Land use and Land cover Mapping Based on Normalized Difference Vegetation Index using Remote Sensing and Geographical Information System in Banjar River Watershed of Narmada Basin

The spatial analysis of land use and land cover (LULC) dynamics is necessary for sustainable utilization and management of the land resources of an area. Remote sensing along with Geographical Information System emerged as an effective technique for mapping the LU/LC categories of an area in an efficient and cost-effective manner. The present study was conducted in Banjar river watershed located in Balaghat and Mandla district of Madhya Pradesh, India. The Normalized Difference Vegetation Index (NDVI) approach was adopted for LU/LC classification of study area. The Landsat-8 satellite data of year 2013 was selected for the classification purpose. The NDVI values were generated in ERDAS Imagine 2011 software and LU/LC map was prepared in ARC GIS environment. On the basis of NDVI values five LU/LC classes were recognized in the study area namely river & water body, waste land & habitation, forest, agriculture/other vegetation, open land/fallow land/barren land. The forest cover was found to be highly distributed in the study area with an extent of 115811 ha and least area was found to be covered under river and water body (4057.28 ha). This research work will be helpful for the policy makers for proper formulation and implementation of watershed developmental plans. Current World Environment Journal Website: www.cwejournal.org ISSN: 0973-4929, Vol. 12, No. (3) 2017, Pg.680-686 CONtaCt Jagriti Tiwari jagritit27@gmail.com Department of Soil & Water Engineering, College of Agricultural Engineering, J.N.K.V.V, Jabalpur (M.P.) © 2017 The Author(s). Published by Enviro Research Publishers This is an Open Access article licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License (https://creativecommons.org/licenses/by-nc-sa/4.0/ ), which permits unrestricted NonCommercial use, distribution, and reproduction in any medium, provided the original work is properly cited. To link to this article: http://dx.doi.org/10.12944/CWE.12.3.19 article history Received: 14 August 2017 Accepted: 4 October 2017


Introduction
The earth resources are key component in the development of a nation as well as of a region.The update information of the landscape is of great significance in the management and monitoring of an environment.LU/LC classification is one of the prime prerequisites for analyzing the information of the existing earth resources and changes occurring on the landscape pattern of an area through time.Land cover data refers to the area of a region covered under natural resources viz.forests, wetlands, agriculture, and other land, rivers and water bodies.Land use documents for the region utilized for carrying anthropogenic activities -whether for development, conservation, mining and mixed uses 1 .In present scenario, the modernization and immense increase in population has imposed pressure on the available earth resources.Overexploitation of the land resources creates an alarming condition for the rational use of these resources.Therefore, to forbid this overuse, the accurate mapping of LU/ LC is very essential.Satellite image classification is an important procedure to categories the different features of the landscape for proper planning of the conservative measures 2,3 .
Observation of Earth surface from space is crucial phenomena to understand the influence of human interferences and there by impacts on changes in the pattern of land over time 4,5 .The introduction of Remote Sensing in the field of advance technology has made it possible to generate landscape information at regional and local scales 6,7 .The Remote Sensing when integrated with Geographical Information System becomes more effective toolset and has considerable potential to efficiently analyse the landscape dynamics 8,9,10,11 .The combination of RS and GIS provides a suitable platform analysing the data to study the transitions of the land landscape in less time and with better accuracy in a cost effective manner 12 .
The classification of multi-spectral satellite images has been successfully applied for classifying the different features of land.But mapping of landscape on the basis of Remote Sensing index i.e.Normalized Difference Vegetation Index (NDVI) by utilizing the satellite data proves to be an effective way of classifying the LU/LC.In order to estimate quantity, quality and changes occurring in vegetation, NDVI takes into account the spectral information of red and near infrared bands 13,14,15 .It is rely upon the reflectance characteristics of vegetation, water, rocks and bare soil 16 .Additionally, the generated NDVI data could be utilized to identify the pattern of changes occurring on vegetation cover on temporal basis.The NDVI data layer is defined as: NDVI = (NIR -red) / (NIR + red) Where NIR refers to spectral reflectance in near infrared band and R refers to red band.The development of NDVI led to worldwide application of it in RS and GIS based studies.17.NDVI values lies between -1.0 and 1.0, where higher values are for higher intensity of dense green vegetation and lower values for other surface materials.Bare soil is associated with NDVI values which are nearest to 0 and water bodies are indicated with negative NDVI values 18 .The important concept of analysising vegetation using NDVI is the high absorption of chlorophyll which is a vegetation pigment in the red spectral region and high reflectance in the near infrared region.NDVI is greatly related with the photosynthetic activity and represents the greenness of dense vegetation.It can significantly separate various spectral classes of LULC.NDVI can be used for comparison of vegetation index information derived from different dates of satellite imagery.This paper concentrates on land use and land cover classification based on NDVI by employing Remote Sensing and Geographical Information System in Banjar River watershed of Narmada Basin.

Overview of Study area
The study focuses on Banjar river watershed which lies in Upper Narmada basin (Fig. 1) geographically located in Balaghat and Mandla districts of Madhya Pradesh, with extent of 22 0 05'N to 23 0 29'N latitudes and 80 0 22'E to 81 0 00'E longitudes.The watershed covers about 246084 ha up to gauging point.Climate of the area is tropical with moderate winter and severe summers and well distributed rainfall received from southwest monsoon.The normal annual rainfall of study area is 1300 mm.Soils of the area are categorized in black grey, red and yellow colours.The shallow soil is found in barren lands.

Materials and Method Database Generation
The satellite data of study area has been acquired in the year 2013.The data set was Landsat-8 OLI-TIRS consist of eight bands downloaded from scientific user's website Earth Explorer (USGS) (https://earthexplorer.usgs.gov)at the resolution of 30 m.This dataset was imported in ERDAS Imagine 2011.This image processing software was used to generate a false colour composite (FCC) (Fig. 2) of study area by selecting the layer stack option available in image interpreter toolset.The study area was extracted from satellite image by taking geo-referenced out line boundary of Banjar river watershed.Universal Transverse Mercator (UTM) coordinate system, Datum WGS 1984, zone 44 North projection system was used for projecting the satellite data using 1:250, 000 topographic map of Banjar river watershed.This data was utilized for LU/ LC classification.

Fig. 2: FCC of study area Land Use and Land Cover Mapping Using Ndvi
The LU/LC features of the study area were classified using Landsat-8 satellite data.The NDVI is found to be an appropriate remote sensing index to understand the variation in the LU/LC pattern.The NDVI tool present in classification menu of ERDAS imagine 2011software was used for creation of NDVI image (Fig 3 1).

Conclusion
The preparation of LU/LC maps is necessary for better understanding of land utilization pattern and its planning.The present study illustrates the usefulness of Remote Sensing and Geographical Information System for classifying study area as per the major LU/LC types on spatial basis.The NDVI method was proposed to identify the LU/LC categories.The higher values of NDVI were found in the forest cover which is the dominant feature of Banjar river watershed.The lower values of NDVI indicate an area under river and water body.It can be concluded that NDVI formed the basis for better classification of different features of an area and can be utilized for effective analysis of the landscape information as compared to any other methods of classification.

acknowledgement
The authors acknowledge all the support received from Department of Soil and Water Engineering, College of Agricultural Engineering, J.N.K.V.V., Jabalpur (M.P.), India for this study.
) using false color composite image of study area.The NDVI images magnified the vegetation class in the satellite images and helped in distinguishing it from other non-vegetation classes.The spectral classes which are function of NDVI are evaluated to assign into suitable LU/LC classes.The various LU/LC classes interpreted in the study area include, river & water body, waste land & habitation, forest, agriculture/other vegetation, open land/fallow land/barren land (Table

Fig. 4 :
Fig. 4: Land use and land cover map of study area of year 2013 (NDVI)