Growth of Dehradun city: An application of linear spectral unmixing (LSU) technique using multi-temporal landsat satellite data sets

• Linear spectral unmixing model for spatiotemporal growth of Dehradun using Landsat.
• Fractional values of endmember of builtup and vegetation used to assess the growth.
• Pixel-wise and ward-wise quantification of spatiotemporal change has been done.
• 20% increase in built-up and 13% decrease in the vegetative cover in just 10 years.
• Higher percentage of changes is observed in the southern most part of the city.

In order to quantify and monitor the spatio-temporal growth of urban areas, remote sensing data and GIS techniques are getting attention for its unparallel utility by the city planners as well as decision makers. The present study aims to analyse the rapid increase of built-up area in Dehradun city, the capital of Uttarakhand, India. The study uses multi-temporal Landsat images to monitor the pattern of urban growth of Dehradun. To classify coarse satellite data the traditional supervised and unsupervised techniques are not very useful because of mixed pixels in the classified data and poor accuracy. In this situation linear spectral unmixing (LSU) is a very useful classification technique because it classifies land use land cover classes by estimating the fractional values of each class. The study shows that built-up area of the city has grown about 20 per cent during 2000–2010 which is also identical with the population growth of Dehradun when compared with the Census of India data. The accuracy of Landsat based LSU model was assessed using high resolution LISS IV data and the RMSE was 0.07 and 0.11 respectively for vegetation and built-up fraction classes. It is seen that built-up area in the Kaulagarh, Kargi, Deep Nagar, Indirapuram and Kawli wards of Dehradun city have experienced more than 30% increase mainly in the South and South-West part of the city. The multi-temporal, multi-scale maps of the Dehradun city generated from remote satellite data will be very useful for planning both at micro- and macro-levels and further to identify the potential areas for urban expansion and future planning keeping in view protecting the urban green landscape of the Himalayan city of Dehradun.