A Multi-Temporal Landsat Data Analysis for Land-use/Land-cover Change in Haridwar Region using Remote Sensing Techniques

Abstract LU/LC (land-use and land-cover) change detection information is essential for monitoring, planning, decision making, and socio-economic development. In the current study, multispectral Landsat-8 OLI, Landsat-7 ETM+, and Landsat-5 TM satellite imagery data has used in the Haridwar region for investigating the utility and possibility of remote sensing (RS) in LU/LC change assessment. After ensuring a ground survey with GPS, discussions with the individuals, and satisfactory literature-review, it has been found that RS can play a vital character in assessing the timely temporary changes in LU/LC of the Haridwar region. The satellite data of the different years 1996, 2003, 2010 and 2017 downloaded from USGS Earth Explorer. The satellite images of the selected study areas are categorized into seven different land use classes, viz. orchards, vegetation, agricultural land, rangeland, urban land, water bodies, and watershed. The pixel-based supervised classification method has performed 81.55%, 86.86%, 80.67%, and 93.00% accuracy using the Maximum Likelihood Classifier (MLC). The result of change detection of the LU/LC proved that the orchards area has been decreased very rapidly by 11806.65 ha (9.82%) and converted permanently in urban and agricultural land due to urbanization and industrialization in Haridwar region from 1996 to 2017 year. While the remaining area of land use such as urban land, rangeland, and watershed have increased up to 3022.11 ha (2.51%), 10088.82 ha (8.39%), and 5191.47 ha (4.32%). The vegetation land and water body have been decreasing since 1996 up to 12856.41 ha (10.70%) and 802.63 ha (0.66%) respectively in 21 years (1996-2017). The overall results presented by this research may be useful for orchard monitoring and socio-economic development of the Haridwar region.