Land Use Fragmentation Analysis Using Remote Sensing and Fragstats

Abstract

AbstractThis study present the results of a set of landscape metrics derived from remotely sensed data aiming to characterize the historical trends of landscape changes in the Allahabad district in the period 1990–2010. However, the identified trends in landscape changes and its effects in the region have potential policy implications. The land use and land cover were estimated from sensors viz. for the period 1990 (LANDSAT TM), 2000 (LANDSAT ETM+) and 2010 (IRS 1D LISS III) through the maximum likelihood classification (MLC) method. The land use land cover change was quantified with the help of ERDAS imagine 9.1. Further, landscape level and class level metrics were derived from the classified satellite images in FRAGSTATS 3.3. Total four metrics for landscape level viz. total area (TA), number of patch (NP), patch density (PD), area mean (AREA MN) and four metrics for class level viz. core area (CA), number of patch (NP), patch density (PD) and percentage of land (PLAND), respectively to uncover the influence of land use change which can be correlated to the degree of urbanization, development and water quality. The different class level metrics of study area has revealed internal exchange of four land use classes given as agricultural land (65.32 % in 1990, 67.13 % in 2000, 68.1 % in 2010), builtup area (9.98 % in 1990,11.63 % in 2000,13.36 % in 2010), cultivable land (4.42 % in 1990, 3.47 % in 2000, 2.1 % in 2010) forest (6.03 % in 1990, 4.47 % in 2000, 5.6 % in 2010), and water body (5.89 % in 1990, 5.82 % in 2000, 5.35 % in 2010). The study showed that the notable changes had occurred in the last 20 years in this landscape, hence there is need of appropriate measures to mitigate these negative impacts of changes.KeywordsLandscape patternFragmentationLISS IIILANDSATEcological metrics