Date of Publication :17th May 2017

Abstract: Forests all over the globe are subjected to enormous pressure resulting in deforestation and degradation due to the anthropogenic activities like urbanization and increase in cultivation. Measuring land-cover change is an essential part of sustainable conservation planning. Hence monitoring change in forest cover has become an important tool for forest management. Remote sensing offers a quick and low cost way for forest cover mapping to monitor deforestation. The objective of the study is to analyze spatial and temporal forest cover changes in Udupi district of Karnataka to understand the deforestation pattern from 1973 to 2016.The study applied supervised classification – Maximum Likelihood Algorithm in Erdas Imagine software using satellite data obtained from Landsat Multispectral Scanner (MSS), Landsat 5 Thematic Mapper(TM),Landsat 7 ETM+,IRS LISS III and Landsat 8 Operational Land Imager(OLI) for the years 1973, 1981, 1991, 2003, 2012and 2016 respectively. The area was classified into 11 land use/land cover classes’ viz. agricultural land, arecanut and coconut plantation, barren land, built up land, dense forest, fallow land, grass land, other vegetation, rubber plantation, sand soil and water body. The overall accuracies obtained are 79.37%, 81.04%, 82.53%, 84.38%, 85.76% and85.76% for the years 1973, 1981, 1991, 2003, 2012 and 2016 respectively. Land use/land cover maps generated through ArcGIS indicated a significant shift from forest cover to agricultural land, plantations and urban settlement areas. The study demonstrated that monitoring of long term forest changes provides critical inputs for management of forest resources.

Reference :