Methods

Temporal analysis of water bodies using remote sensing data for Delhi and Mumbai was done using the data Landsat series Multispectral scanner (57.3m) and Thematic mapper (28.5). This data wereresampled to 28.5m using nearest neighbor interpolation techniques. The base layers were generated from Survey of India (SOI) topo-sheets of 1:50000 and 1:250000 scales. The study area includes the Delhi and Mumbai administrative boundary extended to 10 Km buffer boundary to understand the dynamics at city outskirts due to urbanization process.

Land use analysis was done using maximum likelihood supervised classification technique and classes include: built-up, vegetation, waterbodies and others. The excessive noise in the classified images was removed using 3X3 filter.

Digital elevation models are used in the study for extracting drainage network. DEM images were also resampled to 28.5m using nearest neighbor interpolation techniques. The DEM data is elevation data and helps in assessing the terrain characteristics and widely used in hydrological analysis and modelling.  Wetlands were extracted from the temporal landuse layers, which illustrates the changes during four decades. The wetlands of the Delhi and Mumbai regions are draped on DEM to analyze the impact of urbanization.

Estimation of Land surface temperature:Land surface temperature was calculated from thermal band of Landsat Thematic mapper. Digital number of the band was converted to spectral radiance LETM using eq (1),

            L_ETM = 0.0370588 x DN + 3.2                        (1),

The black body temperature or at satellite brightness temperature was calculated Tk,eq (2), K1 and K2 are considered as K2 and K2 are pre-launch calibration constants.

            T_k = K2/ln (K1/ LETM + 1)                       (2)

Surface emissivity is specified for each of the landuse (Snyder et al., 1998).

Landsurface temperature is calculated, using eq (3), where  is surface emissivity

                                                              (3)