LAND USE LAND COVER CHANGE IN PHEWA WATERSHED AND

ITS LAKE AREA CHANGE STATUS: A GEO-SPATIAL APPROACH

Mr. SUDIP RAJ REGMI

TRIBHUVAN UNIVERSITY
INSTITUTE OF FORESTRY, POKHARA CAMPUS, POKHARA
NEPAL

A THESIS SUBMITTED FOR THE PARTIAL FULFILLMENT OF THE

REQUIREMENTS FOR THE BACHELOR OF SCIENCE IN
FORESTRY

MAY 2019
LAND USE LAND COVER CHANGE IN PHEWA WATERSHED AND
ITS LAKE AREA CHANGE STATUS: A GEO-SPATIAL APPROACH

RESEARCHER
Mr. Sudip Raj Regmi
Institute of Forestry
Tribhuvan University
Pokhara, Nepal

ADVISOR
Mr. Mahendra Singh Thapa
Assistant Professor
Institute of Forestry, Pokhara

TRIBHUVAN UNIVERSITY
INSTITUTE OF FORESTRY
POKHARA, NEPAL

A THESIS SUBMITTED FOR THE PARTIAL FULFILLMENT OF THE

REQUIREMENTS FOR THE DEGREE OF BACHELOR OF SCIENCE IN
FORESTRY

MAY 2019
© SUDIP RAJ REGMI
Corresponding E-mail: sudipregmi7717@gmail.com
Tel: 9846453017

TRIBHUVAN UNIVERSITY
INSTITUTE OF FORESTRY
POKHARA CAMPUS
POST BOX-43
POKHARA, NEPAL
Tel: +977-61-430469/431689
Fax: +977-61-430387
Website: www.iofpc.edu.np

CITATION
Regmi, S.R. 2019. “Land use Land cover change in Phewa watershed and
its Lake Area change Status: A Geo-Spatial Approach” B.Sc. Forestry
Thesis, Tribhuvan University, Institute of Forestry, Pokhara Campus, Nepal.

i
 DECLARATION

I, Sudip Raj Regmi, hereby declare that this thesis entitled " Land use Land
cover change in Phewa watershed and its Lake area change Status: A Geo-
Spatial Approach” is my original work and all the sources of information used
are duly acknowledged. I have not submitted it or any part of it to any other
academic institutions for any degree for any evaluation.

………………………………

SUDIP RAJ REGMI

B.Sc. Forestry
Institute of
Forestry
Pokhara Campus, Pokhara

ii
 ACKNOWLEDGEMENT
I would like to acknowledge a number of people and institutions that have contributed in
different ways toward accomplishing this research study. Without their help and support, this
research would not have been possible to accomplish.

First, I would like to express my cordial thanks and sincere gratitude to my respected Advisor
Assistant Professor Mahendra Singh Thapa for his valuable guidance, constructive
comments and encouragement during the research work and in the preparation of thesis Paper.
Without his regular guidance and supervision, this work would not have been possible. His
suggestions and support are greatly appreciated. .

Likewise, I am grateful to Associate Professor DR. Raju Raj Regmi and Professor Yajna
Prasad Timilsina for their valuable guidance, comments and suggestions. I would also
acknowledge every help and support received from the campus administration and the Institute
Of Forestry during the course of the study.

I would like to extend my profound thankfulness and appreciation to my brother Saroj Poudel
who had supported me during field visit. My special thanks go to my friends Sharad Thapa and
Bibek Subedi for their technical support during data analysis.

I am equally obliged to the residents of the study area who gave me their precious time in
answering the questionnaire and positively responding to many queries.

I am thankful for MOITFE (Ministry Of Industry, Tourism, Forest and Environment) of

Province 4 for providing primary and secondary data related to my research.

My heartfelt of thanks goes to my friends Bhuwan Singh Bist, Milan Budha,Pawan Karki and
Dipendra Pokhrel Dai for their help and continuous support during my study and to my
classmates (2071-2075 batch) for their moral support during my study.

Last, but not the least, my deep gratitude goes to my family members for their constant support
and interest into my studies. Without their encouragement, patience and understanding this
thesis would not have been possible.

iii
 ABREVIATIONS AND ACRONYMS

AOI Area of Interest

CBS Central Bureau of Statistics
DEM Digital Elevation Model
DSC Department of Soil Conservation
ERDAS Earth Resources Data Analysis System
et al., et alli and associates
EMR Electromagnetic Radiation
EMS Electromagnetic Spectrum
GIS Geographical Information System
GPS Global Positioning System
GO Government Organization
Ha Hector
JICA Japan International Cooperation Agency
LRMP Land Resource Mapping Project
LULC Land Use Land Cover
LULCC Land Use Land Cover Change
MLC Maximum Likelihood Classifier
NGO Non-Government Organization
OLI_TIRS Operational Land Imager and Thermal Infrared Sensor
PAN Panchromatic
PCA Principal Component Analysis
PCC Post Classification Comparison
Pixel Picture Element
RS Remote Sensing
SILT Siltation
TM Thematic Mapper
TOA Top of Atmosphere
UTM Universal Transverse Mercator
USGS United States Geological Survey
WGS World Geo-dating System

iv
 ABSTRACT
Natural and anthropogenic processes have resulted in land cover changes that have huge
impact on the global environment and ecosystem. Land use land cover change (LULC)
has thus become a central component of current strategies in managing natural
resources and monitoring environmental changes. RS and GIS have been playing a
significant role in monitoring and detecting LULC dynamics. Hence, this research was
carried out to assess the extent of LULC changes during 2010 and 2018 using
temporal satellite imageries, compute the rate of change in area of Phewa lake and
explore the Drivers of LULC change and its lake area change status of Phewa
Watershed. Landsat images TM and OLI were used for 2010 and 2018 respectively.
Radiometric, Atmospheric and Sun angle correction were carried out followed by
supervised classification using Maximum likelihood algorithm to classify the image
into Forest, Agricultural land, Barren land, Urban areas and Water bodies. Change
detection for two periods was performed using post-classification comparison method.
The area computation for lake area for two different periods was computed in ARC
GIS using Shape files. Purposive Household Survey (N=60), Key informant Survey
(N=5), Focus group discussion (N=4) and Direct field observation was done for
exploring Drivers of LULC and Lake area change status in Phewa Watershed.
Secondary data was collected from newsletters, journals, research papers, published
and unpublished articles, reports from thesis and reports available at internet and library
of Institute of Forestry, Pokhara.Remarkable increase in forest area (from 3882.42
ha/32.18% to 4565.88 ha/38.08%), urban areas (from 600.03 ha/5% to 993.33
ha/8.28%) and barren lands (from 464.31 ha/3.87% to 1065.87 ha/8.89%) while there
is significant decrease of agricultural land (from 6573.78 ha/54.83% to 4914.72
ha/41%) and water bodies (468.99 ha/3.91% to 449.73 ha/3.75%) during 2010 and
2018. The overall classification accuracy for 2010 and 2018 was 92.44% and 88.1%
respectively. The rate of change in area of Phewa Lake between 2010 and 2018 was -
0.61% with Periodic annual increment -2.577 ha. The Drivers responsible for LULC
in Phewa Watershed are Alternative form of energy, Community forestry, Promotion
of private forestry, Migration for foreign employment, Lack of market Price of
agricultural products, Road construction, Soil erosion and Population pressures. Lake
area is found to be decreased due to Sedimentation, Pollutants, Encroachment, Road
construction and Water hyacinth. Thus, bioengineering and other soil conservation
measures, employment opportunities, planned road networks, encroachment
management, boundary demarcation of lake and water hyacinth removal are
recommended for the improvement of the Phewa watershed. The research findings will
not only be useful for formulating effective management policies but also to aware
scientific communities and stakeholders about the Drivers of LULC change and adopt
ways and means to control the negative effects.

Keywords: Land use land cover change, RS and GIS, Phewa Watershed, Landsat, Supervised

v
 Table of Contents
CITATION ............................................................................................................................ i
DECLARATION……………………………………………………………………………...ii
ACKNOWLEGEMENT……………………………………………………………………...iii
ABBREVIATIONS AND ACRONYMS ............................................................................. iv
ABSTRACT…………………………………………………………………………………...v
CHAPTER 1: INTRODUCTION .......................................................................................... 1
1.1 BACKGROUND..................................................................................................................1
1.2 PROBLEM STATEMENT AND JUSTIFICATION ....................................................... 5
1.3 OBJECTIVE OF THE STUDY ....................................................................................... 7
1.3.1 General objective .......................................................................................................... 7
1.3.2 Specific objectives ........................................................................................................ 7
1.4 LIMITATIONS ............................................................................................................... 7
CHAPTER 2: LITERATURE REVIEW ............................................................................... 8
2.1. Concept and definition ................................................................................................... 8
2.1.1. Land Cover and Land Use ........................................................................................... 8
2.1.2. Spatial and Temporal Change ...................................................................................... 9
2.1.3. LULC Mapping Methods using RS and GIS ................................................................ 9
2.1.4 Image Classification ................................................................................................... 10
2.1.5. Maximum likelihood classifier .................................................................................. 11
2.1.6 Assessment of Accuracy of LULC Mapping ............................................................... 12
2.1.7 Change Detection and Analysis .................................................................................. 13
CHAPTER 3: MATERIALS AND METHODS .................................................................. 15
3.1 Description of Study Area ............................................................................................. 15
3.2 Data collection .............................................................................................................. 18
3.2.1 Primary data collection ............................................................................................... 18
3.2.1.1. Reconnaissance survey ........................................................................................... 18
3.2.1.2 GPS Data................................................................................................................. 18
3.2.1.3. Household Survey .................................................................................................. 18
3.2.1.4. Key-Informant Interview ........................................................................................ 18
3.2.1.5. Focus group discussion ........................................................................................... 18

vi
3.2.1.6. Direct Field Observation ......................................................................................... 18
3.2.2 Secondary Data .......................................................................................................... 19
3.2.2.1 Satellite images ....................................................................................................... 19
3.2.2.2 Digital Elevation Model........................................................................................... 19
3.2.2.3 Training samples ..................................................................................................... 19
3.2.2.4. Other Secondary Data Sources ................................................................................ 19
3.3 Data Analysis ................................................................................................................ 19
3.3.1 Computer software used for the Analysis .................................................................... 19
3.3.2 Digital image processing ............................................................................................ 20
3.3.2.1 Sub setting the Satellite images ................................................................................ 20
3.3.2.2 Geometric correction of the satellite images............................................................. 20
3.3.2.3 Radiometric, atmospheric and sun angle correction of the satellite images ............... 20
3.3.3 Land use land cover classes classification ................................................................... 21
3.3.4 Change detection and analysis .................................................................................... 21
3.3.5 Rate of lake area change ............................................................................................. 22
3.3.6 Accuracy assessment .................................................................................................. 23
3.3.7 Social analysis ............................................................................................................ 23
CHAPTER 4: RESULTS AND DISCUSSION ................................................................... 24
4.1 Temporal LULC inventory and Change Analysis .......................................................... 24
4.1.1 Land use land cover of 2010 ....................................................................................... 24
4.1.2 Land use land cover of 2018 ....................................................................................... 25
4.1.3 Land use land cover dynamics of 2010 and 2018 ........................................................ 26
4.1.4 Accuracy assessment .................................................................................................. 28
4.1.5 Change detection of LULC of 2010 and 2018................................................................29
4.2 Computation of the rate of change in area of Phewa Lake .............................................. 31
4.3 Drivers of land use land cover change and lake area change status of Phewa watershed . 32
4.3.1 Drivers responsible for increasing Forest area during 2010 and 2018 ......................... 32
4.3.2 Drivers responsible for decreasing agricultural land during 2010 and 2018 ................ 34
4.3.3 Drivers responsible for increasing barren land during 2010 and 2018 ........................ 36
4.3.4 Drivers responsible for increasing urban areas during 2010 and 2018...........................37
4.3.5 Drivers responsible for decreasing lake area during 2010 and 2018 ............................ 38
CHAPTER 5: CONCLUSION AND RECOMMENDATIONS ........................................... 40
5.1. CONCLUSION ............................................................................................................ 40

vii
5.2. RECOMMENDATION ................................................................................................ 41
REFERENCES ................................................................................................................... 42
ANNEXES.......................................................................................................................... 48
ANNEX-1 ........................................................................................................................... 48
ANNEX-2 ........................................................................................................................... 50
Flow chart of research design .............................................................................................. 50
SOME PHOTO PLATES .................................................................................................... 51

List of Figures
Figure 1 Maximum likelihood classifier……………………………………………………..12

Figure 2 Map of the study area……………………………………………………………….17

Figure 3 Change detection process…………………………………………………………...22

Figure 4 LULC 2010…………………………………………………………………………25

Figure 5 LULC 2018…………………………………………………………………………26

Figure 6 Bar Diagram showing LULC areas during 2010 and 2018......................................27

Figure 7 Change/no change Map……………………………………………………………..30

Figure 8 Lake area…………………………………………………………………………....31

Figure 9 Bar diagram showing drivers of forest area increase and its % of respondents........32

Figure 10 Bar diagram showing drivers of agricultural land decrease and its % of
respondents...............................................................................................................................34

Figure 11 Bar diagram showing drivers of barren land increase and its % of respondents.....36

Figure 12 Bar diagram showing drivers of urban areas increase and its % of respondents….37

Figure 13 Bar diagram showing drivers of lake area decrease and its % of respondents……38

viii
List of Tables
Table 1 Sub watershed wise Area statistics of Phewa lake watershed……………………….15

Table 2 Type of soil in Phewa watershed and respective areas……………………………...16

Table 3 Satellite image used in land use classification………………………………………19

Table 4 LULC classes used in classification…………………………………………………21

Table 5 Attribute data of 2010……………………………………………………………….24

Table 6 Attribute data of 2018……………………………………………………………….25

Table 7 LULC for 2010 and 2108……………………………………………………………26

Table 8 Accuracy assessment for 2010………………………………………………………28

Table 9 Accuracy assessment for 2018………………………………………………………29

Table 10 Change/no change matrix…………………………………………………………..30

Table 11 Change in Parameter of Phewa lake………………………………………………..31

ix
 CHAPTER 1: INTRODUCTION
1.1 BACKGROUND

A watershed is a topographically delineated area that is drained by a stream system i.e. all of
the land draining its rain, snowmelt and ground water into a stream or river (Corn 1993; Naveh,
2000). Land is one of the most precious natural resources on which all-human activities and
functions are based. It is dynamic with respect to seasons and uses under the pressures of
production. The increase in number of populations and their activities exerts pressure on
limited land and soil resources for eking out livelihoods. This is more so in developing
countries whose economy is based on primary products. The rate and kind of changes in the
use of land resources is therefore essential for proper planning, management and to regularize
the use of such resources (Gautam and Narayanan, 1983). Land is a basic resource and has a
unique character in several respects. Land is used for production, for living purpose, for
transportation, for recreation and for disposal of solid wastes. It is non-exhaustible resource
like neither mineral nor it as unlimited resource like air. It is a limited resource in the sense that
once available land is fully used up, it is not further available for other uses (Shrestha and
Malla, 1982).
Land cover corresponds to the physical condition of the ground surface, such as forest,
grassland, agriculture land, water body etc. while land use reflects human activities such as the
use of the land for different purpose such as industrial zones, residential zones, and agricultural
fields. This definition establishes a direct link between land cover and the actions of people in
their environment, i.e. land use may lead to land cover change (Phong, 2004). LULC change
refers to quantitative change in the area (increase or decrease) of a given type of land use or
land cover whereas change detection refers to discerning the changed areas on two registered
remote sensing images at different times (Wafa,Hussein and Irmgard, 2009).
LULCC is a dynamic and continuous process, which causes major environmental changes
globally (Emilio, 2010). So, it has been considered as an important research field for
environmental monitoring researchers, planners, geographers and for policy makers
(Alkharabshen et al., 2013).
The hills and mountains of Nepal are the watershed areas with most eco-fragile areas and poor
agricultural potential due to their steep slopes, fragile mountain geology and poor quality soil.
Studies carried out in various parts of the country (Rimal et al .,2018a, Rimal et al., 2018b)
point out the problems such as loss of water quality, forest depletion, land degradation,

1
improper aquatic ecosystem management, air pollution and food security in the country related
to the sustainable management of the watershed resources.
Many studies found that land use can greatly affect the intensity of runoff and soil erosion
(Mohammad and Adam, 2010). Land use and land cover classes are the outcome of interaction
between man and environment. Land use change is a dynamic process with reference to time
and space. Human induced phenomena such as deforestation, shifting cultivation, and
cultivation in marginal lands are responsible for land use change in the Himalaya region
(Awasthi et al., 2002).
Land use and land cover change has important environmental consequences through their
impacts on soil and water, biodiversity, microclimate, methane emission, reduced CO2
absorption and hence contribute to watershed degradation (Lambin et al., 2001). Changes in
land-use and land cover have impacts on soil and water quality, biodiversity, and global
climatic systems and, thus, have important consequences on natural resources (Awasthi et al.,
2002). Increased consciousness on these impacts enhanced their estimating, forecasting and
modeling at the global, regional or watershed scales (Chen et al., 2001). A change in the land
cover of an area can negatively affect the potential characteristics of the area, which ultimately
leads to land degradation and loss of productivity (Zendu et al., 2016).
Some of the land uses are directly related to cultures, and social and economic conditions of
the people (Vink, 1975). Land Use and Land Cover (LULC) changes are one of the most
important and easily detectable indicators of change in ecosystem and livelihood support
systems (Gilani et al., 2014). Its dynamics vary according to its scale (Keyser and Kaiser
,2010).

RS/GIS technologies can greatly facilitate the collection, analysis and presentation of resource
data (Gautama, 2007). Repeated satellite images and aerial photos are useful for visual
assessment of natural resource dynamics occurring at a particular time and space, physical
features such as land use, soils, vegetation, stream networks, and landforms at different time
scales (Awash, 2004). Satellite images and aerial photos are useful for quantitative evaluation
of LULC changes over time (Ballad et al., 2007).
Interpretation of aerial photographs or satellite imaginary taken at various intervals provides
valuable information of physical features such as land use, soils, vegetation, stream networks
and landforms at different time intervals (Borough and McDonnell, 1998). The capability of
GIS to analyze temporal and spatial data helps in quantifying the land use changes. In areas of
rugged topography and poor accessibility, remote sensing is a valuable tool for monitoring the
2
spatial and temporal changes. Due to spatial nature of watershed parameters, remote sensing
combined with GIS has proved effective for analyzing, storing, retrieving such biophysical and
socio-economic data (Awasthi et al., 2002, Sidhu et al., 2000). Remote sensing for land cover
mapping and change detection, particularly in areas where due to accessibility, spatial
extension or other factors, the conventional means of ground survey are not sufficient, is
considered by several authors as having great potential and as extremely valuable tools (Iowan,
C et al., 1999).
Lake is any relatively large body of slowly moving or standing water that occupies an inland
basin of appreciable size. Definitions that precisely distinguish lakes, ponds, swamps and even
rivers and other bodies of non-oceanic water are not well established. It may be
said,however,that rivers and streams are relatively fast moving; marshes and swamps contain
relatively small in comparison to lakes. Geolgically defined, lakes are temporary bodies of
water (Britannica online,Retrived 2008-06-25).There are a number of natural processes that
form lakes. The advance and retreat of glaciers can scrape depressions that accumulate water
and form lakes. A recent tectonic uplift of a mountain range can create bowl-shaped
depressions in the surface where water accumulates. Lakes can also form by means of
landslides or by glacial blockages (EskoKuusisto and VeliHyvarinen, 2000). Most lakes are
geologically young and shrinking since the natural results of erosion will tend to wear away
the sides and fill the basin. The lake may be infilled with deposited sediments and gradually
become a wetland such as a swamp or marsh (Thomas V.Cech, 2009).
Land use change influenced by human activities plays a significant role in delivering and
depositing sediments in the river and lakes (Sthapit, 1998). Lake and river are the sources of
irrigation, drinking water and most importantly, they are sources for poor people.
Sedimentation of lakes reduces both effective depth and surface area of the lake, which
ultimately reduces the effective life of the lake (Awasthi et al.,2007, Sthapit and Balla, 1998).
Phewa lake watershed is one of the prominent tourist destination of western Nepal. Phewa
Lake, one of the lakes having national significance, and has been one of the prominent tourist
attractions of Pokhara and is one of lake cluster of Pokhara valley (included in Ramsar site in
Feb 2, 2016).Built up land and agricultural lands occupy the majority of the flat and gently
sloped area and forests account for all the remaining land of the watershed (Regmi and saha
2015). The forests and biodiversity reflect the climatic and altitudinal variation of the
catchment area, where subtropical and temperate forests are found in the lower belt and in the
upper catchment respectively (JICA/SILT, 2002).It is believed that the over exploitation and
overused of the resources without proper planning have affected the land use pattern of Phewa
3
lake watershed. Historical LULC pattern provides valuable information for the evaluation of
complex causes and responses in order to predict future trends of LULC dynamics (Fakir
Alemayehu, 2009).
Studies have shown that there remain only few landscapes on the earth those are still in their
natural state. Due to anthropogenic activities, the earth surface is being significantly altered
in some manner and man’s presence on the earth and, his use of land has had a profoudal
effect upon the natural environment thus resulting into an observable pattern in land use land
cover over time. The land use land cover pattern of a region is an outcome of natural and
socio-economic factors and their utilization by man in time and space . The land cover changes
due to human land use activities are regarded as the main reason for global environment
change, so the study on them become the forefront and hot spots of research to scholars (Lum
et al.,2002).

Various factors contribute to LULC change in Nepal including population growth, highly
uneven economic development, and government policies favoring urban-centric economic
growth (Rimal et al., 2018a, Rimal et al. 2018b; Jorgensen et al., 1989). As the issues
associated with the watershed conservation and management in other parts of the county, there
are many concerns associated with the LULC in the Phewa watershed as well. The Major
threats to Phewa watershed include extreme lake-encroachment, accelerated eutrophication,
invasive species, and toxic contaminations, reducing fish density, acidification and climate
change. Thus, it is necessary to conserve the lake from the watershed level. The LULC at the
upstream is adversely affecting the lake lying at the downstream areas. Thus, it is proper time
to think about the conservation of lake from the grass root level.
Therefore, for sustainable development and management of the watersheds, spatial inventories
on trends of land use land cover change and its prediction status are vital (Regmi and Saha,
2015). Scanty research works have been done under these topics by using RS and GIS in the
Phewa watershed of Nepal. So this research has attempt to assess and detect the land use land
cover change rate of lake area change during 2010 and 2018 by using RS and GIS techniques
and exploring the drivers of LULC and its lake area change in Phewa watershed. Thus,
understanding LULC dynamics of Phewa watershed play a vital role to design proper land use
planning for sustainable productivity and benefits.

4
1.2 PROBLEM STATEMENT AND JUSTIFICATION

Changing land use has also been shown to influence weather patterns (Stohlgrenet al., 1998)
and the generation of stream flow (Bronstertet al., 2002; Weng, 2001). LULC change has been
shown to have negative effects on stream water quality (Zampella et al., 2007; Tang et al.,
2005), quantity (White and Greer, 2006) and stream ecosystem health (Wang et al., 2000).
Change in land use and land cover impacts both environmental quality and quality of life.
Changes in habitat, water and air quality and the quality of life are some of environmental,
social and economic concerns associated with land use and land cover change. Sustainable land
resource management can be managed using accurate knowledge of land use land cover
features and relative risk of environmental hazards. Land use change detection and lake area
change is, therefore, a critical requirement for the assessment of potential environmental
impacts and developing effective land management and planning strategies. Assessment of
LULC and it changes over time from a reliable database is crucial for sustainable development.
Land cover change is one of the most important components of global change (Lambin et al.,
2001) and affects many parts of human environment systems. Water quality parameters such
as pH, electrical conductivity, total dissolved solutes, major ions and trace elements in various
aquatic systems have been closely linked to the proportions or types of land use within a
watershed (Lenat and Crawford, 1994).

Phewa lake which belongs to Phewa watershed is subjected siltation due to high bank erosion
in the stream particularly Harpan khola, intense rain during monsoon and intensive land use
practice without considering the soil conservation measures. Soil erosion, deforestation,
unplanned rural road construction and rapid changes in LULC is mainly degrading Phewa lake
watershed (Regmi and saha 2015).
`Lake holds 3% of available water of Nepal. Lake area shrinkage has been severe in Nepal
throughout all ecological regions.Phewa Lake, one of the lakes having national significance.
Despite different conservation program, (Heyojoo et al., 2009) has estimated that annual
siltation rate has a range of about 175,000-225,000 m3. At this rate, the terminal silt trap portion
will be separated from the main lake by next 20-25 years and the lake will be proper “dead” by
next 135-175 years, assuming loss of 80% water volume (Sthapit and Balla, 1998). Different
factors are causing the lake area change. The list out of such casing factors is necessary to
prepare appropriate management strategy. Hence, this research will have attempt to list those
causes and will expect to fulfill the existing gap between data regarding cause of area reduction
due to multiple factors.

5
There are several conservation and management issues of Phewa Lake including restoration of
water quality, reduction of sedimentation load, eco-zoning of lake shoreline, conservation of
aquatic biodiversity, promotion of eco-tourism and improved institutional capability of lake
resource management. Past studies carried out by Brandon and Bottomley (1998), Chen (2000),
Diouf and Lambin (2001), Mendoza S. and Etter R. (2002), Vance and Geoghegan (2002) have
emphasized the importance of investigating landforms dynamics as a baseline requirement for
sustainable management of natural resources.
Remotely sensed data set are emerging as better choice for managers to observe spatially
explicit changes over the time. Satellite based remote sensing offers additionally the possibility
of acquiring information on regular basis, essential in application where a high repeat
frequency is required.
Indeed, attempt has been made to document Land use land cover change and lake area change
in the past with the aid of traditional method of survey and mapping. In recent times, the
assessment of land use land cover change and lake area change requires more powerful and
sophisticated system such as GIS and RS, which provides a general extensive Synopic coverage
of larger areas.
The conservation of Phewa watershed is a serious issue. The watershed has been degrading
year by year. The lake area has been shrinking, agricultural land in the watershed has been
decreasing and encroachment is increasing day by day. At the same time, several institutions
have been working to manage the watershed. However, the exacts data of dynamics of Phewa
watershed has not been seriously assessed especially using RS techniques. In addition, causes
of depletion and degradation of watershed also are not analyzed and management issues are
another important challenge. It is important to continuously monitor the watershed in order to
understand its trend over time and to manage it effectively. Thus, this research is essential to
know the trend of LULC in Phewa lake watershed. Therefore, this study can provide important
scientific reference to formulate policies, planning strategies, information on potential
environmental impacts and effective land management for the sustainability of Phewa
watershed and Phewa Lake at local as well as national level in near future.

6
1.3 OBJECTIVE OF THE STUDY

1.3.1 General objective

The General objective of the study is to assess Land use land cover change of Phewa
Watershed and its Lake Area change status by using Geo-Spatial Approach.

1.3.2 Specific objectives

The Specific Objectives of the study are:

 To assess the extent of LULC changes during 2010 and 2018 using temporal satellite
imageries.

 To compute the rate of change in area of Phewa lake.

 To explore the drivers of land use land cover change and its lake area change status of
Phewa watershed.

1.4 LIMITATIONS
Some limitations are unavoidable in a research, this could be time or access to relevant
information or could be landscape of the study area. Some of the limitations that are faced
during the study are as follows:
 Topographic shadows such as layover and forecasting effect in the satellite images
were a great hindrance in the image classification.
 The satellite imageries of 30 m resolution, which was not good enough to classify
the image more accurately.
 Due to lack of funding, the research was largely constricted with huge limitations.
 Good resolution images made difficult to trace exact area of the lake and so, areas of
lake may slightly differ from true area of the lake.
 The response from the people about drivers of LULC may be biased as per the
situation.

7
 CHAPTER 2: LITERATURE REVIEW
There are several studies conducted in Nepal in relation to land use change and forest
degradation (Awasthi and Balla, 2000; Gautam et al., 2003; Gerrard and Gardner, 2002). An
in-depth study in two watersheds of western Nepal by Awasthi et al., (2002) accounted for land
use changes through the transitions among the land use classes. Interestingly, they reported that
significant area under agriculture in the base year 1978 was abandoned and covered by shrub
and bushes. The Phewa lake watershed constitute forest (44%), agricultural land (39%), urban
and wetland area (5%), pasture and barren land (5%), lake area (4%) and shrub land (3%)
(DSC, 1994).Land use has an impact on the hydrological regime and quality of water
downstream. The importance of this impact varies with the type of land use, the size of the
watershed, climate, soil characteristics, topography, geology, etc. The forest cover in Phewa
Lake Watershed decreased from 1978 to 1998 by annual 2.4% of total forest land
(www.gisdevelopement.net/application/natural resource management).
Remote sensing for vegetation and land cover mapping and change detection, particularly in
areas where due to accessibility, spatial extension or other factors, the conventional means of
ground survey are not sufficient, is considered by several authors as having great potential and
as an extremely valuable tool (Xiuwan, C. et al., 1999; Turker and Derenyi, 2000). The
potential of using satellite data to detect and characterize changes in forest cover depends on
the ability to quantify temporal effects using multi-temporal data sets (Bauer et al
.,2000).Remotely sensed data sets are emerging as a better choice for forest managers to
observe spatially explicit changes over the time. Satellite based remote sensing offers
additionally the possibility of acquiring information on a regular basis, essential in applications
where a high repeat frequency is required (Pathirana, 1999 and Wyatt, 2000). Satellite remote
sensing data has been used in Nepal since past two decades in specified areas with limited
application. Forestry sector is one of the main application areas where this technology has been
using from the beginning. Satellite remote sensing is one of the viable techniques to monitor
the changing pattern of forest cover and Maximum likelihood classification is the most
common supervised classification method used for land cover land use change with remote
sensing image data (Richard, 1995).
2.1 Concept and definition
2.1.1 Land Cover and Land Use
Land cover refers what covers the surface of earth and land use describes how the land
is used. Land cover is the combination of both biotic and abiotic components on the surface of

8
Earth and is one of the basic properties of earth‘s ecosystem. These components are focused in
three aspects (Turner et al., 1994). The first lies in the interaction of land cover with the
atmosphere, which lead to regulation of the hydrologic cycle and energy budget, and as such
is needed in both for weather and climate prediction (DE Fries et al., 2002). Secondly, land
cover plays an important role in the carbon cycle acting both sources and sinks of carbon.
Lastly, land cover provides food, fuel, timber, fiber, and shelter resources for human
population, and serves as a critical indicator of other ecosystem services. Information on land
cover is fundamental to many national/global applications including watershed management
and agricultural productivity. Therefore, the need to monitor land cover is derived from
multiple intersecting needs, including the physical climate, ecosystem health, and social needs.

2.1.2 Spatial and Temporal Change

Land covers undergo changes due to natural or man-made causes over time. Spatial and
temporal dimensions characterize changes. Temporal extent indicates the period of time when
the change in any land cover takes place. The area in which the change happens defines spatial
locations and extent. Spatial-temporal phenomena can be characterized by location, time and
attributes. The change undergoes mainly through three process i.e. basic process,
transformation process and movement process (Langran, 1992; Weir, 2002; Kandel, 2004).
The basic process results in as appearance of new feature, or no change. Transformation
process reveals either expansion of the feature or contraction or deformation of it. Movement
process also results in as translation, rotation, or diffusion. Spatio-temporal data can be used to
detect and analysis change in any land cover feature or its attributes or both. They can be
distinguished in four main phases i.e. change detection, change quantification, change
assessment and change attribution in the analysis of image time series (Henebry et al., 2003).
Reliable information in different spatial and temporal scales can be extracted from the satellite
imagery (Roy, 2003). The basic principle of change detection using remote sensing is that
changes in land cover result in changes in radiance values (Mas, 1999).Analyzing spectral
differences in signatures of an object (Landover), change can be detected. Thus, change
detection through remote sensing can play a key role in providing spatial and temporal change
information resulted by natural and anthropogenic activities in terms of time and cost
effectiveness.
2.1.3 LULC Mapping Methods using RS and GIS
According to James Campbell (1996), RS is the practice of deriving information about
the earth’s land and water surfaces using images acquired from an overhead perspective, using

9
EMR in one or more regions of EMS, reflected or emitted from the earth’s surface. Arnoff
(1989) defined GIS as a computer-based system that provides four set of capabilities to handle
geo referenced data viz data input, data management (data storage and retrieval), manipulation
analysis and data output.
Multi-temporal satellite imagery is the most important data resources of GIS which plays a
crucial role in quantifying spatial and temporal phenomenon which is not possible through
conventional mapping (Rawal and Kumar,2015).Thus, application of RS and GIS data made
possible to study the changes in LULC of Phewa lake watershed in less time with better
accuracy.
It is a scientific technology that can be used to measure, assess and monitor important
biophysical characteristics and human activities in the earth and its surface. People responsible
for managing the earth’s natural resources and planning future development recognize the
importance of accurate, spatial information residing in a digital GIS. Many of the most
important layers of biophysical, land use/ land cover, and socioeconomic information in a GIS
database are derived from an analysis of remotely sensed data (Jensen, 2000). Remote sensing
is the only way to acquire temporal and spatially change data of such large geographic areas
over long periods.
Landsat images are among the widely used satellite remote sensing data and their spectral,
spatial and temporal resolution made them useful input for mapping and planning projects
(Sadidy et al., 2009). Over the past years, data from earth observing satellites has become vital
in mapping the earth‘s features and infrastructure, managing natural resources and studying
environmental change (Zubair, 2006).
2.1.4 Image Classification
The basic principle of multi-spectral classification is that the objects in the earth
surfaces possess different reflectance characteristic in different parts of the electromagnetic
spectrum as digital numbers. Based on this reflectance, the surface features can be categorized
into specified number of classes known as land cover classes through classification software in
terms of new thematic output image. Digital image classification is the popular and challenging
approach of remotely sensed image analysis process. Although, there are different
classification systems in existence throughout the world, they are generally not comparable one
to another and there is no single internationally accepted land cover classification method
(Latham, 2001). Therefore, determination of land cover classification method is decided
considering the purpose of the study and usually it varies according to different research

10
projects (Tateishi, 2002). There are two general classification approaches: supervised and
unsupervised. In the supervised approach, the useful information categories are defined and
examined for their spectral separability where in the unsupervised approach, spectrally
separable classes are determined and defined relative to their informational utility to form a
supervised classification scheme (Kaiser et al., 2008). Supervised classification uses the
independent information from spectral reflectance to define training data for determining
classification categories (Ratanopad and Kainz, 2006). Among the classification procedure,
supervised classification has been widely used in remote sensing applications because in
supervised classification, spectral signatures are collected from specified locations (training
sites) in the image to classify all pixels in the scene by digitizing various polygons overlaying
different land use types (Yüksel et al., 2008). Training sites are the areas defined for each land
cover type within the image. The chosen color composite image is used for digitizing polygons
around each training site for similar land use/cover. Then a unique identifier is assigned to each
known land cover type (Eastman, 2009). In practice, mostly MLC is performed assuming equal
probability of occurrence and cost of misclassification for all classes and output stage has
presented LULC map after the entire data set has been categorized (Lillesand et al., 2008).
2.1.5 Maximum likelihood classifier
The maximum likelihood classifier is one of the most popular methods of classification
in remote sensing. This classifier assigns a pixel with maximum likelihood into a corresponding
class as shown in Figure 1. The likelihood (Lk) is defined as the posterior probability of a pixel
belonging to class k (Japan Association of Remote Sensing, 1996). Supervised classification
with Maximum likelihood classifier was utilized for image classification and for the
preparation of base maps for change detection (Lillesand et al., 2004).

11
 Fig 1: Maximum Likelihood Classifier

2.1.6 Assessment of Accuracy of LULC Mapping

Classification accuracy assessment is the process of comparing the classified image to
another data source that is considered to be accurate or ground truth data. The classified
accuracy is usually measured in terms of overall accuracy, UA and PA. Story and
Congalton(1986) has described about overall accuracy, users accuracy and producers accuracy
in a journal entitled “Accuracy assessment: a user’s perspective “in following ways.
Overall accuracy is total accuracy, which is computed by dividing the total correct pixel by
total number of pixels in the error matrix. It is estimated as:
Overall accuracy=Total number of correct pixel/No.of pixel in the error matrix
User’s accuracy corresponds to error of commission, which indicates that a pixel classified on
the image actually represents the category in the ground. It is called users accuracy because a
map user is interested on how the map represents what really on the ground. It is estimated as:
Users accuracy=Total no, of correct pixel in a category/Total no.of pixel that were classified
in that category
Producer’s accuracy corresponds to error of omission, which represents the probability of a
reference pixel being correctly classified. It is called producers accuracy because the producer
of the classified image is interested on how well a specific area on the earth can be classified.
It is estimated as:

12
Producers accuracy=Total no.of corrected pixel in the category/Total no.of pixels of that
category derived from the reference data
The kappa Coefficient was derived by Cohen (1960) which measures the agreement between
classified and truth-values. It can be computed as:
K=P (A)-P (E)/1-P(E)
Where (A) is the number of times the k raters agree, and P (E) is the number of times the K
raters are expected to agree only by chance.
2.1.7 Change Detection and Analysis
To appreciate the dynamics of the ecosystem, it is necessary to monitor the vegetation
through time and determine what changes in succession are taking place. Relatively medium
to high temporal resolution satellite data is often useful for such type of study. Researchers
involved in change detection studies using satellite images data have conceived a large range
of methodologies for identifying environmental changes. Change detection procedures can be
grouped under three broad headings characterized by data transformation procedures and
analysis techniques used to delimit areas of significant changes: (1) image enhancement, (2)
multi-date data classification and (3) comparison of two independent lands cover classifications
(Mas, 1998). The enhancement approach involves the mathematical combination of imagery
from different dates such as subtraction of bands, rationing, and image regression or
PCA.Thresholds are applied to the enhanced image to isolate pixels that have changed. The
direct multi-date classification is based on the single analysis of a combined data set of two or
more different dates, in order to identify areas of changes. The PCC is a comparative analysis
of images obtained at different moments after previous independent classification.
Change detection techniques using remote sensing techniques involve the use of multi-
temporal satellite data sets to discriminate areas of land cover change between dates of imaging
(Lillesand Kiefer, 2008). It can provide up-to-date spatio-temporal information about forest
resources status that supports in making decision on appropriate intervention (policy
formulation, planning and management). The basic principle of change detection using remote
sensing is that changes in the land cover result in changes in radiance values (Mass, 1999).
Analyzing spectral differences in signatures of an object (land cover change) can be detected.
Thus, change detection in remote sensing play a key role in improving spatial and temporal
change information resulted by natural and anthropogenic activities in terms of time and cost
effectiveness. The applicability of semi-automated and object-oriented approaches for satellite
remote-sensing data has been the subject of many recent studies (Zhan, 2003).

13
Post-classification techniques have significant limitations because the comparison of land-
cover classifications for different date’s does not allows the detection of subtle changes within
land-cover categories (Macleod and Congalton, 1998). Change detection was performed using
post-classification comparison method, which produced acceptable results. The post
classification, change-detection technique of image differencing was applied on subsequent
pairs of the classified single date images so that image difference data was obtained for the
three-time interval. The classified images of the various dates on the ERDAS Imagine was
converted to vector (ESRI Shape file). Again, the vector files will be converted to the raster
grid by using Spatial Analyst extension of the Arc GIS 10.3 .
Phewa lake supplies water to generate 1 megawatt of hydropower and to irrigate
approximately 320 hectares agriculture land (Sthapit and Balla 1998). Efforts to define
shoreline were prescribed sets of government standards to regulate urban development along
shoreline but a physical collapse of Phewa dam occurred in 1974 that drained out much of lake
water causing lake area shrinking to a smaller area (Pokhrel S, 2008) resulting encroachment
of 138 hectares of land of Phewa lake area surveyed illegally during survey in 1977 (The
Kathmandu Post, Aug 4, 2012). According to a survey prepared by the central survey team of
the Ministry of Land Reform,the lake has shrunk from 1,120 hectares in the past to around 507
hectares till 2007. Similarly, an average delta formation is continued at the rate of about 2
hectares annually since 1973 so that silt trap area, depending upon the situation of Harpan
khola, main feeder of the lake, will completely filled up in between 24 to 33 years reducing
16% of the lake area (Sthapit and Balla 1998).
There are mainly two categories for LULC change: direct (proximate) driving forces
and indirect (underlying) driving forces. Direct driving forces include the immediate actions of
local people to fulfill their needs from land use (Geist and Lambin, 2002), such as agricultural
expansion, wood extraction, infrastructure expansion and other causes that change the physical
state of land cover (Meyer and Turner II, 1996). Driving forces mainly operate at the local level
(i.e. individual farms, householders, or communities). In contrast, indirect driving forces are
fundamental socioeconomic and political processes that push ‘direct causes into immediate
action on LULC (Geist and Lambin, 2002). These underlying ‘driving forces include
demographic pressure, economic status, technological and institutional factors, and can
influence LULC in combination (Geist and Lambin, 2002). Land use constantly changes in
response to the dynamic interaction between direct and indirect causes it is non-static (Lambin
et al., 2001).

14
 CHAPTER 3: MATERIALS AND METHODS
3.1 Description of Study Area
Location
The study area, Phewa watershed, is located in the south-western part of Kaski district
covering both rural and urban area. Phewa Lake watershed extends between 28°11'39" North
to 28°17'25" North latitude and 83°47'51" East to 83°59'17" East longitude. The study area
covers 119.89 Km2 with its geometrical east-west length of 18.32 km and north-south width
of 9.53 km. Phewa Lake itself covers about 4.83 km² (483 ha) areas. The variation of altitude
is from (789-2508.81m above msl) in the west at Panchase, the highest summit of the watershed
area.
Geology
The geology of Phewa Lake watershed realm is extremely complicated as mentioned
by (Yamanaka et al., 1982) in their geological explanations of the Annapurna range, the Seti
River and the Pokhara valley. The southern part of the Phewa lake watershed realm possesses
more tale-rich, red phyllitic schist (Mulder, 1978) of metamorphism than the northern part
(Fleming, 1978).
Drainage
Table 1 :Sub Watershed Wise Area Statistics of Phewa Lake Watershed
Sub Watershed Area (ha)
Harpan System 3223.40
Andheri System 2815.10

Mid sub system 2579.20

South Flowing Independent System 2150.00
North Flowing Independent System 738.00
Excluded area 483.83
Total 11989.53
(Source: Topographical Map, Survey Department, Kathmandu)
Climate
The Phewa Lake watershed, locating centrally in the subtropical climate, possesses
moderate subtropical to the cool temperature type of climate. This extreme climate is owing to
its topographical variation.
Vegetation and Wildlife

15
 The lower part the watershed have dominance of Sal (Shorea robusta), Katus
(Castonopsis indica), Chilaune (Schima wallichii), Tooni (Cedrela toona), Sisoo (Dalbergia
sissoo), Pipal (Ficus religiosa), Simal (Bombax ceiba) and Bans (Dandroclamus strictus) etc.
and in upper part Laligurans (Rhododendron aroboratum), Salla (Pinus species), Bamboo
(Dendrocalamus species) etc are the common species found. Shorea robusta occurs in the
tropical belt and Schima wallichii, Castanopsis indica predominate on the sub-tropical hills.
Forests of oak and rhododendron are common in the northern highlands. Natural grasslands
are found on the Seti river terraces and these sometimes overlap with riverrine Acacia catechu
and Dalbergia sissoo. Shifting cultivation, overgrazing, fire and lopping have resulted in the
depletion of forests. In the sub-tropical belt, forests are most vulnerable to destruction due to
human encroachment. Mostly forests have been left only on the steeper slopes. The area is rich
in wildlife. The major wild lives are rabbit, monkey, wild goat, leopard, etc. similarly, major
birds are fowl, eagle, green pigeon, dove, etc. (FRI, 2002; JICA/SILT, 2002).
Soil
The soils in the study area exhibit wide variation due to their texture, depth, stoniness,
color, drainage, moisture, organic matter, capacity exchange etc.
Table 2 Types of soil in Phewa watershed and respective areas
Types of Soil Area (ha) %
Ustochrepts Haplustalfs 991.17 8.27
Ustochrepts Dystrochrepts Haplumbrepts 4497.39 37.51
Typic and Rhodic Haplustalfs Ustochrepts 650.25 5.42
Lithic Subgroups of Typic and Ustorthents 5367.69 44.77
Phewa Lake 483.03 4.03
Total 11989.53 100.00
(Source: LRMP, 1986)

Land use land cover

Forest, Agriculture, Bush/Scrub, Waste Land and Built-up Land are the main land use
categories of the Phewa Lake watershed. The forest land use basically consists of community
forestry whereas Agriculture Land consists of field crop cultivation on both terrace and valley.
The most common gregarious natural vegetation types under tropical to temperate monsoon
climates are Schima wallichii, Castonopsis indica, Alnus nepalensis, Juniper and Pinus
roxburghii (FRI, 2002; JICA/SILT, 2002). However, the watershed is interspersed by a number
of patches of rural settlement and agricultural fields. Agricultural lands are allotted to wet and
dry crops cultivation depending upon prevailing local climatic conditions.

16
 Description of Phewa Lake

The lake is about 2 km North West from Pokhara airport and is more or less leaf-shaped
that lies on the narrow space between the Seti Valley. The lake is extended from latitude of
83°55’44”E to 83°58’10”E and from longitude of 28°11’44”N to 28°13’40”N with average
altitude of 794m from mean sea level. Water is collected from the watershed of about 123
sq.km. area through the different river and tributaries to the Phewa Lake such as Bhumdi,
Marse, Budhimul, Singare, Hadi, Sidhane, Orlang, Andheri, Pokhre byase, Harpan. The Phewa
Lake is formed by complex and rugged ridges, spurs, and valley bottoms. The hilly terrain and
valley bottoms stand out as distinct natural features in the landscape crisscrossed by a number
of irregular ridges and spurs. The climate falls under tropical to sub-tropical monsoon type with
mean annual rainfall of 4160mm and annual temperature ranging from 29.7° C to 32° C
maximum temperatures from April to June in comparison to mean temperature from 23°C to
24°C.

Fig 2: Map of the study area

17
3.2 Data collection
3.2.1 Primary data collection
3.2.1.1 Reconnaissance survey
The reconnaissance survey was carried out at the beginning of fieldwork in order to familiarize
with the study area and selecting sites for ground truth collection.
3.2.1.2 GPS Data
Sufficient GPS points was taken in the entire study area as the primary sources of data. Google
Earth/GPS for Screen digitization and locating the field site for the observation and ground
points collection for accuracy assessment of classified image was taken in the entire study area.
3.2.1.3 Household Survey
The household level questionnaire survey was conducted in the selected area on the basis of
collected official records, which was the main source of data. Both open and close-ended
questionnaires was used to acquire the objective driven result in both qualitative and
quantitative way. Sampling was executed in a purposive manner for the selections of
household after consultation with the various stakeholders in the sense that respondents should
know the terms Land use land cover change and must be living in the Phewa watershed for at
least 8 years. Household survey was conducted with 60 households at the study area in a
representative manner.
3.2.1.4 Key-Informant Interview
Both face-to-face interview and mail questionnaire was used for key informants’ survey. Semi
structured questionnaire with a mix of close and open ended questions will be done with
CFUG’s, local leaders, Teachers, Women Group, local NGO’s officials, GO’s officials to
collect key and unavoidable data about the third objective and its results.
3.2.1.5 Focus group discussion
This was done with Phewa lake Boat association, People whose livelihood dependent on
Phewa lake, People living near of Phewa lake and Phewa lake conservation commitee.The
purpose of this discussion was to cross check the information gathered individually.
3.2.1.6 Direct Field Observation
Direct site observations was made in the respective study area where observation was focused
for land use type, major land use problems including invasion, succession and sedimentation.
This also helped for the major causes of land use change. Direct Field Observation was
conducted to cross check the information that was collected during household surveys, Key-
informant interviews and focus group discussions.

18
3.2.2 Secondary Data
3.2.2.1 Satellite images
In this study, two different types of satellite images were used i.e. Landsat (TM) satellite
image and OLI satellite image. Landsat 5 TM 2010 and Landsat 8 OLI_TIRS 2018 imageries
were freely downloaded from the Earth Resource Observation System Data Center of the
United States Geological Survey (http://www.glovis.usgs.gov).

Table 3 : Satellite image used in land use classification

Satellite Time Sensor Total Temporal Spatial Swath PAN

zone bands resolution Resolution(m) width Band

Landsat 2010 TM 1-7 16 days 30*30 185 km Nil

5

Landsat 2018 OLI_TIRS 1-11 16 days 30*30 185 km 15m,

8 B-8

3.2.2.2 Digital Elevation Model

Aster DEM from earthexplorer.usgs.gov was used for boundary delineation of the
Watershed.
3.2.2.3 Training samples
For acceptable classification results, training data must be both representative and complete.
All the spectral classes constituting each information class must be adequately represented
in the training set statistics used to classify an image (Lillesand et.al. 2004). Training samples
was collected with the help of GPS during field visit and with Google Earth for land use
interpretation for the satellite image of 2018, and quantitative description of the
characteristics of each land use classes for supervised classification.

3.2.2.4 Other Secondary Data Sources

This data was collected from newsletters, journals, research papers, published and unpublished
articles, reports from thesis and reports available at internet and library of Institute Of Forestry,
Pokhara.
3.3 Data Analysis
3.3.1 Computer software used for the Analysis

19
All the statistical data were analyzed using MS excel 2016 and results were interpreted and
presented on Pi-Charts, Graphs, and Tables. The analyses of the satellite images and GIS data
correction were carried out in Arc GIS 10.3 environment. Spatial Analyst Extension tool was
used for change detection. Ground verification data were taken through GPS (GRAMIN-
Etrex30).

3.3.2 Digital image processing

3.3.2.1 Sub setting the Satellite images
The study area was separated out from the whole scene of 120 km² of the Landsat satellite

images of both dates (2010 and 2018) using DEM. Extract by mask tool of ArcGIS was used

for this process. This separated area is used as AOI for the study.
3.3.2.2 Geometric correction of the satellite images
All the images used in research were from same source. Therefore, further rectification was
not done. But the digitized map was re-projected to UTM/WGS 84, Zone 44 to match with
satellite images.

3.3.2.3 Radiometric, atmospheric and sun angle correction of the satellite images
For Landsat 8
Atmospheric correction
Calculating Reflectance value from the Satellite data
ρλ' = M ρ*Q cal + A ρ
ρλ' = TOA planetary reflectance, without correction for solar angle. Note that ρλ' does not
contain a correction for the sun angle.
M ρ = Band-specific multiplicative rescaling factor from the metadata
(Reflectance_Mult_Band_x, where x is the band number)
A ρ = Band-specific additive rescaling factor from the metadata
(Reflectance_Add_Band_x, where x is the band number)
Q cal = Quantized and calibrated standard product pixel values (DN).
Correcting the Reflectance value with sun angle
TOA reflectance with a correction for the sun angle is then

Pλ = Pλ'/sin θSE

Where:

20
ρλ = TOA planetary reflectance

θSE = Local sun elevation angle. The scene center sun elevation angle in degrees is provided
in the metadata (Sun Elevation).

Landsat-5 Thematic Mapper Radiometric Calibration

Conversion to Radiance for L1 Products (Qcal-to-Lλ)

Lλ=(LMAXλ−LMINλ)/255)*Qcal+Lminλ
Correcting the Reflectance value with sun angle

Lλ' = Lλ/Sun angle elevation

Note: All these data were derived from Metadata from respective Landsat images. Different
GIS Tools like Times, Minus and Raster calculator are used for these corrections.

3.3.3 Land use land cover classes classification

Supervised classification was performed to classify the images into different land use
changes. MLC was used for supervised classification (Lillesand et al., 2004). Land cover was
classified to the following 5 classes as Forest, Agricultural land, Barren land, Urban areas
and Water bodies.

Table 4 : LULC classes used for classification

S.N. LULC types Description

1. Forest Shrubs land,trees,grassland bushes

2. Agricultural land Cultivation in sloping mountainous areas in terraced fields

In the downstream of study area, mainly variety of paddy is
growing and irrigation is good.
3. Barren land Sandy areas, Areas exposed after landslides, flash floods
and soil erosion. Quality of soil is poor.

4. Urban areas Urban and rural human settlement areas.

5. Water bodies Lake and rivers with clear water

3.3.4 Change detection and analysis

The raster grids of 2010 and 2018 images were overlaid using Spatial Analyst on the
Arc GIS 10.3. Land use change was calculated by using raster calculator. Finally, the area
converted from each of the classes to any of other classes was computed. The analyses and

21
interpretation of different aspects of the numeric data of land use change was done on Microsoft
excel 2016. The results thus obtained are presented in the easily understandable forms such as
maps, tables and charts. The whole process of change detection is given in the figure below:

Fig 3: Change detection process

3.3.5 Rate of lake area change

• The boundary of the lake in each images was digitized and converted into shape files
and corresponding area was computed using Arc GIS 10.3. The rate of lake area
change was estimated by standard formula.

• The analysis and interpretation of different aspects of the numeric data on lake area
change was done on Microsoft excel 2016.

• The result was presented in the form of maps, tables, graphs and charts.

The Lake area change rate was assessed by the following standard formula.

Rate of Change (%) = [((b/a)^(1/n)-1)*100] (UNDP, RFDTh and FAO cited by

Lamichhane, 2008).

22
PAI=A i+n-Ai/n

Where,

a = base year data area (2010)

b = end time data (2018)

n = number of years (8 years)

A i+n = Area of (i+n)th year (2018)

Ai = Area of ith year (2010)

PAI=Periodic annual increment

3.3.6 Accuracy assessment

It is the most important aspect to assess the reliability of map. No image classification is
said to be complete unless its accuracy has been assessed. To determine the accuracy of
classification, a sample of pixels is selected on the classified image and their class identity
is compared with the ground reference data. Evaluating the quality of a classification result
is of high importance in remote sensing since it gives evidence of how well the classifier is
capable of extracting the desired objects from the image. In this study the classification error
matrix is used which is the common means of expressing classification accuracy. The second
technique used for accuracy assessment is the KAPPA analysis. KAPPA analysis is discrete
multivariate technique which calculates the producer’s and users overall accuracy, as well as
the Kappa accuracy level.
3.3.7 Social analysis
The social data related to Drivers of LULC change and its Lake area change status of Phewa
watershed from Household survey was entered on Excel 2016 and then analyzed. The results
was presented in the easily understandable forms such as tables, graphs and charts.

23
 CHAPTER 4: RESULTS AND DISCUSSION
This section describes the findings obtained from this study addressing both objectives
mentioned in chapter 1.

4.1 Temporal LULC inventory and Change Analysis

The LULC maps of each study period were prepared by digital supervised classification of
Landsat TM and OLI_TIRS satellite data using ground truth points and google earth pro. Five
LULC classes viz. Forest, Agricultural land, Barren land, Water Bodies and Urban areas were
identified and mapped following digital supervised classification.

4.1.1 Land use land cover of 2010

The classification of the Landsat TM for 2010 shows that the Agricultural land was the major
land use covering 6573.78 ha., followed by Forest , Urban areas , Water bodies and Barren
land with 3882.42 ha, 600.03 ha, 468.99 ha, and 464.31 ha respectively.

Table 5: Attribute data of 2010

Land use land cover classes Area (ha) Percentage occupied

Agricultural land 6573.78 54.83
Forest 3882.42 32.38
Urban areas 600.03 5.00
Water bodies 468.99 3.91
Barren land 464.31 3.87
Total 11989.53 100

24
 Fig 4: LULC 2010

4.1.2 Land use land cover of 2018

The classification of the Landsat OLI_TIRS 2018 shows that Agricultural land was the

major land use covering 4914.72 ha followed by Forest, Barren land, Urban areas and

Water bodies with 4565.88 ha, 1065.87 ha, 993.33 ha and 449.73 ha. respectively.

Table 6 Attribute data of 2018

LULC classes Area (ha) Percentage occupied

Forest 4565.88 38.08
Agricultural land 4914.72 41
Urban areas 993.33 8.28
Water bodies 449.72 3.75
Barren land 1065.87 8.89
Total 11989.53 100

25
 Fig 5: LULC 2018

4.1.3 Land use land cover dynamics of 2010 and 2018

Table 7: LULC for 2010 and 2018

LULC classes Area (ha) in Area (ha) in Change in Remarks

2010 2018 LULC area
(ha)
Agricultural land 6573.78 4914.72 -1659.06 Decreased
Forest 3882.42 4565.88 683.46 Increased
Barren land 464.31 1065.87 601.56 Increased
Urban areas 600.03 993.33 393.30 Increased
Water bodies 468.99 449.73 -19.26 Decreased

26
 Area coverage of different LULC classes in 2010 and
2018
7000
6573.78
6000
4914.72
5000 4565.88
Area (ha)

3882.42
4000

3000

2000
1065.87 993.33
1000 464.31 600.03 468.99 449.73
0
Forest Agricultural Barren land Urban areas Water bodies
land
LULC classes

2010 2018

Fig 6: Bar Diagram showing LULC areas during 2010 and 2018

Land use change is very prominent in middle hills of Nepal. Regmi et al; 2014 have revealed
that Urban areas , Forest and Barren land are increasing and the Water bodies and Agricultural
land are decreasing. My result is in line with him showing approximately same result. The
slight variation can be attributed to the difference in LULC class prepared.
Result explained the dynamics on land use change within the study area for the period of 8
years. In general, Agricultural land has been decreased probably due to soil erosion/floods
and shortage of labor as most of the young people migrated to cities or foreign countries.
This led people to shift the land use from Agriculture to plantation of trees in the private land.
Gautam et al., 2003 also observed the increase in forest cover on their study based on Upper
Rosi Watershed of the Middle Mountain Region of Nepal. Similar results were observed by
Balla et al., 2007 from Galaudu and Pokhare Khola Watersheds in Mid-Hill Region of Nepal.
It is worth mentioning here that the private forest has increased throughout the country (NP-
NBSAP, 2014-2020). Thus, the result of this study corresponds with the results from such
previous studies. According to UN DESA (2014) urban growth rate of Pokhara between

27
2010 and 2015 was 5.21% and Muzzini and Apericio 2013 reported that it was the largest
city of central hill which was growing rapidly. Rimal 2011 says after declaration of
Municipality and headquarter of western development region, Rapid urbanization can be
marked in Pokhara Valley. These statements justify that my findings of urban area increase
is with the line of other findings too. In addition, forest had increased due to the community
forest, use of alternative source of energy, awareness on importance of forests, activities of
Forest watchers and decreasing of illegal activities of forest. Barren land had increased due
to foreign employment, excessive use of chemical fertilizers, soil erosion and rural road
construction. Water bodies had decreased due to sedimentation, soil erosion, pollutants and
rural road construction. Urban areas had increased due to Population growth and migration
to downstream areas.
4.1.4 Accuracy assessment

Total 160 ground control points were collected ground truth position using Garmin GPS device
and google earth pro on different land use land cover for 2010 and 2018. Accuracy assessment
was done by using Confusion or Error matrix. The overall accuracy for 2010 was 92.44% with
Kappa coefficient 0.9. So, Kappa of 0.9 means there 90% better agreement than by chance
alone in 2010.The overall accuracy for 2018 was 88.10% with Kappa coefficient 0.84.
Therefore, Kappa of 0.84 means there 84% better agreement than by chance alone in 2018.

Table 8: Accuracy assessment for 2010

LULC classes Producers Users Accuracy Overall Kappa
Accuracy (PA) (UA) Accuracy coefficient
Forest 91.3% 100%
Agricultural 100% 100% 0.9
land
Barren land 74% 94.44% 92.44%
Urban areas 100% 80.56%
Water bodies 95.45% 95.45%

28
Table 9 : Accuracy assessment for 2018

LULC classes Producers Users Accuracy Overall Kappa

Accuracy (PA) (UA) Accuracy coefficient
Forest 100% 100%
Agricultural 81.81% 75% 0.84
land
Barren land 84.61% 55% 88.10%
Urban areas 87.17% 87.16%
Water bodies 70% 100%

4.1.5 Change detection of LULC of 2010 and 2018

Change detection is a statistical technique, which is used to compile a detailed tabulation of

changes between two classified images of the same scene at different times. This technique
provides changes as pixel counts, percentage and areas. In this study ARC GIS 10.3 was used
for change detection statistics. The change detection was done for 5 LULC classes viz.,Forest,
Agricultural land, Water bodies, Barren land and Urban areas.

29
 Fig 7 : Change/no change Map

Table 10: Change/no change matrix

YEARS Forest Urban Barren Water Agricultural Total area

2010/2018 areas land bodies land in 2010
(ha)
Forest 3392.19* 3.51 32.13 ***** 454.59 3882.42

Urban areas ***** 541.08* 46.17 1.44 11.34 600.03

Barren land 5.58 84.06 220.23* 8.19 146.25 464.31

Water 2.61 31.59 30.87 400.05* 3.87 468.99
bodies
Agricultural 1165.5 333.09 736.47 40.05 4298.67 6573.78
land
Total area 4565.88 993.33 1065.87 449.73 4914.72 11989.53
in 2018 (ha)

30
NOTE: ‘*’ represents the no change in area of specific LULC used.

4.2 Computation of the rate of change in area of Phewa Lake

Fig 8: Lake Area

Table 11: Change in parameter of Phewa Lake

Dates Area(ha) PAI (ha) Rate (%) Source

2010 430.535 Not available Not available Not available

2018 409.9223 -2.577 -0.61 Imagery

Heyojoo et al ; 2014 revealed that the annual shrinkage rate of the lake as 0.46% with mean
annual decrease in lake area around 2 ha while my study shows annual shrinkage rate of the
lake as 0.61% with mean annual decrease in lake area as 2.577 ha. The slight variation can
be attributed due to Low resolution images for Digitization, different years considered and
many drivers resulting for decrease in Phewa lake can be intensive now than before.
Sedimentation and siltation has significantly contributed the shrinkage of Phewa Lake. Not
surprisingly, area of the delta in western side has increased which is main evidence for the
shrinkage of the lake. This supports the findings of many studies on estimation of
sedimentation to the lake from watershed, including (Heyojoo et. al, 2009) In the past,

31
 Phewa lake was giant covering large area in comparison to present and part of the Phewa
lake in western side has been isolated from the main lake shows that shrinkage of the lake
area is prominent towards the western part of the lake. However, very little change in area
and shape has been observed on other side of the lake because of well-defined shoreline and
little sediment deposits.

4.3 Drivers of land use land cover change and lake area change status of Phewa
watershed

4.3.1 Drivers responsible for increasing Forest area during 2010 and 2018

People response on Forest area increase

30 27
25

20 18
%

15
15 13
11 11
10
5
5

Causes of forest area increase

Fig 9: Bar diagram showing drivers of forest areas increase and its % of respondents

(a)Community Forest

Respondents response that CF had done many plantation program (especially alnus nepalensis
(utis), Schima wallichi (chilaune), Castanopsis spps (katus) and nursery activities to increase

32
greenery in the study area. Under the community forestry program, the user groups take
responsibility for managing and monitoring forest areas according to a management plan,
which includes measures for fuel wood, timber and fodder extraction on an equitable basis.
This is aided by the natural conditions, which support natural regeneration. An important point
in this respect is the control of livestock grazing in community forests and the promotion of
stall-feeding, which has a remarkable positive impact on vegetation cover. Thus, the forest
cover increase in the study area seems to be mainly supported by natural regeneration and
protection due to the positive influence of the community forestry program on the forest cover
in the area. They have also decrease the illegal felling and forest fire activities in their areas.
They have also conducted awareness programs about importance of forest resources. Various
silvicultural activities like cleaning, pruning and thinning are practiced in their respective areas.
The process of hand over of CF are increasing trend in the study area, which is confirmed by
the fact that the number of CF are increasing in Nepal.

(b)Alternative form of energy

Majority of the respondents (27%) told that previously the local people had used forest as
major source of fuel-wood and fodder, but now they had shifted their consumption pattern.
Nowadays, alternative form of energy such as energy-efficient stoves, LPG and solar panel are
in increasing trend, which is also one of the cause of forest cover increase in the study area.

(c) Promotion of Private forestry

Respondents of the study area claimed that people plant tree species like Alnus nepalensis
(utis), Schima wallichi (chilaune), and Castanopsis spps (katus) in their own land. Some also
claimed that due to increasing of barren land tree species grow due to natural regeneration
itself.

(d)Awareness on importance of forest

According to respondents, people are more aware of the ecological and economical benefits
that forests provide, such as soil erosion/landslides control, employment opportunities and thus
the activities that can damage forests are avoided by the local people in the study area.

(e)Forest watchers/Heralu
Their activities like decreasing illegal activities, Forest fire and grazing animals have
significant impact on increasing forest cover in the study area.

33
(f) Control burning
According to the respondents, this silvicultural practice have been increasing in the study area.
This has helped to increase tree species, which are benefitted by control burning which increase
organic matter (manure) which had beneficial effect on tree growth and development.

(g)Increase of forest on barren land

According to respondents, many youths had migrated to other areas to earn more money
resulting in scarcity of working manpower in the village. So, most of the households changed
their agricultural lands from cultivating agricultural crops to growing trees like Alnus
nepalensis (Utis) ,Schima wallichi (chilaune),Castanopsis speies (Katus) resulting in the
increase in the forest cover and decrease in the agriculture area in Phewa lake watershed. Also
from respondents side it was found that due to increase of barren land and decrease of
agricultural land due to various factors natural regeneration is prominent both barren and
private land which increases forest cover in the study area.

4.3.2 Drivers responsible for decreasing agricultural land during 2010 and 2018

People response on causes of agricultural land decrease

35
33
30
25
20 15 14
%

15 9 10 11
10
8
5
0

Causes of agricultural land decrease

Fig 10: Bar diagram showing drivers of agricultural land decrease and its % of
respondents

34
(a)Soil erosion/landslides

In recent years, the flood level of Harpan and Andheri khola is fluctuating. These two rivers
are major inflow of soil, sand and gravels from upstream areas to downstream areas. This
results in increasing of sediments in downstream areas which results in increasing unproductive
land (barren land increase and agricultural land decrease).

(b)Migration for foreign employment

Majority of respondents (33%) believe that paradigm of livelihood has changed .Large local
populations of Phewa Watershed are attracted to foreign employment for earning more money.
Moreover, Youths migrate to abroad countries and only women, old people and children remain
in homes. This results in decreasing of manpower in Phewa Watershed. So productive
agricultural land had decrease as these lands are left barren. Also due to the topography tractor
cannot be used which results in dependency in human power and oxen. As people in the study
area rear less animals, so agricultural land left barren. Not only for foreign employment, people
also migrate from upstream areas to downstream areas in search of facilities and services. This
also results in agricultural land decrease in upstream areas.

(c) Road construction

Rural road construction in the upstream areas without considering landscape results moving of
sediments from upstream areas to downstream areas for deposition mainly in the monsoon
season. This results in increasing of barren land and decreasing in agricultural and and forest
areas.

(d) Lack of market price of agricultural products

Many respondents claimed that there is no market price of mainly vegetable products. The cost
for growing vegetable products is more than the benefit acquired. This results in demotivation
of farmers for growing vegetable products. Therefore, farmers decrease the dependency on
agricultural land and thus barren land increase.

(e) Urbanization/infrastructures
As increasing of population pressures and migration of people from upstream areas to
downstream areas in search of facilities and services in Phewa Watershed results more
settlements in both agricultural and forest lands. Also different infrastructures development
activities like rural road extension, hospital building, school building etc. has been changing

35
the land use pattern of watershed. This activity results sedimentation from upstream areas to
downstream areas resulting sediments deposition in lower areas during monsoon periods.

(f)Irrigation problem
Many people claimed that due to inconsistency in monsoon rainfall results in unfavorable
climate. Therefore, there occurs less water in time for growing agricultural crops resulting in
less productivity. This demotivated farmers for growing crops and they left barren.

(g) Use of excessive chemical fertilizers

In Phewa Watershed, the trend of rearing animals is decreasing. Therefore, this results in less
organic compost, which are by-product of animals. Therefore, people depend on external
chemical fertilizers Like NPK fertilizers. This results in low productivity in agricultural lands
and thus barren areas increases in study area.
4.3.3 Drivers responsible for increasing barren land during 2010 and 2018

People response on causes of Barren land increase

45 42
40
35
30
25 20
%

20 13 16
15 9
10
5
0

Causes of barren land increase

Fig 11: Bar diagram showing drivers of barren land increase and its % of respondents
(a)Lack of manpower
(b)Road construction
(c)Soil erosion/Landslides
(d) Lack of market price of agricultural products
(e) Lack of irrigation facilities

36
4.3.4 Drivers responsible for increasing urban areas during 2010 and 2018

People response on causes of Urban areas increase

80
69
70
60

50
%

40
31
30
20
10

0
Population pressures Migration to Down streams areas
Causes of urban areas increase

Fig 12: Bar diagram showing drivers of urban areas increase and its % of respondents
(a)Population pressures
Majority of respondents (69%) told that the population is increasing rapidly in Phewa
Watershed, which results in building new houses. This results in urban areas increase.

(b)Migration to down streams areas

Due to more facilities and services in downstream areas than upstream areas, people migrate
in downstream areas resulting in population increase in downstream areas. This results in
increasing houses built in forest and agricultural land resulting urban areas increase.

37
4.3.5 Drivers responsible for decreasing lake area during 2010 and 2018

People response on causes of Lake area decrease

40 34
35
30 26
%

25 19
20 14
15
10 7
5
0

Causes of lake area decrease

Fig13:Bar diagram showing drivers of lake area decrease and its % of respondents
(a)Sedimentation/Soil erosion
Majority of respondents (34%) beleive that in recent years, the flood level of Harpan khola
and Andheri khola is fluctuating.These two rivers are the major inflow of Phewa Lake and they
deposit large sediments of soil, sand and gravel in the bank and in Phewa Lake too. Andheri
khola bring sediments from vadaure, Tamangi, Kristi, Paudur, Dhikurpokhari, serachour and
Thulakhet. Harpan khola feeding the lake near Harapan and Thulakhet brings a lot of stones,
gravels and debris into the lake. The sedimentation has decreased the area of lake and its
volume of water. It has adversely affected the biological diversity especially the water birds,
flora and fauna. This sedimented part is more likely to be encroached in near future.

(b)Pollutants
Sewage from the hotels and urban settlements directly drain to the Phewa Lake. Many people
claimed that animal’s dead remains and crop residues are responsible for the pollution in Phewa
Lake. Farmers use inorganic fertilizers and pesticides in agricultural land to increase their
agricultural production and to protect the crops from insects and diseases. These practices
increase levels of nutrients such as phosphorus and nitrogen in soil. The runoff from
agricultural lands then carries these elements into the lake in dissolved or suspended form. The

38
increase in concentration of these elements in the lake water causes eutrophication.. It promotes
the unwanted growth of algae (floating single-celled plants) and of aquatic weeds.

(c)Encroachment

Sedimentation is seen due to soil erosion mainly by Harpan khola and Andheri khola.Extension
of road construction is one of the source of sedimentation. Local people then convert the area
into rice fields. Local people have also converted the marshy lands along the shore of the lake
into both irrigated Khet and rainfed Bari lands. This encroachment has resulted in a number of
negative impacts, including reduction of wetland areas, loss of wildlife habitat, decrease in
water level, and subsidence and eutrophication caused by runoff containing excess amounts of
chemicals from agricultural lands.

(d)Road construction
Rural road construction without considering the topography has resulted weak landscape in the
study area. During monsoon season sand. Gravel and soil roll directly to the lake area.

(e)Water hyacinth
They absorb water, which results in water volume reduction mainly in the boundary of the lake.
They die in the water and cause water pollution. People claimed that the edge of the lake is
gradually converted to swampy areas (wetland which is also a LULC category) due to
proliferation of invasive alien species mainly by Water Hyacinth.

39
 CHAPTER 5: CONCLUSION AND RECOMMENDATIONS

5.1 CONCLUSION

Changes in land use and land cover dynamics of Phewa Watershed area of western Nepal was
analyzed based on the RS, GIS and Social survey techniques. Digital supervised classification
technique using Landsat TM and OLI_TIRS was found effective for the preparation of
temporal LULC maps with acceptable degree of accuracy. Use of temporal satellite data are
very useful, time saving and cost effective for the preparation of LULC maps and change
analysis.
Last 8 years (2010-2018) significant changes in various LULC are observed in the studied
Phewa watershed of Nepal and this information would provide useful inputs to LULC planners
for effective management of the Watershed. Remarkable increase in forest area (from 32.18%
to 38.08%), urban areas (from 5% to 8.28%) and barren lands (from 3.87% to 8.89%).There is
also remarkable decrease of agricultural land (from 54.83% to 41%) and water bodies (from
3.91% to 3.75%).
Phewa Lake is important natural heritage of our country with economic, cultural and ecological
significance; it is high time to start lake area management from watershed level. The net
reduction of lake area by 20.61 ha over the period from 2010 to 2018 can be attributed to
sediment deposits due to anthropogenic activities such as rural road construction, improper and
inappropriate land use pattern both in upstream and downstream. Considering periodic Area
Increment (PAI) equal to about 2.577 ha/year negative and average rate of change as 0.61 %
negative from the study. Shrinkage is seen prominent in the western part of the lake resulting
isolation of a part of the lake from main body. The shrinkage area has potential to be encroached
as encroachment of lake area has been reported repeatedly. Consequences brought by lake area
shrinkage are not only an environmental phenomenon rather it is a social, political and
economic issue. Country like Nepal needs special attention to mitigate impacts of such
phenomenon as it may bring negative consequences in economy, environment and bio-
diversity in both upstream and downstream.
Various drivers like CF, Alternative source of energy, Promotion of Private forestry,
Awareness, Forest watchers and Control burning bring increased Forest areas in the study
area. In addition, Soil erosion, Migration, Road construction, Lack of market price of
agricultural products, Urbanization, Lack of irrigation facilities and Excessive use of chemical
fertilizers had resulted Agricultural land decrease and Barren land increase. Urban areas are

40
increasing due to Population pressures and Migration from upstream areas to downstream
areas. Lake area has decreased due to Sedimentation, Pollutants, Encroachment, Road
construction and Water hyacinth.

5.2 RECOMMENDATION
Based on the study, following recommendations are made.

 Employment opportunities for the people living in Phewa Watershed is suggested.

 Riverine plantation, Plantation on barren, Private land/Public land is suggested, as soil
erosion is prominent in study area, which ultimately reduces the sedimentation in lake.
 Considering the immense contribution of the upstream catchment for the loss and
degradation of the lake, it is essential to initiate the soil conservation and watershed
management activities in the areas. Application of bioengineering measures will be
effective in terms of cost and feasibility.
 Unfeasible rural road construction in the name of development should be stopped.
IEE/EIA is suggested.
 The boundary of the lake should be demarcated to control the encroachment.

 The invasive plant species in the lake area should be periodically removed.
 Declaration of entire watershed as a protected zone which is provisioned for
implementation help to reduce negative effects on Watershed.
 Land use land cover changes should be monitored time and again. Monitoring such
changes is important for coordinated actions at the national levels.
 Government, NGO’S, INGO’S and local people cooperation and commitments are
needed for finding suitable mitigation measures to lake area shrinkage.
 National level policy and management plan should be made for further proper
management of Phewa watershed.

41
 REFERENCES

 Alkharabsheha,M.M,Alexandridisa,T.K.,G.Bilasb,G.,Misopolinosb,N.,and
Silleosa,N.,(2013). Impact of land cover change on soil erosion hazard in northern Jordan
using remote sensing and GIS.Procedia Environmental Sciences 19,912-921.

 Arnoff (1989). Geospatial Information Systems:A management Perspective,WDL

Publications,Ottawa,Canada,pp.294.

 Awasthi, K.D., (2004). Land Use Change Effects on Soil Degradation, Carbon and
Nutrient Stock Sand Greenhouse Gas Emission in mountain Watersheds. Ph.D Thesis,
Agricultural University of Norway, Norway.
 Awasthi K.D, Regmi R.R, Pradhan B.M, Singh A, Sitaula K and Kafle G, (2007). Land
Use change in Rupa Lake Watershed and Lake Area Shrinkage: A GIS/RS Approach.
 Awasthi, K. D., Sitaula, B. K., Singh, B. R. and Bajracharya, R.M. (2002). Land-use
changes in two Nepalese watersheds: GIS and Geomorphometric analysis. Land
Degradation and Development, 13: 495-513.
 Balla M.K, and Sthapit K.M, (1998). Sedimentation Monitoring of Phewa Lake, Survey
Report submitted to ITTO/IOF Project.
 Balla,M. K., Awasthi , K.D., Singh ,B.K., and Pradhan , B.M., (2007). .Land use changes
and geomorphometric analysis in Galaudu and Phokharekhola watersheds in mid-hill
region of Nepal. International Journal of Ecology and Environment sciences 33(2).
 Borrough, PA and McDonnel, RA. (1998). Principles of Geographical Information
System. Oxford University Press: Oxford.
 Bronstert, A., Niehoff, D. and Bürger, G. (2002). Effects of climate and land use change
on storm runoff generation: Present knowledge and modeling capabilities.
Hydrological Processes, 16(2): 509–529.
 Burroughs, W.J., (2006). Climate Change A Multidisciplinary Approach, Cambridge
University Press, London. Department of Forests (DOF), Babarmahal, Kathmandu.
 CBS, (2005). Statistical Yearbook of Nepal. HMG/N, National Planning Commission
Secretariat, Central Bureau of Statistics, Kathmandu, Nepal.

 Chen, L., Wang, J., Fu, B. and Qiu, Y. (2001). Land use change in small catchments of
northern Loess plateau, China. Agriculture, Ecosystem and Environment, 86(2): 163-
172.

42
 Corn, M. L (1993): Ecosystems, Biomass, and Watersheds. Definitions and Use.
Washington, DC.: National Council for Science and the Environment. DSC
(Department of Soil Conservation), 1994. Sedimentation survey of Phewa Lake.,
Kathmandu.

 DeFries R. S., Bounoua L. and Collatz G. J. (2002). Human modification of the landscape
and surface climate in the next fifty years, Global Change Biology, 8: 438-458.

 DSC, (1994). Sedimentation survey of Phewa Lake. Kathmandu, Nepal.

 Eastman, J. R. (2009). “Idrisi 16 Taiga, Guide to GIS and Image Processing”, Clark
University; Worcester: 01610-1477, USA.

 Emilio, F.M.,(2010). Environmental social science:human-environmental interactions

and sustainability.Chichester,United Kingdom:John Wiley and Sons Ltd.

 Fikir Alemayehu.,Nurhusssen Taha,Nyssen,J.,Atkilt Girma ,Amanuel Zenebe,Mintesinot

Behailu,Deckers,J and Poesen,J.,(2009). The impacts of watershed management on land
use and land cover dynamics in Eastern Tigray (Ethiopia).Resources,Conservation and
Recycling ,53:192-198.

 Fleming, W. M. (1978). Phewa Tal Watershed Management Proposals. IWMP,

Department of Soil and Water Conservation, Ministry of Forests, Kathmandu.

 Gautam, A.P., 2007. Land Use Dynamics and Landscape Change Pattern in a Mountain
Watershed in Nepal. ( Assessed on 18thFeb.2017).
 Gautam, A.P., Webb, E.L., Shivakoti, G.P., and Zoebisch, M.A., (2003). Land Use
Dynamics and landscape Change Pattern in a Mountain Watershed in Nepal.
Agriculture, Ecosystems and Environment, 99:83-96.

 Gautam N. C and Narayanan E. R. (1983). Satellite remote sensing techniques for

natural resources survey. L R Singh, Savindra Singh, R C Tiwari and R P
Srivastava,(Eds:) Environmental Management, Allahabad, pp. 177-181.

 Geist, H. J. and Lambin, E. F. (2002). Proximate Causes and Underlying Driving Forces
of Tropical Deforestation. BioScience, 52(2):143-150.
 Gilani, H., Shrestha, H. L., Murthy, M. S. R., Phuntso, P., Pradhan, S., Bajracharya, B
and Shrestha, B. (2014). Decadal land cover change dynamics in Bhutan. Journal of
Environmental Management. doi:10.1016/j.jenvman.2014.02.014.

43
 Heyojoo B.P, Timilsina Y.P, Sharma R.P and Karki G.B, (2009). An Assessment of
Sediment Sources from Rural Road Construction and Landslide at Phewa Lake
Watershed of Kaski District, Nepal.

 Janssen, M. A. and Jager, W. (2000). The human actor in ecological models. Ecological
Economics, 35(3): 307-310.

 JICA/SILT, (2002). Conservation Development study on the environmental

conservation of Phewa Lake in Pokhara, Nepal. Final report submitted by SILT
Consultants to JICA/ Nepal, Kathmandu.
 Jorgensen, S.E and Volleweider, R.A., (Eds.), (1989). Guidelines of lake management.
Vol.1.International Lake Environment Committee (ILEC), Shiga, Japan.
 Kaiser, M., Aboulela, H. and Ezz Edin, H. (2008). Spectral Enhancement of the SPOT
Imagery Data to Assess Marine Pollution near Port Said, Egypt. In PIERS Proceedings,
2(6): 746-753.

 Keyser JD and Kaiser DA (2010). Getting the Point: Metal Weapons in Plains Rock Art.
Plains Anthropologist 55(214): 111-132.
 Lambin, E. F., Turner, B. L., Geist, H. J., Agbola, S. B., Angelsen, A., Bruce, J. W.,
Coomes, O.T., Dirzo, R., Fischer, G., Folke, C., George, P. S., Homewood, K., Imbernon,
J., Leemans, R., Li X, B., Moran, E. F., Mortimore, M., Ramakrishnan, P. S., Richards,
J. F., Skanes, H., Stefeen, W., Stone, G. D., Svedin, U., Veldkamp, T. A., Vogel, C. and
Xu, J. C. (2001). “The causes of land-use and land-cover change: moving beyond the
myths”. Global Environ. Change, 11(4): 261–269.
 Latham,J.(2001). LCCS: Land covers classification system. (URL:
http://www.fao.org/sd/2001/en0101_en.htm accessed on February 2010).

 Lenat, D.R. and Crawford, J.K., (1994). Effects of land use on water quality and aquatic
biota of three North Carolina piedmont streams. Hydrobiologia 294 (3), 185–200.

 Lillesand, T. M., Kiefer, R. W. and Chipman, J. W. (2008). “Remote Sensing and Image
Interpretation”, John Willy and Sons. Inc. USA.
 Lilisand, T.M. Kiefer, R.W. Andchipman, J.W. (2004). Remote Sensing and Image
Interpretation. John wiley&Sons, New York, USA. 31.
 Liu,J.,Liu,M.,and Zhuang,D.,( 2002). Study on spatial pattern of land use in China in
recent years.Science in China,Ser.D32(12),p.1031-1040.

44
 Meyer, W. B. and Turner II, B. L. (1996). Land-Use/Land-Cover change: Challenges for
Geographers. Geojournal, 39(3): 237-240.
 Mohammad, A. G., & Adam, M. A. (2010). The impact of vegetative cover type on
runoff.

 Mulder, R. P. (1978). A Soil Survey of Phewa Tal Catchment ,Integrated Watershed

Management Project. Department of soil Conservation and Watershed Management,
Ministry of Forests, Kathmandu, Nepal.
 Muzzini,E.,and Aparicio,G.,(2013).Urban Growth and Spatial Transition in
Nepal.Washington D.C:World Bank.http://dx.doi.org/10.1596/978-0-8213-9659-9.
 Naveh, Z., (2000). What is holistic landscape ecology? A conceptual introduction.
Landscape Urban Plan. 50, 7–26.
 Phong, L. T. (2004). Analysis of Forest Cover Dynamics and their Driving Forces in
Bach Ma National Park and Buffer Zone using Remote Sensing and GIS. MSc Thesis
submitted to the International Institute for Geo-information Sciences and Earth
Observation, Enschede, The Netherlands.
 Pokhreal S, (2008). Conservation of Phewa Lake Pokhara of Nepal, NLCDC, Ministry
of Culture, Tourism & Civil Aviation.
 Ratanopad, S. and Kainz, W. (2006). Land Cover Classification and Monitoring in
Northeast Thailand using Landsat 5 TM Data. In ISPRS Technical Commission II
Symposium, Vienna, 12(14): 137-144.

 Rawat,J.S., and Kumar,M.,(2015). Monitoring land use land cover change using remote
sensing and GIS techniques:A case study of Hawabagh block,district
Almora,Uttarakhand,India.The Egyptian Journal of Remote Sensing and Space
Science,18:1.pg.77-84.

 Regmi, R. R., & Saha, S. K. (2015). Impact Assessment of Land Use Land Cover on Soil
Erosion Status in Phewa Lake Watershed,Nepal. International Journal of Current
Engineering and Technology, 5(3), 1708-1717.
 Regmi ,R.R.,S.K.Saha.,M.K. Balla.,(2014). Geospatial Analysis of Land use Land Cover
Change Modelling at Phewa Lake Watershed of Nepal by using Cellular Automata
Markov Model.
 Rimal,B.,(2011). Urban Growth and Land use/land cover .Journal of Theretical and
Applied information Technology,26(2),pp.118-129.
.

45
 Rimal, B.; Zhang, L.; Keshtkar, H.; Haack, B.; Rijal, S.; Zhang, P. , (2018a). Land
use/land cover dynamics and modeling of urban land expansion by the integration of
cellular automata and markov chain. ISPRS Int. J. Geo-Inf. 7, 154.

 Rimal, B.; Zhang, L.; Keshtkar, H.; Sun, X.; Rijal, S., (2018b). Quantifying the
spatiotemporal pattern of urban expansion and hazard and risk area identification in the
Kaski District of Nepal. Land, 7, 37.
 Sadidy, J., Firouzabadi, P. Z. and Entezari, A. (2009). The Use of Radarsat and Landsat
Image Fusion Algorithms and Different Supervised Classification Methods to use Map
Accuracy – Case study: Sari Plain – Iran.

 Shrestha, C. and Mall, U. (1983). Environmental Resource Management in Nepal.

Kathmandu. Published by B. Thapa. Sahayogi Press, Tripureshwar ,Kathmandu, Nepal.

 Sidhu, D.S., Das, T.H., Singh, R.S., Sharma, P.K., Jadaun, S.P.S., and Ravishankar, T.(2000).
Soil resource inventory of watershed for land use planning a GIS approach. Indian journal
of soil conserbvation 128:22-29.
.
 Sthapit, K. M. (1988). Flood control and Watershed Management, Paper presented at the
National Workshop on River Training, Embankment Protection, 4 – 5 May 1988,
HMG/N, UNDP/ILO, Kathmandu, Nepal.
 Stohlgren, T. J., Chase, T. N., Pielke, R. A., Kittel, T.G. F. and Baron, J. S. (1998).
Evidence that local land use practices influence regional climate, vegetation, and stream
flow patterns in adjacent natural areas.Global Change Biology, 4: 495–504.
 Tang, Z., Engel, B. A., Pijanowski, B. C. and Lim, K. J. (2005). Forecasting land use
change and its environmental impact at a watershed scale. Journal of Environmental
Management 76: 35–45.

 Tateishi, R. (2002). Land cover mapping of Asia – problems and solutions of

continental/global land cover mapping [Electronic version]. Tropical Ecology, 43(1):
3-8.

 Turner, B. L., Skole, D., Sanderson, S., Fischer, G., Fresco, L., and Leemans, R., (1995).
Land –Use and Land-Cover Change, Science/Research Plan. IGBP Report No. 35, HDP
Report No. 7. IGBP and HDP, Stockholm and Geneva.

 UN DESA.(2014). World Urbanization Prospects:2014 Revision.New York:United

Nations Department of Economic and SocialAffairs.http://esa.un.org/unpd/wup/Final

46
 Report/WUP 2014-Report.pdf.
 Vink, A. P. (1975). Land use in Advancing Agriculture.New York Spring Verlage
Heibdeberg, Berlin.

 Wafa Nori,Hussein M.Sulieman,Irmguard Niemeyer,(2009). Detection of land cover

changes in EL Rawashda Forest,Sudan :A system comparision.International
Geoscience and Remote Sensing Symposium (IGARSS),vol.1,pp188-191.
 Wang, L., Lyons, J., Kanehl, P., Bannerman, R. and Emmons, E. (2000). Watershed
urbanization and changes in fish communities in southeastern Wisconsin
streams.Journal of the American Water Resources Association,36(5): 1173–1189.
 Wang, L., Lyons, J., Kanehl, P. and Bannerman, R. (2001). Impacts of urbanization on
stream habitat and fish across multiple spatial scales.Environmental Management,
28(2): 255-266.

 White, M. D. and Greer, K. A. (2006). The effects of watershed urbanization on the

stream hydrology and riparian vegetation of Los Peñasquitos Creek,
California.Landscape and UrbanPlanning, 74(2): 125–138.

 Xiuan, C., Heping, H., Tateishi, R., and Ahn, C.H. (1999). A Method on Land cover
Classification by Combining Unsupervised Algorithm and Training Data, Geocarto
International, (14).
 Yamanaka, H., Yoshida, M. and Arrita, K. (1982). Terrace Landform and Quaternary
Deposit around Pokhara Valley, Central Nepal. JI.of Nepal Geological Society, 2: 113-
142.
 Yüksel, A., Akay, A. E. and Gundogan, R. (2008). Using ASTER Imagery in Land
Use/cover Classification of Eastern Mediterranean Landscapes According to CORINE
Land Cover Project [Electronic version]. Sensors, 8 (2): 1237-1251.

 Zampella, R. A., Laidig, K. J. and Lowe, R. L. (2007). Distribution of diatoms in relation

to land use and pH in blackwater coastal plain streams. Environmental Management,
39: 369–384.
 Zendu,S.,Suryabhagavan,K.V.,Balkrishna,M.,(2016). Land –use/land-cover dynamics
in Sego Irrigation Farm,southern Ethiopia :A comparision of temporal soil salinization
using geospatial tools.journal of the Saudi Society of Agricultural Sciences,1:15,91-97.

47
 ANNEXES
ANNEX-1
Questionnaire survey
Land use land cover change in Phewa watershed and its Lake area change status:A
Geospatial approach
The information filled in this questionnaire will be confidential and will be used only for
academic purpose. Please feel free to express your personnel opinion.
Name of respondent: Age/sex:
Location: Caste:
(1)How long have you been living in Phewa Watershed? ………….years
(2)Do you know the terms LULC? Yes or No
(3)Have you experience LULC changes in Phewa Watershed since 8 yrs? Yes or No
(4)If yes, go to question no.5
(5)What are drivers of forest area increase in Phewa Watershed?
(a)CF (c) Awareness
(b)Alternative form of energy (d) Private forestry
(6) Can you tell about other drivers of forest area increase? Yes or No .If yes mention
(7)Why and how these contribute to forest area increase?.....................................................
(8) What are drivers of agricultural land decrease in Phewa Watershed?
(a)Road construction (c) Soil erosion/Landslides
(b)Urbanization (d) Migration for foreign employment
(9) Can you tell about other drivers of agricultural land decrease? Yes or No .If yes mention
(10)Why and how these contribute to agricultural land decrease?..............................................
(11) What are drivers of barren land increase in Phewa Watershed?
(a)Soil erosion/Landslides (c) Road construction
(b)Migration
(12) Can you tell about other drivers of barren land increase? Yes or No .If yes mention
(13) Why and how these contribute to barren land increase?...............................................
(14) What are drivers of urban areas increase in Phewa Watershed?
(a)Population growth (b) Migration
(15) Can you tell about other drivers of urban areas increase? Yes or No .If yes mention

48
(16) Why and how these contribute to Urban areas increase?...............................................
(17) What are drivers of Lake Area decrease in Phewa Watershed?
(a)Sedimentation (b) Encroachment
(c)Pollutants
(18) Can you tell about other drivers of Lake Areas decrease? Yes or No .If yes mention
(19) Why and how these contribute to Lake areas decrease?...............................................
(20) Anything you want to tell at last?................................................................................

THANK YOU

49
 ANNEX-2
Flow chart of research design

50
SOME PHOTO PLATES

51