Mapping fifty global cities' growth using time-series Landsat data

Urban growth and sprawl have drastically altered the ecosystems and ecosystem services. The objectives of this study are to using grid square method to investigate the spatial and temporal dynamics of urban growth in 50 global cities using Landsat ETM/TM imagery from 1985 - 2011. First, MLC classification method were used to produce land cover maps by using Landsat images from 1985a's, 1993a's, and 2007a's (completed); then intersect the land cover maps with 1-km² grid cell maps to represents the proportion of each land cover category within each 1-km² grid cell (ongoing); finally, combining the proportional land cover maps to investigated the relationship between land cover changes based on grid square cells for three intervals (i.e. around 1985, around 1993, and around 2007). Change analysis unveiled large changes in land cover and land use have occurred from 1985a's to 2007as. The case in Tokyo, Japan shows the Settlements area has rapidly expanded to the surrounding sub urban area which was mainly located flat areas or along the transportation lines. The area of Settlements doubled over the past two decades, increasing from 12.5% of the study area in 1987 to 23.5% in 2011. The correlation analysis in Tokyo shows strong, negatively linear relationship between the Settlements change and cropland change (r = - 0.78), suggesting that the vast area of cropland area have been converted to Settlements during the last two decades. In the next step, we will analyze the other 49 cities using 1-km² grid cell approach and calculate the correlation coefficient matrix between the changes of land cover categories from 1985a's - 2007a's for each cities. Furthermore, we expect to compare and contrast the rates and patterns of expansion, and drivers of land cover change in 50 cities..