GeoJournal (2023) 88:425–448

https://doi.org/10.1007/s10708-022-10609-4

Simulating future intra‑urban land use patterns

of a developing city: a case study of Jashore, Bangladesh
Syed Riad Morshed · Md. Abdul Fattah · Md. Mojammel Hoque · Md. Razzakul Islam ·
Fahmida Sultana · Kaniz Fatema · Md. Fazle Rabbi · Asma Amin Rimi · Fahmida Yeasmin Sami ·
F. M. Rezvi Amin · Musfiqur Rahman Chowdhury Seam · Mehedi Hasan Sizan · Mahamudul Hasan ·
Md. Nazmul Haque

Accepted: 9 February 2022 / Published online: 5 March 2022

© The Author(s), under exclusive licence to Springer Nature B.V. 2022, corrected publication 2022

Abstract Increasing urban growth at an unprec- (LULC) changes and simulate future potential intra-
edented rate entails adverse implications for societal, urban LULC growth patterns of Jashore City up to
economic, and environmental sustainability. In the 2050. We used (i) Landsat images to analyze LULC
cities of Bangladesh, the land covers are experiencing change using maximum likelihood supervised image
rapid construction-associated land expansion, popula- classification method; (ii) Markov-CA model to illus-
tion growth, and socioeconomic development. Com- trate the LULC transition matrix during 2000–2020,
prehensive assessment and understanding of the pros- (iii) Multilayer Perception Neural Network Markov
pects for rapid land use/land cover (LULC) changes Chain (MPNNMC) Model to simulate future LULC
are essential for managing land surface resources and patterns. The result shows that built-up area expanded
ensuring sustainable development. Therefore, this quickly, while cropland and water areas have had a
study aims to assess the historical land use/land cover large loss of coverage. The LULC change analysis

S. R. Morshed · M. A. Fattah (*) · M. R. Islam · F. Y. Sami

F. Sultana · K. Fatema · M. F. Rabbi · A. A. Rimi · e-mail: fahmida.s32@gmail.com
F. Y. Sami · F. M. Rezvi Amin · M. R. C. Seam ·
F. M. Rezvi Amin
M. H. Sizan · M. Hasan · M. N. Haque
e-mail: rezviamin.kuet@gmail.com
Department of Urban and Regional Planning, Khulna
University of Engineering and Technology, Khulna, M. R. C. Seam
Bangladesh e-mail: musfiqseam@yahoo.com
e-mail: mafattah.kuet@gmail.com
M. H. Sizan
S. R. Morshed e-mail: mehedihasansizan@yahoo.com
e-mail: riad.kuet.urp16@gmail.com
M. Hasan
M. R. Islam e-mail: offpvl16@gmail.com
e-mail: razzakul152@gmail.com
M. N. Haque
F. Sultana e-mail: nhaque.kuet13@gmail.com
e-mail: fahmiiidatasnim@gmail.com
M. M. Hoque
K. Fatema
Department of Civil Engineering, Chittagong University
e-mail: kanizfatema1013@gmail.com
of Engineering and Technology, Chattogram, Bangladesh
M. F. Rabbi e-mail: hmojammel40@gmail.com
e-mail: azlerabbi.fearless@gmail.com
A. A. Rimi
e-mail: asmaamin97@gmail.com

Vol.: (0123456789)
13
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
 426 GeoJournal (2023) 88:425–448

derived from prior LULC was utilized for future of future potential urban growth (Aburasa et al.,
simulations, where natural and anthropogenic factors 2016; Al-Darwish et al., 2018; Bihamta et al., 2015;
were chosen as the driving variables in the MPN- Feng et al., 2018). Moreover, the complex structure
NMC model. The future LULC modeling shows of the LULC pattern must be understood to accurately
that compared to 2020, the urban area is expected to simulate future urban growth.
increase by 23.64%, whereas cropland, vegetation, City provide many opportunities, such as access
unused land, and water areas are expected to reduce to basic service facilities, health, financial, enter-
by 1.16%, 5.47%, 9.55%, and 7.73% respectively, by tainment, educational, and employment opportuni-
2050. The change analysis shows that urban areas will ties with huge wages (Ahrend et al., 2014; Moore
increase the fastest during 2020–2030. The findings et al., 2003). This has significantly accelerated global
demonstrate that the rapid and unplanned urbaniza- urbanization over the past few centuries. The urban
tion and the rise of the population due to migration growth prediction model helps local authorities and
resulted in the fastest LULC transformation. The policymakers to determine strategies for controlling
study findings contribute to the long-term ecological urban growth and environmental changes and also
development of Jashore City and potentially enhance estimates the investment needed for public infra-
environmental decision making. structure and sustainable urban development (Maar-
seveen et al., 2018; Weng, 2012). Many researchers
Keywords Urban land cover growth pattern · Land around the globe have indicated the study of urban
cover modeler · Jashore city · LULC transition growth for many developed and developing countries
(Aithal et al., 2018; Al-shalabi et al., 2013; Gómez
et al., 2020; Herold et al., 2003; Rimal et al., 2018).
Introduction Cities consume the surrounding lands and areas as
they grow. Therefore, urban planners, policymakers,
The enormous urbanization of the cities around the and engineers need to develop methods to assess and
world is a global concern that creates challenges and simulate future urban growth, regarding the future
opportunities for long-term sustainability (Gómez consequences of LULC, to prevent communities from
et al., 2020). In the last century, continuous urbaniza- building in protected or conserved areas or areas vul-
tion was accomplished under the distinctive influence nerable to hazards.
of population growth, migration, and continued devel- In recent years, several studies have witnessed the
opment. The sustainable development of the cities of rapid urban expansion in both urban and rural areas
Bangladesh has mainly been hindered by uncontrolled of Bangladesh. About 809 ­km2 of cropland in Bang-
and unplanned rapid population growth and urbaniza- ladesh is declining (Gazi et al., 2020). According to
tion (Kafy et al., 2021a). Over the last few decades, Rai et al. (2017) the forest land cover area percentage
the cities in Bangladesh have experienced one of the in Bangladesh declined from 15.7% to 9.5% during
fastest urbanization rates in the world (Faisal et al. 1930–2014, beel and haor area increased from 1.66%
2021a). Migration of people to urban areas acceler- to 1.72% during 1976–2010, urban area increased
ates urban growth and creates problems such as envi- from 23.6% to 28.4% during 2000–2011 (Hasan
ronmental degradation (Kafy et al., 2020), poverty et al. 2013). The average vegetation cover loss was
traps (Grant, 2010), informal settlements, air, and calculated 66.87 k­ m2 between 1989 and 2009. Sev-
noise pollution, fast-spreading disease (Moore et al., eral studies of LULC change analysis revealed that
2003), and the economic costs of increasing public the waterbody area percentage in the urban areas of
infrastructure (Gómez et al., 2020). These phenomena Bangladesh declined rapidly (Gazi et al., 2020; Kafy
have been observed in the major cities of the country et al., 2020; Morshed et al., 2021) while increased in
which are emerging as unplanned cities, mostly domi- the rural areas, especially in the southwestern regions
nated by overcrowded and haphazard settlements. For due to the shifting of rice field to shrimp farms (Fat-
sustainable urban planning, environmental and eco- tah et al., 2021b; Islam et al., 2015; Khan et al.,
logical resource management, it is necessary to con- 2015).
sider the spatiotemporal changes in urban land use Bangladesh is a developing country. According to
and land cover (LULC) and especially the prediction the General Economics Division (2020), Bangladesh

Vol:. (1234567890)
13
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
GeoJournal (2023) 88:425–448 427

will experience a rapid and transformational change development in cities (Al-Hameedi et al. 2021). A
in agriculture, education, trade and industry, trans- large amount of land covers is being transformed
portation, healthcare, and communication. The cur- and substituted into built-up land for fulfilling the
rent government aims to reach the title of developed demands of residential, commercial, industrial and
country by 2041, and in this regard, "Perspective Plan other demands (Somvanshi et al., 2020). The term
2041" has been adopted. Various infrastructures, such "LULC change" is used to describe how human
as roads, bridges, offices, buildings, factories, and activities alter the Earth’s terrestrial surface (Gazi
industries, are being built in the cities and towns of et al., 2020). While humans have been altering land
the country, and this has accelerated the urbaniza- for thousands of years to obtain livelihoods and other
tion rate throughout the country (Hassan & Nazem, necessities, the rate, intensity, and extent of LULC
2016; Shubho & Islam, 2020). Topics related to transformation are much greater now than they were
LULC change, such as simulation of future potential earlier. These changes are causing unprecedented
LULC patterns and their consequences, have recently changes in ecosystems and environmental processes
attracted interest from a wide range of literature at the local, regional, and global levels (Fattah et. al.
(Araya & Cabral, 2010; Han et al., 2015; Kafy et al., 2021b). The study of LULC change is regarded as a
2020; Maduako et al., 2016). Though there have significant aspect of observing parameters that are
been many studies on LULC change and its impact accountable for overall changes (Li et al., 2017; Nath
on the contest of the cities of Bangladesh, there have et al., 2018; Tendaupenyu et al., 2016), landscape
not been sufficient studies on urban growth predic- changes (Li & Liu, 2017; Ying et al., 2017), eco-
tion. Previous studies, related to LULC simulation, system change, landscape fragmentation (Nagendra
have focused on urban growth in Dhaka (Ahmed & et al., 2004; Nurwanda et al., 2015), urbanization
Bramley, 2015; Pramanik & Stathakis, 2015; Shubho (Ahmed, 2011), climate change (Morshed & Fattah,
& Islam, 2020) and Chittagong (Hassan & Nazem, 2021), sustainable development (Boadi, et al., 2005),
2016), but the assessment of urban growth patterns and risk assessment at the city level. Therefore, the
and prediction of other cities has not been studied in effective use and management of land resources is
the previous literature. important for sustainable development from environ-
In this study, we identified the historical urban mental and socioeconomic aspects.
growth pattern, transformation and direction of the Currently, GIS and remote sensing technologies
built-up land expansion of Jashore City during 2000 can provide advanced, more powerful and accurate
– 2020 to model the future intra-urban land use pat- tools to aid in better monitoring and understanding
tern (FILUP) of 2030, 2040, and 2050. The FILUP the spectral, spatial and temporal characteristics of
modelling is performed by using the natural and LULC changes at regional, national and global scales
anthropogenic variables in the MPNNMC model. The (Hasan and Nazem, 2015). The LULC changes are
study contributes to the planning and implementation identified through the image classification process.
of many short- and long-term development plans for Image classification is a broad category of digital
Jashore City and its environs. The simulated spatial image processing techniques that involve the auto-
complex variation in LULC can assist policymakers matic grouping of pixels into specified categories
and planners in designing sustainable plans for the (Kafy et al., 2020). The images have been classified
cities of Bangladesh by providing a holistic under- by (i) supervised image classification method (SICM)
standing of the current scenario in relation to future and (ii) unsupervised image classification method
expansion possibilities and environmental safety. (UICM). In SICM, some training pixels are selected
from each of the classes in the first step, which is
called the training stage. Image classification can be
Literature review achieved by identifying the characteristics of train-
ing pixels and then looking for other pixels with the
Urban population increase has had a massive impact same features (Hassan & Southworth, 2017). There
on the entire world to date, resulting in massive is no need for a training stage in UICM, but differ-
changes in LULC, which has become a key concern ent algorithms are used for clustering. The supervised
in natural resource management and sustainable image classification method is the most popular due

Vol.: (0123456789)
13
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
 428 GeoJournal (2023) 88:425–448

Fig. 1  Study area map of Jashore City, Bangladesh

to more accuracy of image classified data (Hassan Land Transformation Model (LTM) (Pijanowski et al.,
et al., 2016; Simwanda et al., 2021). The classifica- 2014), Conversion of Land Use and its Effects (CLUE)
tion results are influenced by several factors, includ- model (Verburg et al., 2006), Weights of Evidence
ing the classification goal, the spatial and spectral (WE) model (Amato et al., 2014), Logistic Regres-
characteristics of the data, the natural variability of sion Model (Arsanjani et al., 2013), Frequency Ratio
terrain conditions in the geographic area, and the dig- (FR) model (Park et al., 2011), Urban Growth Potential
ital classification technique used. The availability of (UGP) model (Kong et al., 2012), Fuzzy Logic model
high-quality remotely sensed imagery and ancillary (Grekousis et al., 2013), etc. The Artificial Neural Net-
data, the design of a proper classification procedure, work (ANN) based CA model is the first theoretical
and the analysts’ skills and experience are all impor- model to simulate potential urban growth and develop
tant factors in the performance of image classification and increase (i) spatiality, (ii) dynamics, (iii) simplic-
(Gazi, et al., 2020; Hu, et al., 2019). ity, (iv) visualization accuracy, (v) linked macro to
Several models have been introduced to predict micro approaches, and (vi) integrated GIS and remote-
LULC change in the GIS environment, such as Cellu- sensing techniques (Aburasa et al., 2016; Kafy et al.,
lar Automata (CA) model (Clarke & Hoppen, 1997), 2021b). The CA model has a high computational
Vol:. (1234567890)
13
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
GeoJournal (2023) 88:425–448 429

Table 1  Information about Satellite/sensor Path/row Spectral resolution Resolution Date Cloud coverage
the used Landsat images in
the study. Source: USGS, LANDSAT 5 138/44 Multispectral (6 Bands) 30 m 17/03/2000 < 10%
2020
LANDSAT 5 138/44 06/03/2010
LANDSAT 8 137/44 Multispectral (11 Bands) 30/03/2020

efficiency and spatial resolution since models of this cities of Bangladesh, likely Dhaka, Chattogram,
kind are capable of simulating with the assumption that Khulna and Rajshahi. Previous research in this con-
past urban growth will influence future urbanization text has focused on either evaluating the drivers of
patterns based on local land-use interactions (Santé unplanned growth or monitoring and mapping urban
et al., 2010). The main advantages of the CA model are growth. We have considered the study of monitor-
the open scalability of the model structure, its integra- ing and mapping urban growth, urban growth direc-
tion with other models, and its ability to simulate both tion analysis, analyzed the drivers of urban growth,
temporal and spatial patterns (Clarke & Hoppen, 1997; identified the future potential urban LULC scenario
He et al., 2008, 2018). In this study, the Multilayer Per- and future intra-urban LULC growth direction for
ception Neural Network Markov Chain (MPNNMC) 2030, 2040, and 2050. Our study will be useful in
model was used to simulate future potential urban future master plan projects for Jashore City.
land use patterns. The ANN consists of a neural net-
work approach to multilayer perceptron (MLP) that
decides automatically about network parameters and
their changes for an accurate future model (Kafy et al., Methodology
2021; Kafy et al., 2021; Kafy, Rakib, et al., 2021). Sub-
edi et al. (2013) and Simwanda et al. (2021) explained Study area
the MPNNMC model as the most effective model for
simulating future spatiotemporal changes with high- Jashore City, an increasing center of southwest-
precision LULC transformation. This model is also ern region of Bangladesh, under Khulna Division
used to predict future potential environmental and eco- (Fig. 1). The city located on the bank of Bhairob river
logical change (Kafy, Rakib, et al., 2021; Zhang et al., and a major industrialized and commercial center of
2019). Researchers have explained the LULC predic- Jashore District. Jashore district now covers a total
tion model as the most reliable and accurate for urban area of 2578.20 sq km, with Jashore City covering a
growth prediction in the previous literature as spatial, total area of 25.72 sq km, and the population was esti-
environmental, and ecological factors and GIS and mated to be around 237,478 which is half of the entire
remote-sensing techniques can be considered in this district’s population. The annual rainfall is 1537 mm
model (Aburasa et al., 2016; Al-Darwish et al., 2018; and the temperature varies from a minimum of 1­ 10C
Bihamta et al., 2015; He et al., 2018; Kafy et al., 2020). to a maximum of ­370C. Jessore is well-connected to
A few researches have conducted the intra-urban both Bangladesh and India via highways. Jessore is a
LULC simulations in Bangladesh including (Kafy station on the Western Bangla Railway’s broad-gauge
et al., 2020, 2021b; Hoque et al., 2020; Dey et al. network. The Bangladesh Air Force uses the Jes-
2021) and most them are based on the four major sore airport, which is located near Jessore city. This

Table 2  Description of LULC types Description

selected five LULC types in
this study Built-up/ Urban area Constructive lands, industrial, residential and commercial
areas, infrastructures, road network etc
Unused land Bare soils, playground, unused land, open space, landfill sites
Crop land Agricultural/ crop land, grasslands
Greeneries Trees, vegetation and forest areas
Water area Wetlands, river, canals, ponds and lakes

Vol.: (0123456789)
13
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
 430 GeoJournal (2023) 88:425–448

city’s road transportation system includes rickshaws, accuracies (Li et al. 2014). Before image classifica-
automobiles, buses, trucks, and other vehicles (BBS, tion, the preparation of satellite images is essential to
2013)(BBS, 2013). Over the past few decades, rise of avoid these errors and to create a closer link between
temperature and steepness in the northwestern part of the data obtained and biophysical characteristics on
the country have greatly reduced the crop production. the ground (Alawamy, 2020). Radiometric, atmos-
In the last three decades the district’s intercommuni- pheric and geometric corrections have been done to
cation system has been enormously developed. The fill the image gap and enhancement. In this study,
development project and steepness has been influenc- the generation of composite band combinations such
ing the massive LULC change in the study area. as natural color, true color, and false color compos-
ite, etc. is used to identify the land use type in the
Data collection study area (Kafy et al., 2020). During data process-
ing, bands such as red, green, blue, and near-infrared
In this study, remotely sensed Landsat data for a spe- bands were utilized for Landsat images to find true
cific period were used to analyze the LULC changes. color in ERDAS Imagine 2014. The LULC type was
Three Multi-spectral Landsat images were collected classified into five classes (Urban area, unused land,
from the United States Geological Survey (USGS) crop land, water areas, and greeneries) by apply-
for the year 2000, 2010, and 2020 to illustrate the ing the MLC method (Bishta, 2018). Table 2 rep-
LULC dynamics. These three satellite images col- resents the land use types and criteria which was
lected were taken in the same month (March) to considered for classifying the land cover types in
avoid the seasonal effects. Due to less possibility this study. Bands were used 4–3-2 (Color Infrared)
of rainfall in Jashore during March, the accuracy or and 3–2-1 (True Color) for Landsat 5 images, 4–3-2
acceptability of the LST result is higher than in other (True Color), and 5–4-3 (Color Infrared) for Landsat
months. The maximum cloud coverage was set to less 8 images. Then the land cover type change direction
than 10% during collecting the images. The detailed during the study period 2000 to 2020 has been ana-
information of the images is shown in Table 1. lyzed in ArcMap 10.6 version.
To check the accuracy of the classification, more
LULC classification acceptable and best quantitative image classification
accuracy measurement technics named Kappa statis-
LULC classification is one of the key remote sens- tics and confusing matrix were calculated (Janssen &
ing applications to identify the characteristics of Wel, 1994; Saputra & Lee, 2019; Yadav & Congal-
land use by using multispectral satellite imagery. ton, 2019). This study also calculated user accuracy,
Remotely sensed images are classified by supervised producer accuracy, and overall accuracy to check the
and unsupervised methods (Koko, 2020). In this accuracy of the LULC model. In this regard around
study, we have applied the maximum likelihood clas- 50 training sites have been randomly chosen for each
sifier (MLC) method, which is mostly used for image image to ensure that each LULC type is covered by
classification due to its availability, simplicity and all spectral groups. Equation 1–4 is utilized for a dif-
could produce LULC maps with high classification ferent type of accuracy assessment.
∑
Totalsamplenumber × Totalcorrectedsamplenumber − (col.tot × rowtot)
Kappacoefficient = × 100% (1)
Totalsamplenumber2 − col.tot × rowtot
∑

Num. of correctly classified pixels in each category

Useraccuracy = × 100% (2)
Total numb of reference pixels in each category (row total)

Numofcorrectlyclassifiedpixels(diagonal)
Produceraccuracy = × 100% (3)
Totalnumbofreferencepixelsineachcategory(column)

Vol:. (1234567890)
13
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
GeoJournal (2023) 88:425–448 431

Total num of corrected classified pixels (diagonal)

Overallaccuracy = × 100% (4)
Total num of reference pixels

UAj − UAi
The value of the overall accuracy was found Urban growth rate(% ) = × 100% (6)
93.33%, 93.33%, and 96.67% for the year 2000, UAi
2010, and 2020 respectively and the Kappa coef-
where UAi denotes the urban area in the initial year,
ficient values are 0.9143, 0.915, and 0.9577 which
UAj denotes the urban area in the final year.
indicates the good accuracy of the classified data
and suitable for urban expansion detection. Also,
Urban growth prediction using CA based MPNNMC
the results meet the recommended Kappa Statistics
Model
values recommended in the literature (Jansse & van
der Wel, 1994; Saputra & Lee, 2019; Yadav & Con-
The LULC change analysis, simulation and future
galton, 2019).
potential urban growth prediction have been con-
ducted by using Land Change Modeler, an integrated
LULC change analysis using LCM
software presented in IDRISI. The LCM model
includes a set of tools for analyzing and modeling
LULC change detection indicates the process of the
future LULC changes (Tewolde & Cabral, 2011). In
identification and determination of the nature and
this study, the LCM is used to implement the Neu-
extent of the LULC change over a period of time
ral Network Markov (NNM) model. The LCM is one
using remote sensing imagery (Meng et al., 2017). In
of the most widely used simulating systems and is
this study, the LULC map of Jashore was extracted for
widely used in studies related to modeling and simu-
2000, 2010 and 2020 to detect the changes of LULC
lating future potential urban expansion (Mishra &
over two periods of 2000–2010 and 2010–2020 and
Rai, 2016; Iizuka et al. 2017; Ranagalage et al., 2019;
also calculated the increase, decreases in each LULC
Gong et al., 2017). The NNM model requires four
types. The LCM is the most used models for LULC
steps, or phases, to follow for urban growth predic-
change detection and prediction that is accessible in
tion. To begin, (i) factors or model driver variables
the TerrSet software. The LCM is a software-based
are chosen, (ii) transition-potential modeling and
solution, developed to identify the growing challenges
simulation is performed, and (iii) model validation
of LULC and analytic requirements for the conserva-
and calibration of the model and actual pattern with
tion of biodiversity (Kumar et al., 2015). A review of
the Kappa Index is performed (Aburasa et al., 2016;
the recent literatures suggests that the LCM model of
Al-sharif & Pradhan, 2014).and (iv) the creation of
TerrSet that integrates the Markov Chain-based neu-
scenarios and simulations of potential future urban
ral network (MCNN), is very effective in predicting
expansion (Simwanda et al., 2021).
future LULC (Shahi et al., 2020). Based on the LULC
Researchers use LULC change models to detect
change analysis, the transition of one LULC type to
and predict future potential LULC scenarios. The
another was identified, along with gains of losses by
Land-use Change Modeler (LCM) is the most com-
LULC types. The spatial pattern of LULC change
monly used model to illustrate the LULC change
was determined through Eq. 5 (Mannan et al., 2019).
analysis (Shahi et al., 2020). The artificial neural
LULCi,j� × LULCAi 1 network (ANN) method, whose most popular model
LULCCi = × × 100% (5) is the multilayer perceptron algorithm, is a well-
LULCi,j j × j�
known method for analyzing satellite data (Fathol-
where LULCCi = LULC change, LULCi,j = LULC lahi Roudary et al. 2018). The Multilayer Perception
area at an earlier date, LULCi,j′ = LULC area at a later Neural Network Markov Chain (MPNNMC) Model
date, LULCAi = Areas not changed. can manage nonlinear functions, perform model-free
The urban growth rate of the study area has been function estimation, learn from previously unknown
calculated through Eq. 6 (Zhou et al., 2019). data relationships, and generalize unknown cir-
cumstances. As a result, ANNs are useful tools for

Vol.: (0123456789)
13
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
 432 GeoJournal (2023) 88:425–448

Fig. 2  GIS-based land cover distribution in Jashore in A 2000, B 2010, and C 2020

Table 3  Landcover area LULC types 2000 2010 2020 Change rate in %
(in acres) in Jashore City
during 2000–2020 Area % Area % Area % 2000–2010 2010–2020

CL 1143.04 18.54 662.15 10.74 391.41 6.35 -42.07 -40.89

UA 1829.23 29.67 2353.28 38.17 2555.17 41.44 28.65 8.58
UL 534.53 8.67 855.12 13.87 1055.14 17.11 59.98 23.39
CL crop land, UA urban GE 1636.88 26.55 1912.46 31.02 1592.01 25.81 16.84 -16.76
areas, UL unused land, GE WA 1021.58 16.57 382.25 6.2 571.53 9.27 -62.58 49.52
greeneries, WA water area, Total 6165.26 100% 6165.26 100% 6165.26 100%
UG urban growth

modeling intra-urban LULCs, especially when the LULC change depends on several factors. For the
underlying complex relationships in the data are illustration of each LULC transition, choosing the
unknown (Arsanjani et al., 2013). Markov models, on best set of explanatory variables is most important
the other hand, are commonly used as excellent pro- in LULC modelling to ensure the best fit between
jectors of potential LULCs. The Markov model is a the empirical data and the observable reality (Sim-
stochastic model that describes the likelihood of one wanda et al., 2021). As there are no fixed universal
state transforming into another. The Markov model variables, several researchers have shown that the
not only describes how to quantify conversion states selection of LULC modeling drivers varies for differ-
between land-use forms but can also show the tran- ent landscapes and for different study areas (Thappa
sition rate between them (Sang et al., 2011). As a and Murayama, 2012; Zhang et al., 2018; Iizuka
result, combining the ANN and Markov models is a et al. 2017; Simwanda et al., 2021). Although in most
reliable approach for modeling and simulating poten- of the studies, similar variables have been used for
tial intra-urban expansion (Simwanda et al., 2021). which the results differed (Basu et al., 2021). There-
fore, the driver variables are selected on the basis of

Vol:. (1234567890)
13
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
GeoJournal (2023) 88:425–448 433

the characteristics of the study area, observed LULC longitudes were considered to simulate future urban
changes, and based on expert knowledge about the growth. For the prediction of the future LULC pat-
study areas. In this study, factors such as slope, DEM, tern, the LULC maps of the year 2000, 2010, and
LULC patterns of the observed period, latitudes, and 2020 were used as independent variables based on

Fig. 3  LULC Change

analysis during 2000–2020

Fig. 4  Spatiotemporal urban areas’ transition pattern during: A 2000–2010, B 2010–2020

Vol.: (0123456789)
13
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
 434 GeoJournal (2023) 88:425–448

Fig. 5  Spatial illustration of LULC conversion in Jessore during 2000–2020

the recent studies done in different cities in Bangla- the most commonly used simulating model to simu-
desh (Ahmed, 2012; Kafy et al., 2020; Hasan et al. late potential LULC transition (Ahmed, 2012). The
2017). The slope and elevation data used in this study MPNNMC includes 3 layers, one input layer, one
were derived from Google Earth. The urban distur- or several hidden layers, and one output layer. The
bance, the distance of locations from the road were input layer represents input data to the network dur-
calculated from the GIS data, collected from Khulna ing training of the model. The data are received by
City Corporation’s authorities. The data used for the the hidden layers, which extract and express informa-
other factors in this study was derived from land-use tion as weights from the input layers. The output layer
classification. results depend on the back-and-forward interactions
We have used Neural Network Markov Chain between the three layers. In order to create the neu-
(MPNNMC) model to estimate future potential urban ral network–Markov modeling framework, we com-
growth of Jashore City. The MPNNMC is a form of bined the MPNNMC and Markov chain models pro-
ANN that uses back-propagation algorithm and is vided by LCM (Zhang et al., 2018). In the LCM, the

Vol:. (1234567890)
13
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
GeoJournal (2023) 88:425–448 435

Table 4  LULC conversion LULC types CL UA UL GE WA Overall

matrix of Jessore during
2000–2020 2000–2010 (Area in percentage)
CL 4.05% 5.04% 4.96% 5.87% 0.24% 20.16%
UA 2.48% 31.06% 1.68% 1.55% 1.05% 37.81%
UL 1.08% 1.26% 4.73% 1.52% 0.08% 8.67%
GE 0.81% 1.42% 0.59% 15.06% 0.09% 17.97%
WA 0.62% 6.86% 0.26% 3.33% 4.32% 15.39%
Total 9.05% 45.64% 12.21% 27.32% 5.77% 100.00%
2010–2020
CL 1.41% 3.03% 2.45% 1.74% 0.41% 9.05%
UA 1.03% 32.32% 4.99% 5.43% 1.86% 45.64%
UL 2.18% 2.33% 4.88% 2.69% 0.12% 12.21%
GE 1.43% 2.66% 3.73% 15.36% 4.13% 27.32%
WA 0.28% 1.10% 0.67% 0.59% 3.13% 5.77%
Total 6.34% 41.45% 16.72% 25.82% 9.66% 100.00%
2000–2020
CL 2.52% 4.17% 6.66% 6.43% 0.37% 20.16%
UA 0.74% 31.48% 3.21% 1.66% 0.73% 37.81%
UL 1.64% 1.14% 3.93% 1.89% 0.07% 8.67%
GE 0.94% 0.71% 1.51% 12.35% 2.47% 17.97%
WA 0.50% 3.95% 1.41% 3.50% 6.03% 15.39%
Total 6.34% 41.45% 16.72% 25.82% 9.66% 100.00%

multivariable function MPNNMC includes the driver best-fitted prediction model (Shamshirband et al.,
variables to predict the transition potential of LUCs 2020; Shatnawi & Qdais, 2019).
in any location between two points. The intra-urban �
LULC changes of Jashore during 2000–2020 have
∑� �2
Tactual − Tsimulated (6)
been used in the MPNNMC model to simulate future RMSE =
n
potential urban growth.
To check the accuracy of the simulated model, ∑� � � �
Tactual − T � actual × Tsimulated − T � simulated
percentage error, accuracy percentage were esti- R= �
mated and also calculated correlation coefficient (R) � �2 �∑ �
Tactual − T � actual × Tsimulated − T � simulated
�2
and RMSE by Eqs. 8 and 7, respectively. For model (7)
validation, a statistical similarity was performed
by comparing the predicted and observed results of R and RMSE are the most commonly used metrics in
LULC for the year 2020. To check the accuracy of geosciences to evaluate model performance (Mansour
the prediction model, the spatiotemporal matched et al., 2020; Saputra & Lee, 2019). The RMSE value
LULC classes and non-matched LULC classes were close to zero indicates the best performance of the
also identified by comparing the raster data set of the model, while the + 1 value of R represents the perfect
actual and predicted LULC maps for the year 2020. positive correlation between the observed and simu-
The higher the value of accuracy and lower the value lated variables.
of percentage error, the RMSE value indicates the

Vol.: (0123456789)
13
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
 436 GeoJournal (2023) 88:425–448

Fig. 6  Spatiotemporal LULC urban growth direction analysis during: A 2000–2010, B 2010–2020, C 2000–2020

Results and discussion extracted the LULC maps for 2000, 2010, and 2020
by using ERDAS Imagine and ArcGIS software. Fig-
Spatial distribution of LULC types—2000 to 2020 ure 2 shows strong evidence of changing landscape
patterns over the last two decades in Jashore City
The key causes of LULCC are rapid population for each LUCC type. The land cover areas covered
growth in cities, migration from rural to urban areas, by different LULC types such as water areas, urban
conversion of suburban areas to urban areas, a lack areas, greeneries, cropland and unused land cover are
of evaluation of ecological services, ignorance, bio- expressed in acres. Table 3 represents the statistics of
physical limitations, and the use of ecologically mis- different LULC cover areas in the study area during
matched technologies (Hyandye et al., 2015). We 2000–2020.

Vol:. (1234567890)
13
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
GeoJournal (2023) 88:425–448 437

Fig. 7  Simulated future LULC pattern of Jashore city by A 2030, B 2040, and C 2050
In accordance with the LULC analysis in Table 3 LULC type in 2000. About 29.67% (1829.23 acres)

Table 5  Validation of the LULC types LULC area in 2020 RMSE R2 Accuracy (%)
urban growth prediction (acres)
model
Actual Predicted Difference

CL 391.41 388.22 3.19 0.01 0.92 96

UA 2555.17 2569.61 − 14.44 0.03 0.86 86
UL 1055.14 1072.44 − 17.30 0.03 0.83 84
GE 1592.01 1583.94 8.07 0.02 0.89 90
WA 571.53 551.04 20.49 0.04 0.82 82
Total 6165.26 6165.26 – – – 87

shows that urban areas were the most prominent of total areas were extracted under urban areas.

Table 6  Simulated future LULC types Area in percentage Change in % UG rate in %

LULC statistics of Jashore
during 2030–2050 2030 2040 2050 2020–2050 2000–2050 2020–2050

CL 5.64% 5.36% 5.19% − 1.16% − 14.79% − 20.57%

UA 59.84% 62.99% 65.08% + 23.64% 26.52% 39.51%
UL 8.80% 8.02% 7.56% − 9.55% − 0.79% − 108.52%
GE 23.19% 21.59% 20.34% − 5.47% + 2.31% − 23.59%
WA 2.54% 2.04% 1.84% − 7.43% − 13.22% − 292.52%
Total 100% 100% 100% - - -

Vol.: (0123456789)
13
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
 438 GeoJournal (2023) 88:425–448

Sequentially about 26.55% (1636.88 acres), 10.74% greeneries, and unused areas; and secondly, the dec-
(1143.04 acres), 16.57% (1021.58 acres) and 8.67% lination of cropland and water areas at a tremendous
(534.53 acres) of the total areas were for greener- rate over the study period (Fig. 3). Figure 4 represents
ies, cropland, water, and unused land in the year the spatiotemporal urban growth transition, and Fig. 5
2000. A moderate change in different LULC has represents the spatiotemporal gain loss maps of all
been observed during the years 2000–2010. The LULC classes over the study period. The Markov-CA
percentage of areas for LULC types such as crop- LULC transition matrix is presented in Table 4.
land and water were reduced. In 2020, croplands The result shows that the percentage of the urban
were observed at 6.35% (391.41 acres), urban areas area increased by 11.77% (from 1829.23 to 2555.17
were 41.44% (2555.17 acres), vegetative areas were acres) and unused land by 8.44% (from 534.53 to
25.81% (1592.01 acres), water areas were 9.27% 1055.14 acres) during 2000–2020 (Fig. 3). Corre-
(571.53 acres), and unused lands were 17.11% spondingly, the cropland area decreased by 12.19%
(1055.14 acres) of the total study area. The highest (from 1143.04 to 391.41 acres), the greeneries areas
declination rate was found (32.9% per decade) for by 0.74% (from 1636.88 to 1592.01 acres), and the
crop land area over the past two decades, while in water areas by 7.30% (from 1021.58 to 571.53 acres).
the case of water areas, this rate was 22% per dec- Figure 4 shows that during 2000–2010, about 8.50%
ade. The area of urban areas is increasing by an and during 2010–2020, about 3.27% of the cropland
average of 48.7% per decade as compared to the was converted to urban areas. About 10.37% and
past decade. This indicates the huge urban growth 3.07% of water areas were converted to urban areas
trends in the study region for the past two decades. during 2000–2010 and 2010–2020, respectively. The
urban growth rate was 28.65% between the years
Urban growth analysis—2000 to 2020 2000 and 2010 and 8.58% from 2010 to 2020.
Table 4 shows that the LULC transition in Jashore
Two changing trends in LULC classes were iden- City has undergone a significant change since 2000.
tified (Fig. 2); firstly, the increase of urban areas, During 2000–2020, about 15% of water areas have

Fig. 8  The transition map showing the spatiotemporal urban growth during: A 2020–2030, B 2030–2040, C 2040–2050

Vol:. (1234567890)
13
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
GeoJournal (2023) 88:425–448 439

Table 7  Predicted LULC LULC types CL UA UL GE WA Overall

transition probability matrix
for 2020–2050 at Jashore 2020–2030
CL 1.24% 1.92% 1.66% 1.38% 0.13% 5.64%
UA 0.70% 38.11% 1.33% 1.32% 0.06% 59.83%
UL 1.88% 8.38% 3.50% 2.73% 0.25% 8.80%
GE 1.50% 8.04% 2.18% 13.68% 0.31% 23.19%
WA 0.32% 3.45% 0.13% 4.00% 1.78% 2.54%
Total 5.64% 59.91% 8.80% 23.11% 2.54% 100.00%
2030–2040
CL 5.14% 0.22% 0.00% 0.00% 0.00% 5.36%
UA 0.04% 62.17% 0.05% 0.09% 0.13% 62.48%
UL 0.01% 0.49% 7.49% 0.03% 0.00% 8.02%
GE 0.01% 1.34% 0.01% 20.22% 0.00% 21.59%
WA 0.00% 0.04% 0.00% 0.00% 2.51% 2.55%
Total 5.19% 64.26% 7.56% 20.34% 2.65% 100.00%
2040–2050
CL 1.19% 2.31% 0.00% 0.00% 0.00% 5.36%
UA 0.56% 38.96% 0.05% 0.09% 0.13% 62.48%
UL 1.70% 9.85% 7.49% 0.03% 0.00% 8.02%
GE 1.44% 9.64% 0.01% 20.22% 0.00% 21.59%
WA 0.31% 3.60% 0.00% 0.00% 2.51% 2.55%
Total 5.19% 64.35% 7.56% 20.34% 2.65% 100.00%
2020–2050
CL 1.19% 2.31% 1.44% 1.26% 0.14% 6.34%
UA 0.56% 38.96% 1.18% 0.77% 0.07% 41.54%
UL 1.70% 9.85% 2.89% 2.03% 0.27% 16.74%
GE 1.44% 9.64% 1.92% 12.39% 0.33% 25.71%
WA 0.31% 3.60% 0.12% 3.80% 1.84% 9.67%
Total 5.19% 64.35% 7.55% 20.26% 2.65% 100.00%

been transformed, of which 3.95% in urban areas, of the cropland pattern in Jashore, which resulted in
1.41% in unused land, and 3.50% in greeneries. the declination of crop area from 0.45 ha/household
About 6.66% of croplands were converted to unused in 1981–1990 to 0.29 ha/household in 2001–2010.
land, 4.17% to urban areas, and 6.43% to greeneries. This output is consistent with the cropland declina-
While the transformation of greeneries in urban areas tion pattern of this study. The cropland of Jashore
was found to be less, only 0.71% of greeneries was City declined over the past two decades due to flood-
transformed into urban areas during the study period. ing and also for use of lands for commercial and
The LULC transition matrix shows that the croplands housing purposes.
and water areas are experiencing the highest declina- Weather data analysis shows that the annual rain-
tion rates due to urban expansion in Jashore City. fall in Jashore City has decreased significantly over
Several factors are responsible for this increas- the last few decades and the average temperature has
ing trend of urban areas, greeneries and unused land risen (Mondal et al., 2017). Due to inadequate rain-
cover. Rapid urbanization and the large influx of fall, many reservoirs have dried up and converted to
migrants seeking to better educate themselves and unused or vacant land. This adverse climate change
city services account for the construction of more has also affected farming activities and croplands
infrastructures. This increases the amount of urban have become urban areas, unused land and greeneries.
areas. Raufe (2011) found a significant transformation About 77% of ground water has been used in cropping

Vol.: (0123456789)
13
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
 440 GeoJournal (2023) 88:425–448

Fig. 9  Predicted spatial locations of different LULC gain–loss during 2020–2050

in Bangladesh. Due to less rainfall, the ground water This study also performed the spatiotemporal
level of this Jashore has fallen drastically (7.95 m urban growth direction analysis during 2000–2010
from the surface), for which residents face difficul- and 2010–2020 and presented in Fig. 6. The analy-
ties in collecting water for cropping. This resulted in sis shows that urban areas expanded mostly in the
the declination of croplands (DPHE, Jessore, 2012) eastern direction and then in the NW-NNW direc-
and the conversion of croplands to unused lands at tion. The growth rate in the eastern and NNE-NE
Jashore (Mondal & Hossain, 2009). This study also directions was observed 1.32 sq.km and 1.28 sq.km
found that a large percentage (6.66%) of cropland respectively during 2000–2020. While a negative
has been transformed on unused lands during the growth rate has been observed in the WSW-W and
2000–2020 period. Besides many infrastructural and SW-WSW directions. In the NW-NNW direction,
economic activities are influencing this LULC change the urban growth rate varied between 0.82 sq.km
and these are causing an adverse impact on the envi- and 1.05 sq.km over the study period. This indi-
ronment and ecology of the study area. cates that urbanization is not taking place in any

Vol:. (1234567890)
13
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
GeoJournal (2023) 88:425–448 441

A B C
CL CL
0.6 CL 1
0.4
0.4 0.3 0.5
WA 0.2 UA
0.2 WA UA
WA UA
0.1 0
0 0

GE UL
GE UL
GE UL
2020 - 2030 2030 - 2040
2040 - 2050

Fig. 10  Predicted LULC change direction during 2020–2050: A water area, B cropland, C greeneries

structured manner and urban areas are increasing in land (− 0.79%) will decrease slightly in the study area
an unplanned and uncontrolled way. (Table 6).
Table 6 shows that the urban growth rate will
Simulated urban LULC pattern—2030 to 2050 observe 39.51% during the next three decades in
the study area while the water area will possess the
During 2000–2020, the LULC classes underwent highest percentage of declination. Potential expected
a remarkable transformation, which is presented in urban growth may have an adverse impact on the
Figs. 3, 4, 5 and Table 4. Therefore, a simulation to environment, ecosystem, atmosphere, and biodiver-
illustrate future LULC dynamics is essential. The aim sity of the study region. The increase of urban areas
is to forecast urban growth trends, which can pro- moves the additional land to high-temperature areas,
vide the foundation of long-term sustainable urban carbon emitted zones and if this continues, the urban
planning. Based on the LULC change from 2000 to area in the coming years will face significant environ-
2020, the MLP-MC model is used to simulate future mental degradation. Proper implementation of land
potential urban growth patterns of Jashore City. The use planning, natural resource management, conser-
simulated urban LULC pattern of the study area is vation of water areas, and increase of urban green
illustrated in Fig. 7 and LULC statistics in Table 6. To area would make Jashore a sustainable city in the
check the validity of the prediction model, the LULC future.
pattern of the year 2020 was simulated and compared
with the actual LULC pattern of 2020. Table 5 shows Urban growth direction simulation—2030 to 2050
that, with a large number of iterations, all the poten-
tial LULC types showed an overall accuracy of 87% The CA Markov transition matrix was determined to
with an ­R2 value of more than 0.82 and all the land show the future spatial urban area expansion or tran-
cover types with more than 82%. sition from 2030 to 2050 over the study region and is
From the MPNNMC based CA model simula- presented in Fig. 8 and Table 7. The spatial locations
tion, a significant increase was estimated in the urban of gains and losses of different LULC classes during
areas by 18.40%, 21.55%, and 23.64% for the year 2020–2050 are also prepared and presented in Fig. 9.
2030, 2040, and 2050 compared to 2020. Consider- Usually, urban growth took place around the
ing the forecasted years, in 50 years (2000–2050) the city centers. The predicted LULC transition matrix
percentage of the urban area (26.52%) and green- shows that LULC transformation will take place
eries (2.31%) will increase enormously. On the in about 74.79% of total areas during 2020–2030,
contrary, the water area (− 13.22%) and cropland 62.72% areas during 2030–2040 and 64.51% areas
(− 14.49%) area will decrease mostly while unused during 2040–2050. During 2020–2030 10.47%
of unused land and 10.04% of greeneries will be

Vol.: (0123456789)
13
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
 442 GeoJournal (2023) 88:425–448

Fig. 11  Predicted urban area expansion direction during: A 2020–2030, B 2030–2040, and C 2040–2050 in the study region

transformed into urban areas, while it will be 0.50% greeneries will be transformed into urban areas,
and 1.34% respectively during 2040–2050. Figure 8 2.18% to unused lands. About 1.66% of croplands
shows that over time, other LULC conversions into will be converted to unused land. By 2050, the con-
urban areas will reduce gradually because due to version to urban land cover will be 3.60% for water
rapid urbanization, the share of these LULC class areas, 2.31% for croplands, and 9.85% for unused
areas will reduce dramatically. This will transform lands. The extent of urban expansion relative to
the entire study area into urban areas in the future other LUCC varied significantly by period, accord-
and the percentage of eco-friendly LULC will be ing to the previous LUCC trend analysis.
zero and the city will become uninhabitable. The simulated LULC class change rate in Fig. 10
Table 7 shows the predicted LULC transforma- shows that the water area and cropland area will
tion matrix for the years 2020 to 2050. Analysis transform mostly in urban areas rather than greener-
shows that during 2020 – 2030, about 8.04% of ies. Croplands will transform into urban areas and

Vol:. (1234567890)
13
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
GeoJournal (2023) 88:425–448 443

greeneries; the declination rate was found to be 0.38 in effective and sustainable urban planning of infra-
and 0.4 respectively. Due to climate change, such as structure, the environment, and economy in urban
the increase in temperature and drought tendency, and areas. This study adopted Landsat satellite images
the reduction of total rainfall, the percentage of wet- from Landsat 5TM and Landsat 8 OLI to document
lands and cropland will continue to decrease. the historical urban growth pattern of a developing
The simulated intra-urban land expansion showed city, Jashore, and predicted the urban LULC pat-
that if the urban areas continued to expand without tern using the MPNNMC model for the year 2030
proper regulatory measures, the increase in urban to 2050. For the MPNNMC, seven variables such as
areas would be found in NW-NNW direction, with slope, DEM, LULC pattern, latitudes and longitudes,
a growth rate of 1.8 sq.km. The growth rate in the road network, and distance of locations from the road
eastern direction will be the same 1.4 sq.km as the are considered to simulate future potential intra-urban
growth rate observed during 2000–2020. The growth land use change patterns. The LULC change pat-
rate in the south and SSE-S direction will be within tern showed the increase of urban areas by 11.77%
the range of 1 to 1.3 sq.km during 2020–2050. As and unused land by 8.44%, due to the declination of
long as the percentage of cropland is moderately high cropland by 12.19% and water areas by 7.30% dur-
in the ND-NNW direction (Fig. 11), urbanization will ing 2000–2020. The increasing pressure of unplanned
be more in that direction. The density of urban areas urbanization will be reflected on eco-friendly lands,
will be highest in the north and eastern directions causing the declination of croplands, greeneries, and
of the city. This growth direction analysis indicates waterbodies that surround the city. These non-urban
that future urban growth will be in an unplanned and areas have started to disappear gradually, and the
uncontrolled manner like the period from 2000–2020. cropland is expected to decline from 6.35% in 2020
This will be an obstacle to sustainable environmental to 5.19% in 2050 and other LULC classes, such as
and economic development in the future. The overall unused land from 17.11% to 7.56%, greeneries from
analysis suggest that Jashore City had experienced 25.81% to 20.34%, and water areas from 9.27% to
and will experience rapid intra-urban land expansion. 1.84% in 2050. Urban areas are expected to increase
Our results are consistent with the findings of several from 41.44% in 2020 to 65.08% in 2050. The pre-
researches in the context of land use prediction in dicted urban growth reveals an alarmingly rapid
the other major cities of Bangladesh, such as Khulna transformation of cropland and water areas into urban
City (Fattah et al., 2021a), Rajshahi City (Kafy et al., areas. The study demonstrates that the natural land-
2020), Cumilla City (Kafy et al., 2021a), Dhaka City scape of Jashore will be insufficient to achieve urban
(Islam and Ahmed, 2021). The increase in popula- sustainability, as the maximum portion of greeneries
tion and housing demand, along with other service and water-areas will be depleted in the future.
facilities such as commercial and industrial activi- The urban growth simulation model demonstrates
ties and road infrastructure, are the main drivers of that the city of Jashore will go through enormous,
rapid urban expansion in the urban and rural areas of spontaneous, and unplanned urban expansion in the
Bangladesh (Hasan et al. 2017). If this expansion rate future. This will affect the climate, environment, and
and population growth rate cannot be accommodated, ecosystem of the city. So, the city urgently needs the
urban sprawl might occur. Moreover, the declination implementation of an eco-friendly and sustainable
of vegetative land cover will substantially change the urban development plan with the preservation and
climatic factors (Morshed & Fattah, 2021) and these management of existing natural resources and non-
will make the cities uninhabitable. urban areas to ensure a sustainable urban environ-
ment. This research provides useful perspectives for
policymakers, responsible authors, environmental
Conclusion engineers, and urban planners in sustainable decision-
making considering future aspects.
The developing cities around the world are going
through enormous urban expansion, and this is affect- Funding There was no funding for this research.
ing the environment as well as the ecosystem. Simu-
Availability of data and material Data will be available.
lating future potential urban growth patterns helps

Vol.: (0123456789)
13
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
 444 GeoJournal (2023) 88:425–448

Declarations and cellular automata models to simulate urban expan-

sion. International Journal of Applied Earth Observation
Conflict of interest The authors declare that they have no and Geoinformation, 21, 265–275. https://​doi.​org/​10.​
known competing financial interests or personal relationships 1016/j.​jag.​2011.​12.​014
that could have appeared to influence the work reported in this Basu, T., Das, A., Pham, Q. B., et al. (2021). Development
paper. of an integrated peri-urban wetland degradation assess-
ment approach for the Chatra Wetland in eastern India.
Science and Reports, 11, 4470. https://​doi.​org/​10.​1038/​
s41598-​021-​83512-6
References BBS. (2013). Dsitrict Statistics 2011: Jessore. Ministry of
Planning, Government of the People’s Republic of Bang-
Aburasa, M. M., Ho, Y., Ramli, M. F., & Ash’aari, Z. H. ladesh, Statistics and Informatics Division. Dhaka: Bang-
(2016). The simulation and prediction of spatio-temporal ladesh Bureau of Statistics (BBS).
urban growthtrends using cellular automata models: A Bihamta, N., Soffianian, A., Fakheran, S., & Gholamalifard,
review. International Journal of Applied Earth Observa- M. (2015). Using the SLEUTH urban growth model
tion and Geoinformation, 52, 380–389. https://​doi.​org/​ to simulate future urban expansion of the Isfahan met-
10.​1016/j.​jag.​2016.​07.​007 ropolitan area, Iran. Journal of the Indian Society of
Ahmed, S., & Bramley, G. (2015). How will Dhaka grow spa- Remote Sensing, 43, 407–414. https://​doi.​org/​10.​1007/​
tially in future? -Modelling its urban growth with a near- s12524-​014-​0402-8
future planning scenario perspective. International Jour- Boadi, K., Kuitunen, M., Raheem, K., & Hanninen, K. (2005).
nal of Sustainable Built Environment, 4(2), 359–377. Urbanisation without development: Environmental and
https://​doi.​org/​10.​1016/j.​ijsbe.​2015.​07.​003 health implications in African Cities. Environment,
Ahrend, R., Farchy, E., Kaplanis, I., & Lembcke, A. (2014). Development and Sustainability, 7, 465–500. https://​doi.​
WHAT MAKES CITIES MORE PRODUCTIVE? org/​10.​1007/​s10668-​004-​5410-3
EVIDENCE FROM FIVE OECD COUNTRIES ON Chen, M., Zhang, H., Liu, W., & Zhang, W. (2014a). The
THE ROLE OF URBAN GOVERNANCE. Journal of global pattern of urbanization and economic growth:
Regional Science, 57(3), 1–33. https://​doi.​org/​10.​1111/​ Evidence from the last three decades. PLoS ONE, 9(8),
jors.​12334 e103799. https://​doi.​org/​10.​1371/​journ​al.​pone.​01037​99
Aithal, B., Vinay, S., & Ramachandra, T. (2018). Simulat- Chen, S., Chen, B., & Fath, B. (2014b). Urban ecosystem
ing urban growth by two state modelling and connected modeling and global change: Potential for rational urban
network. Modeling Earth Systems and Environment, 4, management and emissions mitigation. Environmen-
1297–1308. https://​doi.​org/​10.​1007/​s40808-​018-​0506-1 tal Pollution, 190, 139–149. https://​doi.​org/​10.​1016/j.​
Al-Darwish, Y., Ayad, H., Taha, D., & Saadallah, D. (2018). envpol.​2014.​03.​032
Predicting the future urban growth and it’s impacts on the Clarke, K., & Hoppen, S. (1997). A self-modifying cellular
surrounding environment using urban simulation models: automaton model of historical urbanization in the San
Case study of Ibb city – Yemen. Alexandria Engineering Francisco Bay area. Environment and Planning B, 24,
Journal, 57(4), 2887–2895. https://​doi.​org/​10.​1016/j.​aej.​ 247–261.
2017.​10.​009 Dewan, A. M., & Yamaguchi, Y. (2009). Land use and land
Al-shalabi, M., Pradhan, B., Billa, L., Mansor, S., & Althu- cover change in Greater Dhaka, Bangladesh: Using
waynee, O. (2013). Manifestation of Remote Sensing remote sensing to promote sustainable urbanization.
Data in Modeling Urban Sprawl Using the SLEUTH Applied Geography, 29, 390–401. https://​doi.​org/​10.​
Model and Brute Force Calibration: A Case Study of 1016/j.​apgeog.​2008.​12.​005
Sana’a City, Yemen. Journal of the Indian Society of Dey, NN., Rakib, A.A.; Kafy, A.A.; Raikwar, V. Geospatial
Remote Sensing, 41, 405–416. https://​doi.​org/​10.​1007/​ modelling of changes in land use/land cover dynamics
s12524-​012-​0215-6 using Multi-layer perception Markov chain model in
Al-sharif, A., & Pradhan, B. (2014). Monitoring and predict- Rajshahi City, Bangladesh. Environmental Challenges.
ing land use change inTripoli Metropolitan City using an 4(100148). https://​doi.​org/​10.​1016/j.​envc.​2021.​100148
integrated Markov chain and cellularautomata models DPHE, (Department of Public Health Engineering), Jessore,
in GIS. Arabian Journal of Geosciences, 7, 4291–4301. Ground Water Level Status of Jessore, 1981–2010, Per-
https://​doi.​org/​10.​1007/​s12517-​013-​1119-7 sonal Communication, 2012
Amato, F., Pontrandolfi, P., and Murgante, B. (2014) Using Faisal, A. A., Haque, S., & Rahman, M. M. (2021). Retriev-
Spatiotemporal Analysis in Urban Sprawl Assessment ing spatial variation of aerosol level over urban mixed
and Prediction. Computational Science and Its Applica- land surfaces using Landsat imageries: Degree of air
tions – ICCSA (pp. 758–773). Springer, Cham. https://​ pollution in Dhaka Metropolitan Area. Physics and
doi.​org/​10.​1007/​978-3-​319-​09129-7_​55 Chemistry of the Earth, Parts a/b/c. https://​doi.​org/​10.​
Araya, Y. H., & Cabral, P. (2010). Analysis and Modeling of 1016/j.​pce.​2021.​103074
Urban Land Cover Change in Setúbal and Sesimbra. Por- Fattah, M., Morshed, S. R., & Morshed, S. Y. (2021a).
tugal. Remote Sensing, 2(6), 1549–1563. https://​doi.​org/​ Multi-layer perceptron-Markov chain-based artifi-
10.​3390/​rs206​1549 cial neural network for modelling future land-specific
Arsanjani, J. J., Helbich, M., Kainz, W., & Boloorani, A. D. carbon emission pattern and its influences on surface
(2013). Integration of logistic regression, Markov chain

Vol:. (1234567890)
13
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
GeoJournal (2023) 88:425–448 445

temperature. SN Applied Science, 3, 359. https://​doi.​ study of Islamabad Pakistan. Springer plus, 5, 1–11.
org/​10.​1007/​s42452-​021-​04351-8 https://​doi.​org/​10.​1186/​s40064-​016-​2414-z
Fattah, M., Morshed, S. R., & Morshed, S. Y. (2021b). He, C., Okada, N., Zhang, Q., Shi, P., & Li, J. (2008). Mod-
Impacts of land use-based carbon emission pattern elling dynamic urban expansion processes incorporat-
on surface temperature dynamics: Experience from ing a potential model with cellular automata. Landscape
the urban and suburban areas of Khulna, Bangladesh. and Urban Planning, 86(12), 79–91. https://​doi.​org/​10.​
Remote Sensing Applications: Society and Environ- 1016/j.​landu​rbplan.​2007.​12.​010
ment., 22, 100508. https://​doi.​org/​10.​1016/j.​rsase.​2021.​ He, Q., He, W., Song, Y., Wu, J., Yin, C., & Mou, Y. (2018).
100508 The impact of urban growth patterns on urban vitality in
Feng, Y., Cai, Z., Tong, X., Wang, J., Gao, C., Chen, S., & Lei, newly built-up areas based on an association rules analy-
Z. (2018). Urban Growth Modeling and Future Scenario sis using geographical ‘big data.’ Land Use Policy, 78,
Projection Using Cellular Automata (CA) Models and 726–738. https://​doi.​org/​10.​1016/j.​landu​sepol.​2018.​07.​
the R Package Optimx. ISPRS International Journal of 020
Geo-Information, 7(10), 387. https://​doi.​org/​10.​3390/​ Herold, M., Goldstein, N., & Clarke, K. (2003). The spatiotem-
ijgi7​100387 poral form of urban growth: Measurement, analysis and
Gazi, M., Rahman, M., Uddin, M., & Rahman, F. (2020). modeling. Remote Sensing of Environment, 86(3), 286–
Spatio-temporal dynamic land cover changes and their 302. https://​doi.​org/​10.​1016/​S0034-​4257(03)​00075-0
impacts on the urban thermal environment in the Chit- Hoque, M. Z., Cui, S., Islam, I., Xu, L., & Tang, J. (2020).
tagong metropolitan area, Bangladesh. Geojournal. Future impact of land use/land cover changes on ecosys-
https://​doi.​org/​10.​1007/​s10708-​020-​10178-4 tem services in the lower Meghna River Estuary, Band-
General Economics Division (GED). (2020). Making Vision ladesh. Sustainability, 12, 2112. https://​doi.​org/​10.​3390/​
2041 a Reality: Perspective Plaon of Bangladesh 2021– su120​52112
2041. Ministry of Planning, Government of the People’s Hossain, M. (2015). Declining productivity of agricultural land
Republic of Bangladesh. in Bangladesh. J Agroecol Nat Resour Manag, 2, 25–30.
Gómez, J., Patiño, J., Duque, J., & Passos, S. (2020). Spati- Hossain, M., Lin, C. K., & Hussain, M. Z. (2001). Goodbye
otemporal modeling of urban growth using machine Chakaria Sunderban: The oldest mangrove forest. The
learning. Remote Sensing, 12(1), 109. https://​doi.​org/​10.​ Society of Wetland Scientists Bulletin, 18, 19–22. https://​
3390/​rs120​10109 doi.​org/​10.​1672/​0732-​9393(2001)​018[0019:​GCSTOM]​
Gong, H., Simwanda, M., & Murayama, Y. (2017). An Inter- 2.0.​CO;2
net-Based GIS Platform Providing Data for Visualization Hu, Y., Batunacun, Zhen, L., & Zhuang, D. (2019). Assess-
and Spatial Analysis of Urbanization in Major Asian and ment of Land-Use and Land-Cover Change in Guangxi,
African Cities. ISPRS Int. J. Geo-Inf., 6, 257. China. Scientific Reports, 9, 2189. https://​doi.​org/​10.​
Grant, U. (2010) Spatial inequality and urban poverty traps. 1038/​s41598-​019-​38487-w
London, UK: Overseas Development Institute. Retrieved Hyandye, C., Mandara, C. G., & Safari, J. (2015). GIS and
from https://​www.​odi.​org/​publi​catio​ns/​4526-​spati​al-​ logit regression model applications in land use/land
inequ​ality-​andur​ban-​pover​ty-​traps cover change and distribution in Usangu Catchment.
Grekousis, G., Manetos, P., & Photis, Y. (2013). Modeling American Journal of Remote Sensing, 3(1), 6–16. https://​
urban evolution using neural networks, fuzzy logic and doi.​org/​10.​11648/j.​ajrs.​20150​301.​12
GIS: The case of the Athens metropolitan area. Cities, Islam, M. R., Miah, M. G., & Inoue, Y. (2016). Analysis of
30, 193–203. https://​doi.​org/​10.​1016/j.​cities.​2012.​03.​ land use and land cover changes in the coastal area of
006 Bangladesh using Landsat imagery. Land Degradation
Han, H., Yang, C., & Song, J. (2015). Scenario simulation and and Development, 27, 899–909. https://​doi.​org/​10.​1002/​
the prediction of land use and land cover change in Bei- ldr.​2339
jing, China. Sustainability, 7(4), 4260–4279. https://​doi.​ Islam, M. S., & Ahmed, R. (2012). Land use change prediction
org/​10.​3390/​su704​4260 in Dhaka City using Gis aided Markov chain modeling.
Hasan, M., Hossain, M., Bari, M., Islam, M. (2013). Agri- Journal of Life and Earth Science, 6, 81–89. https://​doi.​
cultural Land Availability in Bangladesh; SRDI, Min- org/​10.​3329/​jles.​v6i0.​9726
istry of Agriculture: Dhaka, Bangladesh (p. 42). ISBN Islam, G. M., Islam, A. K., Shopan, A. A., Rahman, M. M.,
978-984-33-6141-7. Lazar, A. N., & Mukhopadhyay, A. (2015). Implications
Hassan, M., & Nazem, M. N. (2016). Examination of land use/ of agricultural land use change to ecosystem services in
land cover changes, urban growth dynamics, and envi- the Ganges delta. J. Environ. Manag., 161, 443–452.
ronmental sustainability in Chittagong city, Bangladesh. Jansse, L., & van der Wel, F. (1994). Accuracy assessment of
Environment, Development and Sustainability, 18, 697– satellite derived land-cover data: A review. Photogram-
716. https://​doi.​org/​10.​1007/​s10668-​015-​9672-8 metric Engineering and Remote Sensing, 60(4), 419–426.
Hassan, M. M., & Southworth, J. (2017). Analyzing land cover Janssen, L. L., & Wel, F. J. (1994). Accuracy assessment of
change and urban growth trajectories of the mega-urban satellite derived Land-Gover data: A review. Photogram-
region of Dhaka using remotely sensed data and an metric Engineering and Remote Sensing, 60(4), 419–426.
ensemble classifier. Sustainability, 10(10), 1–24. https://​ Kafy, A.-A., Faisal, A.-A., Hasan, M. M., Abdullah-Al- Faisal,
doi.​org/​10.​3390/​su100​10010 Islam, M., & Rahman, M. S. (2020) Modelling future
Hassan, et al. (2016). Dynamics of land use and land cover land use land cover changes and their impacts on land
change (LULCC) using geospatial techniques: A case

Vol.: (0123456789)
13
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
 446 GeoJournal (2023) 88:425–448

surface temperatures in Rajshahi, Bangladesh. Remote Mannan, A., Liu, J., Zhongke, F., Khan, T., Saeed, S.,
Sensing Applications: Society and Environment, 18(2). Mukete, B., & Shah, S. (2019). Application of land-
Kafy, A.-A., Faisal, A.-A., Rahman, M., Islam, M., Rakib, A., use/land cover changes in monitoring and projecting
Islam, M., & Sattar, G. S. (2021a). Prediction of seasonal forest biomass carbon loss in Pakistan. Global Ecology
urban thermal field variance index using machine learn- and Conservation, 17, 535. https://​doi.​org/​10.​1016/j.​
ing algorithms in Cumilla, Bangladesh. Sustainable Cit- gecco.​2019.​e00535
ies and Society, 64, 102542. https://​doi.​org/​10.​1016/j.​scs.​ Mansour, S., Al-Belushi, M., & Al-Awadhi, T. (2020). Moni-
2020.​102542 toring land use and land cover changes in the mountain-
Kafy, A.-A., Faisal, A.-A., Shuvo, R. M., Naim, M. N., Sik- ous cities of Oman using GIS and CA-Markov modelling
dar, M. S., Chowdhury, R., & Kona, M. (2021b). Remote techniques. Land Use Policy, 91, 104414. https://​doi.​org/​
sensing approach to simulate the land use/land cover and 10.​1016/j.​landu​sepol.​2019.​104414
seasonal land surface temperature change using machine Meng, B., Ge, J., Liang, T., Yang, S., Gao, J., Feng, Q., Cui,
learning algorithms in a fastest-growing megacity of X., Huang, X., & Xie, H. (2017). Evaluation of remote
Bangladesh. Remote Sensing Applications: Society and sensing inversion error for the above-ground biomass of
Environment, 21, 100463. https://​doi.​org/​10.​1016/j.​rsase.​ Alpine Meadow grassland based on multi-source satellite
2020.​100463 data. Remote Sens, 9(4), 372.
Kafy, A. A., Rakib, A. A., Akter, K. S., et al. (2021). Moni- Mishra, V.N.; Rai, P.K. (2016) A remote sensing aided multi-
toring the effects of vegetation cover losses on land sur- layer perceptron-Markov chain analysis for land use and
face temperature dynamics using geospatial approach in land cover change prediction in Patna district (Bihar),
Rajshahi City, Bangladesh. Environmental Challenges. India. Arab. J. Geosci. 9.
https://​doi.​org/​10.​1016/j.​envc.​2021c.​100187 Mohammad, M., Sahebgharani, A., & Makeipour, E. (2013).
Khan, M. M. H., Bryceson, I., Kolivras, K. N., Faruque, F., Urban growth simulationthrough cellular automata (CA),
Rahman, M. M., & Haque, U. (2015). Natural disasters anaiytic hierarchy process (AHP) and GIS, case study of
and land-use/land-cover change in the southwest coastal 8th and 12th municipal districts of Isfahan. Geographia
areas of Bangladesh. Regional Environmental Change, Technica, 8(2), 57–70.
15, 241–250. Mondal, M. S. and Hossain, M. M. A. 2009. Characterizing
Khanal, N., Uddin, K., Matin, M., & Tenneson, K. (2019). Long-term Changes of Bangladesh Climate in Context of
Automatic detection of spatiotemporal urban expansion Agriculture and Irrigation. Bureau of Research, Testing
patterns by fusing OSM and landsat data in Kathmandu. and Consultation of Bangladesh University of Engineer-
Remote Sensing, 11(19), 2296. https://​doi.​org/​10.​3390/​ ing & Technology, Climate Change Cell, DoE, MoEF;
rs111​92296 Component 4b, CDMP, MoFDM. June 2009, Dhaka
Kong, F., Yin, H., Nakagoshi, N., & James, P. (2012). Simu- Mondal, K. K., Akhter, M. A., Mallik, M., & Hassan, S. (2017)
lating urban growth processes incorporating a potential Study on Rainfall and Temperature Trend of Khulna
model with spatial metrics. Ecological Indicators, 20, Division in Bangladesh. DEW-DROP, 4.
82–91. https://​doi.​org/​10.​1016/j.​ecoli​nd.​2012.​02.​003 Morshed, S. R., Fattah, M. A., Rimi, A. A., & Haque, M. N.
Kumar, K. S., Bhaskar, P. U., & Padmakumari, K. (2015). (2020). Surface temperature dynamics in response to
Application of land change modeler for prediction of land cover tranformation. Journal of Civil Engineering,
future land use land cover: A case study of Vijayawada Science and Technology, 11(2), 94–110.
city. International Journal of Advanced Engineering Morshed, S. R., & Fattah, M. A. (2021). Responses of spa-
Science and Technological Research, 3(01), 773–783. tiotemporal vegetative land cover to meteorological
Li, H., Xiao, P., Feng, X., Yang, Y., Wang, L., Zhang, W., & changes in Bangladesh. Remote Sensing Applications:
Wang, X. (2017). Using Land Long-Term Data Records Society and Environment. https://​doi.​org/​10.​1016/j.​rsase.​
to Map Land Cover Changes in China Over 1981–2010. 2021.​100658
IEEE Journal of Selected Topics in Applied Earth Morshed, S. R., Fattah, M. A., Haque, M. N., & Morshed, S.
Observations and Remote Sensing, 10(4), 1372–1389. Y. (2021). Future ecosystem service value modeling with
https://​doi.​org/​10.​1109/​JSTARS.​2016.​26452​03 land cover dynamics by using machine learning based
Li, Y., & Liu, G. (2017). Characterizing spatiotemporal pat- Artificial Neural Network model for Jashore city, Bang-
tern of land use change and its driving force based on ladesh. Physics and Chemistry of the Earth, Parts a/b/c.
GIS and landscape analysis techniques in Tianjin dur- https://​doi.​org/​10.​1016/j.​pce.​2021.​103021
ing 2000–2015. Sustainability, 9(6), 894. https://​doi.​ Moore, M., Gould, P., & Keary, B. (2003). Global urbanization
org/​10.​3390/​su906​0894 and impact on health. International Journal of Hygiene
Maarseveen, M., Martinez, J., & Flacke, J. (2018) GIS in and Environmental Health, 206(4–5), 269–278. https://​
sustainable urban planning and management: A global doi.​org/​10.​1078/​1438-​4639-​00223
perspective (1st Edn). CRC Press. https://​doi.​org/​10.​ Nagendra, H., Munroe, D., & Southworth, J. (2004). From pat-
1201/​97813​15146​638 tern to process: Landscape fragmentation and the analy-
Maduako, I., Yun, Z., & Patrick, B. (2016). Simulation and sis of land use/land cover change. Agriculture, Ecosys-
prediction of land surface temperature (LST) dynamics tems & Environment, 101(2–3), 111–115. https://​doi.​org/​
within Ikom City in Nigeria using artificial neural net- 10.​1016/j.​agee.​2003.​09.​003
work (ANN). Journal of Remote Sensing and GIS, 5(1), Nath, B., Niu, Z., & Singh, R. (2018). Land use and land cover
158–165. https://​doi.​org/​10.​4172/​2469-​4134.​10001​58 changes, and environment and risk evaluation of Duji-
angyan City (SW China) using remote sensing and GIS

Vol:. (1234567890)
13
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
GeoJournal (2023) 88:425–448 447

techniques. Sustainability, 10(12), 4631. https://​doi.​org/​ Saputra, M. H., & Lee, H. S. (2019). Prediction of Land Use
10.​3390/​su101​24631 and Land Cover Changes for North Sumatra, Indone-
Nurwanda, A., Zain, A., & Rustiadi, E. (2015). Analysis of sia, Using an Artificial-Neural-Network-Based Cellular
land cover changes and landscape fragmentation in Automaton. Sustainability, 11(11), 3024. https://​doi.​
Batanghari Regency, Jambi Province. International Con- org/​10.​3390/​su111​13024
ference, Intelligent Planning Towards Smart Cities, CIT- Simwanda, M., Murayama, Y., Phiri, D., Nyirenda, V. R., &
IES 2015. Surabaya, Indonesia: In Proceedings of the Ranagalage, M. (2021). Simulating Scenarios of Future
Social and Behavioral Sciences, CITIES 2015. Intra-Urban Land-Use Expansion Based on the Neu-
Oliva, F. E., Dalmau, O. S., and Alarcón, T. E. (2014) A Super- ral Network–Markov Model: A Case Study of Lusaka.
vised Segmentation Algorithm for Crop Classification Zambia. Remote Sens., 13, 942. https://​doi.​org/​10.​
Based on Histograms Using Satellite Images. Mexican 3390/​rs130​50942
International Conference on Artificial Intelligence. 8856, Shahi, E., Karimi, S., & Jafari, H. R. (2020). Monitoring
pp. 327–335. Mexico: Springer, Cham. https://​doi.​org/​ and modeling land use/cover changes in Arasbaran
10.​1007/​978-3-​319-​13647-9_​30 protected Area using and integrated Markov chain and
Park, S., Jeon, S., Kim, S., & Choi, C. (2011). Prediction and artificial neural network. Model. Earth Syst. Environ.,
comparison of urban growth by land suitability index 6, 1901–1911.
mapping using GIS and RS in South Korea. Landscape Shamshirband, S., Hashemi, S., Salimi, H., & Samadianfard,
and Urban Planning, 99(2), 104–114. https://​doi.​org/​10.​ S. (2020). Predicting Standardized Streamflow index
1016/j.​landu​rbplan.​2010.​09.​001 for hydrological drought using machine learning mod-
Pijanowski, B., Tayyebi, A., Doucette, J., Pekin, B., Braun, D., els. Engineering Applications of Computational Fluid
& Plourde, J. (2014). A big data urban growth simulation Mechanics, 14(1), 339–350. https://​doi.​org/​10.​1080/​
at a national scale: Configuring the GIS and neural net- 19942​060.​2020.​17158​44
work based Land Transformation Model to run in a High Shatnawi, N., & Qdais, H. A. (2019). Mapping urban land
Performance Computing (HPC) environment. Environ- surface temperature using remote sensing techniques
mental Modelling & Software, 51, 250–268. https://​doi.​ and artificial neural network modelling. International
org/​10.​1016/j.​envso​ft.​2013.​09.​015 Journal of Remote Sensing, 40(10), 3968–3983. https://​
Pramanik, M., & Stathakis, D. (2015). Forecasting urban doi.​org/​10.​1080/​01431​161.​2018.​15577​92
sprawl in Dhaka city of Bangladesh. Environment and Shubho, M., & Islam, I. (2020). An integrated approach to
Planning b: Urban Analytics and City Science, 43(4), modeling urban growth using modified built-up area
756–771. https://​doi.​org/​10.​1177/​02658​13515​595406 extraction technique. International Journal of Envi-
Ranagalage, M., Wang, R., Gunarathna, M. H. J. P., Dissanay- ronmental Science and Technology, 17, 2793–2810.
ake, D., Murayama, Y., & Simwanda, M. (2019). Spatial https://​doi.​org/​10.​1007/​s13762-​020-​02623-1
forecasting of the landscape in rapidly urbanizing hill Somvanshi, S. S., Bhalla, O., Kunwar, P., et al. (2020). Moni-
stations of South Asia: A case study. Remote Sensor, 11, toring spatial LULC changes and its growth predic-
1743. tion based on statistical models and earth observation
Reddy, C. S., Pasha, S. V., Jha, C. S., Diwakar, P. G., & Dad- datasets of Gautam Budh Nagar, Uttar Pradesh. India.
hwal, V. K. (2016). Development of national database Environ Dev Sustain, 22, 1073–1091. https://​doi.​org/​
on long-term deforestation (1930–2014) in Bangladesh. 10.​1007/​s10668-​018-​0234-8
Global Planet Change, 139, 173–182. https://​doi.​org/​ Subedi, P., Subedi, K., & Thapa, B. (2013). Application of a
10.​1016/j.​glopl​acha.​2016.​02.​003 Hybrid Cellular Automaton – Markov (CA-Markov)
Rimal, B., Zhang, L., Keshtkar, H., Haack, B., Rijal, S., & Model in Land-Use Change Prediction: A Case Study of
Zhang, P. (2018). Land use/land cover dynamics and Saddle Creek Drainage Basin. Florida. Applied Ecology
modeling of urban land expansion by the integration of and Environmental Sciences, 1(6), 126–132.
cellular automata and Markov chain. ISPRS Int. J. Geo- Talukdar, S., Ghose, B., Shahfahad, Salam, R., Mahato, S.,
Inf, 7(4), 154. https://​doi.​org/​10.​3390/​ijgi7​040154 Pham, Q., Avand, M. (2020) Flood susceptibility mod-
Salam, R., Islam, A. T., Shill, B., Alam, G., Hasanuzzaman, eling in Teesta River basin, Bangladesh using novel
M., Hasan, M., & Shouse, R. (2021). Nexus between ensembles of bagging algorithms. Stochastic Environ-
vulnerability and adaptive capacity of drought-prone mental Research and Risk Assessment, 34, 2277–2300.
rural households in northern Bangladesh. Natural Haz- https://​doi.​org/​10.​1007/​s00477-​020-​01862-5
ards. https://​doi.​org/​10.​1007/​s11069-​020-​04473-z Tendaupenyu, P., Magadza, C. H., & Murwira, A. (2016).
Sang, L., Zhang, C., Yang, J., Zhu, D., & Yun, W. (2011). Changes in landuse/landcover patterns and human popu-
Simulation of land use spatial pattern of towns and vil- lation growth in the Lake Chivero catchment. Zimbabwe.
lages based on CA-Markov model. Mathematical and Geocarto Internationa, 32(7), 797–811. https://​doi.​org/​
Computer Modelling, 54, 938–943. https://​doi.​org/​10.​ 10.​1080/​10106​049.​2016.​11788​15
1016/j.​mcm.​2010.​11.​019 Tewolde, M. G., & Cabral, P. (2011). Urban Sprawl analysis
Santé, I., García, A., Miranda, D., & Crecente, R. (2010). and modeling in Asmara, Eritrea. Remote Sensor, 3,
Cellular automata models for the simulation of real- 2148–2165.
world urban processes: A review and analysis. Land- Thapa, R. B., & Murayama, Y. (2012). Scenario based urban
scape and Urban Planning, 96(2), 108–122. https://​doi.​ growth allocation in Kathmandu Valley, Nepal. Land-
org/​10.​1016/j.​landu​rbplan.​2010.​03.​001 scape and Urban Planning, 105, 140–148.

Vol.: (0123456789)
13
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
 448 GeoJournal (2023) 88:425–448

Ullah, S., Tahir, A., Akbar, T., Hassan, Q., Dewan, A., Khan, Ying, C., Ling, H., & Kai, H. (2017). Change and Optimiza-
A., & Khan, M. (2019). Remote sensing-based quantifi- tion of Landscape Patterns in a Basin Based on Remote
cation of the relationships between land use land cover Sensing Images: A Case Study in China. Polish Journal
changes and surface temperature over the lower Himala- of Environmental Studies, 26(5), 2343–2353. https://​doi.​
yan Region. Sustainability, 11(19), 5492. org/​10.​15244/​pjoes/​70007
Verburg, P., Overmars, K., Huigen, M., de Groot, W., & Veld- Zhang, D., Liu, X., Wu, X., Yao, Y., Wu, X., & Chen, Y.
kamp, A. (2006). Analysis of the Effects of Land Use (2018). Multiple intra-urban land use simulations and
Change on Protected Areas in the Philippines., 26(2), driving factors analysis: A case study in Huicheng,
153–173. https://​doi.​org/​10.​1016/j.​apgeog.​2005.​11.​005 China. Gisci. Remote Sensor, 56, 282–308.
Weng, Q. (2012). Remote sensing of impervious surfaces in the Zhang, F., Yushanjiang, A., & Jing, Y. (2019). Assessing
urban areas: Requirements, methods, and trends. Remote and predicting changes of the ecosystem service values
Sensing of Environment, 117, 34–49. https://​doi.​org/​10.​ based on land use/cover change in Ebinur Lake Wetland
1016/j.​rse.​2011.​02.​030 National Nature Reserve, Xinjiang, China. Science of the
World Bank. (2017). Cities, slums, and early child growth: Total Environment, 656, 1133–1144. https://​doi.​org/​10.​
Empirical evidence from Bangladesh. World Bank 1016/j.​scito​tenv.​2018.​11.​444
Group. Zheng, H. W., Shen, G. Q., Wang, H., & Hong, J. (2015). Sim-
Wu, J., Li, R., Ding, R., Li, T., & Sun, H. (2017). City expan- ulating land use change in urban renewal areas: A case
sion model based on population diffusion and road study in Hong Kong. Habitat International, 46, 3–34.
growth. Applied Mathematical Modelling, 43, 1–14. https://​doi.​org/​10.​1016/j.​habit​atint.​2014.​10.​008
https://​doi.​org/​10.​1016/j.​apm.​2016.​08.​002 Zhou, Y., Varquez, A., & Kanda, M. (2019). High-resolution
Yadav, K., & Congalton, R. (2019). Correction: Yadav. K. and global urban growth projection based on multiple appli-
Congalton. R. Accuracy Assessment of Global Food cations of the SLEUTH urban growth model. Scientific
Security-Support Analysis Data (GFSAD) Cropland Data, 6, 34. https://​doi.​org/​10.​1038/​s41597-​019-​0048-z
Extent Maps Produced at Three Different Spatial Reso-
lutions. Remote Sensing, 11(6), 630. https://​doi.​org/​10.​ Publisher’s Note Springer Nature remains neutral with regard
3390/​rs110​60630 to jurisdictional claims in published maps and institutional
Yadav, K., and Congalton, R. G. (2019) Correction: Yadav. K. affiliations.
and Congalton. R. Accuracy Assessment of Global Food
Security-Support Analysis Data (GFSAD) Cropland
Extent Maps Produced at Three Different Spatial Resolu-
tions. Remote Sens. (2018). 10, 1800. Remote Sensing,
11(6), 630. https://​doi.​org/​10.​3390/​rs110​60630

Vol:. (1234567890)
13
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”),
for small-scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are
maintained. By accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use
(“Terms”). For these purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or
a personal subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or
a personal subscription (to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the
Creative Commons license used will apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data
internally within ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking,
analysis and reporting. We will not otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of
companies unless we have your permission as detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that
Users may not:

1. use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to
circumvent access control;
2. use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil
liability, or is otherwise unlawful;
3. falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by
Springer Nature in writing;
4. use bots or other automated methods to access the content or redirect messages
5. override any security feature or exclusionary protocol; or
6. share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer
Nature journal content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates
revenue, royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain.
Springer Nature journal content cannot be used for inter-library loans and librarians may not upload Springer Nature journal
content on a large scale into their, or any other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any
information or content on this website and may remove it or features or functionality at our sole discretion, at any time with or
without notice. Springer Nature may revoke this licence to you at any time and remove access to any copies of the Springer Nature
journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express
or implied with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or
warranties imposed by law, including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be
licensed from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other
manner not expressly permitted by these Terms, please contact Springer Nature at

onlineservice@springernature.com