TRIBHUWAN UNIVERSITY

INSTITUTE OF ENGINEERING

PULCHOWK CAMPUS

Thesis No: PUL074MSDRM017/108

Flood Hazard Mapping and Vulnerability Assessment of

West Rapti River Basin

by
Gayatri Joshi

A THESIS

SUBMITTED TO THE DEPARTMENT OF CIVIL ENGINEERING

IN PARTIAL FULFILLMENT OF THE REQUIREMENT FOR THE

DEGREE OF MASTER OF SCIENCE IN ENGINEERING IN

DISASTER RISK MANAGEMENT

DEPARTMENT OF CIVIL ENGINEERING

LALITPUR, NEPAL

SEPTEMBER, 2022
 COPYRIGHT

The author has agreed that the library, Department of Civil Engineering Pulchowk
Campus, Institute of Engineering may make this thesis freely available for inspection.
Moreover, the author has agreed that permission for extensive copying of this thesis for
scholarly purpose may be granted by Professor(s) who supervised the work recorded
herein or, in their absence, by the Head of the Department wherein the thesis was done.
It is understood that the recognition will be given to the author of this thesis and to the
Department of Civil Engineering, Pulchowk Campus, and Institute of Engineering in
any use of the material of this thesis. Copying or publication or the other use of this
thesis for financial gain without approval of the Department of Civil Engineering,
Pulchowk Campus, Institute of Engineering and author’s written permission is
prohibited.

Request for permission to copy or to make any other use of the material in this thesis in
whole or in part should be addressed to:

_____________________

Head

Department of Civil Engineering

Pulchowk Campus, Institute of Engineering

Lalitpur, Nepal

2
 TRIBHUVAN UNIVERSITY
INSTITUTE OF ENGINEERING
PULCHOWK CAMPUS
DEPARTMENT OF CIVIL ENGINEERING

The undersigned certify that they have read, and recommended to the Institute of
Engineering for acceptance, a thesis entitled “Flood Hazard Mapping and Vulnerability
Assessment of West Rapti River Basin” submitted by Gayatri Joshi in partial
fulfillment of the requirements for the Degree of Master of Science in Disaster Risk
Management.

Supervisor, Suraj Lamichhane, Ph.D.

Associate Professor
Director of Centre for Disaster Studies,
Pulchowk Campus

External Examiner, Dr. Divas Basnet

Program Coordinator, Dr. Basanta Raj Adhikari

MSc. in Disaster Risk Management
Department of Civil Engineering

September, 2022

3
 ABSTRACT

Flood occurs repeatedly in Nepal and causes tremendous loss in terms of property and
life, particularly in the lowland areas. The majority of flood victims are poor people
who resides in floodplain (Englhardt et al., 2019). The frequency of flood events and
associated risk with respect to built-up area and agriculture land is the critical issue for
developing countries, such as Nepal. This study, develop the flood hazard mapping for
various return period and assess the flood vulnerability in building typology and land
use along the West Rapti River (WRR) reach.

In this study, 2D hydrodynamic flood model is constructed to determine the inundation

area of the study area of West Rapti river in Nepal. Total catchment area of this basin
is around 6500 sqkm and average annual rainfall is about 1,500 mm. It is a rainfed
river, more than 80 % of annual precipitation occurs in monsoon season. The
topography of this basin is flat at downstream where the settlement is dense, exposed
to the risk of flooding. Flood frequency analysis for 2,5,10,25,50 and 100 years return
period was carried out by Gumbel, Log Pearson Type III and Log Normal method based
on maximum instantaneous flow recorded at Bagasotigaun station. Two-dimensional
hydraulic model HECRAS with RAS-MAPPER was used for two-dimensional flood
modeling and Geographic Information System (GIS) for flood hazard mapping. The
terrain was imported to the Ras mapper and a mesh of 30*30 m size was created in
geometry file. In HEC-RAS, boundary condition for upstream and downstream were
defined. Similarly, flood flows were also entered for different return periods. Unsteady
flow analysis was done for the results.

The results of flood frequency analyzed by Gumbel’s method showed discharges of

1793 m3/s, 2870 m3/s, 3583 m3/s, 4484 m3/s, 5153 m3/s and 5816 m3/s for 2-,5-,10-,25-
,50- and 100-years return period floods, which were higher among the results obtained
using different methods. The output was used to prepare inundation maps of study area.
The map shows that, people and infrastructure near the banks on both sides are at
heavily exposed to the flooding. Moreover, the agricultural area (53%) is more
vulnerable alarming the risk of food security. The marginalized people with low income
are observed to live in adobe house near river corridor. The assessment shows that the
adobe houses are the most vulnerable building typologies with maximum numbers of
existing typologies followed by wooden and brick masonry. The existing vulnerability

4
curves shows that the adobe houses may damage 50 % at the inundation of 50 cm and
were fully damage at the depth of 1.5 m. Similarly, wooden houses and brick mortar
houses were to damage 50 % at the inundation depth of 80 cm and 1 m, respectively.
These typologies of houses are more vulnerable to flooding. Reinforce cement
buildings were damage 40 % at 2.5 m inundation depth, indicating most resilient house
typology for flooding. The majority of flood disasters victims are poor people living in
floodplain.

Thus, finding of this study can help in planning and management of flood plain area to
mitigate future probable disaster through technical approach.

5
 ACKNOWLEDGEMENT

I am incredibly indebted to my respected thesis supervisor Dr. Suraj Lamichhane for

his constant encouragement, motivation, support and guidance. This report is the
outcome of continuous support in terms of research ideas, proper guidance during
literature review and support in editing my report. I express my sincere thanks to
program Ex-coordinator Assistant Prof. Nagendra Raj Sitoula for his guidance and
present program coordinator Dr. Basanta Raj Adhikari for his support.

I am also thankful to Department of Hydrology and Meteorology (DHM), Department

of water resources and Irrigation (DoWRI), Water and Energy Commission Secretariat
(WECS) for their support in providing the necessary data for the thesis work.

Most importantly, I would like to thank my beloved family: my mother Mrs. Jayanti
Devi Joshi, my father Mr. Keshab Datt Joshi, my husband Dr. Bashu Dev Pant and my
Son Suyog Pant supporting me as usual in any situation. I would also like to thank all
the people who contributed to this thesis work. My special thanks go to my seniors who
helped me to access some important literatures.

I would like to thank all of my colleagues and support staff at Pulchowk Campus, IOE,
who very diligently supported my research endeavors.

Gayatri Joshi
074/MSDRM/017

6
 TABLE OF CONTENTS

COPYRIGHT ............................................................................................................... 2

ABSTRACT .................................................................................................................. 4

ACKNOWLEDGEMENT ........................................................................................... 6

TABLE OF CONTENTS............................................................................................. 7

LIST OF TABLES ..................................................................................................... 10

LIST OF FIGURES ................................................................................................... 11

LIST OF ANNEXES .................................................................................................. 13

LIST OF ABBREVIATIONS ................................................................................... 14

CHAPTER ONE: INTRODUCTION ...................................................................... 15

1.1 Background........................................................................................................ 15

1.2 Problem Statement ............................................................................................. 16

1.3 Objectives of the Study ...................................................................................... 18

1.4 Limitations of Study........................................................................................... 19

CHAPTER TWO: LITERATURE REVIEW ......................................................... 20

2.1 Floods in Global Context ................................................................................. 20

2.2 Floods in Nepalese Context .............................................................................. 22

2.3 Flooding Problem in West Rapti River .......................................................... 24

2.4 Causes of Flooding in Nepal ............................................................................ 25

2.6 Flood Hazard and Vulnerability Assessment ................................................ 26

2.7 Flood Hazard Maps in Context of Nepal ....................................................... 28

2.8 Causes of Increasing Vulnerability of People and Structure ....................... 29

2.9 Flood Frequency Analysis ............................................................................... 31

2.9.1 GUMBEL’S DISTRIBUTION ............................................................................... 32

2.9.2 LOG PEARSON TYPE III................................................................................... 34
2.9.3 LOG-NORMAL DISTRIBUTION ......................................................................... 35
2.10 Flood Modeling ................................................................................................ 35

7
2.11 Available Tools for Modeling .......................................................................... 36

2.12 Hydrodynamic Model: HEC RAS .................................................................. 38

2.13 Calibration and Validation ............................................................................. 39

2.14 Geographic Information System (GIS) .......................................................... 39

2.15 Policy Context ................................................................................................... 40

CHAPTER THREE: METHODOLOGY ................................................................ 42

3.1 Study Area .......................................................................................................... 42

3.2 Acquisition and Processing of Data .................................................................. 44

3.2.1 DIGITAL ELEVATION MODEL .......................................................................... 44

3.2.2 LAND USE DATA ............................................................................................ 45
3.2.3 HYDRO-METEROLOGICAL DATA .................................................................... 46
3.3 Flood Frequency Analysis ................................................................................. 47

3.4 Manning’s Roughness (n) .................................................................................. 48

3.5 Hydraulic Modeling ........................................................................................... 49

3.5.1 THEORETICAL BACKGROUND OF HEC-RAS................................................... 49

3.5.1.1 Governing Equations................................................................................ 49
3.5.1.2 Boundary Conditions ............................................................................... 52
3.6 Model Inputs....................................................................................................... 53

3.6.1 GEOMETRY FILE ............................................................................................. 53

3.6.2 PLAN FILE ....................................................................................................... 53
3.6.3 UNSTEADY FLOW DATA FILE ......................................................................... 53
3.6.4 THE PROJECT FILE .......................................................................................... 54
3.7 HEC-RAS Model Construction, Calibration, and Validation ....................... 54

3.8 Flood Hazard Analysis....................................................................................... 56

3.9 Vulnerability Assessment .................................................................................. 57

CHAPTER FOUR: RESULTS AND DISCUSSION .............................................. 60

4.1 Flood Frequency Analysis ................................................................................. 60

4.2 Flood Hazard Analysis....................................................................................... 62

4.2.1 INUNDATION AREA........................................................................................... 62

8
 4.2.2 EXTENT AND DEPTH OF INUNDATION ............................................................. 63
4.3 Flood Vulnerability Analysis............................................................................. 65

CHAPTER FIVE: CONCLUSION AND RECOMMENDATIONS..................... 70

5.1 Conclusion............................................................................................................. 70

5.2 Recommendation .................................................................................................. 71

ANNEXES................................................................................................................... 75

9
 LIST OF TABLES

Table 1: Losses due to flood in year 2021 compared to 2001-2020 ............................ 21

Table 2: Comparison of different models .................................................................... 37

Table 3: Study of flood modeling and analysis using HEC-RAS and ArcGIS in Nepal
...................................................................................................................................... 38

Table 4: Hydro-meteorological stations in the West Rapti Basin (Source: DHM_2021)

...................................................................................................................................... 47

Table 5: Manning’s n Values (Source: HECRAS user manual) .................................. 48

Table 6: Flood hydrograph used on the model ........................................................... 55

Table 7: Calibrated value of Manning’s n .................................................................. 56

Table 8: Classification of hazard level according to flood depth ................................ 57

Table 9: Flood discharge for different return period.................................................... 60

Table 10: Examining the goodness of fit distribution .................................................. 61

Table 11: Extent of Inundation for different return period .......................................... 62

Table 12: Area under hazard in terms of depth (m) ..................................................... 64

Table 13: Vulnerability analysis with building data .................................................... 66

Table 14: Number of houses Inundated for different year return period flood ............ 66

Table 15: Damage factor of different house typology ................................................. 67

10
 LIST OF FIGURES

Figure 1: Frequency of occurrence of natural hazards ................................................. 21

Figure 2: Human loss due to different disaster in 2017-2018 ...................................... 23

Figure 3: Map of West Rapti River Basin Showing Study Area ................................. 43

Figure 4: Digital Elevation Model ............................................................................... 45

Figure 5: Land Use Map of West Rapti river basin (Source: IMP_2019) ................... 46

Figure 6: HEC-RAS graphical user interface, the main window. ................................ 49

Figure 7: Vulnerability curves for different building Typologies [a) (Englhardt et al.,
2019), b) (Dutta et al., 2003)] ...................................................................................... 58

Figure 8: Comparison of floods by different methods ................................................. 61

Figure 9: Trend of inundation area for high hazard class ............................................ 63

Figure 10: Return period -Flood depth relationship..................................................... 64

Figure 11: Impacts of different return period flood on various types of landcover ..... 68

Figure 12: Trend of vulnerable agricultural area for different return period ............... 69

Figure 13: Flood inundation map for 2 years return period ......................................... 82

Figure 14: Flood inundation map for 10 years return period ....................................... 83

Figure 15: Flood inundation map for 25 years return period ....................................... 84

Figure 16 : Flood inundation map for 50 years return period ...................................... 85

Figure 17: Flood inundation map for 100 years return period ..................................... 86

Figure 18: Flood hazard map for 2 years return period ............................................... 87

Figure 19: Flood hazard map for 10 years of return period ......................................... 88

Figure 20: Flood hazard map for 25 years return period ............................................. 89

11
Figure 21: Flood hazard map for 50 years return period ............................................. 90

Figure 22: Flood hazard map for 100 years return period ........................................... 91

Figure 23: Numbers of building affected by 2 years of return period ......................... 92

Figure 24: Numbers of building affected by 10 years of return period ....................... 93

Figure 25: Numbers of building affected by 25 years of return period ....................... 94

Figure 26: Numbers of building affected by 50 years of return period ....................... 95

Figure 27: Numbers of building affected by 100 years of return period ..................... 96

Figure 28: Flood vulnerability map for 2 years return period ...................................... 97

Figure 29 :Flood vulnerability map for 10 years return period .................................... 98

Figure 30: Flood vulnerability map for 25 years return period .................................... 99

Figure 31: Flood vulnerability map for 50 years return period .................................. 100

Figure 32: Flood vulnerability map for 100 years return period ................................ 101

12
 LIST OF ANNEXES

Annex 1:Maximum Instantaneous Discharge of Bagasoti Hydrological Station ........ 76

Annex 2: Flood Frequency Analysis using different method ...................................... 78

Annex 3: Computation Mesh created in the Study area ............................................... 81

Annex 4: Flood Inundation Map of different year return period ................................. 82

Annex 5: Flood Hazard Map of different year return period ....................................... 87

Annex 6:: Flood Vulnerability Map of buildings affected by different years return
period ........................................................................................................................... 92

Annex 7: Flood vulnerability map for different return period ..................................... 97

13
 LIST OF ABBREVIATIONS

ArcGIS Aeronautical Reconnaissance Coverage Geographic Information

System

CBS Central Bureau of Statistics

CRED Centre for Research on the Epidemiology of Disasters

DEM Digital Elevation Model

DHM Department of Hydrology and Meteorology

EM-DAT Emergency Events Database

HEC-RAS Hydrologic Engineering Center's (CEIWR-HEC) River Analysis

System

ICHARM International Centre for Water Hazard and Risk Management

ICIMOD International Centre for integrated Mountain Development

INSEC Informal Sector Service Centre

MOHA Ministry of Home Affairs

NRCS Nepal Red Cross Society

RBP River Basin Plan

SWAT Soil and Water Assessment tool

UNDP United Nations Development Programme

WECS Water and Energy Commission Secretariat

YRP Years Return Period

14
 CHAPTER ONE: INTRODUCTION

1.1 Background

Flood is one of the striking water induced disaster that hits most of the part of the world.
It can be explained as flows exceeding the transporting capacity of river channels, lakes,
ponds, reservoirs, drainage system, dam and any other water bodies whereby water
inundates the area outside water bodies. It causes damages to the crops and property, a
lot of difficulties to the people and causes death of people also. It is one of the major
natural hazards in terms of fatality and economic loss.

Flooding, a global phenomenon occurring most frequently among all-natural hazards

cause major economic losses, devastation and human lives losses (CRED, 2018).
According to Talchabhadel & Sharma, 2014, flooding is responsible for one third of
total economic loss and two thirds of people affected by natural disasters. Bertilsson et
al., 2019b) describes floods as the most pervasive hydrometeorological hazard and
mention that climate change is likely to alter the hydrological cycle in many regions
increasing the probability of extreme weather events such as droughts and floods.
According to the center for research in the epidemiology of disasters (CRED) through
its emergency event database (EMDAT) which has been recording statistics of losses
and damages from the disasters events globally indicated that floods have affected more
people than any other hazards in the 21st century. In 2018, four major flood events that
occurred in India, Japan, Nigeria, and Korea are categorized among the top 10 deadliest
disaster events (CRED, 2019).

Nepal is considered one of the most disaster-prone countries in the world. Nepal is the
second highest country at risk of floods in South Asia (UNDP, 2009). In addition to
other natural hazards, such as earthquakes and landslides, floods represent a recurring
risk for large sections of the population. River flooding represents the most common
global hazard causing significant number of losses. Throughout human history, river
floods have caused excessive damages on economy and lives and caused more
economic losses than any other hazard and despite substantial measures that have been
enacted to prevent floods, resultant loss of human life and property persists at high
levels. Flooding has a particular impact on communities residing within the flood plain.
It indicates that the majority of flood disasters’ victims are poor people. The increase

15
of population and squatter settlements of landless people living at the bank of the river
has tremendous pressure in encroachment of flood plain making them vulnerable to the
flood damage (Dangol & Bormudoi, 2015). The historical data showed that the country
witnessed major floods in Tinau; 1978, Koshi; 1980, Tadi; 1985, Sunkoshi; 1987,
Kulekhani; 1993 and many more. The flood frequently affects approximately 30,000
people based on CBS, 2011 and during medium to large flood events this number rises
to over 100,000 people.

The West Rapti River of Nepal is considered as one of the most flood-prone rivers. The
major causes of flooding in this basin are natural causes like high rainfall ,soil erosion,
flat topography, debris flows and sedimentation, river coarse shifting, deforestation,
poor planning, design and construction of roads, massive increase of settlements
along the highway, lack of participatory approach in disaster management, lack of
long-term flood management plan covering whole flood prone areas and socio- political
causes like lack of public sensitivity and awareness, attitude of people and culture of
abandonment (Gautam & Phaiju, 2013). This basin usually suffers from flash floods as
the catchment responds to high intensity and short duration precipitation. Recent
extreme flood events were reported in 1984, 1989, 1998 and 2007, 2013 and 2017.

Flood hazard mapping is beneficial in aspects of land and water resources management
such as preventing unwise land use in flood prone areas. Through a proper flood plain
hydraulic analysis, design of structures like bridge and culvert openings for roadway
crossings of streams and flood reduction measures like dams, levees can be achieved
(Salimi et al., 2008). Among various nonstructural measures needed for disaster
mitigation, hazard and vulnerability and risk mapping is one of the important
nonstructural measures (Dangol & Bormudoi, 2015). Flood hazard assessment is the
estimation of overall adverse effects of flooding. It depends on many parameters such
as depth of flooding, duration of flooding, flood wave velocity, and rate of rise of water
level. One or more parameters can be considered in the hazard assessment (Aryal et al.,
2020).

1.2 Problem Statement

Flooding has been a serious problem for the communities of the West Rapti River basin
for years. The deposition of sands in the farmland by the torrents originated from the

16
Chure/Siwalik range, inundation due to flooding, and bank cutting at various locations
due to rapid geomorphological changes are the major problems affecting lives and
livelihoods of the people living on this basin. The settlement near the river have been
suffering from drainage congestion and inundation problems due to unplanned growth,
poor drainage system design and poor waste disposal practices (ICHARM, 2006). So,
the impact of the flood in this basin seems to be very high with greater change of human
and economic loss.

Flood event occurred in 2013 was the worst flooding to hit Nepal’s western region in a
decade-caused by heavy monsoon rains which killed at least 50 people. Flood waters
from the Rapti River affected as many as 60,000 villagers in 13 districts. The flooding
has destroyed the farms, killed nearly 2,000 animals, and damaged more than 10,000
tons of food stocks. Nepal has sought international help worth US $3.58 million to
provide emergency relief. Similarly, the flood event occurred in 2017 was also a highest
level of flooding at Kusum gauge station, recorded 8.9 m. If proper actions are not taken
in time, the result of similar flooding would be increased in future.

Many studies are carried out in this basin due to availability of long historical data. But
it is seen that most of the researches are carried out near Indo-Nepal border to see the
effect of Laxmanpur Barrage, Kalkaluwa embankment which lies downstream of Rapti
River in Indian territory and other flood controlling structures such as spurs and dykes
constructed in Nepalese territory. The selected section of basin has wide flood plain
due to comparatively flat topography which results in sedimentation and ultimately
results the shifting of river coarse. This makes the nearby settlement vulnerable to
floods.

Encroachment of the settlement to the flood plain area has been induced by fertile soil
and availability of irrigation water from the river. Presence of the river has enabled
development of broad agriculture in the region, and today 85% of the sources of
livelihood arises from cultivation(ICHARM, 2006). Researchers had found out that
increase in build-up area is one of the major causes for flooding as well as exposure to
the flood impacts and therefore settlement near the river is also at high risk of flooding.
Past records of flood shows that the impacts on human and economic loss might be
much more if it happens now or on near future. So, to minimize the loss, flood

17
 forecasting and warning system is must, for this detail flood hazard and risk mapping
is required. The hydrodynamic models are capable of providing the products required
for assessing the flood hazard mapping and flood vulnerability assessment.

DHM data shows that the flooding in the recent years are becoming more frequent and
more intense. Hence apart from structural measures, non-structural measures like flood
prediction and analysis of vulnerable areas are a must to minimize the damage. As,
floods events are inevitable, this research focuses on the analysis of the potential
impacts and vulnerability along the Rapti river through non-structural measures which
are inundation mapping. The hydraulic model can be used to delineate the flood plain
and quantify the potential impacts of flooding. The maps prepared during the study is
helpful for decision makers to plan for the built-up areas along the river as well as
identification of the evacuation sites during the flooding events. The maps also can be
used to disseminate hazard information to the public so that human and property loss
due to flood is reduced. This shows the essence of 2D hydrological model which is
capable of providing the products required for assessing the flood inundation mapping.

1.3 Objectives of the Study

The major objective of the study is to develop flood hazard mapping for various return
period and assess the flood vulnerability in building typology and land use along the
West Rapti River (WRR) reach and more specifically:

• To assess the flood frequency for the quantitative assessment.

• To analyze the flood hazards, assess the flood vulnerable sites and mapping the disaster-
prone areas along the reach for various return period,

• To quantify the damage of the vulnerable area.

18
 1.4 Limitations of Study

The main limitations during the course of this research are listed below:

• The availability of hydrological data of the Bagasoti hydrological station was only up
to 2015 AD. Therefore, the study has been carried out using the data up to 2015 AD.

• The topographical data is limited to the freely available DEM (i.e ASTER DEM of 30
m resolution).

19
 CHAPTER TWO: LITERATURE REVIEW

2.1 Floods in Global Context

Flooding is a major natural disaster that affects different parts of the world. Due to this
disaster, billions of dollars in infrastructure and property damages and hundreds of
human lives are lost each year (Demir & Kisi, 2016). The driving factors of this risk
are the persistent increase in global population, the concentration of people in high-risk
areas such of coastlines, floodplains and hill slopes, increased vulnerability of assets,
infrastructure and social systems and the consequences of environmental and climate
change. Flood risk is high in urban centers, and the rapid growth of cities, especially
those settlements located along rivers and coasts which increases the exposure of people
and assets to flooding (A. K. Jha et al., 2012). According to WHO, more than 2 billion
people are affected worldwide in between 1998-2017. People who live in floodplains
or non-resistant buildings (i.e., temporary houses) or lack of warning systems and
awareness of flooding hazard are most vulnerable to floods. According to the center for
research in the epidemiology of disasters (CRED) through its emergency event database
(EMDAT) which has been recording statistics of losses and damages from the disasters
events globally indicated that floods have affected more people than any other hazards
in the 21st century. In 2021, three major flood events that occurred in India, China, and
Afghanistan, are categorized among the top 10 deadliest disaster events (CRED, 2021).
Every year floods take thousands of lives, affect large number of people and cause
significant economic losses all over the world (Table 1).

20
Year Flood Casualties Economic losses
events (Billion US$)

2021 223 Death-4143, affected (in million)-29.2 74.4

2001-2020 163 Death-5185, affected (in million)-82.7 34.1

Table 1: Losses due to flood in year 2021 compared to 2001-2020

(Source: CRED, 2021)

Figure 1: Frequency of occurrence of natural hazards

(Source: CRED, 2021)

21
2.2 Floods in Nepalese Context

Nepal is considered as one of the most disaster-prone countries in the world. It is the
second highest country at risk of floods in South Asia (UNDP, 2009). It has been
affected by the extreme weather events that results in natural disasters such as floods
and landslides. Due to fragile geological conditions, topographical extremities, climatic
extremities and seismic activities along with population growth, poverty, illiteracy,
deforestation, unscientific agricultural practices, land use changes, and developmental
activities such as construction of roads, irrigation systems, hydro-powers and
urbanization make it vulnerable to several natural disasters. River flooding signifies the
most common global hazard causing significant number of losses. Throughout human
history, river floods have caused extreme damages on economy and lives and caused
heavy economic losses than any other hazard. Flooding has a particular impact on
communities residing within the flood plain. It indicates that the most of flood disasters’
victims are poor people. The increase in population and squatters of landless people
living on the river bank and encroachment of flood plain making them vulnerable to the
flood damage (Dangol & Bormudoi, 2015).

Globally, Nepal lies in the 4th rank in terms of climate change vulnerability, 11th rank
in terms of earthquake and 13th in terms of flood vulnerability (UNDP, 2019). Also,
Nepal is ranked as the second highest country at risk of floods in South Asia. Most of
the rivers originates from the Himalayas and more than 6000 snow-fed and perennial
rivers and rivulets. Research by Dewan 2015 stated accelerated development works,
rapid urbanization, accelerated retreat of glaciers and increased intensity of monsoon
precipitation as the cause for frequent flooding in Nepal.

Extreme precipitation events during monsoon periods are common in Nepal. Extreme
rainstorms in the past 120 years (17-18 September 1880, 28-30 September 1924, and
19-21 July 1993) were caused by abnormal behavior of monsoon depression paths
originating from Bay of Bengal when associated with low-pressure systems in Nepal
(Yogacharya & Gautam, 2008). The historical data showed that the country witnessed
major floods in Tinau; 1978, Koshi; 1980, Tadi; 1985, Sunkoshi; 1987, Kulekhani;
1993, Seti; 2012, Mahakali; 2013, Karnali; 2014 Hanumante; 2018, Melamchi; 2021
and many more.

22
 Figure 2: Human loss due to different disaster in 2017-2018

(Source: Nepal Disaster Report,2019)

Heavy rainfall during 19-21 July 1993 in central and eastern regions of Nepal had
disastrous consequences with heavy loss of life and property as well as damages to
infrastructures by floods, landslides and debris flows. Within the country, more than
500,000 people were directly affected, 1336 people dead and 163 injured and 17,113
houses destroyed. In the agricultural sector, more than 57,584 hectares of arable land
were damaged; and 67 small and large irrigation projects along with thousands of
farmer-managed irrigation schemes were seriously damaged (Yogacharya & Gautam,
2008). 1998 floods and landslides that severely affected the terai and middle hill region
claimed 273 human lives, injured 80 people, affecting 33,549 families, damaging
13,990 houses, and ruining 45,000 hectares of land and agricultural crops resulting in a
total loss of about NRs. 2 billion. Also, August 2008 flood of Koshi River affected 4
village development committees causing more than 107,000 people homeless in
Nepalese side. About 6000 hectares of agricultural land was inundated; and agricultural
products worth more than US $ 3.7 million have been damaged. Heavy rainfall in far

23
western Nepal in September, 2008 had also resulted in devastating floods in
Kanchanpur, Kailali and Bardiya districts caused extensive damage to lives and
property.. 2014 flood in Surkhet also killed 24 people, and more than 12, 385 people
from 2327 families were displaced.

In 2017 flood, terai region of Nepal, the Ministry of Home Affairs declared that 135
people lost their lives. Nearly 80,000 houses were completely destroyed and many more
suffered damages (NRCS, 2016). The incessant rainfall for almost 10 hours in July 12,
2018 caused Hanumante River of Bhaktapur to swell causing waterlogging in urban
areas of Thimi and Bhaktapur.

It is seen that the number of deaths due to extreme events is decreasing but at the same
time the number of families affected, agricultural areas impacted and monetary
damages is steadily increasing. This shows that while the exposure and even
vulnerability are increasing, systems and institutions are able to limit losses, but failed
to reduce risk or prevent increased impact on livelihoods.

Nepal comes at the top among the developing countries in Asia where the flood impacts
are high. Every year floods hit the districts of terai very hard and the impacts are that
high that another flood occurs before that could be recovered.

2.3 Flooding Problem in West Rapti River

According to Yogacharya & Gautam, 2008, the medium size rivers originating from
middle mountain ranges such as Kankai, Kamala, Bagmati, East Rapti and West Rapti
cause more flooding problems than major rivers like Koshi, Narayani and Karnali. The
major causes of flooding in this basin are natural causes like. high rainfall ,soil erosion,
flat topography, debris flows and sedimentation, river coarse shifting, deforestation,
poor planning, design and construction practices of roads, massive increase of
settlements along the highway, the institutional causes like lack of participatory
approach in disaster management, lack of long-term comprehensive flood
management plan covering whole flood prone areas and socio- political causes like lack
of public sensitivity and awareness, attitude of people and culture of negligence
(Gautam & Phaiju, 2013). This basin usually suffers from flash floods as the catchment

24
 responds to high intensity and short duration precipitation. Recent extreme flood events
were reported in 1984, 1989, 1998 and 2007, 2013 and 2017.

2.4 Causes of Flooding in Nepal

Floods are a natural phenomenon during the monsoon. MoHA, 2010, in Nepal Disaster
Report 2009, states that the main causes of flooding in Nepal as follows

a) Rainfall Variability

Floods result from the uneven distribution of precipitation in time and space. More than
80% of the rainfall in the country occurs during the monsoon from June to September.
The average annual rainfall is 1627 mm (Alford, 1992). Torrential rainfall and
cloudbursts that bring over 400 mm rainfall in a single day are common in the
Mahabharat Range and often cause heavy floods. Rainfall intensity during a single hour
is equally high, with over 40 millimeters common in the lower Mahabharat and Chure
ranges. In 1989, Pokhara recorded a tremendous 88 millimeters in one hour, and Bhusal
et al. (1993) report that the 45 millimeter/hour rainfall on 29 September, 1993 triggered
massive landslides and debris flows in South Central Nepal.

b) Topography

The slopes of Nepal’s hills and mountains range from steep to very steep (more than 30
per cent), while the terai is flat (less than 10 per cent). Due to this change in gradient,
rivers in the highlands flow at very high speeds and have a large sediment carrying
capacity, but when they enter the terai and slow down, all this material—stones, gravel,
sand and silt—is deposited as an alluvial fan. As this deposit raises the level of the river
bed, it increases the likelihood of flooding even during moderate rainfall.

c) Deforestation
As vegetation intercepts rainfall, it reduces runoff and therefore the possibility of soil
erosion. As agriculture reduces vegetation cover, it has contributed to local flooding.
Humans also alter the naturally existing conditions in the watershed through the
construction of hydraulic structures, urbanization, deforestation and quarrying.
Compared to urban watersheds, rural watersheds have greater interception of
precipitation. The many impervious surfaces such as roofs, streets and paved zones in

25
a city channel water and thereby accelerate runoff; they also impede surface drainage
and contribute to localized flooding. In research of Upadhaya, 2007 attributes the
flooding of the Dhobi Khola of Kathmandu to increasing urbanization because it
increases peak flow and surface runoff, but reduces base flow.

2.6 Flood Hazard and Vulnerability Assessment

A hazard is defined as a process, phenomenon or human activity that might cause loss
of life, injury or other health impacts, damage in property, social-economic disruption
or environmental degradation. Hazards can be natural, anthropogenic or socio natural
in origin (UNISDR, 2013).

Flood hazard mapping is beneficial in aspects of land and water resources management
such as preventing unwise land use in flood prone areas. Through a proper flood plain
hydraulic analysis, design of structures like bridge and culvert openings for roadway
crossings of streams and flood reduction measures like dams, levees can be achieved
(Salimi et al., 2008). According to Dangol & Bormudoi, 2015 among various non-
structural measures of disaster mitigation, hazard and vulnerability and risk mapping is
one of the important nonstructural measures. Flood hazard assessment is the estimation
of overall adverse effects of flooding in a particular area. Many parameters can be
considered such as depth of flooding, duration of flooding, wave velocity of flood, and
rate of rise of water level (Aryal et al., 2020).

Vulnerability is a concept which describes factors or constraints of an economic, social,

physical or geographic nature, which reduce the ability to prepare for and cope with the
impact of hazards. UNDP ranked Nepal as 30th country with respect to relative
vulnerability to flood in a study. According to Dhakal, 2013 of the total death by any
type of natural disaster in 2010, 29.02% were by flood and 24.55% were by landslides;
and 71.35% of the total affected families by any type of disasters in 2010 are by flood.
Risk is the probability that negative consequences may arise when hazards interact with
vulnerable areas, people, property and environment & it is the product of hazard and
vulnerability.

One of the major challenges in flood prone area is managing floods, people's safety and
environment. So, reducing vulnerability and increasing resilience are important

26
methods to overcome from above challenges. For this, measuring vulnerability is one
of the key elements, which also involves identifying susceptible regions and taking
appropriate action to minimize the effects. Generally, vulnerability assessment is
carried out studies using three major approaches: vulnerability matrices, vulnerability
curves and vulnerability indicators.

Vulnerability Matrices: -

This method of vulnerability assessment involves the development of vulnerability or

damage matrices and is generally known as a qualitative method for vulnerability
assessment. The development of a matrix is based on empirical data or expert
judgement. Buildings that have not been damaged by the event are given a vulnerability
of 0 and the ones that are totally damaged a vulnerability of 1. All the other buildings
are organized within in-between categories by an expert. Vulnerability matrices are a
qualitative method which presents the possible damages on elements at risk.

Vulnerability Curves:

Vulnerability curves can be defined as a curve associating the intensity of the hazard to
the damage response of a building. The physical vulnerability of individuals can be
quantitatively evaluated using these curves.

A plot of floodwater depth against the percentage of damage for different types of
houses is known as depth damage curve. It is widely used for determining structure
damage caused by floods. This curve relates the percent damage or estimated economic
loss of the building with respect to depth of water level.

Depth damage curve is generally used for determining damages that are directly caused
by flood which denote the flood damage that would occur at specific water depths per
asset. Assessment of potential damage that can be caused by flood events is an
important component in vulnerability assessment. Vulnerability curves focuses on the
degree of loss. It provides information on loss and hazard magnitude/intensity for a
building.

27
Vulnerability Indicators: -

(Birkmann & Wisner, 2005) defined vulnerability indicators as “variables which are
operational representations of a characteristic or quality of the system able to provide
information regarding the susceptibility, coping capacity and resilience of a system to
an impact of ill-defined event linked to a hazard of a natural origin”. These indicators
help to assess the impact of disasters on social, economic and environmental conditions
and disseminate the results to decision-makers, the public and populations at risk”. This
approach includes the selection of relevant indicators, the identification of variables,
their weighting and, finally, their aggregation in a vulnerability index.

2.7 Flood Hazard Maps in Context of Nepal

Flood is a major disaster in case of Nepal which affects many lives and causes economic
losses every year and the vulnerability to flood damage has increased with increase in
population and squatter settlement at the riverbanks increasing river encroachment.

Many researchers have carried out the study regarding preparation of hazard maps in
different rivers of Nepal. Some of them are listed below:

Manandhar, 2010 carried out a study in Lothar Khola basin using HEC RAS with the
help of recorded precipitation and river flow data of the basin. Dangol, 2017 assessed
the flood inundation in Balkhu Khola using steady flow analysis and the result shows
barren area near the river is susceptible to flood hazard, which indicates future human
lives are more prone to disasters as those lands have gone through planning for future
settlement. Gautam & Kharbuja, 2006 carried out flood hazards maps of Bagmati River
of Kathmandu valley using unsteady hydrodynamic modelling.

Khanal, 2004 studied floods in Madi River watershed central Nepal. It tries to highlights
type, magnitude, recurrence interval of floods in Madi River and damages caused by it.
This study shows that high magnitude destructive flood in the mountain areas is
triggered by landslide and debris flow and downstream areas are highly affected. Karki,
et al. (2011) assessed flood hazards, their impacts and the resilience of communities at
the watershed level. danger level and warning level of flood were identified using
maximum instantaneous discharge data and gauge height. The results show that

28
agriculture system of the study area in a geographically vulnerable position.
(Talchabhadel & Sharma, 2014) studied West Rapti River based on real time data
analysis. The real-time flood early warning system together with the development of
water management and flood protection schemes play a crucial role in reducing the loss
of lives and properties in the basin. Basically, this paper presents an overview of flood
problems in the West Rapti River basin, causes and consequences of recent floods and
the applicability and effectiveness of the real time data to flood early warning in Nepal.
(Thapa et al., 2020)have studied on Khando catchment in eastern Nepal and focuses to
quantify the hazard and vulnerabilities across one of the frequently flooding catchments
Khando River and conducted flood hazard assessment for 20, 50, 100, and 200 years
return periods.

Based on the measured inundation depths, they have created vulnerability and fragility
functions. Tamang et al., 2017 studied Malekhu khola in central Nepal and prepared
flood hazard maps using GIS and HEC-RAS. It was concluded that the downstream
portions are comparatively more affected by the flood rather than the upstream section.

Many research and flood hazard mapping has been done in urban areas. Flood hazard
mapping and vulnerability analysis of Bishnumati River was done by Dangol et al.,
(2015) using HEC RAS version 4.0, GIS and HEC-geoRAS as an interface between
HEC-RAS and GIS. The hazard map as prepared showed the inundation coverage of
the area which showed urban land and cultivated land as the most vulnerable areas.

2.8 Causes of Increasing Vulnerability of People and Structure

The number of flood events and its impacts on people and economy has been increasing
each year. Though the fatalities have decreased in some of the regions of Nepal, the
number of affected people due to flood has an increasing trend. Many people at the terai
areas are vulnerable to floods. The main cause of vulnerability of those people during
flood events are listed below:

a) Encroachment of people towards the river corridor

Many civilizations started at the bank of the river in the history. People has been
attracted towards the riverbanks for settlement from the very past. At present informal

29
settlements also known as slums has increased rapidly along many rivers’ basin in
Nepal. Due to this, those people are vulnerable to the flood impacts. Moreover, many
infrastructures have been built along the sides of river and there are many examples that
they have been affected during the flooding. For examples, Various places near
Hanumante River like Radheradhe, Jagati, Madhyapur Thimi, Sallaghari, Sirjananagar
was affected anf flood water also entered houses in Duwakot, Nikosera, Liwali,
Changunarayaan, Beahmayani, Nayabato and also affected the Arniko Highway due to
incessant rainfall on 11 July, 2018.

b) Distributing natural pathway of rivers

The river starts flooding when its pathway is distributed. The natural path of the river
might be distributed by the formation of natural dam due to landslide and other is the
anthropogenic activities. Anthropogenic activities like construction of hydraulic
structures like dams, bridges, embankment walls and other structures blocks and natural
route of the rivers. However, when the discharge is high, the river finds its path and
results in flooding to the nearby lowland areas and making the life of people there prone
to flood effects.

c) Excessive extraction of sediments from riverbanks and near bridges

The rivers originating from the Churia hills bring a lot of sediment which results in
eroding the hill slopes as well as river banks. The deposition of sediment significantly
increases the bed level of the river. The extraction of such sediment from river and near
brides without any limitation is causing problem in flow dynamics.

d) Change in land use

Increased and unmanaged urbanization has converted the natural green lands to the
urban concrete areas. A study showed that landcover projection for 2021 has an
increasing trend threating the forest and agriculture lands (Rimal,2011). The urban
concrete areas instead of forest and agricultural areas trigger flash flood during heavy
rainfall. Due to which the people are vulnerable to the impacts of that flash floods.

30
2.9 Flood Frequency Analysis

Flood frequency analysis is the procedure for estimating the frequency of occurrence
(return period) of a hydrological event such as flood. The technique involves using
observed annual peak flow discharge data to calculate statistical information such as
mean values, standard deviations, and skewness and recurrence intervals. These
statistical data are then used to construct frequency distributions, which are graphs and
tables that tell the likelihood of various discharges as a function of recurrence interval
or exceedance probability. Though the nature of most hydrological events such as
rainfall is erratic and varies with time and space, it is commonly possible to predict
return periods using various probability distributions.

Flood frequency analysis is used for making probabilistic estimates of a future flood
event based on the historical streamflow record, with probability often expressed as the
average length of time between floods and called the return period or average
recurrence interval (T).

For the quantitative assessment of flood problems, flood frequency estimation is very
essential. We can extrapolate the future possibilities floods using the past record of
flood events (Manandhar, 2010). After a detailed study of the gauge data and its
descriptive parameters such as mean and standard deviation, etc. and applying
probability theory, one can reasonably predict the probability of occurrence of any
major flood events in terms of discharge or water level for a specified return period
(Singh, 2004). The use of historical data in the estimation of significant flood events
has grown in recent years, although extended data series are necessary for accurate
estimations for high floods.

The selection of a return period is linked to the level of associated risks. The size of
catchment, importance of project area, expected benefits and economic viability are
some of the criteria that need to be considered while selecting return periods. Low risk
is associated with the protection of agricultural land and rural areas whereas high risk
is linked with the protection of populous urban and industrial areas. In general, the Soil
Conservation Service under the US government recommends the use of a 25-year
frequency for minor urban drainage design and a 100-year frequency is recommended
when extensive property damage may occur (WECS, 2019).

31
There is no exact method to compute the right value of discharge which cause flooding
while there are various methods that use empirical and probability approach. Gumbel’s,
Log Normal and Log Pearson III type, etc. are based on probability theory (Manandhar,
2010)

2.9.1 Gumbel’s Distribution

This extreme value distribution was introduced by Gumbel (1941) and is usually known
as Gumbel’s distribution. In general practice extreme value type I distribution is known
as Gumbel’s distribution to fit flood discharges of various rivers. It is one of the most
commonly used probability distribution functions for extreme values analysis in
hydrologic and meteorologic studies for prediction of peak flood, maximum rainfalls,
maximum wind speed, etc.

Gumbel defined a flood as the largest of 365 daily flows and annual series of flood
flows constitute a series of largest values of flows. According to his theory of extreme
events, the probability of occurrence of an event equal to or larger than a value x0 is

−𝑦
𝑃(𝑋 ≥ 𝑥0 ) = 1 − 𝑒 −𝑒 Eq. 1

In which y is a dimensionless variable given by

1.2825
𝑦 = 𝛼(𝑥 − 𝑎) 𝑎 = 𝑋̅ − 0.45005𝜎𝑥 𝛼= 𝜎𝑥

1.285(𝑥−𝑥̅ )
Thus, 𝑦 = + 0.577 Eq 2
𝜎𝑥

Where, 𝑥̅ = 𝑚𝑒𝑎𝑛 𝑎𝑛𝑑 𝜎𝑥 = 𝑠𝑡𝑎𝑛𝑑𝑎𝑟𝑑 𝑑𝑒𝑣𝑖𝑎𝑡𝑖𝑜𝑛 𝑜𝑓 𝑡ℎ𝑒 𝑣𝑎𝑟𝑖𝑎𝑡𝑒 𝑋

In practice, it is the value of X for a given P that is required as and as such eq 1 is

transposed as,

𝑦𝑝 = − ln[ln(1 − 𝑃)]

1
Noting that the return period T = 𝑃 and designating

𝑦𝑝 = 𝑡ℎ𝑒 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝑦, 𝑐𝑜𝑚𝑚𝑒𝑛𝑙𝑦 𝑐𝑎𝑙𝑙𝑒𝑑 𝑡ℎ𝑒 𝑟𝑒𝑑𝑢𝑐𝑒𝑑 𝑣𝑎𝑟𝑖𝑎𝑡𝑒, 𝑓𝑜𝑟 𝑎 𝑔𝑖𝑣𝑒𝑛 𝑇

32
 𝑇
𝑦𝑇 = − [ln. ln ]
𝑇−1

𝑇
𝑦𝑇 = − [0.834 + 2.303 log log ]
𝑇−1

Now, rearranging the Eq. 2, the value of variate X with a return period T is

𝑋𝑇 = 𝑥̅ + 𝐾𝜎𝑥 Eq 3

(𝑦𝑇 −0.577)
𝑊ℎ𝑒𝑟𝑒 𝐾 = Eq 4
1.285

Eq 3 and 4 constitute the basic Gumbel’s equations and are applicable to an infinite
sample size (i.e. N → ∞)

Gumbel’s Equation for Practical Use:

The Eq 3 can be write as,

𝑋𝑇 = 𝑥̅ + 𝐾𝜎𝑛−1 Eq 5

∑(𝑥−𝑥̅ )2
Where𝜎𝑛−1 = standard deviation of the sample size N=√ 𝑁−1

K= frequency factor represent as

(𝑦𝑇 −𝑦̅𝑛 )
𝐾= Eq 6
𝑆𝑛

𝑇
𝑊ℎ𝑒𝑟𝑒 𝑦𝑇 = − [ln. ln 𝑇−1] Eq 7

𝑇
𝑂𝑟, 𝑦𝑇 = − [0.834 + 2.303 log log ]
𝑇−1

𝑦̅𝑛 = reduced mean, a function of sample size N,

for N → ∞, 𝑆𝑛 →1.285 and when y→∞ , 𝑦̅𝑛 → 0.577

(𝑦𝑇 −0.577)
𝑇ℎ𝑒𝑛 𝐾 = 1.285

33
 Chow has given a formula to find frequency factor for large data as:

𝑇
𝐾 = − [0.45 + 0.78 ln ln ]
𝑇−1

2.9.2 Log Pearson Type III

This distribution is extensively used in US for projects sponsored by US for frequency

analysis of annual maximum floods. In this method, the variate is first transformed into
logarithmic form base 10. lf X is the variate of a random hydrologic series, then the
series of Z variates

Where,

Z = log X Eq 1

are first obtained. For this Z series, for any recurrence interval T,

ZT= 𝑍̅+KZ𝜎𝑧 Eq 2

Where, KZ = a frequency factor which is a function of recurrence interval T and the

coefficient of skew Cs,

𝜎𝑧 = standard deviation of the Z variate sample

̅ 2 /(𝑁 − 1)
=√∑(𝑧 − 𝑧)

Where Cs = Coefficient of Skewness

𝑁 ∑(𝑧−𝑧̅ )3
=(𝑁−1)(𝑁−2)(𝜎 3
𝑧)

𝑧̅= mean of the z values

N=sample size = number of years of record

After finding ZT by using Eq 2, the corresponding value of XT is obtained by Eq 1 as

XT =antilog (ZT)

34
2.9.3 Log-Normal Distribution

A log-normal distribution is a continuous distribution of random variable y whose

natural logarithm is normally distributed. Thus, if the random variable X is log-
normally distributed, then Y = ln(X) has a normal distribution

The Log Pearson-III distribution converges to Log-Normal Distribution when

coefficient of skewness (Cs) = 0

2.10 Flood Modeling

Recently, with the advancement of science and technology, many computer models
have been developed in the world. The hydrological models are capable of estimating
the flow based on given rainfall at any location of the river reach. Due to the capability
of such models to generate the flow hydrograph in the basin, such models are now used
for the forecasting the flood based on either real time or near real time surface weather
observation or using the weather forecast data. Flooding is a natural hazard that can
cause devastating damage. To protect people and property from this hazard, many kinds
of flood mitigation practices have been developed to lower either the damage caused
by floods or the likelihood of their happening. Generally, flood mitigation practices
involve construction of various kinds of man-made structures such as dams, detention
basins, levees, floodwalls, embankments, diversions, etc and the flood inundation
models that are available and capable of estimating the hydro-dynamic characteristics.
The additional part is that such models can incorporate the structures such as dams,
levees or embankment or any other infrastructure such as bridge.

There are different types of hydrological and hydrodynamic model like Hydrologic
Engineering Center's Hydrologic Modeling System (HEC-HMS), Soil and Water
Assessment tool (SWAT), Mike SHE, Mike 11, Hydrologic Engineering Center's River
Analysis System (HEC-RAS) etc. being used in different parts of the world. Its efficient
uses are becoming appropriate for water resource management, planning and decision-
making process. In table 2 below there is mentioned about the various types of models.
Also, these models are well tested in different basins in Nepal.

35
2.11 Available Tools for Modeling

There are a number of hydrologic and hydraulic tools or models that can be used for
flood modeling. Many computer-based models have been developed in the last
three/four decades. Among them few are commonly used mathematical models
(Jodhani et al., 2022), (WECS, 2020) as shown in table below:

Name Uses Advantages Disadvantages Availability

MIKE Mainly used for flow Includes Limited to river Commercial

11 simulation and water almost all and fluvial
(WECS level identification, aspects of related flood
, 2020) water quality and river events
sediment transport in modelling
rivers, flood plains,
irrigation canals,
reservoirs and other
inland water bodies

HEC- Shallow water and Wide range of Instability and Open-source

RAS unsteady flow parameters, limitations
1D,2D user friendly environments
(WECS in
, 2020) multidimension
al modelling

Telema Used in open channel Can perform Conditional Open-source

c-2D and floodplain simulation in stability
modelling transient and
permanent
conditions

36
 (Jodhan
i et al.,
2022)

LISFL Works on GIS based Extensive Requires high Open-source

OOD- techniques documentation resolution
FP , flexible and topographic
(Jodhan user friendly data for
i et al., for setting up simulation
2022)

ISIS- Modelling level and Wide range of Requires high Open-

FREE,1 water flow in open application resolution source,
D,2D, channels including topographic Commercial
FAST urban areas, data for
(Jodhan coastal and simulation
i et al., river channels
2022)

TUFLO Simulation of Dynamic Slow and Commercial

W- complex linking uncertainties in
1D,2D hydrodynamics of capability solution
(Jodhan flood between
i et al., domains
2022)

Table 2: Comparison of different models

According to (Dangol & Bormudoi, 2015) the application of HEC RAS and ArcView
with HEC-GeoRAS provide effective results within less time consumption and little
resources. Also, Banstola and Sapkota (2019) used Hydrodynamic Model HEC-RAS
for Flood Risk Mapping and Analysis of Daraudi River, Gorkha, Nepal. A direct

37
relationship between the inundation area and the flood discharge of various return
periods was identified. Similarly, (Acharya et al., 2011) prepared flood risk map of
upper Bagmati Basin using GIS and a hydrological model (HEC-RAS) and concluded
that most of the places of Kathmandu NP is highly vulnerable to flood. (Bhattarai et
al.,) prepared inundation map of a section of Babai Basin and to find out the inundated
areas for different return periods using GIS and HECRAS. The assessment of flood
vulnerability with relation to the return period of floods and their water depth. Bhattrai
et.al (2009) studied the effects of the Laxmanpur Barrage and the Kalkaluwa Bandh
(dyke) near the Indo-Nepal border using HECRAS in his study Flood Hazard Modeling
of West Rapti River Basin of Banke District of Nepal. Hence, this shows that the most
widely and frequently used hydaulic model in Nepal is HECRAS which can be defined
as wide acceptable model in context of Nepal.

2.12 Hydrodynamic Model: HEC RAS

HEC-RAS is a hydraulic model developed by the Hydrologic Engineering Center

(HEC) of the U.S. Army Corps of Engineers. This software can be used to perform one
dimensional steady flow, one and two-dimensional unsteady flow calculations for flood
plain areas, natural and constructed channels etc. (Brunner, 2016). Within steady flow
analysis, it can be modeled for sub critical, supercritical or mixed flow regime (Acharya
et al., 2011). The model is applicable for steady flow water surface profile computation,
unsteady flow simulation, movable boundary sediment transport computation and water
quality analysis.

Research Title Research Author

Year
Flood Risk and Vulnerability Mapping Using 2004 Shrestha et.al.,
GIS: A Nepal Case Study
Flood Risk Mapping of Upper Bagmati Basin 2011 Acharya and
Deuja
Flood hazard mapping and vulnerability analysis 2015 Dangol
of Bishnumati river, Nepal

Table 3: Study of flood modeling and analysis using HEC-RAS and ArcGIS in
Nepal

38
HEC-RAS can perform inundation mapping of water surface profile results directly
from HEC-RAS. The geometric data can be created in RAS mapper. The DEM of the
study area can be directly imported to RAS mapper and using this terrain as base map,
the geometric data can be created. Through RAS mapper, inundation depth and
floodplain boundary dataset are created.

This method has been successfully used worldwide for flood modelling and analysis.
Table 3 above shows use of ArcGIS and HEC-RAS method for flood hazard and flood
risk mapping in case of Nepal.

2.13 Calibration and Validation

Model validation requires calibration of model parameters. According to Horritt, 2005

the calibration is a process where the model parameters are adjusted for fine-tuning of
the model outputs in order to obtain different model behavior and thus validate the
model by its performance. Calibration is generally a trial-and-error process of multiple
iteration, consisting adjustments and comparisons to the real system before the
acceptable level of accuracy is achieved. Calibration can be done with any model
parameter that have unambiguous value, such as the Manning roughness coefficient,
which can hardly be univocally defined. In inundation mapping, models are
traditionally calibrated by adjusting the Manning roughness coefficients of the channel
and the floodplain. (Horritt 2006; Di Baldassarre 2012).

According to Horritt (2005), the environmental models are generally validated through
a field and ground-based measurements. Di Baldassarre (2012a: 52) and Bates et. al.
(2007: 294) have presented three commonly utilized data for the model validation: 1)
at-a-point bulk flow measurements, such as stage and discharge time series from gauge
station, 2) spatially distributed binary data like flood extent maps, and 3), spatially
distributed continuous point data like measured water surface levels at certain locations
or high-water marks.

2.14 Geographic Information System (GIS)

A Geographic Information System (GIS) is a computer system for capturing, storing,

querying, analyzing and displaying geographic referenced data (Chang, 2006). It is an

39
instrument (often a software package) that is increasingly used to study and analyze
spatial data. Some of the major issues facing society today have a geographical
component. This explains why GIS is used for a wide variety of applications and in a
wide variety of sectors, e.g., production of (topographic) maps.

GIS may be used to make interactive map overlays for flood hazard management that
quickly and clearly show which parts of the area are at risk of flooding. Such maps can
then be used to coordinate mitigation efforts before an event and recovery after. In this
study, Arc GIS 10.2.2 is used for generating inundation maps.

2.15 Policy Context

One of the key components of Disaster Risk Management (DRM) is the institutional
and policy framework of the government. The success of disaster management
activities mostly depends on systematic and organized formulation of policy strategies,
legal provisions, institutions and its roles and responsibilities in dealing and responding
with disasters (Nepal et al., 2018). Generally, the management actions and
arrangements are divided into four phases of a disaster cycle and they are preparedness,
response, relief/recovery and mitigation (Noji, 2005). Unsound and unmanaged disaster
management policy and practices in national level might increase disaster risk and
disaster losses (Hasan et al., 2013). International bodies like UNFCC, IPCC, Hyogo
framework, Kyoto protocol etc. concern about disaster events.

Government of Nepal has also made effort to formulate and implement various legal
and policy provisions to create favorable environment to disaster risk management.
Integrated flood risk management is guided by a large number of strategies, policies,
guidelines, action plans and framework. The national plans and acts regarding disaster
are noted below:

1. Natural Calamity Relief Act, 1982

2. National Action Plan on Disaster Management in Nepal, 1996
3. National Strategy for Disaster Risk Management, 2009
4. District Disaster Management Plan, 2010
5. Local Disaster Risk Management Planning Guidelines, 2011
6. Nepal Risk Reduction Consortium,2011

40
7. Nepal Disaster Response Framework,2013
8. National Early Warning Strategy,2013
9. Water Induced Disaster Management Policy, 2016
10. Disaster risk reduction and Management Act, 2017
11. National Policy on Disaster Risk Reduction, 2018
12. Disaster risk Reduction National Strategic Plan of Action (2018-2030)
13. Private Housing Rebuilding Grant for the Flood and Landslide Victim, 2017
14. Guidelines for the Relocation and Rehabilition of High Risked Settlements,
2018

Disaster Risk Reduction and Management Act, 2017 has made provision of effective
disaster risk management throughout the disaster management cycle-preparedness,
response and rehabilitation and mitigation. Also, this Act decentralizes the role of
disaster management activities and provides detailed action plan right from the central
government to the district and to the local levels to implement and execute a disaster
management plan. Most importantly the Act has made provision of recommendation to
the government of Nepal for the declaration of disaster-prone zones and also the
streamlining of responsibilities and involvement of local communities for risk
management.

Despite the huge growing concerns, increasing flood events continue to cause damages
worldwide even in case of developed countries and demand high investment capacity
thus demanding a new strategy and policy for flood management (Bertilsson et al.,
2019a). The same research states that flood policy should consider risk management
along with structural and non-structural measures to prevent, mitigate, prepare, respond
and recover from flooding.

41
 CHAPTER THREE: METHODOLOGY

3.1 Study Area

The West Rapti River basin is located in Lumbini Province of Nepal. Geographically
this basin extends from 27˚56'50'' to 28˚02'30'' North latitudes and 81˚45'00'' to
81˚40'00'' East longitudes. This river originates from the middle mountains of Nepal
and then enters the lowlands and finally drains to the Ghagra (Karnali) River. It has
several tributaries. Major tributaries are Jhimruk River, Madi River, Arun River, Lungri
River, Dunduwa River, Sotiya and Gandheli rivulets. Downstream of the confluence of
the Jhimruk and Madi Rivers, the river is named the West Rapti River. This river
streams through 4 different districts of Nepal: Rolpa, Pyuthan, Dang and Banke (from
upstream to downstream). The length of the river in the Nepalese side is about 280
kilometers, and the approximate size of the catchment area up to the state border is 6250
square kilometers (Gautam & Phaiju, 2013).

The discharge in the river is mainly regulated by the monsoon season from June to
September. Other times of the year the river discharge is relatively low. The study
report of ICHARM (2008) states that the increment in the upstream erosion and debris
flow has caused the sediment accumulation in the downstream thus, conveyance
capacity of West Rapti river is declining. This, may cause floods with even smaller
discharges than before.

For this study, the study area is Bagasoti-Kusum section of West Rapti river basin. The
length of the stretch is 85km and the average gradient is 0.001285 and within study area
altitude ranges from 381m to 235m. The lower basin of the river is gauged on three
hydrometric stations, which are located in Kusum, Jalkundi and Bagasotigaon. Since
the study area is from bagasotigaun, the discharge data from that station is used in the
hydraulic modelling. Figure 3 shows the Map of West Rapti river basin.

42
 Figure 3: Map of West Rapti River Basin Showing Study Area

The most characteristic features of the lower basin of West Rapti river is the shifting of
the main river channel. (Bhusal, 2004) states that some of the geomorphological
conformities that are common in the alluvial river plains, such as the lower basin of
West Rapti, the large stream rivers tend to have a relatively smooth slope and the width
of the channel is significantly greater than the depth. This phenomenon is well suited
to the study area circumstances and are main aggravators for the shifting of the West
Rapti river channel. Also, bank erosion has intensified the poor conditioning of the river
environment and lead to an increased flood risk (Bhusal 2004).

The delineation of the study region is based on three vital points related to the
characteristics of the region. First, one of the most vulnerable and affected area of this
basin due to dense settlements, livelihood of the people (which mainly depends on
agriculture and livestock) and plain region of the river. Second, the quality of available
topography and hydrological data. Third, the physical characteristics i.e., river
environment of the river, In the process of creating a decent representation of the river
channel by the GIS-methods, the heavily braided upstream reach becomes a notable

43
issue, whereas the downstream end of the reach was significantly easier to construct
due to its lighter meandering and negligible rate of braiding. For the abovementioned
reasons, only the downstream part up to Kusum station of the river was modelled.

3.2 Acquisition and Processing of Data

The data were collected from the multiple sources. Hydro-met data were taken from
DHM whereas other spatial data such as Digital Elevation Model, Land use,
topographic map and other data required for the models were obtained from freely
available sources.

Literature review was done throughout the course of the study.

3.2.1 Digital Elevation Model

A Digital Elevation Model (DEM) is a digital representation of topography. It is also

widely known as a Digital Terrain Model (DTM). A DEM can be represented as a raster
(a grid of squares) or as a Triangular Irregular Network (TIN). DEMs are one of the
most important spatial data sources for digital hydrologic analyses as they describe the
topography that drives surface flow. The DEM used for the study was ASTER dem
with 30m resolution that was downloaded from USGS Earth Explorer site. Shuttle
Radar Topographic Mission SRTM (90m), SRTM (30m), ASTER (30m) which were
compared to get better DEM among all. The comparison shows that the elevation from
ASTER (30m) is closer than others in the area. So, ASTER (30m) was selected in the
study. The plot of the elevation from the ASTER DEM is shown in Figure 4. The map
shows the highest elevation of 3646 m at upstream part of the catchment and lowest of
about 109 m near Indo Nepal border.

44
 Figure 4: Digital Elevation Model

3.2.2 Land Use Data

Land use data was collected from the irrigation master plan 2019 as shown in Figure 5.
Basically, eight different land use types are seen on this basin. In the study area forest
is the dominant land use in the upper part whereas agricultural area is dominant in the
lower part. Agricultural land use is dominant on the study area as well because of
availability of water for irrigation. It is also seen that the sand/gravel/boulder is
dominant after agricultural this is because the area has wide flood plain and the regular
inundation of land leaves the residuals from the river.

45
 Figure 5: Land Use Map of West Rapti river basin (Source: IMP_2019)

3.2.3 Hydro-Meterological Data

The hydro-metrological data was taken from Department of Hydrology and

Meteorology (DHM). Published hydrological data for year 1976 to 2015 of Bagasoti
Gaun hydrological station (station No. 350) was used in this study. Table 4 shows the
hydro-meteorological stations in and around the West Rapti Basin.

46
 No. St. No Name of River Location Latitude Longitude Elevation (m)

1 327 Lungri Khola Khungri 28.23 82.71 695

2 330 Mari khola Nayagaon 28.07 82.80 536

3 333 Arun Khola Devisthan 28.03 82.76 575

4 350 Rapti River BagasotiGaun 27.90 82.85 381

5 350.5 Ransing Khola Tinkunne 27.79 82.82 250

6 360 Rapti River Jalkundi 27.56 82.13 218

7 361 Khari khola Agaiya 28.14 81.89 222

8 364 Dundawa River Masurikhet 28.20 81.70 162

9 375 Rapti River Kusum 28.00 82.05 235

Table 4: Hydro-meteorological stations in the West Rapti Basin (Source: DHM_2021)

3.3 Flood Frequency Analysis

Flood frequency analysis was carried out on the output from the model using statistical
distributions such as Gumbel, Log-normal, and Log-Pearson. The estimated flood flows
at various return periods were form the input to the hydrodynamic model for flood
hazard mapping.

The calculation of frequency analysis for the three different methods was done in MS
Excel sheet and the output of the three distributions was compared and one distribution
(Gumbel) was selected for use in modeling as described in another section.

47
3.4 Manning’s Roughness (n)

The Manning formula is an empirical formula estimating the average velocity of a liquid
flowing in a conduit that does not completely enclose the liquid, i.e., open channel flow
The Manning’s equation is

1 2 1
𝑉= 𝑅3𝑆 2
𝑛

Where,

V is the cross-sectional average velocity; n is the Gauckler–Manning coefficient;

R is the hydraulic radius;

S is the bed slope;

Manning’s n is adopted from HECRAS user manual which has extensive compilation
of n values for streams and floodplains following Chow’s Book “Open-Channel
Hydraulics” [ Chow,1959] and then calibrated for the study region according to the land
use.

Table 5: Manning’s n Values (Source: HECRAS user manual)

48
3.5 Hydraulic Modeling

Hydrologic Engineering Center's River Analysis System (HEC-RAS) was used to

perform analysis of unsteady flow water surface profile and unsteady flow simulation.
HEC- RAS have been applied in the current study as a part of hydrodynamic model. It
is freely available, very well documented and has been applied in many river basins of
Nepal.

Figure 6: HEC-RAS graphical user interface, the main window.

3.5.1 Theoretical Background of HEC-RAS

Hydrologic Engineering Center's River Analysis System (HEC-RAS) was developed

by the US Army Corps of Engineers to perform 1D analysis of steady flow water
surface profile, unsteady flow simulation, movable boundary sediment transport
computation and water quality analysis. HEC- RAS is applied in the current study as a
part of hydrodynamic model. Unsteady flow case was followed by this study as
explained below.

3.5.1.1 Governing Equations

HEC-RAS has the ability to perform two-dimensional unsteady flow routing with either
the full saint venant equations or the diffusion wave equations. Within HEC-RAS the
diffusion wave equations are set as the default, however the user should always test if
the full saint venant equations are needed for their specific application. A general
approach is the use of diffusion wave equations while developing the model and getting
all the problems worked out. Once the model is in good working order, then switching
the computational method to the full momentum equation option (full momentum will
generally require a smaller computation interval than the diffusion wave method to run

49
in a stable manner) in another copy of the model. Run this second plan and compare
them throughout the system. If there are significant differences between the two runs,
the user should assume the full momentum (saint venant equations) answer is more
accurate and proceed with that equation set for model calibration and other events
simulations. It is strongly recommended that the full momentum equation sets should
always be used in the cases of highly dynamic flood waves, abrupt contractions and
expansions, tidally influenced conditions, general wave propagation modeling, super
elevation around bends, detailed velocities, water surface elevations at structures and
mixed flow regime. In general, the diffusion wave equations are more forgiving
numerically than the saint venant equations. This means that larger time steps can be
used with the diffusion wave equations (than can be with the saint venant equations)
and still get numerically stable and accurate solutions.

Mass Conservation

The unsteady differential form of the mass conservation (continuity) equation is:

q =∂h/∂t+∂(hu)/∂x+∂(hv)/∂y

where t is time, u and v are the velocity components in the x- and y- direction
respectively.

In vector form,

∂h/∂t+∇⋅(hV)=q

where ,

V=(u,v)

T is the velocity vector and

∇ is the gradient operator given by

∇=(∂/∂x,∂/∂y)T

Momentum Conservation

50
The shallow water equations along the x and y directions are:

∂u ∂u ∂u ∂H ∂2 u ∂2 u
+ (𝑢 ∂x + 𝑣 ∂y)=-g ∂x + 𝑣𝑡 (∂𝑥 2 + ∂𝑦 2) − 𝑐𝑓 𝑢 + 𝑓𝑣 ………… 2a
∂t

∂v ∂v ∂v ∂H ∂2 v ∂2 v
+ (𝑢 ∂x + 𝑣 ∂y)=-g ∂y + 𝑣𝑡 (∂𝑥 2 + ∂𝑦 2) − 𝑐𝑓 𝑣 + 𝑓𝑢 ………… 2b
∂t

Where,

∂u ∂u ∂u
+ 𝑢 ∂x + 𝑣 ∂y =partial differential form of the flow acceleration
∂t

∂H
g ∂x = hydrostatic pressure gradient terms

∂2 u ∂2 u
𝑣𝑡 (∂𝑥 2 + ∂𝑦 2 ) = Viscosity

𝑐𝑓 𝑢= bed friction

𝑓𝑣= Coriolis parameter

Numerical Methods

A hybrid discretization scheme combining two numerical methods, namely, finite

difference approximations and finite volume approximations are used to take advantage
of orthogonality in grids and solve the equations at each grid point. Diffusion wave and
saint venant equations are solved differently applying different boundary conditions
and simplifications.

Conditions for Model Stability

The conditions shown below need to be satisfied for successful convergence of the
solution and numerical stability of the model.

For Saint Venant Equations (full momentum):

51
 𝑉𝛥𝑇
𝐶= ≤ 1.0 (With a max C=3.0)
𝛥𝑋

Or,

𝛥𝑋
𝛥𝑇 ≤ With C=1.0
𝑉

Where, C= Courant Number

V= Flood wave velocity (wave celerity) (ft/s)

ΔT= Computational time step (s)

ΔX= Average cell size (ft)

For Diffusive wave Equations:

𝑉𝛥𝑇
𝐶= ≤ 2.0 with a max C=5.0
𝛥𝑋

Or,

2𝛥𝑋
𝛥𝑇 ≤ With C=1.0
𝑉

3.5.1.2 Boundary Conditions

Boundary condition is always need to be defined in HECRAS.

• For subcritical flow, both upstream and downstream boundary conditions are required.

• For supercritical flow, only downstream boundary condition is required

Upstream boundary condition (depending upon the availability)

• Flow hydrograph

• Stage hydrograph

Downstream boundary condition (depending upon the availability)

52
• Flow hydrograph

• Stage hydrograph

• Single valued rating curve

• Normal depth

3.6 Model Inputs

Required data files for unsteady flow simulation are listed below:

3.6.1 Geometry File

The geometric data file used in 2D unsteady simulation consists of the terrain data, land
use data and modeling approach information. For basin characteristics manning’s
friction coefficient ‘n’ data set within HEC-RAS mapper was build, using land cover
data layers. The manning’s ‘n’ value is highly variable and depends on a number of
factors including surface roughness, vegetation, channel irregularities, channel
alignment, scour and deposition, obstructions, size and shape of the channel, stage and
discharge, seasonal changes, suspended material and bed-load (Samarasinghe & Sharp,
2010).

3.6.2 Plan File

The plan file contains: a description and short identifier for the plan; a list of files that
are associated with the plan (e.g., geometry file and steady/unsteady flow file); and a
description of all the simulation options that were set for the plan.

3.6.3 Unsteady Flow Data File

Unsteady flow data files contain: flow hydrographs at the upstream boundaries; starting
flow conditions; and downstream boundary conditions.

53
3.6.4 The Project File

The project file contains: the title of the project; the unit’s system of the project; a list
of all the files that are associated with the project; and a list of default variables that can
be set from the interface.

The HEC-RAS 2D model was run in unsteady mode. For this, the upstream boundary
flow data was required to be provided as a continuous timeseries (flood hydrograph).
As real-time hourly discharge data is not available at the gauging stations, historical
instantaneous maximum flood values were obtained from DHM. This data was then
used for flood frequency analysis to generate peak floods for 2-year, 10-year, 25-year,
50-year and 100-year floods respectively. Such peak flood values were adjusted with
the time of concentration to generate the flood hydrographs of each of these return
periods.

3.7 HEC-RAS Model Construction, Calibration, and Validation

After preparing the data for the modeling, following procedure was followed. Firstly,
new project was set up for modelling and then a geometry file was created for the
project. The filled DEM was imported into the project via RAS Mapper and associated
to new geometry. Next, the computational mesh was determined by boundaries of
previously described flood area mask with the cell size of 30 meters. Possible errors in
cells reported by HEC-RAS were manually fixed by adjusting cell geometry. The final
mesh contained 356966 computational cells. Then, the land cover data was imported to
the model via RAS Mapper and the default roughness values were selected for all eight
classes by the tabulation provided by in HECRAS manual (table 5). The boundary
condition (BC) line was drawn for the upstream end of the reach over the channel
section. Similarly, the BC line was drawn for the downstream end.

Next, the model hydrograph was defined for the upstream boundary condition in the
unsteady flow data editor. The hydrograph was constructed from the daily discharge
data from the period of 12 July 2013 – 30 July 2013 (table 6). The hydrograph was
defined to begin from the 12 July 2013 since the flood peak occurred between 18-July
2013- 23-July-2013 and there was not any initial water stage defined in the model (i.e,
the channel was empty at the beginning). Thus, the model had to fill the empty channel

54
with normal discharge (12.07 – 13.07) before the flood wave in order to achieve a
natural representation of the flood wave propagation. The hydrograph was extended
until 30 July 2013. The assumption of normal depth with value of 0.001706 was used
as a downstream boundary condition due to unavailability of more accurate data, such
as stage & flow hydrograph or a rating curve.

Flood Hydrograph from 12-Jul-2013 to 30-Jul- 2013

Date Daily mean

Discharge
12-Jul-13 544
13-Jul-13 390
14-Jul-13 361
15-Jul-13 465
16-Jul-13 407
17-Jul-13 425
18-Jul-13 392
19-Jul-13 1058
20-Jul-13 745
21-Jul-13 564
22-Jul-13 1734
23-Jul-13 609
24-Jul-13 619
25-Jul-13 777
26-Jul-13 567
27-Jul-13 593
28-Jul-13 494
29-Jul-13 410
30-Jul-13 410

Table 6: Flood hydrograph used on the model

Finally, the plan file for performing unsteady flow computations was created for the
model. Diffusive wave equation was chosen as a calculation equation option. The time
step was defined as 30 seconds, hydrograph output interval, the mapping and detailed
output interval settings for 1 hour. Hence, the conveyance changes in all the mesh cells
were solved once in every 30 seconds throughout the 19-day hydrograph, and the
detailed results were saved for every hour. In addition to unsteady flow simulation, the
geometry pre-processor, post-processor, and the floodplain mapping programs were
selected to achieve pre-calculated depth raster layers to be exported into ArcGIS.

55
After this model calibration and validation was to be performed. Calibration was
performed by simply adjusting the Manning’s roughness coefficient in the model
geometry and validation was done by comparing the simulated depth and observed
depth taken from DHM. This methodology was applied mainly due to the unavailability
of data and insufficiency for field visit and due to costly satellite data. The model
parameter i.e., manning’s n was calibrated and model was run for several times in order
to get the ground result. The result was validated with the data observed for Kusum
hydrological station. The final manning’s n value that was obtained after the calibration
of the model and was used for other calculation is tabulated below:

Calibrated Model Parameter

Land Use Manning's n
Agriculture- Valley, Tar, Terai 0.05
Agriculture- Sloping Terrace 0.035
Agriculture- Level Terrace 0.055
Residential 0.08
Shrubland/Grassland 0.11
Forest 0.12
Barrenland 0.04
Sand/Gravel/Boulder 0.05

Table 7: Calibrated value of Manning’s n

3.8 Flood Hazard Analysis

The flood effect varies from one place to another. With greater hazard, the vulnerability
also increases and hence the effect is high when the exposure is also more. The hazard
due to flood can be classified for a better understanding of vulnerability and risk. The
hazard level depends upon flood depth and the flood discharge. Hence effect of the
flood can be quantified using water depth as a primary measure.

Flood depth is considered as the most important indicator of intensity of flood hazard
(Islam & Sado, 2002). So, for quantifying the flood hazard, three levels of hazard are
categorized in this study as below.

56
 S.N. Flood Depth Hazard Level

1 Less than 0.5 m Low hazard

2 0.5-1m Minor hazard
3 1-2m Major hazard
4 2-3m Significant hazard
5 Greater than 3m Extreme hazard

Table 8: Classification of hazard level according to flood depth

It is assumed that lesser flood depth has less effect on the people and properties and
higher ones has larger effect. Flood hazard maps for various flood scenarios were
prepared by overlapping depth grid with open street map.

3.9 Vulnerability Assessment

A point shape file for buildings is created in GIS using open street map. The settlement
area of the study area is clipped by the flood depth boundary polygon. The attribute
table is exported to excel and the count of building was done according to the depth of
the flood for 2-,10-, 25-,50- and 100- YRP.

To obtain flood vulnerability functions, most of the researchers are seen to use the
inundation depth as the intensity measure (Fuchs et al., 2019)(Thapa et al., 2020).
Bhochhibhoya & Maharjan, 2022 in their study adopted the damage ratio of building
from literature for different building typology for seismic risk assessment. In the study
of Li et al., 2022 has used the existing curves to calculate the comprehensive
vulnerability score of building’s ability to withstand disaster. Here, in this study also a
existing vulnerability curves are used to know the performance of different building
typologies. Depth damage curve is commonly used for determining damages that are
directly caused by flood which denote the flood damage that would occur at specific
water depths per asset or per land-use class. Assessment of potential damage that can
be caused by flood events is an important component in vulnerability assessment.

57
 Using Krejcie & Morgan method of sample size calculation and raosoft sampling tool
a sample size of 342 houses was selected and the random sampling was conducted
during field survey in the affected areas. The survey was focused to observe the existing
house typology, assessing the damage levels in affected structures. The adobe houses
with thatched roof were mostly seen near the river corridor. It was analyzed that the
people living near river corridor in adobe houses were the marginalized group of people
with weak economic condition, the people who don’t have legal land they priotize to
live near river where land is available free of cost and the farmers who came nearby for
agriculture purpose are seen to live in temporary house.

a)

b)

Figure 7: Vulnerability curves for different building Typologies [a) (Englhardt et al.,
2019), b) (Dutta et al., 2003)]

58
In this study the vulnerability assessment is carried out using an existing vulnerability
curve for different building typologies. From literature, well developed vulnerability
curve was chosen for respective construction type and building material. (Englhardt et
al., 2019) has developed the vulnerability curve for different classes they are (I) non-
structured buildings built of materials such as mud and adobe or informal buildings;
(II) wooden buildings; (III) unreinforced masonry/concrete buildings; and (IV) steel
buildings. As the construction material and typologies of adobe, wooden and brick
masonry of the study area resembles with the typologies in the literature so the damage
factor from this literature is adopted from developed vulnerability curve. Also, for RCC
building typology developed vulnerability curve is adopted from the literature of Dutta,
et al., 2003

59
 CHAPTER FOUR: RESULTS AND DISCUSSION

4.1 Flood Frequency Analysis

The instantaneous maximum flow at Bagasoti station for year 1976-2015 was used to
estimate the return year floods. Flood magnitude of 2, 10-, 25-, 50- and 100-year return
period was computed using three different widely used statistical distribution;

Gumbel’s distribution, Log Pearson – III & Lognormal distribution. In Nepal, Log
Pearson Type III and Gumbel Extreme Value frequency distributions are fitted in most
of the cases (Jha, 2015).

Flood discharge in m3/s

Return Period Gumbel’s Log Pearson Log normal

(Years) Distribution type III Distribution
Distribution

2 1793 1717 1726

10 3583 3325 3315

25 4484 4253 4209

50 5153 4991 4911

100 5816 5769 5642

Table 9: Flood discharge for different return period

Above table shows the flood discharge for different return period by using statistical
distribution; Gumbel’s distribution, Log Pearson – III & Lognormal distribution. Here,
in this study discharge calculated from Gumbel’s distribution is used as it has higher
discharge value than others and we are analyzing for the flood cases. The comparison
of floods by different method using graphical representation is shown below.

60
 Frequency Analysis Chart
7000

6000

5000
Discharge (m3/s)

4000

3000

2000

1000

0
0 20 40 60 80 100 120
Return period (Years)
Lognormal Distribution Logpearson Type 3 Distribution
Gumbel's Distribution

Figure 8: Comparison of floods by different methods

In addition, Easy fit was run to get fit distribution among them all. The rank and
statistics, obtained from Easy fit is shown in the Table 10 below. From the table it is
seen that Gumbel max has good rank in comparison of all three test in Chi-square test.
So, Gumbel distribution was thus adopted for further analysis.

Distribution Kolmogorov Anderson Darling Chi-Squared

Smirnov

Statistics Rank Statistics Rank Statistics Rank

Gumbel Max 0.15763 9 0.7466 6 3.6201 4

Log-normal 0.12674 4 0.49353 2 21.441 14

Log-Pearson 3 0.12526 2 0.49438 3 21.437 13

Table 10: Examining the goodness of fit distribution

61
4.2 Flood Hazard Analysis

After the successful run of HEC-RAS 2D model, results were extracted to understand
the flood dynamics and impact of inundation at different parts of the study reach.

4.2.1 Inundation Area

After the successful run of HEC-RAS 2D model, the results were extracted to generate
flood maps. Inundation depth main dataset was extracted from the model to generate
the inundation maps and of the various scenario runs (plans). Table 11 illustrates the
inundation area due to flood of different return period. Nearly, 2329 ha land gets
inundate due to 2-year return period flood. Similarly, 4017 ha, 4934 ha and 5438 ha of
land gets inundate due to flood of 10, 25 and 50 years return period. Highest area i.e.,
5849 ha of land gets inundate due to flood of 100 years return period. It can be clearly
stated that as the flood discharge increases the inundation area also increases. Flood
maps have been prepared for 2-, 10-, 25-, 50- and 100-years The flood Inundation map
for different year return period is shown below in figure 13-17:

Return Period Discharge (m3/s) Inundation Area (ha)

2-year flood 1793 2329.18
10-year flood 3583 4017.4
25-year flood 4484 4934.14
50-year flood 5153 5438.59
100-year flood 5816 5849.26

Table 11: Extent of Inundation for different return period

The trendline was drawn for the inundation area with return period flood for extreme
hazard class i.e., >3 m and the logarithmic trend of increment of the inundation area
was seen with increase in discharge. The plot of trendline is shown in figure 9 below.

62
 Trend of inundation area for different return period
(Extreme Hazard Class i.e >3 m)
Inundation Area (ha) 2500

2000

1500

1000

500

0
0 20 40 60 80 100 120
Return Period (Years)

Figure 9: Trend of inundation area for high hazard class

4.2.2 Extent and Depth of Inundation

RAS mapper in HEC RAS model and ArcGIS tool were used for inundation analysis.
The results were further analyzed for flood zonation. The inundation maps provide
crucial information about the vulnerability of the areas. There are five depth categories
in the map and they are given below.

• <0.5 m (low hazard)

• 0.5-1 m (moderate hazard)
• 1-2 m (major hazard)
• 2-3 m (significant hazard)
• >3 m (extreme hazard)

For 2-year return period flood, very few sites lie in the higher hazard category except
main river channel. For remaining return period floods, it has been seen that inundation
occurs with slightly higher depth around that area. More than 3m depth was observed
generally around the center of the river for the main river channel. For other higher
return period floods, more inundation extent and depth have been observed along main
river channel. Table 12 below shows the total area inundated for different return period
floods for different depth categories.

63
 Flood Area (ha)
Water
Depth 2-YRP 10-YRP 25-YRP 50-YRP 100-YRP
(m)
Area % Area % Area % Area % Area %

<0.5 493 21.18 871 21.68 1083 21.95 1189 21.87 1220 20.86

0.5-1 417 17.92 493 12.28 601 12.17 675 12.40 685 11.71

1-2 899 38.60 1053 26.20 1068 21.64 1073 19.73 1125 19.23

2-3 220 9.46 628 15.64 721 14.62 802 14.74 900 15.39

>3 299 12.84 972 24.20 1461 29.62 1699 31.25 1919 32.81

Total 2329 100 4017 100 4934 100 5439 100 5849 100

Table 12: Area under hazard in terms of depth (m)

Depth of Inundation and the Respective Inundated

Area

6000.00

5000.00

>3
Inundation Area (Ha)

4000.00 Water depth (m)

2-3
3000.00 1-2

2000.00 0.5-1

<0.5
1000.00

0.00
2 years 10 years 25 years 50 years 100 years

Return Period (Years)

Figure 10: Return period -Flood depth relationship

64
From above table it seems that, the percentage of area under low hazard is 21.18 % and
20.86 % respectively for 2-, and 100-year return period. The area under extreme hazard
for 2, 10, 25, 50 and 100 return period are 12.84 %, 24.24 %, 29.62 %, 31.25 % and
32.81% respectively. The percentage of hazard increases with the increase of flood
depth. It concluded that the hazard rate increases with increase of flood magnitude

The above table 12 shows that the total area under classified water depths of <0.5 m
and > 3 m increased with increase in intensity of flood. The flood hazard map for
different year return period is shown below in figure 18-22:

4.3 Flood Vulnerability Analysis

Mainly two data sets were used for assessment of vulnerability of flood prone areas.
The societal/flood vulnerability was considered as a function of land use and settlement
(buildings). The stretch from Bagasoti hydrological station to 45 km downstream with
wide flood plain and heavy settlement was selected for vulnerability assessment of the
building.

Table 13 is about the number of households inundated for different return year period
per depth. From the table 13, it is clear that the number of household inundated
increases with the increase of flood magnitude. For 2-year return period, maximum
households inundated at <0.5 m whereas the number of building inundated decrease for
other categories of depth.

65
 Building Count

2year 10year 25years 50years 100years

Hazard Numbers of Buildings under hazard

<0.5 415 809 983 1051 1090

0.5-1 347 390 552 608 621

1-2 241 530 633 729 816

2-3 33 156 211 268 354

>3 16 67 151 169 256

Total 1052 1952 2530 2825 3137

Table 13: Vulnerability analysis with building data

In sample size out of 342 houses, it was found that 120 numbers of houses were adobe
houses (i.e., ~ 35%), 69 numbers of houses were wooden houses (i.e., ~ 20%), 102
numbers of houses were made of brick- masonry (i.e., ~ 30%) and 51 numbers of houses
were RCC building (i.e., ~ 15%). This sample data was then projected for inundated
target houses for different return period. Table 14 shows the number of inundated
houses with different existing house typologies for different return periods. The
assessment of the houses in flood areas shows that large numbers of adobe houses are
vulnerable to flood.

Building Typology
Return Brick-
period Wooden Cement RCC-
Adobe building mortar Building Total

2 years 368 210 316 158 1052

10 years 683 390 586 293 1952

25 years 886 506 759 380 2531

50 years 987 564 848 424 2823

100 years 1098 627 941 471 3137

Table 14: Number of houses Inundated for different year return period flood

66
The flood hazard map predicts that 368 numbers of adobe houses get affected by the 2
years return period flood. Whereas, the highest numbers of 1098 houses get affected by
flood of 100 years of return period. This is because maximum numbers of adobe type
houses were seen to build near the river corridor. For wooden building typologies the
numbers of houses affected are 210, 390, 206, 564 and 627 numbers as the flow
increases from 2-, 10-, 25-, 50 to 100 years return period. Also, for brick – cement
mortar houses highest numbers of houses affected are 941 for 100 years return period.
Affected numbers of RCC buildings increases from 158 to 471 as the flood discharge
increases from 2- years to 100 years return period. The map showing the building
affected by different years return period is shown below in figure 23-27. Using the well-
developed vulnerability curves from figure 7, the damage factor of different building
types is calculated and tabulated below:

Damage Factor

Brick-Masonry
Depth (m) Adobe houses Wooden houses RCC buildings
houses

0 0 0.00 0 0

0.2 0.25 0.23 0.2 0.08

0.5 0.5 0.44 0.38 0.15

0.8 0.75 0.58 0.45 0.22

1 0.9 0.65 0.5 0.25

1.5 1 0.82 0.7 0.3

2 1 0.93 0.75 0.38

2.5 1 1.00 0.8 0.4

Table 15: Damage factor of different house typology

With reference to existing vulnerability curve, it is seen that adobe houses have higher
damage ratio on same inundation depth with respect to other building typologies this is
because they do not have specific foundations. The wooden buildings are seen more
resilient than adobe. The brick masonry building is seemed to damage 50% at 1m depth
this is related to the fact that unreinforced walls are less able to resist pressure of water
exerted on walls (Englhardt et al., 2019). RCC building are likely to damage 40% at the
inundation depth of 2.5 m so these typologies of building can be defined as most

67
resistant class to flooding. Literature shows that mud walls will collapse when flooded
by about a meter of water (Maiti, 2007). Same as stated it is seen that adobe houses can
dissolve and collapse easily when impacted by flood waters.

Moreover, the land use vulnerability assessment was also carried out. The vulnerability
maps for the flooded areas were prepared by intersecting land use map of the study area
with the polygon of flooded area. The result shows that major vulnerable area according
to the land use class was agricultural area for every return period flood and 3965 ha of
agricultural land is vulnerable due to 100 years return period flood. Figure 11 illustrates
the area inundated with respect to land use for different year return period. The flood
vulnerability map according to land use is shown below in figure 28-32:

Inundation of landuse class for different years

return period flood
7000

6000 Landuse
Sand/Gravel/Boulder
5000
Inundation Area (ha)

Barrenland
4000
Forest
3000
Shrubland/Grassland
2000 Residential

1000 Agriculture- Valley, Tar,

Terai
0
2year 10year 25years 50years 100years
Return Period (Years)

Figure 11: Impacts of different return period flood on various types of landcover

68
The trendline was drawn for the total vulnerable agricultural area with return period
flood and the logarithmic trend of increment of the vulnerable agricultural area was
seen with increase in discharge. The plot of trendline is shown in Figure 12 below:

Trend of vulnerable agricultural area for different

Total vulnerable Agricultural area (ha)

return period flood

4500
4000
3500
3000
2500
2000
1500
1000
500
0
0 20 40 60 80 100 120
Return Period (Years)

Figure 12: Trend of vulnerable agricultural area for different return period

69
 CHAPTER FIVE: CONCLUSION AND RECOMMENDATIONS

5.1 Conclusion

Two-dimensional unsteady flow model HEC-RAS was used for the flood hazard
analysis of the floodplain. ArcGIS and RAS Mapper were used for the preparation of
flood hazard maps. Flood hazard maps for floods with probable return period of 2 years,
10 years, 25 years, 50 years, and 100 years, were prepared. Low hazard (<0.5m),
moderate hazard (0.5-1m) major hazard (1-2m) significant hazard (2-3m) and extreme
hazard (>3m) were the five classes of flood depth analysis used in this study. The water
depth for different return-period floods showed that the levels of inundation increase
for increase in return period floods. The numbers of affected building increases as the
flood intensity increases. The study can be concluded as follows:

• The assessment of hazard maps shows that greater the value of discharge greater is the
affected areas. The affected area with low hazard class has decreased from 21.18% to
20.86% whereas extreme hazard class has increased from 12.84% to 32.81%.
• The assessment of vulnerability maps prepared shows that vulnerability of agricultural
lands i.e., the fertile land is increasing due to increasing discharge which will directly
effect on the food security.
• The trend of inundation area for extreme hazard class as well as land use vulnerability
shows that the hazard and vulnerability both are increasing with respect to the flood
intensity as shown in figure 9 and 21. Since risk is a function of hazard and vulnerability
and greater the hazard and vulnerability higher will be the risk. Hence the study
suggests that flood risk for the study area is also increasing which indicates the need of
effective land use planning.
• With reference to existing vulnerability curve, it is seen that adobe houses have higher
damage ratio on same inundation depth with respect to other building typologies. So,
these building typologies should not be allowed for construction in the specific areas
which are susceptible to exposure due to flood.
• The result presented in the form of maps provides a new perspective for visualization
and quantification of flood hazard and vulnerability which can help the decision makers
to understand the problems and take necessary action.

70
 The result of this study can be helpful for public and decision makers. Once people are
informed about the possible depth of flooding and the performance of building to the
depth of inundation, people would be less inclined to live near the river course. The
maps can be used to improve public flood awareness and helps in acceptance of
government flood related programs.

5.2 Recommendation

Based on this study following recommendations can be made to mitigate the flood
effects:

• The results obtained from the hydraulic model can be used to determine warning and
danger level for the river.
• Right of way of river should be defined according to 50 years of return period flood for
proper river management to reduce the impacts of flood induced by human activities.
• Flood inundation and hazard maps can be used by local authorities for effective land
use planning as well as determining evacuation sites.

71
 REFERENCES
Aryal, D., Wang, L., Adhikari, T. R., Zhou, J., Li, X., Shrestha, M., Wang, Y., &
Chen, D. (2020). A model-based flood hazard mapping on the southern slope of
Himalaya. Water (Switzerland), 12(2). https://doi.org/10.3390/w12020540
Bertilsson, L., Wiklund, K., de Moura Tebaldi, I., Rezende, O. M., Veról, A. P.,
& Miguez, M. G. (2019a). Urban flood resilience – A multi-criteria index to
integrate flood resilience into urban planning. Journal of Hydrology, 573, 970–
982. https://doi.org/https://doi.org/10.1016/j.jhydrol.2018.06.052
Bertilsson, L., Wiklund, K., de Moura Tebaldi, I., Rezende, O. M., Veról, A. P.,
& Miguez, M. G. (2019b). Urban flood resilience – A multi-criteria index to
integrate flood resilience into urban planning. Journal of Hydrology, 573, 970–
982. https://doi.org/10.1016/j.jhydrol.2018.06.052
Bhattarai, P., Khanal, P., Tiwari, P., Lamichhane, N., Dhakal, P., Lamichhane, P.,
Panta, N. R., & Dahal, P. (n.d.). Flood Inundation Mapping of Babai Basin using
HEC-RAS & GIS. http://www.inseconline.org
Bhochhibhoya, S., & Maharjan, R. (2022). Integrated Seismic Risk Assessment in
Nepal. https://doi.org/10.5194/nhess-2021-354
Bhusal, J. K. (2004). Consequences of river training structures: A case study of
Laxmanpur barrage and afflux bunds near Indo-Nepal border.
Birkmann, J. 1972-, & Wisner, B. (2005). Measuring the un-measurable the
challenge of vulnerability ; report of the Second Meeting of the UNU-EHS Expert
Working Group on Measuring Vulnerability, 12 - 14 October 2005, Bonn,
Germany.
CRED. (2018). 2018 EM-DAT file dated 02/07/2018. In Brussels.
CRED. (2019). United nations , UN-SPIDER.
CRED. (2021). 2021: Disaster in numbers. https://doi.org/10.1787/eee82e6e-en
Dangol, S. (2017). Use of Geo-Informatics in Flood Hazard Mapping: A Case of
Balkhu River. Journal on Geoinformatics, Nepal, 13, 52–57.
https://doi.org/10.3126/njg.v13i0.16937
Dangol, S., & Bormudoi, A. (2015). Flood Hazard Mapping and Vulnerability
Analysis of Bishnumati River. In Nepal Nepalese Journal on Geoinformatics.
https://www.researchgate.net/publication/282505448
Demir, V., & Kisi, O. (2016). Flood Hazard Mapping by Using Geographic
Information System and Hydraulic Model: Mert River, Samsun, Turkey. Advances
in Meteorology, 2016. https://doi.org/10.1155/2016/4891015
Dev Acharya, T., Banjara, S., & Deuja, S. (2011). Flood Risk Mapping of Upper
Bagmati Basin. https://doi.org/10.13140/RG.2.2.12449.10087
Dhakal, S. (2013). Flood Hazard in Nepal and New Approach of Risk Reduction.

72
Dutta, D., Herath, S., & Musiake, K. (2003). A mathematical model for flood loss
estimation. Journal of Hydrology, 277(1–2), 24–49.
https://doi.org/10.1016/S0022-1694(03)00084-2
Englhardt, J., de Moel, H., Huyck, C. K., de Ruiter, M. C., Aerts, J. C. J. H., &
Ward, P. J. (2019). Enhancement of large-scale flood damage assessments using
building-material-based vulnerability curves for an object-based approach.
https://doi.org/10.5194/nhess-2019-32
Fuchs, S., Keiler, M., Ortlepp, R., Schinke, R., & Papathoma-Köhle, M. (2019).
Recent advances in vulnerability assessment for the built environment exposed to
torrential hazards: Challenges and the way forward. In Journal of Hydrology (Vol.
575, pp. 587–595). Elsevier B.V. https://doi.org/10.1016/j.jhydrol.2019.05.067
Gautam, D. K., & Kharbuja, R. G. (2006). Flood Hazard Mapping of Bagmati
River in Kathmandu Valley Using Geoinformatics Tools.
Gautam, D. K., & Phaiju, A. G. (2013). Community Based Approach to Flood
Early Warning in West Rapti River Basin of Nepal. Journal of Integrated Disaster
Risk Management, 3(1), 155–169. https://doi.org/10.5595/idrim.2013.0060
Hasan, Z., Akhter, S., Ahmed, S., & Kabir, A. (2013). Challenges of Integrating
Disaster Risk Management and Climate Change Adaptation Policies at the
National Level: Bangladesh as a Case. In Global Journal of HUMAN SOCIAL
SCIENCE Geography, Geo-Sciences, Environmental Disaster Management (Vol.
13).
ICHARM. (2006). ICHARM Strategies and Action Plan for 2006-2008.
Islam, M. M., & Sado, K. (2002). Development Priority Map for Flood
Countermeasures by Remote Sensing Data with Geographic Information System.
https://doi.org/10.1061/ASCE1084-069920027:5346
Jha, A. K., Bank, W., Lamond, J. E., Proverbs, D., & Bhattacharya-Mis, N. (2012).
Cities and Flooding: A guide to integrated urban flood risk management for the
21st Century Improving resilience to flooding View project Adaptation of Urban
Infrastructure to Enhance Climate Resilience in Nigeria View project.
https://www.researchgate.net/publication/272793713
Jha, P. chandra. (2015). Applications of Simulation in Flood Frequency Analysis
in Nepal.
Jodhani, K. H., Patel, D., & Madhavan, N. (2022). A review on analysis of flood
modelling using different numerical models. Materials Today: Proceedings.
https://doi.org/10.1016/j.matpr.2021.07.405
Khanal, N. raj. (2004). Floods in mountain watershed: A case of Madi River.
Li, Z., Wang, L., Shen, J., Ma, Q., & Du, S. (2022). A Method for Assessing Flood
Vulnerability Based on Vulnerability Curves and Online Data of Residential
Buildings—A Case Study of Shanghai. Water, 14(18), 2840.
https://doi.org/10.3390/w14182840

73
Maiti, S. (2007). Defining a Flood Risk Assessment Procedure using Community
Based Approach with Integration of Remote Sensing and GIS-Based on the 2003
Orissa Flood.
Manandhar, B. (2010). Flood Plain Analysis and Risk Assessment: A Case Study
of Lothar Khola, Nepal.
Nepal, P., Narendra, ;, Khanal, R., Prasad, B., & Sharma, P. (2018). Policies and
institutions for disaster risk management in Nepal: A review. In The Geographical
Journal of Nepal (Vol. 11).
Noji, E. K. (2005). Disasters: Introduction and State of the Art. Epidemiologic
Reviews, 27(1), 3–8. https://doi.org/10.1093/epirev/mxi007
NRCS. (2016). Nepal Red Cross Society.
Salimi M. Shokoufeh, Reza Ghanbarpour, Karim Solaimani, & Mirkhalegh Z.
Ahzemi. (2008). Flood Plain Mapping using Hydraulic Simulation Model in GIS.
Journal of Applied Sciences.
Samarasinghe, O., & Sharp, B. (2010). Flood prone risk and amenity values: A
spatial hedonic analysis. Australian Journal of Agricultural and Resource
Economics, 54(4), 457–475. https://doi.org/10.1111/j.1467-8489.2009.00483.x
Talchabhadel, R., & Sharma, R. (2014). Real Time Data Analysis of West Rapti
River Basin of Nepal. Journal of Geoscience and Environment Protection, 02(05),
1–7. https://doi.org/10.4236/gep.2014.25001
Tamang, N. B., Tamrakar, N. K., Magar, M., & Raut, M. (2017). Fluvial
morphology and sediment transport of the Malekhu Khola, Central Nepal Lesser
Himalaya. Bulletin of the Department of Geology, 18, 35–48.
https://doi.org/10.3126/bdg.v18i0.16455
Thapa, S., Shrestha, A., Lamichhane, S., Adhikari, R., & Gautam, D. (2020).
Catchment-scale flood hazard mapping and flood vulnerability analysis of
residential buildings: The case of Khando River in eastern Nepal. Journal of
Hydrology: Regional Studies, 30. https://doi.org/10.1016/j.ejrh.2020.100704
UNDP. (2009). NEPAL COUNTRY REPORT GLOBAL ASSESSMENT OF RISK
ISDR Global Assessment Report on Poverty and Disaster Risk 2009.
http://www.undp.org.np
UNDP. (2019). Human development report 2019 : beyond income, beyond
averages, beyond today: inequalities in human development in the 21st century
WECS. (2019). Flood Control and Management Manual
ACKNOWLEDGEMENT WECS A ACKNOWLEDGEMENT.
WECS. (2020). Flood Control and Management Manual.
Yogacharya, K. S., & Gautam, D. K. (2008). Floods in Nepal: Genesis, Magnitude,
Frequency and Consequences. International Conference on Hydrology and
Climate Change in Mountainous Areas, At Kathmandu, Nepal, May 2015.

74
ANNEXES

75
Annex 1:Maximum Instantaneous Discharge of Bagasoti Hydrological Station

Station number: 350

Location: Bagasoti Gaon Latitude: 27 51 12
River: Rapti River Longitude: 82 47 34

EXTREME DISCHARGES
==================

MAXIMUM INSTANTANEOUS
Year Discharge(m3/s) Gauge height(m) Date
1976 970 3.70 23/08/1976
1977 3480 7.50 26/08/1977
1978 1990 5.68 16/07/1978
1979 1560 5.02 24/07/1979
1980 2290 6.10 06/09/1980
1981 6030 9.84 11/09/1981
1982 2090 5.82 12/09/1982
1983 3330 7.35 11/09/1983
1984 1790 5.40 07/09/1984
1985 940 3.75 22/08/1985
1986 1240 4.42 19/06/1986
1987 1030 3.85 30/07/1987
1988 1200 4.20 05/07/1988
1989 1180 4.22 15/07/1989
1990 1180 4.14 28/07/1990
1991 1760 5.44 14/06/1991
1992 930 4.20 07/08/1992

76
1993 2080 5.90 06/09/1993
1994 2090 5.92 12/07/1994
1995 1130 4.54 30/08/1995
1996 1660 5.30 19/07/1996
1997 2080 5.90 21/07/1997
1998 1790 5.48 01/09/1998
1999 3710 7.74 13/06/1999
2000 4170 8.20 10/08/2000
2001 1690 5.35 24/08/2001
2002 858 4.08 20/08/2002
2003 3810 7.85 18/08/2003
2004 535 3.49 18/07/2004
2005 1740 5.41 24/08/2005
2006 1730 5.40 10/08/2006
2007 1900 5.65 31/08/2007
2008 1940 5.70 20/09/2008
2009 2040 5.85 28/07/2009
2010 1730 5.40 20/07/2010
2011 1280 4.84 03/08/2011
2012 2570 6.50 03/08/2012
2013 1920 5.80 22/07/2013
2014 2390 6.40 15/08/2014
2015 625 3.67 21/08/2015

(Source: Department of Hydrology and Meteorology)

77
 Annex 2: Flood Frequency Analysis using different method
a) Using Gumbel’s Method
Ranked p= Return
Year Q Q LnQ m m/(n+1) Period T yT KT QT
1976 970 6030 8.7045 1 0.0244 41.00 3.7013 2.7667 4962
1977 3480 4170 8.3357 2 0.0488 20.50 2.9955 2.1484 4291
1978 1990 3810 8.2454 3 0.0732 13.67 2.5772 1.7818 3894
1979 1560 3710 8.2188 4 0.0976 10.25 2.2764 1.5183 3608
1980 2290 3480 8.1548 5 0.1220 8.20 2.0398 1.3110 3383
1981 6030 3330 8.1107 6 0.1463 6.83 1.8437 1.1392 3197
1982 2090 2570 7.8517 7 0.1707 5.86 1.6755 0.9918 3037
1983 3330 2390 7.7790 8 0.1951 5.13 1.5276 0.8621 2896
1984 1790 2290 7.7363 9 0.2195 4.56 1.3950 0.7460 2771
1985 940 2090 7.6449 10 0.2439 4.10 1.2744 0.6404 2656
1986 1240 2090 7.6449 11 0.2683 3.73 1.1636 0.5432 2551
1987 1030 2080 7.6401 12 0.2927 3.42 1.0605 0.4529 2453
1988 1200 2080 7.6401 13 0.3171 3.15 0.9640 0.3683 2361
1989 1180 2040 7.6207 14 0.3415 2.93 0.8729 0.2885 2274
1990 1180 1990 7.5959 15 0.3659 2.73 0.7864 0.2128 2192
1991 1760 1940 7.5704 16 0.3902 2.56 0.7038 0.1404 2114
1992 930 1920 7.5601 17 0.4146 2.41 0.6245 0.0709 2038
1993 2080 1900 7.5496 18 0.4390 2.28 0.5480 0.0039 1966
1994 2090 1790 7.4900 19 0.4634 2.16 0.4740 -0.0610 1895
1995 1130 1790 7.4900 20 0.4878 2.05 0.4019 -0.1242 1827
1996 1660 1760 7.4731 21 0.5122 1.95 0.3315 -0.1858 1760
1997 2080 1740 7.4616 22 0.5366 1.86 0.2625 -0.2463 1694
1998 1790 1730 7.4559 23 0.5610 1.78 0.1946 -0.3058 1630
1999 3710 1730 7.4559 24 0.5854 1.71 0.1274 -0.3646 1566
2000 4170 1690 7.4325 25 0.6098 1.64 0.0608 -0.4230 1503
2001 1690 1660 7.4146 26 0.6341 1.58 -0.0055 -0.4811 1440
2002 858 1560 7.3524 27 0.6585 1.52 -0.0719 -0.5393 1377
2003 3810 1280 7.1546 28 0.6829 1.46 -0.1386 -0.5977 1313
2004 535 1240 7.1229 29 0.7073 1.41 -0.2059 -0.6567 1249
2005 1740 1200 7.0901 30 0.7317 1.37 -0.2744 -0.7167 1184
2006 1730 1180 7.0733 31 0.7561 1.32 -0.3443 -0.7780 1118
2007 1900 1180 7.0733 32 0.7805 1.28 -0.4163 -0.8411 1049
2008 1940 1130 7.0300 33 0.8049 1.24 -0.4911 -0.9066 978
2009 2040 1030 6.9373 34 0.8293 1.21 -0.5697 -0.9754 904
2010 1730 970 6.8773 35 0.8537 1.17 -0.6533 -1.0487 824
2011 1280 940 6.8459 36 0.8780 1.14 -0.7439 -1.1281 738
2012 2570 930 6.8352 37 0.9024 1.11 -0.8447 -1.2164 642
2013 1920 858 6.7546 38 0.9268 1.08 -0.9612 -1.3185 531
2014 2390 625 6.4378 39 0.9512 1.05 -1.1054 -1.4448 394
2015 625 535 6.2823 40 0.9756 1.03 -1.3120 -1.6259 198
2.00 0.3665 -0.1552 1793
5.00 1.4999 0.8379 2870
10.00 2.2504 1.4955 3583
25.00 3.1985 2.3262 4484
50.00 3.9019 2.9426 5153
100.00 4.6001 3.5543 5816

78
 b) Using Log Pearson Type -3 Method
Return
Ranked p= Period
Year Q Q LnQ m m/(n+1) T w z k KT ZT QT
1976 970 6030 8.7045 1 0.0244 41.00 2.725 1.971 0.010 1.999 8.471 4776
1977 3480 4170 8.3357 2 0.0488 20.50 2.458 1.657 0.010 1.674 8.306 4048
1978 1990 3810 8.2454 3 0.0732 13.67 2.287 1.453 0.010 1.464 8.199 3636
1979 1560 3710 8.2188 4 0.0976 10.25 2.157 1.296 0.010 1.302 8.117 3349
1980 2290 3480 8.1548 5 0.1220 8.20 2.051 1.165 0.010 1.169 8.049 3129
1981 6030 3330 8.1107 6 0.1463 6.83 1.961 1.052 0.010 1.053 7.990 2950
1982 2090 2570 7.8517 7 0.1707 5.86 1.880 0.951 0.010 0.950 7.937 2800
1983 3330 2390 7.7790 8 0.1951 5.13 1.808 0.859 0.010 0.856 7.889 2669
1984 1790 2290 7.7363 9 0.2195 4.56 1.741 0.774 0.010 0.770 7.845 2554
1985 940 2090 7.6449 10 0.2439 4.10 1.680 0.694 0.010 0.688 7.804 2450
1986 1240 2090 7.6449 11 0.2683 3.73 1.622 0.618 0.010 0.611 7.765 2356
1987 1030 2080 7.6401 12 0.2927 3.42 1.568 0.545 0.010 0.538 7.728 2270
1988 1200 2080 7.6401 13 0.3171 3.15 1.516 0.475 0.010 0.468 7.692 2190
1989 1180 2040 7.6207 14 0.3415 2.93 1.466 0.408 0.010 0.400 7.657 2116
1990 1180 1990 7.5959 15 0.3659 2.73 1.418 0.342 0.010 0.334 7.623 2046
1991 1760 1940 7.5704 16 0.3902 2.56 1.372 0.278 0.010 0.269 7.591 1979
1992 930 1920 7.5601 17 0.4146 2.41 1.327 0.215 0.010 0.206 7.558 1917
1993 2080 1900 7.5496 18 0.4390 2.28 1.283 0.153 0.010 0.143 7.527 1857
1994 2090 1790 7.4900 19 0.4634 2.16 1.240 0.092 0.010 0.082 7.495 1799
1995 1130 1790 7.4900 20 0.4878 2.05 1.198 0.030 0.010 0.021 7.464 1744
1996 1660 1760 7.4731 21 0.5122 1.95 1.198 -0.030 0.010 -0.040 7.433 1691
1997 2080 1740 7.4616 22 0.5366 1.86 1.240 -0.092 0.010 -0.101 7.402 1639
1998 1790 1730 7.4559 23 0.5610 1.78 1.283 -0.153 0.010 -0.163 7.371 1589
1999 3710 1730 7.4559 24 0.5854 1.71 1.327 -0.215 0.010 -0.225 7.339 1540
2000 4170 1690 7.4325 25 0.6098 1.64 1.372 -0.278 0.010 -0.287 7.307 1491
2001 1690 1660 7.4146 26 0.6341 1.58 1.418 -0.342 0.010 -0.351 7.275 1444
2002 858 1560 7.3524 27 0.6585 1.52 1.466 -0.408 0.010 -0.416 7.242 1397
2003 3810 1280 7.1546 28 0.6829 1.46 1.516 -0.475 0.010 -0.483 7.208 1350
2004 535 1240 7.1229 29 0.7073 1.41 1.568 -0.545 0.010 -0.552 7.173 1303
2005 1740 1200 7.0901 30 0.7317 1.37 1.622 -0.618 0.010 -0.624 7.136 1257
2006 1730 1180 7.0733 31 0.7561 1.32 1.680 -0.694 0.010 -0.699 7.098 1210
2007 1900 1180 7.0733 32 0.7805 1.28 1.741 -0.774 0.010 -0.777 7.058 1162
2008 1940 1130 7.0300 33 0.8049 1.24 1.808 -0.859 0.010 -0.861 7.015 1113
2009 2040 1030 6.9373 34 0.8293 1.21 1.880 -0.951 0.010 -0.952 6.969 1063
2010 1730 970 6.8773 35 0.8537 1.17 1.961 -1.052 0.010 -1.051 6.919 1011
2011 1280 940 6.8459 36 0.8780 1.14 2.051 -1.165 0.010 -1.162 6.862 956
2012 2570 930 6.8352 37 0.9024 1.11 2.157 -1.296 0.010 -1.289 6.798 896
2013 1920 858 6.7546 38 0.9268 1.08 2.287 -1.453 0.010 -1.442 6.720 829
2014 2390 625 6.4378 39 0.9512 1.05 2.458 -1.657 0.010 -1.640 6.619 749
2015 625 535 6.2823 40 0.9756 1.03 2.725 -1.971 0.010 -1.942 6.465 642
2.00 1.177 0.000 0.010 -0.010 7.449 1717
5.00 1.794 0.841 0.010 0.838 7.880 2645
10.00 2.146 1.282 0.010 1.288 8.109 3325
25.00 2.537 1.751 0.010 1.771 8.355 4253
50.00 2.797 2.054 0.010 2.086 8.515 4991
100.00 3.035 2.327 0.010 2.370 8.660 5769

79
 c) Using Log Normal Distribution Method
Return
Ranked p= Period
Year Q Q LnQ m m/(n+1) T w z k KT ZT QT
1976 970 6030 8.7045 1 0.0244 41.00 2.725 1.971 0.010 1.9709 8.4569 4707
1977 3480 4170 8.3357 2 0.0488 20.50 2.458 1.657 0.010 1.6572 8.2972 4013
1978 1990 3810 8.2454 3 0.0732 13.67 2.287 1.453 0.010 1.4529 8.1932 3616
1979 1560 3710 8.2188 4 0.0976 10.25 2.157 1.296 0.010 1.2958 8.1132 3338
1980 2290 3480 8.1548 5 0.1220 8.20 2.051 1.165 0.010 1.1654 8.0468 3124
1981 6030 3330 8.1107 6 0.1463 6.83 1.961 1.052 0.010 1.0523 7.9893 2949
1982 2090 2570 7.8517 7 0.1707 5.86 1.880 0.951 0.010 0.9512 7.9378 2801
1983 3330 2390 7.7790 8 0.1951 5.13 1.808 0.859 0.010 0.8590 7.8909 2673
1984 1790 2290 7.7363 9 0.2195 4.56 1.741 0.774 0.010 0.7736 7.8474 2559
1985 940 2090 7.6449 10 0.2439 4.10 1.680 0.694 0.010 0.6935 7.8066 2457
1986 1240 2090 7.6449 11 0.2683 3.73 1.622 0.618 0.010 0.6176 7.7680 2364
1987 1030 2080 7.6401 12 0.2927 3.42 1.568 0.545 0.010 0.5452 7.7311 2278
1988 1200 2080 7.6401 13 0.3171 3.15 1.516 0.475 0.010 0.4755 7.6956 2199
1989 1180 2040 7.6207 14 0.3415 2.93 1.466 0.408 0.010 0.4080 7.6613 2125
1990 1180 1990 7.5959 15 0.3659 2.73 1.418 0.342 0.010 0.3424 7.6279 2055
1991 1760 1940 7.5704 16 0.3902 2.56 1.372 0.278 0.010 0.2783 7.5952 1989
1992 930 1920 7.5601 17 0.4146 2.41 1.327 0.215 0.010 0.2153 7.5632 1926
1993 2080 1900 7.5496 18 0.4390 2.28 1.283 0.153 0.010 0.1531 7.5315 1866
1994 2090 1790 7.4900 19 0.4634 2.16 1.240 0.092 0.010 0.0916 7.5002 1808
1995 1130 1790 7.4900 20 0.4878 2.05 1.198 0.030 0.010 0.0305 7.4691 1753
1996 1660 1760 7.4731 21 0.5122 1.95 1.198 -0.030 0.010 -0.0305 7.4381 1699
1997 2080 1740 7.4616 22 0.5366 1.86 1.240 -0.092 0.010 -0.0916 7.4070 1647
1998 1790 1730 7.4559 23 0.5610 1.78 1.283 -0.153 0.010 -0.1531 7.3757 1597
1999 3710 1730 7.4559 24 0.5854 1.71 1.327 -0.215 0.010 -0.2153 7.3440 1547
2000 4170 1690 7.4325 25 0.6098 1.64 1.372 -0.278 0.010 -0.2783 7.3120 1498
2001 1690 1660 7.4146 26 0.6341 1.58 1.418 -0.342 0.010 -0.3424 7.2793 1450
2002 858 1560 7.3524 27 0.6585 1.52 1.466 -0.408 0.010 -0.4080 7.2459 1402
2003 3810 1280 7.1546 28 0.6829 1.46 1.516 -0.475 0.010 -0.4755 7.2116 1355
2004 535 1240 7.1229 29 0.7073 1.41 1.568 -0.545 0.010 -0.5452 7.1761 1308
2005 1740 1200 7.0901 30 0.7317 1.37 1.622 -0.618 0.010 -0.6176 7.1392 1260
2006 1730 1180 7.0733 31 0.7561 1.32 1.680 -0.694 0.010 -0.6935 7.1006 1213
2007 1900 1180 7.0733 32 0.7805 1.28 1.741 -0.774 0.010 -0.7736 7.0598 1164
2008 1940 1130 7.0300 33 0.8049 1.24 1.808 -0.859 0.010 -0.8590 7.0163 1115
2009 2040 1030 6.9373 34 0.8293 1.21 1.880 -0.951 0.010 -0.9512 6.9694 1064
2010 1730 970 6.8773 35 0.8537 1.17 1.961 -1.052 0.010 -1.0523 6.9179 1010
2011 1280 940 6.8459 36 0.8780 1.14 2.051 -1.165 0.010 -1.1654 6.8604 954
2012 2570 930 6.8352 37 0.9024 1.11 2.157 -1.296 0.010 -1.2958 6.7940 892
2013 1920 858 6.7546 38 0.9268 1.08 2.287 -1.453 0.010 -1.4529 6.7140 824
2014 2390 625 6.4378 39 0.9512 1.05 2.458 -1.657 0.010 -1.6572 6.6100 742
2015 625 535 6.2823 40 0.9756 1.03 2.725 -1.971 0.010 -1.9709 6.4503 633
2.00 1.177 0.000 0.010 0.0000 7.4536 1726
5.00 1.794 0.841 0.010 0.8415 7.8819 2649
10.00 2.146 1.282 0.010 1.2817 8.1061 3315
25.00 2.537 1.751 0.010 1.7511 8.3450 4209
50.00 2.797 2.054 0.010 2.0542 8.4993 4911
100.00 3.035 2.327 0.010 2.3268 8.6381 5642

80
Annex 3: Computation Mesh created in the Study area

81
Annex 4: Flood Inundation Map of different year return period

Figure 13: Flood inundation map for 2 years return period

82
Figure 14: Flood inundation map for 10 years return period

83
Figure 15: Flood inundation map for 25 years return period

84
Figure 16 : Flood inundation map for 50 years return period
85
Figure 17: Flood inundation map for 100 years return period
86
Annex 5: Flood Hazard Map of different year return period

Figure 18: Flood hazard map for 2 years return period

87
Figure 19: Flood hazard map for 10 years of return period

88
Figure 20: Flood hazard map for 25 years return period

89
Figure 21: Flood hazard map for 50 years return period
90
Figure 22: Flood hazard map for 100 years return period

91
Annex 6:: Flood Vulnerability Map of buildings affected by different years return period

Figure 23: Numbers of building affected by 2 years of return period

92
Figure 24: Numbers of building affected by 10 years of return period

93
Figure 25: Numbers of building affected by 25 years of return period

94
Figure 26: Numbers of building affected by 50 years of return period

95
Figure 27: Numbers of building affected by 100 years of return period

96
Annex 7: Flood vulnerability map for different return period

Figure 28: Flood vulnerability map for 2 years return period

97
Figure 29 :Flood vulnerability map for 10 years return period

98
Figure 30: Flood vulnerability map for 25 years return period

99
Figure 31: Flood vulnerability map for 50 years return period
100
Figure 32: Flood vulnerability map for 100 years return period

101