Study of land use/land cover dynamics through classification algorithms for Harangi catchment area, Karnataka State, INDIA

Industrialization and urbanization are mainly responsible for the conversion of significant amount of vegetation rich lands such as forests and wetlands into agricultural lands and human settlements. This land use and land cover change in turn has influence on various hydrological processes such as soil erosion rate, sediment deposition in rivers and dams and also stream flow pattern in a catchment. Qualitative and quantitative analysis of Land use and Land cover (LU/LC) changes is necessary to assess the impacts of change in the pattern of natural vegetation on the earth's environment. Satellite images can provide useful information regarding spatial and temporal variation of LU/LC in an area. The purpose of this paper is to analyse the dynamics of LU/LC changes using LISS-III data for the years 2007, 2010 and 2013 of Harangi catchment, Coorg District, Karnataka State, India. Also, study compares three classification techniques such as Parallelepiped Algorithm, Minimum Distance to Mean Algorithm and Maximum Likelihood Algorithm based on the value of overall accuracy and kappa coefficient to check the reliability of the methods in classifying images. The overall accuracy and kappa coefficient obtained by using Parellelepiped, Minimum distance to mean and Maximum likelihood algorithm for the year 2013 are 81.47% & 0.71, 78.67 & 0.68 and 89.36% & 0.81 respectively. Based on these results, it can be concluded that, among all the three techniques, the Maximum Likelihood Algorithm gave higher accuracy with high kappa coefficient and Minimum Distance to Mean Algorithm gave lower accuracy. The LU/LC change detection results between years 2007 and 2013 indicated a drastic change in forest area, plantation and waste land among all other classes. The forest area decreased dramatically from 183.12 sq. km to 131.02 sq. km. Also, fallow land and water body are decreased from 68.89 sq. km to 42.63 sq. km and 6.71 sq. km to 3.82 sq. km respectively. In contrast, the plantation and waterlogged area are observed to have an increasing trend of about 56.07 sq. km to 146.55 sq. km and 17.99 sq. km to 23.81 sq. km. The urban area has been increased from 13.06 sq. km to 13.72 sq. km. The urbanization and agricultural activities are the major reasons for increase in plantation and urban area in an expense of decrease in forest area. (C) 2015 The Authors. Published by Elsevier B.V.