Land Use and Land Cover change detection

using Geospatial techniques in Meghalaya,

India

ABSTRACT
Mapping and monitoring of Land Use Land Cover (LULC) changes in a region
like Umiam river basin is vital for sustainable development planning and
management. Based on Remote Sensing (RS) and Geographic Information
System (GIS) techniques, this study is an initiative to study the changes in LULC
patterns of regions of Umiam river basin of Meghalaya for the period 2013-2023.
Images from Landsat 8 were used for this purpose. Supervised classification
using Maximum Likelihood Classification (MLC) was applied to prepare LULC
map. LULC in the study area has undergone changes over the decade. Five major
classes for LULC viz; river, builtups/barren, very dense forest, vegetation, and
small water bodies has been identified. The results has shown an increase and
decrease in area of some of the classes.

INTRODUCTION
In the heart of northeastern India’s lush highlands, Meghalaya emerges as a state
of enchanting beauty and climatic marvels. ‘Meghalaya’, means ‘the abode of
clouds’. It is a poetic statement given to its mist-laden hills and the monsoon
clouds that lovingly cover the landscape. This region is not just a relief for those
who cherish the natural world, rather it stands as an important study area for
meteorologists and environmental researchers due to its distinction as one of the
planet’s rainiest places, particularly in places like Mausynram and Cherrapunji.

Mausynram’s importance is not solely due to its extraordinary precipitation level.

It also represents a dynamic ecosystem where studies on tropical forests, water
conservation, and climate change converge. The abundant rainfall supports a wide
range of biodiversity and sustains a network of waterfalls and streams, building
an environment that is both lush and ever-evolving. The neighboring areas, each
with their unique charm, contribute to the rich natural and cultural mosaic that
characterizes Meghalaya. From Mawlynnong’s unspoiled setting (Asia’s cleanest
village) and adherence to Khasi traditions to Shillong’s lively fusion of historical
architecture and contemporary energy, Meghalaya offers diverse experiences.

Within this captivating state flows the Umiam River basin, our focus area that
illustrates the delicate balance between nature’s offerings and human pursuits.
The basin is crucial for generating hydroelectric power, agriculture, fisheries, and
supplying drinking water to communities. It stands as an example model for
exploring land use trends, watershed stewardship, and the socio-economic ties to
natural endowments.

In this background, remote sensing technology becomes an indispensable asset

for environmental analysis. It facilitates discreet surveillance of these
ecologically sensitive territories, yielding data on land use evolution, vegetation
vitality, and water resources while maintaining the integrity of the habitats. The
technology’s ability to process extensive imagery over periods is crucial for
observing environmental transformations and fostering sustainable resource
utilization. For Meghalaya, where rugged landscapes and intense rainfall pose
access challenges, satellite imagery coupled with GIS capabilities create novel
pathways for conservation initiatives and policy formulation.

The incorporation of remote sensing into Meghalaya’s environmental assessment

not only enriches our comprehension of its ecological mechanisms but also
accentuates the significance of technological innovation in preserving this Indian
gem. Specifically within the Umiam River basin, remote sensing plays a pivotal
role in monitoring sediment deposition patterns, appraising water quality, and
devising sustainable hydropower strategies that honor both the ecological sanctity
and the dependent human livelihoods.

STUDY AREA DESCRIPTION

The following map(Fig 1) provides an overview of the outline of the terrain of
Meghalaya, with a focused lens on the East Khasi district, leading to an in-depth
visual analysis of a region of Umiam River basin, which is the selected study
area.

The Meghalaya State Map, in shade of blue, outlines the expansive geography of
the state, offering a wide-angle view of its position within the region. The East
Khasi District Map, highlighted in purple, zooms in on this particular district
within Meghalaya’s borders, a place of significant interest for its unique climatic
conditions and diverse ecosystems.

The Study Area Map provides a detailed topographic representation of our

selected Area of Interest(AOI) which is a region of the Umiam river basin, using
a color gradient to distinguish between various elevations and landforms. The
legend breaks down the landscape by degrees of slope from ‘Very gentle slope’
to ‘Very steep slope,’ with precise angle measurements. This granularity is
essential for evaluating water dynamics, soil erosion potential, and ecological
sustainability.

From a remote sensing and environmental monitoring perspective, this map is

pivotal for, Monitoring shifts in land use patterns over time, Measuring vegetation
health through indicators like NDVI and NDWI, Crafting water conservation
strategies by examining hydrological formations, Directing ecological
preservation efforts by locating critical or vulnerable ecosystems.

Moreover, the Land Use Land Cover (LULC) Map is a critical complement to
these maps, offering an illustrative overview of the various human and natural
land uses and vegetation types across Meghalaya. It plays a key role in
understanding the impact of human interventions on natural terrains and is
essential for developing policies that safeguard Meghalaya’s environmental
health.

These maps serve as vital instruments for academics and decision-makers alike,
fostering well-informed strategies for sustainable resource management within
Meghalaya’s distinctive landscape.
Fig1: Meghalaya state map(top left), East khasi district map(top right),
Study area map:region of Umiam basin(down)
Fig2: Google earth imagery of the study area

Fig 3: Toposheet from Survey of India with AOI

 OBJECTIVE
This study aims to create a detailed Land Use and Land Cover (LULC) change
detection map for the year 2013-2023 of a targeted area through remote sensing
techniques. This process involves a thorough analysis of satellite and aerial
imagery to distinguish and categorize the various natural and man-made features
of the land, such as vegetation, waterways, and urban developments, as well as
their uses in agriculture, living spaces, and business.

The LULC map will provide a crucial visual representation of the region’s current
landscape, offering insights into how the land is being used and its natural state.
Additionally, by comparing imagery from various time points, a change detection
map will be developed that highlights areas where significant environmental
changes have occurred, including changes in forest cover, urban growth, or
farming activities. This map is essential for monitoring how the landscape
changes over time.

Land Use Land Cover map is useful for those involved in environmental planning
and resource management, aiding in the development of policies that support
sustainable growth and conservation. They will enhance our understanding of
how land use and land cover are changing and support our commitment to
maintaining ecological balance.

DATA SOURCE
Landsat 8 imageries of 2013 and 2023 were used for mapping LULC classes of a
small region of Umiam river basin. Digital Elevation Model (DEM) dataset was
also used to generate the boundary of the study area.
All the images were downloaded from USGS Earth Explorer website. Also an
India shapefile for the creation of state and district map and a Survey of India
topographic map was also used to obtain additional info about the region.

METHODOLOGY

The whole process is showed in the following flowchart:

Chart 1

I) Pre-processing
DEM data was used to extract boundary of the study area. Landsat imageries of
2013 and 2023 were downloaded from USGS Earth Explorer and Supervised
classification was done using ArcGIS 10.8 Software.The 7 bands of satellite
imagery downloaded were undergone band composition and standard false colour
composite(FCC) was created for mapping. This was also done in ArcGIS 10.8
Software.

Fig 4: FCC of the composited bands

II) Land cover classification scheme:
To prepare LULC map a classification scheme consisting of five classes was
considered. Those are: River, builtups/barren land, very dense forest, vegetation,
and small water bodies.

CLASS NAME DESCRIPTION

River Areas covered by river

Builtups/ Barrenland Areas with settlements or devoid of

vegetation
Very dense forest Land with more canopy density

Vegetation Land with normal vegetation

Small water bodies Areas with small water bodies

Table 1: Description of different LULC classess

III) Post-Processing:
Supervised classification:- After the preparation of the classes Maximum
likelihood classification was done using ArcGIS 10.8 Software to map all classes.
Google Earth imagery and toposheets were observed carefully for greater detail.
This classification was done for both 2013 and 2023 satellite imagery so that it
can produce 2 maps showing LULC change detection.

Fig 5: Raster image of study area after classification (2013)

Fig 6: Vector image of study area after classification (2013)

Fig 7: Raster image of study area after classification (2023)

Fig 8: Vector image of the study area after classification (2023)

RESULTS AND ANALYSIS

LULC change detection map covering 5 major classes: river, builtups/barren land,
Very dense forest, vegetation, and small water bodies of year 2013 and 2023 are
shown in Fig 9.
Fig 9: LULC 2013 and 2023
The percentage of distribution of each class in the AOI for 2013 and 2023 is
shown in Fig 10.
Fig 10: Percentage of distribution of classes in AOI

Results from the classified maps indicate that in 2013 area occupied by river was
about 2%, builtups/barrenland was 16%, very dense forest covered 39%,
vegetation was 41% and small water bodies was 1%. On the other hand, in 2023
the river was 2%, buitups/barrenland was 20%, very dense forest was 43%,
vegetation was 30% and small water bodies were 5%.

From moving on from 2013 to 2023, there has been an increase in the area were
builtips/barrenland occurred (16% to 20%). Also there has been an increase in
small water bodies (1% to 5%). This can mean that there has been an increase in
small water bodies since the settlements or builtups increased. It might be because
of the construction of artificial water systems like manmade ponds, lakes etc. As
the biltups/barrenland increased we can see a decrease in the area covered by
normal vegetation (from 41% to 30%). Whereas, there has been a slight increase
in the cover of very dense forest (39% to 43%). This may be happening due to
the presence of 2 major villages nearby, i.e., Mausynram and Cherrapunji.
Recently, they have been facing a lot of urbanization.

REFERENCES
➢ Mishra, P. K., Rai, A., Rai, S. C. (2019). Land use land cover change
detection using geospatial techniques in Sikkim Himalaya, India. The
Egyptian Journal of Remote Sensing and Space Sciences, volume 23, issue
2, 133-143. Land use and land cover change detection using geospatial
techniques in the Sikkim Himalaya, India - ScienceDirect