Literature Review

Urban research during the last decade has made consider-able efforts in bringing into focus urbanizing pattern of cities and challenges [12]. Studies have also focused on peripheral areas emerging as employment centers and transforming the immediate rural neighborhood [13].
However, earlier research focus were mainly into urban core and urban–rural neighborhood and not much emphasis given to the corridors that connect two major urban centers. The corridors have always been attributed to improved functional properties of a growing city, however very minimal attention has been provided especially the land use change dynamics and associated environmental impacts. Hence, more often industrial corridors exhibit or develop a tendency to represent fragmented urban growth connected to cities with extremely poor environmental conditions.
Spatial analyses for inventorying, mapping and monitoring land cover using remote sensing data acquired through space borne sensors have been proved to be the quickest and economical technique for mapping large areas. The multi-resolution (spatial, spectral and temporal) data and robust classification algorithms with the best accuracy are useful in land use mapping and to understand landscape dynamics. Now, Earth-observation-based moni-toring of urban growth has been widely accepted and implemented by local, regional and national governments [14]. Satellite technology helps urban mapping at finer scales and thereby making it easier for policy makers and planners to understand the growth dynamics and sprawl [15]. GIS (Geographic information system) with facilities to capture, manage, store, retrieve, analyze and display geo-spatial data on a real-time basis, has entered the majority of the service sectors. Further strength to GIS comes from a built-in database, decision support system and application-specific plugins, which makes GIS a more reliable tool for urban studies [16]. Studies in India [10, 12, 13, 17] and worldwide by researchers [2, 6, 14, 18], mainly address land cover changes and its implications. Assessment of spatial patterns of land use dynamics and identification of agents is a prerequisite to understanding urban growth dynamics and modelling the urbanisation process. Further, this will aid in understanding and visualizing the specific pockets of growth and influ-ence of urban corridors in the buffer regions. There has been extensive research in urban modelling in recent years. Urban modelling is the process of identifying a theory which could be translated into a mathematical model as well as developing a specific computer-aided programs to feed the model with data so as to calibrate, validate, verify and predict future urban trends [19]. During the past four decades, there have been significant contributions to urban growth models and visualization with a common goal to study land use dynamics and simulate urban growth using geospatial techniques [20]. Theoretical assumptions, the method followed, spatial, temporal aspects and geograph-ical extents might vary with each model types, but the final outcome of these models is to understand the complex interrelationships between natural ecosystem and urban environment by observing irreversible heterogeneous spa-tial patterns of changes [21]. Cellular automata (CA) based modelling framework has been one of the widely accepted technique of urban simulation and modelling due to its simplicity, flexibility and intuitiveness [22]. CA models were gradually improved on the aspects of increasing complexity of transition rules [23]. Despite all these, CA models do not consider the change probability. This opens up an opportunity to couple Markov Chains (MC) and CA providing a potent modelling structure. The CA–Markov develops based on time series and spatial predictions of the Markov and CA, making it possible for Spatial–Temporal simulation. Modelling based on CA depends on five ele-ments namely, the spatial arrangement, states, neighbour-hood, rules of transition, temporal scale of cells [24]. Among other approaches, regression modelling has been effectively applied in urban research [25]. These models are extremely useful in knowledge development if inte-grated with the process of physical and socioeconomic patterns. These socio-economic drivers are called ‘Agents of Development’ and modelling approaches are referred to as ‘Agent Based Modelling (ABM)’. ABM has proved to be reliable, individual decision-making tool to capture spatial dynamics by incorporating socio-economic and environmental factors [26]. Potential usage of ABM in research areas like LULCC (land use land cover changes) simulation, design, development and implementation has been explored earlier [27]. Over the last two decades, ABM has evolved with various factors and understanding inter-actions contributing to urban growth and has proved to be superior over conventional modeling techniques [28]. ABM is evolving and flexible, considers bottom-up approaches studying behavior of factors in urban evolution, while allowing complex, heterogeneous nonlinear interac-tions. ABM exhibit an extensive knowledge base on how the agents of changes interact with each other to be responsible for the social and physical environment of urban areas and their immediate vicinity. Numerous ABM techniques have been used to understand/predict urban sprawl/land use changes in a region such as Sleuth, SLUCE/Some, Land Change Molder (LCM), Artificial Neural Networks, Dynamic Urban Evolution Model, Multi Regression, CLUES, Dyna-CLUES, CAPRI-Spat, Fuzzy-AHP-CA etc. [27].
Evaluation of a process was earlier restricted to precise values [29], and the process of using fuzzy logic has freed the boundary conditions by a range of values [30]. Inte-gration of Fuzzy into the modelling allows the decision maker to compare two variables (pair wise comparison) which allows accessing the relative importance of one variable over the other in AHP—analytical hierarchical process [31]. Analytical hierarchical process formulates and analyses the decisions made individually or as a group [32]. AHP with multi-criteria evaluation approach has been useful in evaluating the site suitability for landscape development [33]. Thus, this integrated agent-based approach would help in visualizing development scenarios aimed at decision making for a sustainable development goal [23]. The current study analyses landscape dynamics using temporal remote sensing data, identifies agents of local growth poles and analyse spatial patterns using spatial metrics through segment (zones) approaches and predict future land uses using ABM in the two major Industrial corridors in India.
The paper is organized in five sections: Sect. 1 brief introduction of urban growth. Section 2 reviews the related studies, Sect. 3 details out the study area considered for the analysis, and explains the method and data used in the analysis. This is followed by the Sect. 4 explaining the results. Section 5 presents the outcome of the research with suggestions.