The development of urbanization and industrialization led to the change in land use and land (LULC) in the last two centuries resulting in the degradation of the sustainable conditions for the future. This intense urbanization does not follow a linear trend (Bose & Chowdhury, 2020). Few cities in the world have established planned layouts for urban distributions. Over time, low-density built-up regions become high-density and later extremely density regions because of the increasing land demand due to changes in the desirability of the communities (Saxena & Jat, 2020). LULC transition is a critical issue while analyzing global trends as they impact groundwater infiltration, evapotranspiration, and natural disaster (Tewabe & Fentahun, 2020). The controlling mechanism of urban development involves assessing the time-space relationship between nonlinear interactions such as culture, economy, topography, population, land use, and river systems. (Thapa & Murayama, 2011).
India is the second-populous country with 1.3 billion in 2015 and the seventh-largest country in the world. Over the last 140 years, India has seen drastic shifts in LULC, including a reduction in the forest, change in cropland, and an increase in urbanization). This practice continued till the 1960s, when, for the first time, the ‘Green Revolution’ limited itself to increasing crop production, irrigation, and the use of fertilizer and pesticides (Abrol et al., 2002; Roy et al., 2015). UN estimates show that 60 percent of the world’s rural villages will be covered by large cities in 2050. This is due to unplanned urban growth caused by a lack of planning and development (Bose & Chowdhury, 2020). Spatial modeling can be used to investigate the complexities of potential agricultural production, associated LULC change, and environmental consequences. Anthropogenic and environmental mechanisms need to be thoroughly understood concerning the temporal dynamics and possible changes in land cover (Chughtai et al., 2021; Silva et al., 2020). Urban growth would be beneficial for urban planners and engineers; furthermore, recognizing the future trends and extent of urban development would be crucial for open space and natural habitat protection (L. Singh et al., 2021).
The emergence of remote sensing (RS) has led to various land use studies, such as the evolution of LULC in a global environment and evaluating the transition of various land types (Ghosh et al., 2017). Satellite remote sensing has the advantages of comprehensive coverage, extensive collections of data, and ongoing observations (Hu et al., 2018). RS and Geographical Information Systems (GIS) are effective methods for obtaining precise and temporal information on the spatial distribution of LULC over large areas. When analyzed in RS, ground surfaces have a distinct spectral signature, referred to as spectral reflectance patterns (Becker et al., 2021). GIS offers a robust framework in which digital change detection information is collected, stored, displayed, and analyzed. Landsat 5, 7, and Landsat 8 were commonly employed for LULC analysis because of their moderate or high-resolution images and continuous global coverage (since 1972) (Alam et al., 2020).
Over the last decade, RS data processing has shifted from conventional workstations to cloud-based platforms that enable users to access and analyze geospatial data through user-friendly cloud interfaces and programming languages (Tassi & Vizzari, 2020). Google Earth Engine (GEE) is a cloud-based platform to extensively compute satellite data for research, education, and non-profit applications. GEE can be accessed using a JavaScript code editor platform, and it simplifies the processing of satellite imagery. A quick sign-in to your Google account is enough to gain access to the GEE (Gorelick et al., 2017a; Midekisa et al., 2017; Sidhu et al., 2018a; Tsai et al., 2018; Wagle et al., 2020). Earth Engine has a variety of supervised and unsupervised classifications, including machine-based learning algorithms for the mapping implementation. The data available covers a wide range of satellites, including the Sentinel series; Moderate Resolution Imaging Spectrometer (MODIS); Landsat series; Advanced Land Observing Satellite (ALOS); National Oceanographic and Atmospheric Administration Advanced very high-resolution radiometer (NOAA AVHRR), etc., (Kumar & Mutanga, 2018). The data is relatively accessed and can be conveniently downloaded or stored in the cloud. GEE is utilized in several disciplines to solve different environmental issues such as forest fire and changes in the forest (Seydi et al., 2021; Ye et al., 2021), surface water area change (Yang et al., 2020), estimation of crop yield (Cao et al., 2021; Tian et al., 2020), urban mapping (Ji et al., 2020; Zhang et al., 2021), flood mapping (Pourghasemi et al., 2021; Tiwari et al., 2020) and few studies for fire recovery (Huang et al., 2017; Soulard et al., 2016), mangrove mapping (de Jong et al., 2021), drought analysis (Khan & Gilani, 2021).
Idrisi TerrSet is an automated geospatial information framework that researchers commonly use to analyze and model earth system processes for sustainable development (Nath et al., 2020). Land change modeler (LCM) is an important tool for environmental and other learning scenarios on LULC change in TerrSet. It is used to analyze the historical and forecast potential LULC changes. The extent and direction of LULC are two major factors that have been considered in the modeling. Models for assessing LULC transition may be static or dynamic, inductive or deductive, agent-based or pattern-based, spatial or non-spatial (V. N. Mishra & Rai, 2016; Noszczyk, 2019; Ozturk, 2015; Zadbagher et al., 2018). The Cellular Automata (CA)-Markov model is stochastic and assesses the likelihood that a state will switch to a different state that is, the condition time t2 is predicted from the time t1. LCM analyses the changes in LULC over various periods, calculates, visualize, and displays the changes using several maps and graphs (Aburas et al., 2017; Noszczyk, 2019).
CA-Markov model using TerrSet software is used in this analysis to simulate LULC changes and urban growth prediction on the northern Tamil Nadu (TN) coast, India. According to the international scientific community, there is a chance that by 2050, few centimeters of sea-level increase will occur in the districts of Tiruvallur, Chennai, Kanchipuram, Villupuram, Cuddalore, and several other districts. This research aims to detect the urban growth pattern and recognize the future growth scenario using simulations in the LCM. These simulations were used to predict the urban growth in 2019, 2025, 2030 using LULC from 2009 and 2015. Transition directions and spatial variations of potential changes from 2009 to 2030 were calculated.