Detection Of Change In Land Cover In Jabalpur District From 1991-2021 Using Remote Sensing

Land use/ land cover is an important component in understanding the interactions of human activities with the environment and thus it is necessary to monitor and detect the changes to maintain a sustainable environment. In this paper, an attempt has been made to study the changes in land use and land cover of Jabalpur district in the last 4 decades from 1991 to 2021 classifying majorly in Forest (Medium to Dense), Trees, Waterbody, Settlements & Agricultural elds. The study was carried out through the Remote Sensing and GIS approach using High-resolution Imagery from Google Earth, and LANDSAT 8, 7, 5 imagery of 2021, 2011, 2001, 1991 respectively. The land use/land cover classication was performed based on the Supervised Classication approach available in ArcGIS. GIS software is used to prepare the thematic maps and ground truth observations were also performed to check the accuracy of the classication. The present study has brought out that the Tree cover has been decreased from 12.97–5.40% during 1991-2021 showing a considerable decrease in Forest area as well. The settlement area increased from 4.26% in 1991 to 12.46% in 2021. The areas under natural streams, have shown no signicant change and can be considered as a positive sign for sustainable development. MNDWI-CBI-SVM 90.5% 0.87. classied LUC maps validated satellite performed using satellite meter/pixel in LANDSAT and 23 meters in LISS III image. Digital land use land cover classication through supervised classication method to perform the LULC classication in ERDAS IMAGINE 9.1 software They classied land into forest”, “Sand Bar”, “Water Body”, Forest”, grassland”, & Their study showed major changes occurred in cropland and scrubland The study conducted by Sarma, G Murali, in Godavari Delta region that land use/land cover alterations might be responsible for local level climatic changes. their study, MSS image 1973 scale and IRS-LISS-II 1992, LISS III


Introduction
Satellite-based remote sensing data are extensively used in the mapping of Land Use/Cover (LUC) of the earth's surface. The impact of LUC changes is noticed globally with signi cant effects on urban areas visa-vis rural areas. LULC mapping is one of the most important applications of remote sensing. Land use is widely used in the development of groundwater resources. The hydrogeological processes such as In ltration, evapotranspiration, surface runoff, etc. were controlled by land use. The surface cover provides roughness to the surfaces, reducing discharge thereby increasing the in ltration. In the forest areas, in ltration will be higher and runoff will be less whereas in urban areas rate of in ltration may decrease. Remote sensing techniques will give detailed information with respect to the spatial distribution of land use and vegetation type in minimum time and low cost in comparison to other conventional data.

Related Studies
A similar conducted by N.C.Anil, G.Jai Sankar, M. Jagannadha Rao in the southern part of West Godavari district. Their objective was to determine the trend, nature, rate, location, and magnitude of land use/land cover change. GIS software is used to prepare the thematic maps and ground truth observations were also performed to check the accuracy of the classi cation. Their study showed that the aqua-culture tanks have been decreased from 33.02% to 25.66% during 2000-2010 with a net decrease of 7.34%.
Agriculture was also decreased from 43.32% to 37.75% with a net decrease of 5.56% during 2000-2010 (Anil 2011). They classi ed the land into "Agriculture Fields", "Aquaculture Tanks", "Drains", "Fallow Land", "Mangroves", "Mud Flats", "Plantation mixed with the crop", "Rivers", "Settlements". Their study concluded that one decade the signi cant positive observations as per environment is concerned is the natural systems represented by natural drains, mud ats, mangroves, and river systems indicated signi cant change. Another study was conducted by Anitha Selvaso a, S Shrividya, S Karunya, in Coimbatore District in Tamil Nadu state in 2008. Their paper describes adjustments in the Land use/land cowl pattern of Coimbatore District. Their study found that the cropland is decreasing at the cost of hazard growth of plantations and settlements (Anitha, 2021). The Study conducted by Vineela Nandam & P. L. Patel on the western coast of Gujarat state in India proved to be helpful in a wide range of applications including the development of ood risk maps, systematic planning of urban infrastructure, analysis of urban heat islands, and averting the coastal disasters in the future. They did a critical comparison of SVM and RFC classi ers, nally, a hybrid approach is proposed as a combination MNDWI-CBI-SVM for mapping of the study area with Overall Accuracy of 90.5% and a Kappa value of 0.87. The performance of classi ed LUC maps was validated using reference data of high-resolution images of March to about the middle of June. The period is the middle of June to September in the southwest monsoon season. October and November form the post-monsoon or transition period. The average annual rainfall of Jabalpur District is 1279.50mm. Jabalpur received maximum rainfall received during the southwest monsoon period i.e. June to September. About 90% of the annual rainfall is received during the monsoon season. Only 10% of the annual rainfall takes place between October to May period. Thus, surplus water for groundwater recharge is available only during the southwest monsoon period.

Remote Sensing
Remote sensing refers to a wide range of technologies used to detect Earth's surface, usually using aerial or satellite platforms. The earliest use of remote sensing in soil science was the development of aerial photographs as base maps for soil survey in the United States in the 1920s and 1930s (Bushnell, 1929), which represented a major advance over creating base maps using plane tables and odometers (Worthen, 1909) or using topographic maps when they were available as was common prior to the use of aerial photography (Miller and Schaetzl, 2014 Imaging in remote sensing can be carried out from both satellite and aircraft platforms. In many ways, their sensors have similar characteristics although differences in their altitude and stability can lead to very different image properties. There are essentially two broad classes of satellites: those satellites that sit at geostationary altitudes above the earth's surface and which are generally associated with weather and climate studies, and those which orbit much closer to the earth's surface and that are generally used for earth surface and oceanographic observations. Usually, the low earth orbiting satellites are in a sunsynchronous orbit, in that their orbital plane processes around the earth at the same rate that the sun appears to move across the earth's surface. In this manner, the satellite acquires data at about the same local time on each orbit. Lower earth-orbiting satellites can also be used for meteorological studies.

LANDSAT:
Landsat is part of a global research program known as NASA's Mission to Planet Earth, a long-term program that is studying changes in Earth's global environment. The goal of Mission to Planet Earth is to provide people with a better understanding of natural environmental changes. Mission to Planet Earth data, which will be distributed to researchers worldwide at the cost of reproduction, is essential to people making informed decisions about their environment.

Methodology
To perform change analysis of the study area, images from the satellite LANDSAT are used which are obtained from USGS. The images are acquired for the same study period to retain consistency. Digital land use land cover classi cation through the supervised classi cation method is done to perform the LULC classi cation in ArcGIS 10.8.0, and the results were compared. Area statistics of each land use category are calculated in square kilometers in the attribute table in ArcGIS.
Following are the steps followed: 1. The very rst step is to download the LANDSAT Image tile, for the study area. For this study, our study area is Jabalpur District & the data is downloaded from USGS Earth Explorer. we go to the study area for which we need to download the LANDSAT Data.
2. After zooming, up to a level, where our study area is roughly visible, using the coordinates make a polygon covering the study area.
3. Now, when our study area is marked, the next step is to select what kind of data we, are looking for?
Here in this study, we are looking for LANDSAT Data, so using Dataset Tab, we will select Landsat > Landsat Collection 1 > Landsat 4. Now using the additional Criteria Option, data with less than 20% cloud cover and for the speci c time (see Table.1) the period for which data is required is selected. We need data from 1991-2021, speci cally recorded in the rst two months of the year to increase uniformity.
5. Now Results, show a list of tiles available for various dates, the necessary tiles, covering the full study area with speci c data which is required is selected & downloaded.
Performing Supervised Classi cation: 1. The downloaded dataset consists of 12 processed Black & white .TIF images representing each band/sensor from which data was recorded.
2. Using the Composite tool available in the ArcTool Box, Band 1 to Band 7, images are merged into one RGB image.
3. If there are more than 2 images tiles area covering the study area, using Composite to a New Raster tool available in ArcTool Box, 2 tiles are combined into 1 image. The results clearly indicate a reduction in individual trees and medium forest area in the Jabalpur district.

Conclusion
Supervised classi cation using LANDSAT satellite imagery on ArcGIS is a very quick, easy & powerful way to detect changes in land features. In this study, we classi ed land use majorly in Forest, Trees, Waterbody, Settlement Agricultural elds. The results clearly show major increment in Settlements in Jabalpur district & reduction in Trees & Forest see gure 3. Agricultural elds clearly show major increment as more and more land is being used for farming to support the food demand of the population.
In the future, the imagery will be processed using Arti cial neural networks, with more land classes, including smaller features like roads natural streams/ rivulets, mangroves, etc.

Declarations
Con ict of Interest Statement: Figure 1 Study area Jabalpur District. Jabalpur District Land Use Classes in 2021