Study of land use classification in an arid region using multispectral satellite images

Abstract

Rapid urbanization and deforestation have led to increased areas of wasteland in the northern region of the Akola district of Maharashtra, India. This study investigates land use variations in the arid region with the help of multi-temporal images. Land use maps were employed for analysis of different classes using image classification tools in ArcGIS software. Multispectral satellite imagery data were used to create land cover variation maps and land use forecast maps for the study area. The land use classification change maps were produced from LISS-III satellite images and Landsat Enhanced Thematic Mapper Plus (2008 and 2015) using supervised classification techniques. Land use was divided into five major classes, i.e. agricultural land, developed land, wasteland, water bodies, and forestland. We observed significant changes in agricultural and forestland as a result of many factors including population growth, drought conditions, road infrastructure development, flooding, and soil erosion in the arid area. The overall accuracy of the supervised classification was 94.10% for 2008 and 88.14% for 2015, using the kappa method, which was a satisfactory result. The analysis of land use maps in the arid region revealed different patterns of use between 2008 and 2015. The results of this study may be useful for developing and implementing valuable management strategies for resource protection in the study area. These results show the potential for land use planning and development in arid regions using remote sensing and GIS technology.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 100 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.