Analysis

1. Preprocessing: The remote sensing data obtained were geo-referenced, rectified and cropped pertaining to the study area. Landsat ETM+ bands of 2010 were corrected for the SLC-off by using image enhancement techniques, followed by nearest-neighbour interpolation.

2. Land use analysis: 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  v) 60% of the training data has been used for  classification of the data, while the balance is used for validation or accuracy assessment.

Land use  classification was carried out using supervised pattern classifier - Gaussian maximum likelihood algorithm. This classifier is superior as it uses various classification decisions using probability and cost functions (Duda et al., 2000). Mean and covariance matrix are computed using estimate of maximum likelihood estimator. Land use was computed using the temporal data through open source program GRASS - Geographic Resource Analysis Support System (http://wgbis.ces.iisc.ernet.in/grass/index.php). Four major types of land use classes were considered: built-up area, forestland, open area, and water body. Application of this method resulted in accuracy of about 88% using Landsat data, 91% accuracy using IRS-P6 data, 94% accuracy using Ikonos data and 74% using Modis data. For the purpose of accuracy assessment, a confusion matrix was calculated.

3. Landscape Metrics: Landscape metrics were computed for each of chosen multi-resolution data - MODIS data (500 m) was resampled to 250 m and 100 m. Landsat resampled to 30 m and 15m, Ikonosof 4m resampled to 3m 2m and 1m respectively. The resampled data were considered for further analysis. Classified land use data (data and also for resampled data) was converted to ASCII format and metrics at the landscape level were computed with FRAGSTATS [32]. Fragstat is open-source software that can be freely downloaded (http://www.umass.edu/landeco/research/fragstats/fragstats.html). The spatial metrics include the patch area, edge/border, shape, compact/contagion/ dispersion and are listed in Appendix 1.