Materials and Methods

Data: Survey of India (SOI) toposheets of 1:50000 and 1:250000 scales were used to generate base layers. Field data was collected with a handheld GPS. The time series remote sensing data acquired from Landsat Series Multispectral sensor (57.5m) and Thematic mapper (28.5m) sensors for the period 1973 to  2010 and Landsat ETM+ (2000 and 2009) were downloaded from public domain (http://glcf.umiacs.umd.edu/data).Google Earth data (http://earth.google.com) served in pre and post classification process and validation of the results. Latest data for 2010 (IRS – Indian remote Sensing) was procured from the National remote Sensing Centre (http://www.nrsc.gov.in), Hyderabad. The methods adopted in the analysis involved:

• Pre-processing of data: The remote sensing data were geo-referenced, rectified, and cropped pertaining to the study area. Geo-registration of remote sensing data (Landsat data) was done using ground control points collected from the field using pre calibrated GPS (Global Positioning System) and also from known points (such as road intersections, etc.) collected from geo-referenced topographic maps published by the Survey of India. Geo-referencing of acquired remote sensing data to latitude-longitude coordinate system with Evrst 56 datum: Landsat bands, IRS LISS-III MSS bands, MODIS bands 1 and 2 (spatial resolution 250 m) and bands 3 to 7 (spatial resolution 500 m) were geo-corrected with the known ground control points (GCP’s) and projected to Polyconic with Evrst 1956 as the datum, followed by masking and cropping of the study area. The Landsat satellite 1973 images have a spatial resolution of 57.5 m x 57.5 m (nominal resolution) were resampled to 28.5m comparable to the 1989 - 2010 data that are 28.5 m x 28.5 m (nominal resolution). Landsat ETM+ bands of 2010 were corrected for the SLC-off by using image enhancement techniques, followed by nearest-neighbour interpolation. As the accuracy of the classified output of LANDSAT ETM+ was relatively low, the analysis for 2010 was repeated with IRS (LISS III) data procured from NRSC coinciding with the dates of field data collection.

• Land use analysis: The analyses of land use were carried out using supervised pattern classifier - Gaussian maximum likelihood classifier (GMLC) for Landsat and IRS data, and Bayesian Classifier (MODIS data). The method involves: i) generation of False Colour Composite (FCC) of remote sensing data (bands – green, red and NIR). This helped in locating heterogeneous patches in the landscape ii) selection of training polygons (these correspond to heterogeneous patches in FCC) covering 15% of the study area and uniformly distributed over the entire study area, iii) loading these training polygons co-ordinates into pre-calibrated GPS, vi) collection of the corresponding attribute data (land use types) for these polygons from the field. GPS helped in locating respective training polygons in the field, iv) supplementing this information with Google Earth (latest as well as archived data),  v) 60% of the training data has been used for  classification, while the balance is used for validation or accuracy assessment.

• Recent remote sensing data (2010) was classified using the collected training samples. Statistical assessment of classifier performance based on the performance of spectral classification considering reference pixels is done which include computation of kappa (κ) statistics and overall (producer's and user's) accuracies. For earlier time data, training polygon along with attribute details were compiled from the historical published topographic maps, vegetation maps, revenue maps, etc.

• Normalised Difference Vegetation index (NDVI) was computed to understand the changes in the vegetation cover during the study period. NDVI is the most common measurement used for measuring vegetation cover. It ranges from values -1 to +1. Very low values of NDVI (-0.1 and below) correspond to soil or barren areas of rock, sand, or urban built-up. Zero indicates the water cover (Ramachandra and Kumar, 2009). Moderate values represent low-density vegetation (0.1 to 0.3), while high values indicate thick canopy vegetation (0.6 to 0.8).

• Density Gradient Analysis: Urbanisation pattern has not been uniform in all directions. To understand the pattern of growth vis a vis agents, the region has been divided into 4 zones based on directions - Northwest (NW), Northeast (NE), Southwest (SW), and Southeast (SE) respectively based on the Central pixel (Central Business district). Further, each zone was divided into concentric circle of incrementing radius of 1 km radius from the centre of the city that would help in visualizing and understanding the agents responsible for changes at local level. These regions are comparable to the administrative wards ranging from 67 to 1935 hectares.The growth of the urban areas in respective zones was monitored through the computation of urban density for different periods.