Global cities, which have big populations and are the engines of economic development (Sherbinin

et al., 2007), are especially susceptible to the effects of changes in land use and land cover (LULC).
The LULC change is the main driver of Global environmental change fueled by factors including the
release of habitat loss/fragmentation, biodiversity loss (Lambin et al., 2000; Sala et al., 2000), and
reduce greenhouse gases (MEA, 2005),human wellbeing quality (Lambin, 1997; Boissiere, et al.,
2009). Changes in land use and land cover (LULC) have a direct impact on how land resources are
used in a region, and they are one of the major elements influencing the biological environment
(Saleem et al. 2018). Human influences on LULC change, such as urbanisation, agricultural decline,
have all contributed to scarcity of natural resources, as well as pervasive and worldwide, permanent
losses in biodiversity (Pickard et al. 2017; Bounoua et al. 2018). The proper planning of the use of the
land and other natural resources, as well as the inventory of these resources, depend on the
knowledge of land use and land cover (LULC). Consequently, it is believed to be becoming more
significant in sustainable and effective environmental and natural resource management Sheeja et
al. (2011) Rapid population expansion and the subsequent economic development accelerate these
changes, especially in rapidly urbanising areas (Liping et al., 2018), endangering the sustainability of
those regions's growth. The UN (2016) reported that, in 2016, more than 54.4% of the world's
population resided in cities, and that number is expected to rise to 60% by 2030, placing additional
strain on urban resources. Around the world, 55% of people live in urban areas, and by 2050, that
number is projected to rise by 68%. By 2050, the number of urban dwellers is predicted to rise by 2.5
billion. The region of Asia and Africa will see a growth in urban population of almost 90%. However,
it is anticipated that just China, India, and Nigeria will account for 35% of the additional urban
population between 2015 and 2050. By 2050, there will be 416 million, 255 million, and 189 million
additional urban residents in India, China, and Nigeria, respectively, to accommodate the growing
urban population (UNDESA 2018). In developing nations like India, China, Brazil, and Indonesia, the
population is growing at a very rapid rate (Dadras et al. 2014). India's population has hardly doubled
during the past 50 years (presently 1.25 billion). In the meantime, urban populations have multiplied
by five. Numerous metropolises, including Mumbai, Chennai, Kolkata, Delhi, Bangalore, and
Hyderabad, are predicted to grow by up to six times by 2021 (Taubenbock et al., 2009). By 2021,
India will continue to have the highest population density in major cities (Chakrabarti and Dhar
2001).The dynamics of land use and land cover (LULC) have been significantly altered by humans. In
recent decades, LULC modelling has become a focus of research to address the issue that results
from the modification and conversion of LULC (Lambin et al. 2001). A significant amount of
information regarding the Earth's surface and living habitats is necessary for effective study and
monitoring of land cover (Schneider and Woodcock 2008). According to Niemelä et al. (2000) and
Srivastava et al. (2012b), anthropogenic factors typically initiate the development of LULC change
(LULCC), which is then followed by natural processes. Urbanization, a global trend that has greatly
increased since the turn of the 20th century (Srivastava et al. 2010), is a key factor in the alteration
of the structure and functions of the landscape (Antrop 2004; Banerjee and Srivastava 2013).
Researchers (Lambin et al. 2001; Satterthwaite 2009; Csorba and Szabó 2012; Srivastava et al. 2012c;
Vaz et al. 2012; Patel and Srivastava 2013) have also identified a number of landscape pattern
scenarios as being responsible for urbanisation. These scenarios take into account changing
environmental conditions for, among other things, climate change, land use change, the creation of
new road networks, etc.

The primary drivers of urban growth are biophysical and socio-economical factors. Important
biophysical factors for urban growth include climatic and topographic variation, soil type and
geomorphic processes, landforms, volcanic eruption, drainage pattern, and natural resources (Fu et
al. 2006; Serra et al. 2008; Thapa and Murayama2010; Oprsal and colleagues 2016). However,
population expansion, widespread rural-urban migration, economic disparities between rural and
urban areas, fast industrialization, economic growth, and government policies that support urban-
centric economic growth. Urban expansion is also fueled by economic possibilities and globalisation
(Rimal et al. 2018).

Different statistical methods, together with remote sensing and GIS techniques, have been
employed as an effective substitute for the old surveying and mapping procedures because they
were time-consuming and expensive for the assessment of urban sprawl (Yeh and Li 2001; Punia and
Singh 2011; Sudhira et al. 2004). For environmental and geomatics research on LUCC, land change
models are crucial instruments (Olmedo, M.T.C. et al. 2014;)In order to provide decision-makers
with information on current land use patterns and changes . (Liu, Y. et al. 2015;) and to encourage
sustainable development . (Fan, F. et al. 2007;) monitoring and analysis of LULC changes are
required. LULCC models are used to enhance and/or better understand how human activity-induced
changes in land use affect the environment . (Brown, D.G. et al. 2004;).

Numerous LULC change and prediction studies have been conducted as a result of the South Asian
region's rapid population expansion, migration rate, industrial growth, and economic growth. For the
need analysis and policy formulation necessary to assure a sustainable urban future, accurate,
consistent, and current information on the urbanisation trend is essential. Land-use change maps
throughout time offer vital data for land-use planning ( Shafizadeh Moghadam et al, 2013; Rimal et
al. 2018;) that can aid in understanding the dynamics of changing land-cover and forecasting future
economic and environmental consequences (Eastman, J.R. et al. 2005;). For land-cover monitoring,
urban/regional planning (Dewan, A.M. et al.2009; Thapa and Murayama2010; ),and examining
spatiotemporal changes in LULC from a local to a global scale, GIS and remote sensing are the ideal
instruments (Yuan, F. et al. 2005;). Spatiotemporal analysis of urban expansion is therefore of
utmost importance. All detail information on the spatial location, characteristics, and effects of
urban growth foundations are (1) the formulation of urban development plans,(2) the development
and modification of urban morphology theories, and the (3) definition of the boundary between the
urban area and environment for some environment models(Jiang, B. et al. 2010; Jokar Arsanjani, J.
2013;)

LULC models examine change vs. rates of change, can be static and geographical, and include a
variety of analytical techniques (Mas, J.-F. et al. 2014;). A change demand sub-model, a transition
potential sub-model, and a change allocation sub-model are typically included in land-cover change
models. These sub-models calculate the amount and spatiotemporal location of LULC change as well
as the shifting land-cover type from one to the other (Eastman, J.R. et al. 2005;). Due to low cost and
capacity to collect long-term data sets over a wide geographic area, remote sensing has enormous
potential for managing land resources (Chandra et al. 2003). Several investigations of LULC changes
have been conducted using remote sensing data sets, including MODIS (Alhamdan et al. 2017),
Landsat (Chen et al. 2013a, b; Osgouei and Kaya 2017), and SPOT (Halmy et al. 2015). For Chennai,
Tamil Nadu, India, a neural network coupled agents-based Cellular Automata model using Landsat
imagery was used to anticipate urban expansion (Aarthi & Gnanappazham, 2018). Through the use
of Landsat, spatially precise studies have been made, including the spatiotemporal evolution
research of LULC change (Behera 2007), urbanisation (Fu and Weng 2016), vegetation coverage (Liu
et al. 2017a, b), and flood mapping (Mueller et al. 2016; Mohammadi et al.2017)

In order to comprehend the stochastic nature and stability of the land use/land cover, the Markov
model is excellent and very helpful (LULC). Due to improvements in GIS and remote sensing
technology, the Markov model has gained in popularity.
The Markov model is commonly used to simulate changes in the landscape (Baker, 1989; Muller and
Middleton, 1994), assess different forms of land use and trends, and quantify changes (Weng, 2002;
Huang et al., 2008). The Markov chain model and the CA (Cellular Automata) Markov model are two
examples (Muller and Middleton, 1994). (Clarke, 1997). The later state (land cover type) of a pixel in
the Markov chain model is only related to its immediately preceding state and not to any other prior
states The result of a transition probability is Markov chain model. however, the CA Markov model
the other hand, made tremendous progress in merging when developing forecasts, consider spatial
contingency information. As There has been study using the Markov chain, which is a step ahead
model for increased accuracy. Guan et al. (2011) examined the spatial distribution and temporal
change in land use for Saga, Japan, between 1976 and 2006. Additionally, LULC change prediction
has been done from 2015 to 2042. For the Kathmandu valley between 1988 and 2016, Rimal et al.
(2018) conducted LULC change research. They also predicted urban sprawl for the years 2024 and
2032. For the Chaoyang district of Beijing, Liu and Zhou (2004) used the Post-classification technique
to calculate land-use change trajectories. For the purpose of analysing watershed change between
the years 1999 and 2004, Zhou et al. (2008) used high-resolution temporal data along with object-
based categorization and Post-classification change detection. Alboody et al. 2018; have suggested
combining spatial thinking and the outcomes of post-classification comparison in order to achieve
theme shift (2008). In addition, the post-classification technique for land use change analysis was
taken into consideration by Ahlqvist (2008), Fan (2008), Yang and Wen (2011), Frojse (€ 2011), and
Kafi et al. (2014). Additionally, Raja et al. (2013) used this technique to identify newly formed urban
outskirts. Madanian et al. (2012) claim that the best method for 'from-to' change detection is a mix
of Post-classification comparison and ID. Markov Transition Matrix or Transition Probability Matrix
used for Quantities Change Analysis for Land Use (Petit et al. 2001). Transition Matrix and Kappa
indices were utilised by Biondini and Kandus (2006) to analyse the incidence of LULC and estimate
changes in LULC. (Zhang et al. 2011; Petit et al. 2001; Biondini and Kandus 2006) TPM (Markov
Chain) was used to illustrate the likelihood of each LULC class transforming into the others. Only a
non-spatial representation of LULC is provided by TPM. Therefore, cellular automata (CA) have been
employed to represent the temporal and spatial information of LULC in order to ignore the
shortcomings of the transition matrix. Many earlier studies used a CA-Markov model to simulate the
pattern of changing land use. The CA-Markov model was successfully applied in the Arasbaran
biosphere by (Parsa et al. 2016;).reserve-Iran to forecast the LULC, which assists planners and
decision-makers in land use make the proper choices to address upcoming land use issues. They
mentioned utilising a CA-Markov model can be utilised as an early warning system, the model can be
helpful in designing land use policies.

For the Atakum urban growth simulation, Ozturk (2015;) compared CA-Markov chain and Multi-layer
Perceptron-Markov Chain (MLP-MC) models to forecast future change in [Link]'s Samsun. The
authors claim that the MLP-MC model produced improved outcomes for anticipated as opposed to
the CA-Markov model. However, Regmi et al. (2014;) examined the LULC dynamics in the Phewa
lake watershed in Nepal using the CA-Markov and GEOMOD models. When they projected a future
LULC scenario, they discovered that CA-Markov chains worked pretty well as an operational model.
Many driving forces, including infrastructure, socioeconomic drivers (road network and human
settlement), and physical terrain drivers (Digital Elevation Model), have been included in those
simulations (DEM derived slope). The outcomes show that the spatial distribution of the watershed
LULC has been impacted by the driver forces.
METHDOLOGY

In this study, the Baghpat district of Uttar Pradesh, LULC classes were mapped using digital remote
sensing data from the Landsat series of satellites. The Landsat images are widely used for remote
sensing, and because of their high spectral, spatial, and temporal resolutions essential information
for planning and mapping (Sadidy et [Link]. 2009). Geometric correction of TM, ETM+, and OLI Landsat
images was confirmed, and all images were processed in an ArcGIS environment. During pre-
processing, these multispectral pictures were corrected for atmospheric conditions and geometric
errors.