Skip to content


The evolution of urban form and global urban land expansion

How can we best quantify and describe changes in urban land-use patterns beyond extent and cumulative growth rates? Researchers have been trying to capture essential elements of urban morphology for decades and landscape ecology has provided us with good metrics for characterizing changing urban landscapes. In a paper published in Landscape Ecology in 2005 (co-authored with Karen Seto) and titled “Quantifying spatiotemporal patterns of urban land-use change in four cities of China with time series landscape metrics” I conducted the first comparative analysis of a system of rapidly developing cities using landscape pattern metrics. The study was an intra- and inter-city assessment of landscape pattern changes across time and space to identify characteristics of the evolving urban form in four Chinese cities in the Pearl River Delta. I used spatial pattern metrics (essentially measures of urban patch size, shape, fragmentation and connectedness) for a time series of Landsat TM images that span 11 years (1988 to 1999). I showed that urban form can be malleable over a relatively short period of time. Although di®erent policies guided the development of the cities in the time period of the study, we observe similar at trajectories in the pattern of growth.

The question of the quantification of urban form has also been addressed in studies of complexity empirics. Approximately 100 years ago, researchers identified regularities in the size distributions of cities within national urban systems, branded power laws. For a little more than 10 year now, we have known that cities themselves also present such regularities in the distributions of their urban clusters. In research published in Computers, Environment and Urban Systems in 2009, I have advanced the understanding of evolving urban spatial structure through a novel time-series view of urban cluster rank-size distributions across three cities in the Pearl River Delta, China. Several propositions can explain the phenomenon based on theories of complex economic and social interactions within and between cities. All past examination of regularities in rank-size distributions – identified as power laws – has focused on Western cities. My research is the first to reveal that these regularities hold considerably well for the three Asian cities across a time span of eleven years (1988 to 1999), an era of rapid political and economic reform. The implications of this research are important: planning that fails to take into account the city as a complex system will not likely be particularly e®ective in bringing about solutions to urban pathologies and sustainable urban landscapes. The fact that the majority of urban clusters in a metropolitan area obey power laws with parameters following an oscillating behavior across time as the distribution of the clusters progresses from more even to less even and vice versa – should be strongly considered before planning or designing “good” or sustainable urban form.

At the global scale, the conversion of Earths land surface to urban uses is one of the most irreversible human impacts on the global biosphere. Urban land expansion drives the loss
of farmland, affects local climate, fragments wildlife habitats, and threatens biodiversity. However, there is little understanding of global patterns of urban land expansion, with most
studies focusing on individual metropolitan areas. In recent research (together with Karen Seto, Burak Guneralp and Michael K. Reilly), we present a meta-analysis of studies that have used satellite images to map urban land expansion. We report a massive worldwide monitored increase in urban land area that has more than doubled in three decades, with significant di®erences across geographic regions. India, China, and Africa have experienced the highest rates of urban land transformation, and North America has the largest total urban extent. In most cases, rates of urban land transformation are higher than rates of urban population growth, indicating that urban areas are becoming less compact. Growth in GDP per capita has twice the e®ect on urban land expansion than urban population growth, and agricultural land is more likely to be converted than non-farm land. Coastal regions are also more likely to become urban. Rates of urban land expansion over the last 30 years have been faster in low lying coastal elevation zones. We conclude that urban land transformation will accelerate in the coming decades, primarily in coastal and agricultural regions, with serious consequences for agricultural land, climate change impacts and response capacity.