primary transitional probability
of four landuse/cover types (bare soil, built-up area, green coverage and water body) during 2016-
2022.
12 ORIENTAL GEOGRAPHER

Table 6: Transitional Matrix of Landuse/Cover between 2011 and 2016

Transitional Probability
Landuse Category
Bare soil Built up area Green coverage Water body
Bare soil 0.0042 0.8846 0.0539 0.0572
Built up area 0.0054 0.5189 0.247 0.2287
Green coverage 0.0047 0.3591 0.2913 0.345
Water body 0.0209 0.6948 0.1545 0.1298
Source: Result of Landsat image analysis

Table 7: Cellular Automata (CA) Markov Matrix of Probability to Simulate Landuse/Cover

Change from 2016 to 2022
Transitional Probability
Landuse Category
Bare soil Built up area Green coverage Water body
Bare soil 0.070 0.814 0.050 0.066
Built up area 0.011 0.901 0.056 0.032
Green coverage 0.009 0.770 0.135 0.085
Water body 0.007 0.641 0.123 0.229
Source: Result of Landsat image analysis

Simulated Landuse/Cover in Dhaka Metropolitan Area

The simulated result shows that the built-up area of DMA has been continuously
increasing. There will be 84.78 percent built-up area in the City in 2022. All other land
covers will decrease, such as 1.14 percent bare soil, 7.49 percent green coverage and the
6.6 percent water body in 2022 (Table 8 and Figure 6). Similarly, the simulated result of
the year 2028 and 2034 was also calculated using Markov model, which shows, the built-
up area will increase to 87.40 percent in 2028 and 88.46 percent in 2034.

Table 8: Landuse/Cover Change Prediction Using Markov Probability Matrix

Yea Landuse/Cover Category

r Bare soil Built-up area Green coverage Water body
Area ( sq
Area ( sq km) % Area ( sq km) % % Area ( sq km) %
km)
1991 24.34 7.93 100.43 32.73 58.64 19.11 123.43 40.22

2001 20.24 6.59 130.57 42.55 72.00 23.46 84.05 27.39

2010 14.87 4.84 192.32 62.68 49.27 16.06 50.39 16.42

2016 6.32 2.06 220.53 71.87 38.16 12.44 41.83 13.63

2022 3.49 1.14 260.15 84.78 22.98 7.49 20.24 6.60

Source: Result of Landsat image analysis

LANDUSE AND LAND COVER CHANGE IN DHAKA METROPOLITAN AREA 13

Figure 6: Trend of Landuse/Cover Change during 1991-2022

Source: Landsat image analysis and Markov simulation model

Markovian Model for Sectoral Change of Landuse/Cover

According to Markovian model, the sectoral movement of landuse change is always
towards built-up area (Table 9). The change of bare soil to bare soil, built-up area, green
coverage and water body has decreased during 1991 to 2016 and the same trend will
follow in 2022 (Table 9). Similarly, water body to water body, bare soil, and green
coverage has decreased during 1991 to 2016 and will decrease in 2022. The green
coverage has also decreased in all sectors (green to green, green to bare soil and green to
water) except green to the built-up area. The change between built-up to built-up was
very high during 1991-2016 which was not usual and this trend will continue in 2022.
Among this period the changing rate of built-up to built-up is 98 percent. The trend of
changing built-up to green coverage has been slightly increasing. In the recent time
people are motivated to planting trees in and around the compound of houses and
courtyards. May be this is the main reason to increase of built-up to green coverage. The
scenario of the landuse change within 32 years (between 1991 and 2022) gives us a
direction that the land of all categories is moving significantly. If this trend continues in
future, the City will be losing its ecological balance.

Table 9: Each Type of Landuse Change in Dhaka Metropolitan Area for Each Time Period
1991-2001 2001-2011 2011-2016 2016-2022
Landuse change Area Area Area Area
% % % %
(sq km) (sq km) (sq km) (sq km)

Bare soil  bare soil 10.7963 3.52 8.6241 2.81 5.0073 1.63 0.4419 0.14

8.3576 2.72 9.139 2.98 6.7359 2.20 5.1458 1.68

Bare soil  built-up area
4.0974 1.34 1.7373 0.57 0.9386 0.31 0.3191 0.10
Bare soil  green coverage
Bare soil  water body 1.0925 0.36 0.7355 0.24 2.1844 0.71 0.4173 0.14

28.18 113.2691 36.91 170.135 55.45 198.7833 64.78

Built-up area  built-up area 86.4617
14 ORIENTAL GEOGRAPHER

1991-2001 2001-2011 2011-2016 2016-2022

Landuse change Area Area Area Area
% % % %
(sq km) (sq km) (sq km) (sq km)

Built-up area  bare soil 2.8206

0.92 2.5576 0.83 0.8163 0.27 2.375 0.77

Built-up area  green coverage 9.2216 3.01 7.4575 2.43 10.3746 3.38 12.3752 4.03

Built-up area  water body 1.9296 0.63 7.2839 2.37 10.997 3.58 7.0001 2.28

Green coverage  green coverage 25.0659

8.17 17.365 5.66 12.1713 3.97 5.158 1.68

Green coverage  bare soil 2.9279 0.95 1.9548 0.64 0.265 0.09 0.359 0.12

Green coverage  built-up area 21.4854 7.00 38.9409 12.69 25.5703 8.33 29.4011 9.58

Green coverage ` water body 9.1657 2.99 13.7357 4.48 11.2673 3.67 3.2423 1.06

Water body  water body 71.8613 23.42 28.6315 9.33 17.3829 5.66 9.5776 3.12

Water body  bare soil 3.6912 1.20 1.7297 0.56 0.2354 0.08 0.3099 0.10

Water body  built-up area 14.2633 4.65 30.9739 10.09 18.0923 5.90 26.8166 8.74

Water body  green coverage 33.6116 10.95 22.7141 7.40 14.676 4.78 5.1276 1.67

Total 306.8496 100 306.8496 100 306.8496 100 306.8498 100

Source: Analysis result of Markov matrix

CONCLUSIONS
Dynamics of landuse/cover changes and the quantification of land cover patterns in the
Dhaka Metropolitan Area were shown in this paper by using geospatial techniques and
Markov model. Markov model helps to calculate sifting of sectoral landuse and
landuse/cover. The results show that City’s built-up area has been increasing from
1991and continuing in accelerating rate. In 1991, the built-up area was 100 km2, but in
2016 it was 220.53 km2 and it will increase to 260 km2 in 2022. On the other hand, during
1991-2016, the green coverage has been decreased from 58 km2 to 38 km2 and will stand
at 23 km2 in 2022. Similarly, the water body has decreased unusually. In 1991, there was
123.4 km2 water body in the City, which stood at 41.8 km2 in 2016 and it will be 20.2 km2
in 2022. The result indicates that the land cover, especially, the water bodies, green
coverage is decreasing continuously, and Dhaka is losing its sustainability. Dhaka is
growing without any guidance of Master Plan or Detail Area Plan (DAP). If this
continues, Dhaka will be uninhabitable within a few years. It is necessary to find a way
by which the City development can be made sustainable.

REFERENCES
Alves, D. S., and Skole, D. L. (1996). Characterizing land cover dynamics using multitemporal
imagery. International Journal of Remote Sensing, 17(4), 835–839.
Amin, S. M. N., Dewan, A. M., & Kabir, M. H. (2008). Urban development and environmental
degradation: the case of Dhaka metropolitan, Bangladesh. Poster presented at Annual
Meeting of the Association of American Geographers, Boston, April 15–19.
Baker, L. A., Hope, D., Xu, Y., Edmonds, J., and Lauver, L. (2001). Nitrogen balance for the
Central Arizona-Phoenix (CAP) Ecosystems. Ecosystems, 4(6), 582–602.
LANDUSE AND LAND COVER CHANGE IN DHAKA METROPOLITAN AREA 15

BBS. (1996). Bangladesh Census of Agriculture. Bangladesh Bureau of Statistics (BBS), Statistics
and Informatics Division (SID), Ministry of Planning, Government of the People’s
Republic of Bangladesh.
BBS. (1998). Statistical Year Book, 1998. Bangladesh Bureau of Statistics (BBS), Statistics and
Informatics Division (SID), Ministry of Planning, Government of the People’s Republic
of Bangladesh.
BBS. (2001). Bangladesh Population Census 2001. Bangladesh Bureau of Statistics (BBS),
Statistics and Informatics Division (SID), Ministry of Planning, Government of the
People’s Republic of Bangladesh.
BBS. (2011). Bangladesh Population and Housing Census 2011. Bangladesh Bureau of Statistics
(BBS), Statistics and Informatics Division (SID), Ministry of Planning, Government of
the People’s Republic of Bangladesh.
BBS. (2014). Union Statistics, 2014. Bangladesh Population and Housing Census 2011.
Bangladesh Bureau of Statistics (BBS), Statistics and Informatics Division (SID),
Ministry of Planning, Government of the People’s Republic of Bangladesh.
Berger, J. (1987). Guidelines for landscape synthesis: some directions–old and new. Landscape
and Urban Planning, 14, 295–311.
Blodget, H., Taylor, P., and Roark, J. (1991). Shoreline changes along the Rosetta-Nile
Promontory: monitoring with satellite observations. Marine Geology, 99(1–2), 67–77.
Civco, D. L., Hurd, J. D., Wilson, E. H., Arnold, C. L., and Prisloe, S. (2002). Quantifying and
describing urbanizing landscapes in the northeast United States. Photogrammetric
Engineering and Remote Sensing, 68(10), 1083–1090.
Dadhich, P. N. and Hanaoka, S. (2010). Remote sensing, GIS and Markov’s method for landuse
change detection and prediction of Jaipur district. Journal of Geomatics and Geosciences.
4:9–15.
Dewan, A. M. and Yamaguchi, Y. (2009), Landuse and Land Cover Change in Greater Dhaka,
Bangladesh: Using Remote Sensing to Promote Sustainable Urbanization. Journal ofApplied
Geography, 29, pp.390-401.
Dongjie, G., Weijun, G., Kazuyuki, W. and Hidetoshi F. (2008). Landuse change of Kitakyushu
based on landscape ecology and Markov model. International Journal Geographical
Information Science 18:455–468.
Dramstad, W. E., Fry, G., Fjellstad, W. J., Skar, B., Helliksen, W. and Sollund, M. L. B. (2001).
Integrating landscape based values- Norweigenian monitoring of agricultural landscapes.
Landscape and Urban Planning, 57(3), 257–268.
Egan, D. and Howell, E. A. (2001). The historical ecology handbook. Washington DC: Island
Press.
FAP 8A. (1991). Flood Action Plan. Master Plan Study for Greater Dhaka Protection Project.
Dhaka: Japan International Cooperation Agency.
Geoghegan, J., Klepeis S.C.P., Mendoza, P.M., Yelena, O.R., Chowdhury, R., Turner II, B.L., and
Vance, C. (2001). Modeling tropical deforestation in the southern Yucatan peninsular region:
comparing survey and satellite data. Agriculture, Ecosystems and Environment, 85, pp. 25-46.
Ghimire, S., Higaki, D. and Bhattarai, T. (2013). Estimation of soil erosion rates and eroded
sediment in a degraded catchment of the Siwalik Hills, Nepal. Land 2:370–391.
Glade. T. (2003). Landslide occurrence as a response to landuse change: a review of evidence
from New Zealand. Catena 51:297–314.
16 ORIENTAL GEOGRAPHER

Green, D. M., and Baker, M. G. (2003). Urbanization impacts on habitat and bird communities in
a Sonoran desert eco-system. Landscape and Urban Planning, 63(4), 21–37.
Haack, B. N. and Rafter, A. (2006). Urban growth analysis and modeling in the Kathmandu
valley, Nepal. Habitat International, 30(4), 1056–1065.
Hargis, C., Bissonette, J., and David, J. (1998). The behavior of landscape metrics commonly used
in the study of habitat fragmentation. Landscape Ecology, 13(3), 167–186.
Herold, M., Clarke, K. C., and Scepan, J. (2002). The use of remote sensing and landscape metrics
to describe structures and changes in urban landuse. Environmental Planning A, 34(8),
1443–1458.
Houghton, R. A., and Skole, D. L. (1990). The earth as trans-formed by human action. In W. C.
Clark, R. W. Kates, J. F. Richards, J. T. Mathews, W. B. Meyer, & B. L. Turner (Eds.),
The earth as transformed by human action (pp. 393–408). Cambridge: Cambridge
University Press.
Huang, W., Liu, H., Luan, Q., Bai, M. and Mu, X. (2008). Monitoring urban expansion in
Beijing, China by multi temporal TM and spot images. IEEE Proc, IGARSS. 4:695–698.
Islam, M. S. and Ahmed, R. (2011). Landuse change prediction in Dhaka city using GIS aided
Markov chain modeling. Journal of Life Earth Science. Vol. 6, pp. 81-89.
Jat, M. K., Garg, P. K., and Khare, D. (2008). Modeling urban growth using spatial analysis
techniques: A case study of Ajmer city (India). International Journal of Remote Sensing,
29(2), 543–567.
Jensen, J. R., and Cowen, D. C. (1999). Remote sensing of urban/suburban infrastructure and
socioeconomic attributes. Photogrammetric Engineering and Remote Sensing, 65(5),
611–622.
Jensen, J.R. (2004). Introductory Digital Image Processing: A Remote Sensing Perspective. 3rd
Edition, Prentice-Hall, Inc., New York, 316.
Jianping, L,I., Bai, Z. and Feng, G. (2005). RS-and-GIS-supported forecast of grassland
degradation in southwest Songnen plain by Markov model. Geo-spatial Information
Science. 8:104–06.
Kamusoko, C., and Aniya, M. (2007). Landuse/cover change and landscape fragmentation
analysis in the Bindura District, Zimbabwe. Land Degradation and Development, 18(2),
221–233.
Khondker, H. H. (2010). “Dhaka: Megacity Despair”, Global Dialogue, Vol. 1, Issue 2,
Bangladesh
Korner C, Ohsawa M, Spehn E et al (2005) Mountain systems. In: Hassan R, Scholes R, Ash N
(eds) Ecosystem and human well being: current state and trends: findings of the
Conditions and Trends working group. Millennium Ecosystem Assessment, pp 681–716.
Kumar, S., Radhakrishnan, N. and Mathew, S. (2014). Landuse change modeling using a Markov
model and remote sensing, in Geomatics, Natural Hazards and Risk, 2014,Vol. 5, No. 2,
145–156.
Lambin, E. F., Turner, B. L., Geist, H. J., Agbola, S. B., An-gelsen, A., Bruce, J. W., Coomes, O.
T., Dirzo, R.,Fischer, G., Folke, C., George, P. S., Homewood, K., Imbernon, J.,
Leemans, R., Li, X., Moran, E. F., Morti-more, M., Ramakrishnan, P. S., Richards, J. F.,
Ska˚nes, H., Steffen, W., Stone, G. D., Svedin, U., Veldkamp, T. A., Vogel, C., and Xu,
J. (2001). The causes of landuse and land cover change moving beyond the myths.
Global Environmental Change, 11(4), 261–269.
LANDUSE AND LAND COVER CHANGE IN DHAKA METROPOLITAN AREA 17

Lausch, A., and Herzog, F. (2002). Applicability of landscape metrics for the monitoring of
landscape change: Issues of scale, resolution and interpretability. Ecological Indicators,
2(1–2), 3–15.
Leitao, A. B., and Ahern, J. (2002). Applying landscape ecological concepts and metrics in
sustainable landscape planning. Landscape and Urban Planning, 59(2), 65–93.
Li, Z., Li, X., Wang, Y., Ma, Y., and Wang, J. (2004). Landuse change analysis in Yulin
prefecture, northwest China using remote sensing and GIS. International Journal of
Remote Sensing, 25(24), 5691–5703.
Logsdon, M. G., Bell, J, E. and Westerlund, V,F. (1996). Probability mapping of landuse change:
a GIS interface for visualizing transition probability. Computer Environment Urban
Systems. 20:389– 398.
Luck, M., and Wu, J. (2002). A gradient analysis of urban landscape pattern: A case study from
the Phoenix metropolitan region, Arizona, USA. Landscape Ecology, 17(4), 327–339.
McDonnell, M. J., Pickett, S., Groffman, P., & Bohlen, P. (1997). Ecosystem processes along an
urban-to-rural gradient. Urban Ecosystem, 1(1), 21–36.
McGarigal, K. and B. Marks. (1995). FRAGSTATS: Spatial analysis program for quantifying
landscape structure. USDA Forest Service General Technical Report. PNW-GTR–351.
Miah, M. M., and Bazlee, B. L. (1968). Some aspects of geomorphology of the Madhupur Tract.
Oriental Geographer, 12, 39–48.
Muller, R.M. and Middleton J. (1994). A Markov model of land-use change dynamics in the
Niagara region, Ontario, Canada. Landscape Ecology. 9:151–157.
Nagendra, H., Munroe, D. K., and Southworth, J. (2004). From pattern and process: Landscape
fragmentation and the analysis of landuse/cover change. Agriculture, Ecosystem and
Environment, 101(2–3), 111–115.
Narumalani, S., Mishra, D. R. M., and Rothwell, R. G. (2004). Change detection and landscape
metrics for inferring anthropogenic processes in the greater EFMO area. Remote Sensing
of Environment, 91(3–4), 478–489.
O’Nell, R. V., Ritters, K. H., Wickham, J. D., and Jones, K. B. (1999). Landscape pattern metrics
and regional assessment. Ecosystem Health, 5(4), 225–233.
Papathoma-Köhle M, Glade, T. (2013). The role of vegetation cover change for landslide hazard
and risk. In: Renaud FG, Sudmeier-Rieux K, Estrella M (eds) The role of ecosystems in
disaster risk reduction. UNU-Press, Tokyo, pp 293–320.
Ridd, M. K. (1995). Exploring a VIS (vegetation-impervious-surface soil) model for urban
ecosystem analysis through remote sensing: Comparative anatomy for cities. Inter-
national Journal of Remote Sensing, 16(12), 2165–2185.
Rouse, J. W. Haas, R. H., Schell, J. A. and Deering, D. W. (1974). Monitoring vegetation systems
in the Great Plains with ERTS. Third Earth Resources Technology Satellite-1,
Syposium. Volume I: Technical Presentations, NASA SP-351(I), pp. 309-317.
Seto, K. C. and Fragkias, M. (2005). Quantifying spatiotemporal patterns of urban landuse change
in four cities of China with time series landscape metrics. Landscape Ecology, 20(7),
871–888.
Southworth, J., Nagendra, H., and Tucker, C. (2002). Frag-mentation of a landscape:
Incorporating landscape metrics into satellite analysis of land cover change. Landscape
Research, 27(3), 263–269.
 18 ORIENTAL GEOGRAPHER

Tasser E, Mader M, Tappeiner U (2003) Effects of landuse in alpine grasslands on the probability
of landslides. Basic and Applied Ecology. 4:271–280.
Tzanopoulos, J., and Vogiatzakis, I. N. (2010). Processes and patterns of landscape change on a
small Aegean island: The case of Sifnos, Grecce. Landscape & Urban Planning, 99(1),
58–64.
UN. (2014). World Urbanization Prospects, United Nations, Department of Economic and
Social Affiars, Population Devision. New York. USA.
Verburg, P.H., Chen, Y., Soepboer, W. and Veldkamp, A. (2000).GIS-based modeling of human-
environment interactions for natural resource management.Applications in Asia. Proc. 4th
International.conference.Integration. GIS Environment. Modeling (GIStEM4): Problems,
Prospectsand Research Needs, Banff, Alberta, Canada, Sept. 2 - 8, 2000, pp. 1-18.
WASA. (1991). “Dhaka regional groundwater and subsidence model,” Dhaka Water Supply and
Sewerage Authority (WASA), 1991.
World Bank. (2007). Dhaka: Improving living conditions for the urban poor. South Asia Region:
World Bank, Sustainable Development Unit.
World Bank. (2013). World Bank Annual Report 2013. The World Bank. Washinton, D.C. USA.
Xu, H. (2006). Modification of normalised difference water index (NDWI) to enhance open water
features in remotely sensed imagery.International Journal of Remote Sensing.2006, 27,
pp. 3025-3033.
Yang, X., and Lo, C. P. (2002). Using a time series of satellite imagery to detect landuse and cover
changes in the Atlanta, Georgia. International Journal of Remote Sensing, 23(9), 1775–
1798.
Yu, X., and Ng, C. (2006). An integrated evaluation of landscape change using remote sensing and
landscape metrics. International Journal of Remote Sensing, 27(6), 1075–1092.
Yuan, F. (2010). Urban growth monitoring and projection using remote sensing and geographic
information systems: A case study in the Twin Cities Metropolitan Area, Minnesota.
Geocarto International, 25(3), 213–230.
Zha, Y. Gao, J and NI, S. (2003). Use of normalized difference built-up index in automatically
mappingurban areas from TM imagery. International Journal of Remote Sensing.Vol. 24,
N0, 3, pp.583 -594.
Zhang, R., Tang, C., Ma, S., Yuan, H., Gao, L. and Fan, W. (2011). Using Markov chains to
analyze changes in wetland trends in arid Yinchuan plain, China. Math Computer
Modelling 54:924–930.

View publication stats