Analysis of spatio-temporal pattern and driving force of land cover change using multi-temporal remote sensing images

Landuse and land cover change is regarded as a good indicator that represents the impact of human activities on earth's environment. When the large collection of multi-temporal satellite images has become available, it is possible to study a long-term historical process of land cover change. This study aims to investigate the spatio-temporal pattern and driving force of land cover change in the Pearl River Delta region in southern China, where the rapid development has been witnessed since 1980s. The fast economic growth has been associated with an accelerated expansion of urban landuse, which has been recorded by historical remote sensing images. This paper reports the method and outcome of the research that attempts to model spatio-temporal pattern of land cover change using multi-temporal satellite images. The classified satellite images were compared to detect the change from various landuse types to built-up areas. The trajectories of land cover change have then been established based on the time-series of the classified land cover classes. The correlation between the expansion of built-up areas and selected economic data has also been analysed for better understanding on the driving force of the rapid urbanisation process. The result shows that, since early 1990s, the dominant trend of land cover change has been from farmland to urban landuse. The relationship between economic growth indicator (measured by GDP) and built-up area can well fit into a linear regression model with correlation coefficients greater than 0.9. It is quite clear that cities or towns have been sprawling in general, demonstrating two growth models that were closely related to the economic development stages.