Method and Data

4.1 Method

Understanding the urban dynamics and evolution of spatial patterns of urbanization along the gradients (of 1 km incrementing radii) is illustrated in figure 3. The approach involves understanding decadal land use and built-up dynamics,
(ii) zone wise (based on directions) gradient analysis (with 1 km gradients),
(iii) analyses of spatial patterns of urbanisation through computation of metrics in in each gradient,
(iv)Comparing spatial metrics to understand trajectories of urbanization at local levels,
(v) analyses of overall patterns of urbanisation through Principal Component Analysis to bring out the intra and interregional similarities and variability of urban growth.

4.2 Data

Time series spatial data acquired from the data archive (http://glcf.umiacs.umd.edu/data) of Landsat Series Multispectral sensor (57.5m) and thematic mapper (28.5m) sensors for the period 1973 to 2010. The IRS 1C data (23.5m) of Chennai for the year 2012 was acquired through NRSC (http://nrsc.gov.in), and the field data for the study regions was collected through pre-calibrated Global Positioning System (GPS). In order to rectify geometric errors of the data acquired through space borne sensors the Ground Control Points (GCP’s) were collected from field through GPS and also from geo-registered topographic maps of the Survey of India. The data were geometrically corrected and were resampled to 30 m for uniformity in the spatial resolution of Landsat and IRS 1C data. The study region (administrative boundary with 10 km buffer) were cropped from the respective scenes of RS data.