Conclusion

This work explores the potential of implementing cellular automata to model the historical urban growth over Bangalore city from 1973 to 1992 and 1992 to 2006. The main goal was to design the model as a function of local neighbourhood structure to minimise the input data to the model. Satellite imagery represents the medium over which the model works taking into account spatial and temporal calibration based on transition rules. Spatial calibration fits the model on a directional basis to its site specific feature while temporal calibration adapts it to the urban growth dynamic change over time, producing a good spatial match between the real and simulated image data.

The technique demonstrated here was found effective in predicting urban growth and visualising them through pixels in images. Such studies are important for relating pixels in RS data and people in society for sustainable development, pollution prevention, global environmental change, and issues of human-environment interaction at different spatial and temporal scales. The limitation of this work is that the present CA model takes into account only the land use categories, population density, and distance from the city center. There are many other factors such as distance from rail / roads, levels of services available in different locations, education and employment opportunities, economic flow, etc. that are major triggering factors for urbanisation in developing countries like India. Therefore, future work involves considering agents and external factors (such as foreign direct investment flow in the city, decisions of setting up of special economic zones (SEZ), etc.) that act as catalyst and driving forces in the expansion of cities. Land use category such as builtup land may be further classified into their different forms using very high spatial resolution imageries such as Quickbird, IKONOS, etc. The detailed land use practices may further improve the modelling result and give better prediction accuracy.