|dc.description.abstract||Landscape is a heterogeneous collection of visibly distinct features of various elements of land and its various forms on the earth surface. Its pattern is subjected to disturbances and undergo rapid alterations in its grain sizes. The evolving patterns of landscape define and decide various parameters for the planning and management of resources. These dynamic systems possess both spatial and temporal complexity. Exploitation of natural resources and drastic land cover changes have given rise to significant impacts on ecosystem structure and dynamics. The functional abilities (bio-geo chemical cycling, hydrological cycling, etc.) of the landscape are basically dependent on the structure and its complexity. This necessitates inventorying, mapping and modeling of landscape dynamics. Patterns and scale are central issues that are essential to understand complex interactions and driving forces. Large scale changes have been rapid and occurring since industrialization and urbanisation in the last century. The exponential growth of cities has been noticed since the industrial revolution and as transport sector changed the mobility of the masses drastically. Urbanisation interacts with the neighboring landscape structures in the form of commuter’s flow, pollution, obtaining food grain, which create dispersed growth or sprawl in between the metropolis and the semi urban area, and these areas are often devoid of basic amenities due to lack of prior information and necessitates predictions of such growth while planning, policy and decision-making. Planning determines appropriate future action through a sequence of choices that tend to occur. To understand uncertain conditions, planners and city managers need vital comprehensive information about the temporally evolving landscape and try to predict the future, for effective decisions. The quality of planning and its decision processes can be substantially improved when the required information is handled appropriately and efficiently. This explains that an effective planning requires descriptive, predictive, and prescriptive information inputs for sustainable resource management. Therefore, modeling future trends becomes a necessary part of planning. Urban growth models help in modelling future trends that can be an efficient and effective support tool. In recent years, the confluence of developments in Remote sensing, Geographic Information System and Image processing, Computational Urban Growth and Urban Land-use Modeling has made possible in timely provision of information inputs to planners.
In the context of Indian cities, this research attempts to study the patterns of urban growth and the rate of change of that growth using various techniques such as Land use, land cover models, Gradient and zonal approach, spatial metrics and urban growth models. Indian cities are divided based on population into various categories. These categories were considered separately and dealt with sample number of cities. This works helps in understanding the change pattern of rapidly urbanising, moderately urbanising and rural landscape is accomplished using various metrics and gradients. The research, is mainly aimed at understanding the pattern of growth and device computational urban growth model using well known techniques and develop a suitable technique in order to understand the context of agents and their role in modelling future urban growth and estimate the rate of loss of other land use categories due to urban growth. Satellite images for different time series was used to study the pattern of urban growth in the study areas. Well know indicators were derived from the data. This was further used to model one of the rapidly urbanising cities based on scenario no agents/factor and with agents of growth using city development plans and in absence of it. This adaptation to Indian context will help in gaining better understanding of the urban growth system in various levels of cities classified, and thus help in providing inputs and specific information of future growth for urban planners and city managers to provide better basic amenities and for sustainable growth of cities.
The objective of the proposed research is to understand and model the spatio temporal patterns of landscape dynamics. This involves
i. Analysis of Landscape dynamics using multi-resolution (spatial, temporal and spectral) data.
ii. Quantifying landscape dynamics using landscape metrics and associated landscape parameters.
iii. Modeling and geo-visualisation of landscape dynamics in rapidly urbanizing, moderately urbanising and rural landscape using these parameters.
iv. Model the landscape dynamics using soft computing techniques.
The thesis consists of nine chapters. Chapter 1 introduces the basic concepts such as landscape, landscape dynamics, use of spatio-temporal data to monitor landscape dynamics, geo-visualisation of landscape dynamics, research gaps and motivation for taking up the research in this domain.
Chapter 2 presents the study region, which are broadly grouped as (i) Rapidly urbanizing landscapes (corresponding to Tier I Cities in India), (ii) Moderately urbanizing landscapes (Tier II cities, chosen select Tier II cities in Karnataka), and
(iii) Landscape experiencing minimal urbanisation (rural landscape).
Chapter 3 discusses the material and method adopted for understanding landscape dynamics and geo-visualisation of landscape dynamics
Chapter 4 presents the landscape dynamics in rapidly urbanizing landscape (Bangalore) in India. Spatial pattern analyses are done through metrics using zonal- gradient approach.
Chapter 5 analyses the environmental sustainability aspects considering one case study of rapidly urbanizing landscape – Bangalore
Chapter 6 discusses urbanisation process and patterns across macro cities in India. Similarly Chapter 7 discusses the urbanisation pattern in Tier II cities (in Karnataka) and Chapter 8 presents the rural landscape dynamics
Geo-visualisation of a rapidly urbanizing landscape (Bangalore) through techniques such as Cellular Automata – Markov Chain, land change modeler (LCM), Geographical land use change modeler (GEOMOD), Markov Cellular automata based process of deriving agent’s behavior using Fuzziness in the dataset and Analytical Hierarchal process. Further research in progress in this domain focusses on integration of various agents and evaluation of proposed development plans and likely scenario of integrating land use with mobility.
Keyword: landscape, landscape dynamics, urbanisation, urban growth, urban sprawl, urban footprint, modelling, geo-visualisation||en_US