Identify the flood hazard index in the huong river basin hue city area

The objectives of this research were i to simulate flood flow in the city by using 2D hydrodynamic model MIKE 21 FM, ii to develop a hierarchical structure through the analytic hierarchy

ACKNOWLEDGEMENT

I am indebted to my respected Advisors, Dr Pham Thanh Hai and Ass - Prof Hoang Thanh Tung who work as lecturers in Department of Hydrology and Water resources in Thuy Loi University for their continuous guidance, advice and expedience from the proposal preparation to thesis finalization Their constructive comments, untiring help, guidance and practical suggestions inspired me to accomplish this work successfully

Besides, I am especially grateful to other lecturers in the Department of Hydrology and Water resources who supported me in terms of the data collection and gave me useful advices for my thesis

I remember all those who have contributed directly or indirectly to successfullycompleting my study.

Finally, I must express my very profound gratitude to my family for providing

me with unfailing support and continuous encouragement throughout my years of study and through the process of researching and writing this thesis This accomplishment would not have been possible without them Thank you

Hanoi, November 11th 2016

Vu Hoang Tung

Ha Noi, November 11th 2016

Vu Hoang Tung

ABSTRACT

Flooding is one of the major natural hazards in the city of Hue This city is frequently affected by flooding and most of the low-lying areas in the city are flood-prone areas Annually, the losses of people and property caused by flooding in Hue city are very much This has a great influence on the local’s life and inhibits the socio-economic development of the city Therefore, in order to minimize losses of life and economic, a detailed and comprehensive flood hazard assessment is necessary for both flood control and mitigation works The objectives of this research were (i) to simulate flood flow in the city by using 2D hydrodynamic model MIKE 21 FM, (ii) to develop a hierarchical structure through the analytic hierarchy process (AHP) to define and qualify parameters that contribute to flood hazard, (iii) to map the flood components using the geographic information system (GIS), and (iv) to integrate these three methodologies and apply them to the Huong river basin in the Hue city to create flood hazard index map In addition, based on the sea level rise scenarios for Hue city in

2030, this study also calculated and created flood hazard index maps corresponding to B1, B2 and A1 scenarios Three flood components were considered, including flood depth, flood flows velocity and flood duration Flood maps were thenc drawn based on the data collected from institutes, inheriting the results of studies in the past, and documents related to historical flood events, climate change in Hue city The results show that high level of flood hazard tends to broaden over the low, medium and high emission scenarios In the high emission scenario (A1), the high flood hazard zone covers 45.3% of the study area While the medium and low hazard zones covers 19.6% and 17.5%, respectively It is concluded that integration of hydrodynamic model, AHP and GIS in flood hazard assessment can provide useful detailed information for flood risk assessment, and the method can be easily applied to other areas where necessary

data is readily available

Abbreviation

WRI World Resources Institute

GDP Gross Domestic Production

CCFSC Central Committee for Flood and Storm Control

AHP Analytical Hierarchy Process

GIS Geographical Information System

IPCC International Panel on Climate Change

UNFCCC United Nations Framework Convention on Climate Change GDP Gross Domestic Product

FDI Foreign Direct Investment

WMO World Meteorological Organization

DHI Danish Hydraulic Institute

DEM Digital Elevation Model

CBDRM Community-Based Disaster Risk Management

ADPC Asia Disaster Preparedness Center

TABLE OF CONTENTS

CHAPTER I INTRODUCTION……… 8

1.1 General introduction 8

1.2 Description of the study area 9

1.3 Description of the Huong River 10

1.4 Hue city in the context of climate change 11

1.5 Problems and need of study 18

1.6 Objectives of the study 21

1.7 Scope of study 21

CHAPTER II LITERATURE REVIEW……… 23

2.1 Flood hazard mapping 23

2.2 AHP method 25

2.3 Flood hazard index 27

CHAPTER III METHODOLOGY……… 30

3.1 Conceptual framework 30

3.2 Overview of the research 31

3.3 Flood hazard mapping 31

3.4 Flood hazard index identification 37

CHAPTER IV DATA COLLECTION AND ANALYSIS……….44

4.1 Data collection 44

4.2 Data analysis 44

CHAPTER V RESULTS AND DISCUSSION……… 49

5.1 Hydrodynamic model parameters 49

5.2 Flood hazard mapping 57

5.3 Flood hazard index 61

5.4 The impacts of flood on Hue city in the contexts of climate change 62

5.5 Community-based disaster risk management (CBDRM) 71

CONCLUSION AND RECOMMENDATION……… 75

References……… 78

LIST OF TABLES

Table 1.1: Flood season in the Huong river 11

Table 1.2: The change of average temperature in the recent decades 13

Table 1.3: Scenario of temperature change in future in Hue 13

Table 1.4: Scenario of rainfall change in future in Hue city 17

Table 1.5: Scenarios of sea level rise in the future of Hue city (cm) 17

Table 3.1: Saaty Rating Scale 38

Table 3.2: Random inconsistency indices (RI) for different number of criteria 40

Table 4.1: Pairwise comparison of flood depth categories respect to flood hazard 47

Table 4.2: Pairwise comparison of flood duration categories respect to flood hazard 47 Table 4.3: Pairwise comparison of flood velocity categories respect to flood hazard 48

Table 4.4: Pairwise comparison of components respect to flood hazard 48

Table 5.1: Result of Mike21 FM calibration flood event in 1983 52

Table 5.2: Hydrodynamic parameters after calibration process 54

Table 5.3: Result of Mike21 FM verification, flood event in 1999 55

Table 5.4: Flooded area in districts in Hue City 58

Table 5.5: The change of flood depth between climate change scenarios with 64

Table 5.6: The change of flood hazard levels by area 70

Table 5.7: What community should do and should not do in each stage of flood management 73

LIST OF FIGURES

Figure 1.1: Trend of average temperature in July (1986-2006) 12

Figure 1.2: Trend of average annual temperature in period 1986-2006 12

Figure 1.3: Trend of average rainfall change in September – November 15

Figure 1.4: Trend of average rainfall change in July 15

Figure 1.5: The maximum daily rainfall in 10 past decades 16

Figure 1.6: Administration Map of Thua Thien Hue Province 22

Figure 2.1: Structure of FHI study methods 23

Figure 3.1: Framework for flood risk assessment and risk management 30

Figure 3.2: Overview of the research 31

Figure 3.3: Study area 34

Figure 3.4: Steps for model calibration and verification 35

Figure 3.5: Diagram for converting qualitative indexes to quantitative value 40

Figure 3.6: Applying AHP in identifying flood hazard index at the Huong river 42

Figure 4.1: Surface topography of the study area 45

Figure 5.1: Topographic Mesh 50

Figure 5.2: Checking cross-sections 51

Figure 5.3: Observed and Calculated discharge at Cross-section 1 53

Figure 5.4: Observed and Calculated discharge at Cross-section 2 53

Figure 5.5: Observed and Calculated discharge at Cross-section 3 54

Figure 5.6: Observed and Calculated discharge at Cross-section 1 55

Figure 5.7: Observed and Calculated discharge at Cross-section 2 56

Figure 5.8: Observed and Calculated discharge at Cross-section 3 56

Figure 5.9: Flood depth map at the Huong river – Hue city in 1999 57

Figure 5.10: Flood flows velocity map at the Huong river – Hue city in 1999 59

Figure 5.11: Flood duration map at the Huong river – Hue city in 1999 60

Figure 5.12: Flood Hazard Index map at the Huong river – Hue city in 1999 61

Figure 5.13: The change of flood depth in climate change scenarios 64

Figure 5.14: The change of flood velocity in climate change scenarios 66

Figure 5.15: The change of flood duration in climate change scenarios 67

Figure 5.16: The change of flood hazard index in climate change scenarios 69

Figure 5.17: Disaster management cycle 72

- in Vietnam, Pakistan, Indonesia, Egypt, Myanmar, Afghanistan, Nigeria, Brazil, Thailand, Democratic Republic of Congo, Iraq, and Cambodia - account for 80% of the people at risk world-wide In addition, an average of $96 billion in global Gross Domestic Product (GDP) is exposed to river flooding each year And these numbers are expected to increase gradually in the next years because of population growth, urbanization, and climate change As a result, it will increasingly put people at risk Floods in Viet Nam are well-known phenomena and occur in all regions of the country, especially in the Central Coast region (CCFSC 2006) As an example, the Central Viet Nam’s flood of November 1999 killed 780 people, affected around 1 million residents, and sunk and damaged more than 2,100 boats This flood caused damage worth US$364 million (CCFSC 2006) Being a coastal province in Central of Vietnam, Thua Thien Hue province has been suffering from floods impacts annually Especially, in the context of climate change, catastrophic floods are increasing in term

of frequency and magnitude, and taking a high death toll, assets and infrastructures Therefore, the measures in flood risk management and mitigation for Thua Thien Hue province are very indispensable and need to be researched strictly One of the effective approaches which are being used widely in flood risk management is flood hazard assessment This approach showed its capacity to apply in practice and it is a useful tool to facilitate in flood risk management and mitigation

Flood hazard assessment in a river basin can be performed by overlaying maps or/and identify indexes Each certain area has a hazard value The value can be utilized in analyzing, estimating and comparing among different areas in order to support for making decision Thus, realizing the impacts of floods on Thua Thien Hue province in general and Hue city in particular, this research studies “Identify the flood hazard index in the Huong river basin – Hue city area” The result of this project will be foundation for identifying the flood risk index and evaluating flood risk in the area, and support to help decision makers in making flood prevention plans for Hue city

1.2 Description of the study area

Thua Thien Hue is a province in the North Central Coast region of Vietnam The province is located at the latitudes 16°14'- 16°15' north, longitudes 107°02' - 108°11' east Area of the province is 5,053.990 km2; population is 1.115.523 people according

to statistics in 2012 It borders Quang Tri province to the North and Da Nang to the South, Laos to the West and the East Sea to the East The province has 128 km of coastline, 22,000 ha of lagoons and over 200,000 ha of forest The province comprises

4 different zones: a mountainous area, hills, plains and lagoons separated from the sea

by sandbanks The mountains, covering more than half of the total surface of province, with height ranges from 500 to 1480 m The hills are lower, between 20 to 200 m, and occupy about a third of the province’s area, between the mountains and the plains The plains account for about a tenth of the surface area, with a height of only up to 20m above sea level Between the hills are the lagoons which occupy the remaining 5% of the province’s surface area

The climate in Thua Thien Hue province is similar to Central Vietnam in general – a tropical monsoon climate In the plains and in the hills, the average annual temperature

is 25oC, but in the mountains only 21oC (statistical yearbook 2004) The annual precipitation in the province is 3200 mm but there are important variations Depending

on the year, the annual average may be 2500 to 3500 mm in the plains and 3000 to

4500 mm in the mountains In some years the rainfall may be much higher and reach more than 500 mm in the mountains

The sale of goods and services in the province is 10930.6 billion VND accounting for 0.9% of total sale of goods and services in the whole country This is compared with 12.7% of Hanoi and 23.5% of Ho Chi Minh City The province has more than 120 km

of coastline, which provides for a seafood industry which produces over 40,000 tons/year consisting of over 500 species of fish

Hue is the city of Thua Thien Hue province The city is the center of culture, politics, education, science, tourist… Area of the city is 71.68 km2 Population in 2012 is estimated as 344,581 people Hue city is located in downstream of the Huong River and Bo River, average height is about 3-4 m above sea level and is often submerged when a heavy rain occurs in the upstream of the Huong River

1.3 Description of the Huong River

The Huong river basin is located from 15o29’ – 16o35’ of the North latitude to

105o07’-107o52’ of the East longitude, with the basin area of 2830 km2 The river length is 86.5km included 28 distributaries The upstream of the Huong River is called

as Ta Trach River which derives from a high mountain area of Bach Ma mountain range The Ta Trach River connects with the Huu Trach River at the Tuan confluence From the Tuan confluence, the main flow is called the Huong River

The river then flows in the general direction of southeast to northwest, passing the Hue city, and before flowing into the sea, the Huong River goes through the Tam Giang – Cau Hai lagoon Tam Giang - Cau Hai is the largest lagoon in the South East Asia, with an area of 22,000 ha, and a length of 68 km along the coastline of the province Finally, the river flows to the sea at the Thuan An and Tu Hien mouths Besides the Thuan An and Tu Hien estuaries, the lagoon systems have some smaller river mouths linking to the sea

As same as other rivers in the Central of Vietnam, flood season in the Huong river basin is not so long, only about 4 months: from September to December with the amount of water accounting for 70% - 75% of total volume of annual flow Therein, November has the largest amount of flood and often account for 30% - 35% of total volume of annual flow Although flood season is short but many large floods occurred

in the Huong River such as floods happened in 1953, 1975, 1983 and 1999 These floods often lasted in 5-7 days

Table 1.1: Flood season in the Huong river

Station Flood discharge in flood season (m 3 /s) with P = 75% Average

1953, 1975 and 1983 corresponding to 5.81m, 5.50m, 5.32m and 4.92m of water level

In summary, the Huong River plays an important role in development of livelihood, economy and society in Thua Thien Hue province However, the Huong river basin is also vulnerable and susceptible to natural disaster (especially to flood inundation) and impacts of climate change In recent years, Thue Thien Hue province and the Huong river basin has been affected by many natural disasters such as: storm, heavy rain, flood and drought with high intensity and frequency, caused many losses of people and socio-economy, damaged cultural heritage and property of local residents

1.4 Hue city in the context of climate change

1.4.1 Climate change in Hue city

a The change of temperature from the past to now

The trend of temperature change is estimated based on series of observed data from

1931 to now Analytic results showed that in this period, the average annual and

monthly temperature has inconsistent change and does not show a clear trend Basically, average annual temperature tends to increase slightly (0.10C – 0.20C) from period 1931 – 1940 to 1971 – 1980, however, from the late periods until now, temperature tends to decrease 0.20C – 0.30C (Phong, 2014)

Figure 1.1: Trend of average temperature in July (1986-2006)

(Source: Climate action plan Responding to Climate Change From 2014 – 2020)

Figure 1.2: Trend of average annual temperature in period 1986-2006

(Source: Climate action plan Responding to Climate Change From 2014 – 2020)

The scenario for the temperature change in the future

In general trend, average seasonal and annual change is likely to increase in the future with minimum increase of 10C in 2050 (corresponding to low emission scenario B1) occurs in the summer The maximum increase of average seasonal and annual temperature can reach to 3.70C in 2100 (corresponding to high emission scenario A1)

Thus, the average temperature tends to rise more in the spring and winter while it rise

a little in the summer In terms of extreme value, the lowest temperature in winter (corresponding to B2 scenario) increases 1.20C in 2050 and 2.20C in 2100, while the highest temperature increases 2.20C in 2050 and 3.20C in 2100 Besides, in 2100, the number of days with maximum temperature above 350C may increase 10 days –

20days per year (corresponding to medium emission scenario B2) (Phong, 2014)

Table 1.2: The change of average temperature in the recent decades

Average temperature in January

Average temperature in July

Average annual temperature

Seasonal average Extremely temperature (B2)

Winter (XII-II) 1.4-1.80C 1.6-3.70C Minimum: 1.0-1.20C

Maximum: 1.2-2.20C

2.0-2.20C 2.2-3.20C Spring (III-V) 1.2-1.60C 1.6-3.70C

Summer (VI-VIII) 1.0-1.40C 1.0-3.10C Minimum: 1.7-20C 2.7-3.20C

Maximum: 1.0-1.20C 2.2-3.20C Autumn (IX-XI) 1.0-1.60C 1.3-3.70C

Average annual 1.2-1.60C 1.6-3.70C Minimum: 1.0-1.70C

Maximum: 1.0-1.70C

2.2-3.00C 2.0-3.20C

b The change of precipitation from the past to now

According to observed data, annual rainfall in Hue city is relatively high compared with other regions with the annual rainfall from 2700 mm to 2800 mm The extremely high total rainfall occurs in some years (for example: rainfall in 1999 was up to 3093 mm) Regarding to distribution of rainfall over time, rainfall usually concentrates mainly in October and November In some years, rainfall in one of 2 months accounts for 60% to 80% annual rainfall (for example the rainfall in November 1999 is 2452

mm while the annual rainfall is 3093 mm)

According to statistics, rainfall in Hue has a considerable variation over the decades and it does not show a clear trend The average annual rainfall tends to decrease from the decade 1961-1970 to 1981-2000(from 2842 mm down to 2575 mm), but then increase gradually in the next 2 decades The most significant increase is more than

500 mm in 1991-2000 compared with the previous decade It is worth noting that even though the average annual rainfall increases but rainfall in July (dry season) in the period 1991-2000-2010 has a strong downtrend, and rainfall in September, October and November (rainy season) tends to increase compared with 2 previous decades Compared with the period 1961-1970, the average rainfall on July in the decade 2001-

2010 decreases 23% while the rainfall in November increases 27%

Regarding to rainfall intensity, in the past few decades, the intense rainfalls appear more and more and always happen in October and November Heavy rainfall occurs in some days, for example, on 2nd November 1999, a rainfall with 978 mm happened, accounted for 20% the total rainfall that year

In summary, the average annual rainfall in the decade 2001-2010 is larger than the previous decades since 1961 but we cannot confirm about the trend of average annual

value Only one thing we can confirm is that the rainfall in October and November has always reached extremely levels, accompanied by intense rainfalls and tend to rise in October and November while the rainfall on July – dry season tends to decrease

(Phong, 2014)

Figure 1.3: Trend of average rainfall change in September – November

(Source: Climate action plan Responding to Climate Change From 2014 – 2020)

Figure 1.4: Trend of average rainfall change in July

(Source: Climate action plan Responding to Climate Change From 2014 – 2020)

Figure 1.5: The maximum daily rainfall in 10 past decades

(Source: Climate action plan Responding to Climate Change From 2014 – 2020)

The scenario for the precipitation change in the future

According to forecast, the average annual rainfall in Hue city may increase 3% - 4% in

2050 and 6% - 10% in 2100 The average rainfall in the spring, summer and autumn may increase while the rainfall in winter may decrease

It is worth noting that the rainfall decreases in the dry season with the maximum decrease is 6% in the middle and 10% in the end of the decade The rainfall in autumn (from October to December) has the largest increase with the maximum increase is up

to 16% in 2100 while the highest rainfall in the year concentrates in this stage As a result, flood and drought in Hue city is likely to become more serious in the future In addition, the maximum daily rainfall in Hue may increase about 20% compared with corresponding value in the period 1980-1999 and even the abnormal rainfall can appear with the rainfall rises twice as the record rainfall at present (Phong, 2014)

Table 1.4: Scenario of rainfall change in future in Hue city

(Source: Climate action plan Responding to Climate Change From 2014 – 2020)

c Sea level rise

Sea level rise scenario for Hue city is taken to the forecasting figures for the region from Ngang mountain pass to Hai Van mountain pass The change of sea level in the

future is compared with the average sea level in the period 1980-1999

According to the table below, sea level tends to increase in the future In 2020, 2050 and 2100, sea level could rise at the highest level of 9 cm, 28 cm and 94 cm Besides, the error in forecasting between low emission scenario and high emission scenario tends to increase over time This shows that the uncertainty of the forecast in future is

larger and larger (Phong, 2014)

Table 1.5: Scenarios of sea level rise in the future of Hue city (cm)

1.4.2 The impacts of flood on Hue city in the contexts of climate change

The main types of natural disaster in Hue city includes typhoon, flooding, drought, whereas flooding is considered as the most dangerous natural disaster and causes the most damage to Hue In recent years, under the influences of variations regarding temperature, rainfall, floods tend to serious, complicated and unpredictable

Floods usually occur in rainy season and mostly concentrated in the period from September to December annually Total of flow during flood season accounts for about 65% of the total annual flow According to observed data, there are 3.5 floods in average annually which are equal or higher than flood alarming level II occurred in the Huong river

Normally, the flood duration is about 3-5 days in average The longest period of a flood ups to 6-7 days The average time for transferring flood from upstream (Thuong Nhat) to downstream (Kim Long) with the distance of 51 km is about 5-6 hours The severity of a flood (flood duration and flood depth) depends on many factors such as rainfall in upstream, rainfall in Hue city, tide and sea level rise (due to storms or the rising of earth’s temperature)

Thus, under the impacts of climate change, in recent decades, floods in Hue city tend

to become more complex, less predictable and more dangerous (Phong, 2014)

1.5 Problems and need of study

Vietnam is a coastal country with a long coastline and located in the tropical monsoon climate region, Vietnam has suffered impacts of floods annually Since ancient times, Vietnamese people regarded flooding as one of the four biggest dangers to people, along with fires, robbers and invaders

In order to control flooding, a large system of river and coastal dykes has been constructed For many centuries, these flood control measures achieved results all over the country However, this structure approach is now under pressure because the conditions inducing flooding are intensifying, both at local and global level (CCFSC, 2006) For example, increasing population, rapid urbanization, high demand for natural resource exploitation, environmental pollution, and degradation are coupled

with global threats, such as climate change (Tran et al 2008) In addition, due to the

limitation of current fund in Vietnam, non-structure measures are more concentrated and given preference over the structure measures One of the most effective non-structure measures currently is flood risk assessment Flood risk assessment will be a background for planning and coping with floods

Being the central of Thua Thien Hue province, Hue city is located in intersection of traffic routes from North to South and East-West economic corridor connecting Thailand, Vietnam and Laos It is also where has interference about society – economy – culture of 3 regions North – Central – South With the importance of a central urban where experienced many historical events, Hue city has been had outstanding development in socio – economy, infrastructure, culture, tourism,… In recent years, infrastructure, new urban areas have been invested continuously and this made the appearance of Hue city become more modern and civilized

In the context of climate change, Thua Thien Hue province in general and Hue city in particular is suffering from many impacts of climate change The ultra-weather events tend to become more severe in terms of both frequency and magnitude The direct consequences caused by climate change in Hue city in recent years are the occurrence

of many events such as flood, storm, drought and deep freeze These events have influences on social and economic development Whereas, flooding is considered as a top threat to coastal cities like city of Hue This was demonstrated via the historical flood events, such as the flood events in 1953, 1975, 1983, 1999… And now, the severe flood events are still happening more and more powerful under the impacts of climate change

The flood event started from 20 to 26/9/1953 caused 500 casualties, swept away 1290 houses, 80% of the crops were lost

A big flood occurred from 15-20/10/1975 took a heavy toll of people and property The flood event in 1983 lasted for 8 days with the flood peak discharge observed at Co

Bi, Binh Dien and Thuong Nhat station corresponding to 2850 m3/s, 4020 m3/s and

1470 m3/s The flood caused 252 dead, 115 people were wounded, 2100 houses were collapsed, and 1511 houses were swept away

The flood event 11/1999 happened for 6 days (1/11 – 6/11), was a very large flood The flood peak discharge observed at Ta Trach, Binh Dien and Co Bi stations corresponding to 7000-8000 m3/s, 5500-6000 m3/s and 3500-4000 m3/s The flood caused widespread inundation and flood damage was enormous Flood level in the upstream up to 1.4 m, more than 90% residential areas living in the delta were submerged for 4-9 days From 1/11 to 12/11, there were 372 dead and missing, total damage was estimated as 1,762 billion VND Transportation sector had the most damage of 600 billion VND, followed by Agriculture sector of 307 billion VND and Fishery sector of 110 billion VND

According to predictions of experts, Hue city will be one of areas where has to suffer from many impacts of climate change in the process of socio-economic development Vulnerability caused by climate change to Hue city is considered more serious than other regions in the province due to population density and the level of infrastructure investment is very high, especially, Hue city is planning to become a nuclear urban in the future However, the problems related to climate change are still new and have not been perceived deeply and implemented specifically to local authorities and residents Although, the local authority has made great efforts in urban management, environmental protection and disaster damage mitigation, but the role of the city authority in adaptation and mitigation of climate change impacts is not clear Facing with negative effects of climate change requires the local authority need to assess properly the situation and existing capacity in responding, then recommends suitable solutions to minimize the negative impacts caused by climate change

Facing with the problems as mentioned above, this study will identify one of the most important components in Flood risk assessment in the Huong river basin: Flood hazard index Flood hazard index represents the level of flooding impacts It is combination of all hazard parameters such as flood depth, flood duration, velocity of flood flow… Base on the index, the hazard zones are determined The impacts of flood on local people, social-economic development in whole study area can be reduced by the

structure and non-structure measures However, the structure and non-structure measures also need detailed calculations and estimations in order to mitigate the influences of flood in the economical and efficient way

1.6 Objectives of the study

The general objective of this study is to support to flood control and mitigation works

in the Hue city

The specific objectives of the study are:

1 Simulate flood flow: Simulation and estimation magnitudes, drainage process, and flooding flow along the river

2 Create flooding maps: Developing flood depth, flood velocity, flood duration maps for the river basin

3 Identify flood hazard index: quantify factors which contribute to the damaging potential of flood hazard to serve for flood risk assessment

socio Selecting suitable model for simulating a flood processes in the Huong River

- Calibrating and verifying the selected model

- Developing flood hazard mapping based on results of the above model and analysis

- Applying Analytical Hierarchy Process (AHP) method and ArcGIS software to develop flood hazard maps and calculate flood hazard index

- Assuming scenarios in context of climate change to estimate the impacts of flood to Hue City in the future

- Recommending some non-structure measures to help local residents in coping with flooding

Figure 1.6: Administration Map of Thua Thien Hue Province

CHAPTER II

LITERATURE REVIEW

Flood hazard assessment including the integration of GIS and hydraulic model as well

as identifying indexes related to flooding by applying weight methods have been described and studied by many researchers These approaches are also estimated as effective approaches in flood hazard assessment currently In order to determine theoretical base for this research, some available references have been reviewed This chapter summarizes the review of related literature as following structure:

Figure 2.1: Structure of FHI study methods

2.1 Flood hazard mapping

Gardiner (1990), as cited by Omran et al (2011), indicated that, the morphometric

characteristics of basins have been used to predict and describe flood peaks and estimation of erosion rate Indeed, the relationship between basin morphometric and flooding impact have also been investigated Morphometric studies include the evaluation of streams through measurement of stream network properties, which are

Literature review

calculated based on such characteristics as drainage density, water allocation ratio, stream frequency and overland flow However, it is difficult to measure the details of drainage elements in the field due to their extent throughout rough terrain over vast area Thus, in order to solve the task, the author applied GIS techniques to extract the stream networks as well as analysis morphologic characteristics of the basin The aim

of this research is to produce a potential flood hazard map based on geomorphic parameters and to estimate the risk degree of individual sub-basin in the study area According to Marfai and Njagih (2002), flood is recognized as one of two hazardous phenomenon which has the most serious damages to people and economy in Turialba City in Costa Rica annually Therefore, the authors found that it’s necessary to do risk assessment in order to know how much would be the damage if the flood hazard occurs In the research, the authors considered flood hazard assessment as an indispensable part in flood risk assessment, beside flood vulnerable assessment In this part, he used ILWIS software to generate the flood hazard maps corresponding to various return periods These maps then were used to serve for risk assessment and cost estimation for study area

Karagiozi et al (2011) conducted their research in Laconia Prefecture in Peloponneus,

Greece In the research, flood hazard assessment was implemented by using hydrological models in a GIS environment taking into account the geomorphologic characteristics of the study area For each basin, the morphologic characteristics such

as area, mean slope, mean elevation and total relief were calculated These factors then were combined by using GIS to produce a final flood hazard map

According to Ripendra (2000), flood hazard mapping and risk assessment in Nepal is still rudimentary Most of the flood protection works were carried out at the local scale without proper planning or without considering the problem at river basin scale Apart from piecemeal approaches on a limited scale, no pragmatic efforts in comprehensive flood risk assessment and flood hazard mapping have been done Therefore, in his own study, he prepared flood vulnerability, flood hazard and flood risk maps by integrating the hydraulic model HEC RAS and GIS with the case study of Lakhandei River basin The results of the research are the flood vulnerability, flood hazard and flood risk

maps with the assessment for each elements The author then emphasized to the important role of community as well as stakeholders in flood management and made the action plan for flood mitigation in study area

2.2 AHP method

Siddayao et al (2014) combined Analytical Hierarchy Process (AHP) method with

Geographical Information System (GIS) for flood risk analysis and evaluation in the town of Enrile, a flood-prone area located in northern Philippines AHP results showed the relative weights of three identified flood risk factors, and these results were validated to be consistent, using a standard consistency index Using the GIS software, the factor weights from the AHP were incorporated to produce a map with 5 levels of estimated flood risks Using such a GIS weighted overlay analysis map as guide, local residents and other stakeholders can act to prepare for potential flooding, promote appropriate land-use policy that will minimize threat to lives due to flooding

According to Chen et al (2011), flooding is one of the major natural hazards in

Taiwan and most of the low-lying areas in Taiwan are flood prone areas Thus, a comprehensive decision making tool for flood control planning and emergency service operation is necessary in order to reduce losses of life and economy A research about flood risk assessment was then carried out by him The research objectives were to develop a hierarchical structure through the Analytic Hierarchy Process (AHP) to provide preferred options for flood risk analysis; map the relative flood risk using the Geographic Information System (GIS), and integrate these two methodologies in flood risk assessment The results of research indicated that integration of AHP and GIS in flood risk assessment can provide useful detailed information for flood risk management, and the method can be easily applied to most areas in Taiwan where required data sets are readily available

A flood hazard assessment using AHP and mapped by GIS has also been applied for

the Yasooj River, Iran (Rahmati et al 2015) The aim of this research is to identify

potential flood hazard zones The decision factors for flood hazard of the AHP matrix include distance to river, land use, elevation and land slope The set of criteria were integrated by weighted linear combination method using ArcGIS 10.2 software to

generate flood hazard prediction map The efficiency of AHP method in identifying potential flood hazard zones was then accessed by comparing with the results of the hydraulic model HEC-RAS Finally, the results showed that the AHP technique is promising of making accurate and reliable prediction for flood extent Therefore, the AHP and geographic information system (GIS) techniques are suggested for assessment of the flood hazard potential, specifically in un-gauged regions

The integration between AHP and GIS was also applied in flood hazard assessment in

Kujikuri Plain, Chiba Prefecture, Japan (H Chen et al 2014) In the research, six

factors were selected including river system, elevation, depression area, ratio of impermeable area, detention ponds, and precipitation The method of analytic hierarchy process was applied to calculate the weighting values of each factor The flood hazard map was also obtained The hazard map was then compared with the actual flood area, and good coincidence was found between them In conclusion, the flood hazard assessment method presented in the research is meaningful for the flood management and environment protection in the area under the similar condition as this study

Ajin et al (2013) conducted flood hazard assessment in Ghaggar basin, India by using

Multi criteria evaluation methods In the research, the authors showed some criteria which are characteristics of study area and have the strongest effects in triggering flood in the area The maps of Rainfall distribution, micro watershed size, slope drainage density, soil type, land use land cover, and Roads/micro watershed were created using Equal Interval Method in ArcGIS software by assigning weightage for each class The final result of this research is the Flood hazard zone map with flood prone areas was identified after the author gives suitable rank for each contributing factors based on its estimated significance in causing flood

Until now, many studies applied advanced technologies such as the integration between GIS and hydrodynamic models in order to analysis and assess flood hazard very well However, most of studies used the 1D hydrodynamic models In order to make the assessment results more accurate, the combination between GIS tools and 2D model should be applied in researches

2.3 Flood hazard index

Flood hazard assessment is considered as an essential part in the flood risk assessment for Phrae flood plain of the Yom River basin in northern Thailand by Tingsanchali and Karim (2010) The study was carried out by applying a hydrologic-hydrodynamic model in association with a geographic information system (GIS) Flooding scenarios were estimated in term of flooding depths for 25, 50, 100, 200 years return periods The authors then select critical depths based on the guide-lines in the flood-plain development manual to create four hazard categories A hazard index was introduced

in the research to represent degree of hazard corresponding to each category The results showed that 78% of the Phrae flood-plain area of 476 km2 in the upper Yom River basin lies in the hazard zone of the 100 years return period flood

Masood (2011) carried out flood hazard assessment for the mid-eastern Dhaka, Bangladesh In the research, the inundation simulation is conducted by using HEC-RAS model for 100 years flood A Flood Hazard map then was prepared using the inundation status which was found from hydrologic simulation Based on the inundation depth, the hazard index was assigned for the study area The index then was associated with vulnerable index to calculate flood risk index with the purpose of assessing flood risk for the area

According to Elkhrachy (2015), flash flood in the cities led to high levels of water in the streets and roads, causing many problems such as bridge collapse, building damage and traffic problems It is impossible to avoid flood risk or prevent flood occurrence, but it is plausible to reduce their effects and the losses The author supposed that flash flood mapping to identify sites in high risk flood zones is one of the powerful tools for this purpose Therefore, the objective of this paper is to generate flash flood map for Najran city, Saudi Arabia, using satellite images and GIS tools Analytical Hierarchical Process (AHP) is also used to determine relative impact weight of flood causative factors to get a composite flood hazard index (FHI) The causative factors in this study are runoff, soil type, surface slope, surface roughness, drainage density, distance to main channel and land use

Flood hazard mapping is also considered as a vital component for appropriate land use planning in flood-prone areas (Bapalu, 2006) The research was carried out to identify flood hazard index for Kosi river basin, North Bihar, India in a GIS environment The authors used one of the multi-criteria decision-making techniques, Analytical Hierarchical Process (AHP) which provides a systematic approach for assessing and integrating the impact of various factors A composite index of flood hazard derived from topographical, land cover, geomorphic and population related data All data are finally integrated in a GIS environment to prepare a final Flood Hazard map This flood hazard index computed from AHP method not only considers susceptibility of each area to be inundated but also takes into account the factors that are inherently related to flood emergency management

The role of flood hazard map is emphasized one more time in the research of Forkuo (2011) because of its advantages such as easily read and rapidly accessible This study addresses the need for an efficient and cost-effective methodology for preparing flood hazard maps in Ghana A composite flood hazard index of the study area was created by incorporating variables of near distance to the White Volta River, population density, number of towns in each district, area of cultivated savanna (crops), and availability of high ground (Shelter) Also, maximum flood hazard zones were mapped in a GIS environment

Conclusion:

The references mentioned above at first indicated the necessary of flood hazard assessment while flood has strong impacts on people, socio-economy in regions or areas where could be affected easily by flood There are many approaches or methodology employed to do this assessment However, one of the methodologies applied widely in flood hazard assessment nowadays is the integration of hydrodynamic models and Geographical information system This methodology demonstrated its capabilities and applicability in flood hazard assessment via references mentioned above Therefore, this research will use a hydrodynamic model

to simulate the flooding process and then show the specific factors of a flood by creating visual maps in GIS environment

Besides flood simulation and flood map creation, flood hazard index identification also has a great significance in flood hazard assessment Flood hazard index is determined based on the factors related to flood and the weight of each factors Analytical Hierarchy Process (AHP) method applied in references mentioned above is a powerful tool in group decision – making and it is used to quantify the qualitative preferences among components or subcomponents as well as indicators or categories Thus, the research will applied this method to assess the importance of each flood factor, and then identify the flood hazard index in GIS environment

The integration of GIS and AHP method has been applied in many researches regarding flood hazard assessment and flood hazard identification In some researches mentioned in literature review, the components of a flood are selected based on the topographic and geomorphological characteristics However, the collection of these data is very difficult in reality due to the limitations of system of measurements in Vietnam In other researches, flood hazard index was just identified based on flood depth Flood depth is a critical component of a flood when we consider to the flood impacts on people and socio-economy But, if only this component is considered in identifying flood hazard, it is unable to access comprehensively the flood influences Therefore, this study will select 3 crucial components of a flood: flood depth, flood duration and flood velocity and used them as 3 criterions in flood hazard assessment and flood hazard index identification

In addition, the reports and documents of international organizations and Vietnam agencies indicated that flood properties will tend to increase significantly under the effects of climate change in the future in Central of Vietnam in general and city of Hue

in particular Thus, this study will base on the climate change scenarios which built for Hue city to identify flood hazard index for this area in 2030 Finally, the study will have comparisons between the historical floods with the flood which is predicted in the future for the purpose of assessing the changes of flood by time

CHAPTER III

METHODOLOGY

3.1 Conceptual framework

Figure 3.1: Framework for flood risk assessment and risk management

Source: Adapted from WMO, 1999

In the Flood risk assessment and flood risk management, flood hazard assessment plays an important role This research focuses on the assessment and quantification of the indicators of flood hazard

Flood risk assessments starts with an assessment of the flood hazard, which indicates

the probability and intensity of a possible event (de Moel et al, 2015) Flood hazard

Natural System Observations

Inventory Accounting Data Thematic Event Maps

Establish Flood Hazard Potential

Stream versus Probability of Occurence

Assess Vulnerability

Value (Material, People) Injuries Resillency

Risk Assessment

Risk * function (Hazard, Vulnerability, Value)

Protection Goals/ Risk Acceptance

Unacceptable Risk Acceptable Risk

Implementation and Periodic Review

Planning / Mitigation Measures

Reduce Risk Through Best Possible

Management Practices

> Land use Planning (Flood Plain Mapping)

> Structural Measures (Dams, BuildingCodes)

> Flood Forecasting and Warning Systems

could be estimated through many indicators, however, three basis characteristics are flood depth, flood velocity, flood duration The flood depth and flood velocity have direct impacts on infrastructure, houses, loss of life, public health Flood duration causes indirect and intangible damages such as diseases, life disruption stress This research will base on these factors to quantify and identify hazard index in the Huong river basin

3.2 Overview of the research

Figure 3.2: Overview of the research

3.3 Flood hazard mapping

Flood hazard mapping is the basic document, the scientific basis for flood prevention planning, selection of measures, design of flood control constructions and the

Hydraulic model (MIKE 21)

Simulating and estimating flood

inundation process

Data analysis

GIS tools (MIKE-21, ArcGIS) Estimating the depth, duration and

velocity of the flood

Quantify the factors and Identify Flood Hazard index

necessary information to inform to people about the potential damage caused by flood

in local areas

Flood hazard mapping determines the boundaries of inundation areas caused by a flood event The boundary of an inundation area relies on flood water level, topographic and geomorphologic factors in the area; while topographic factors have fewer changes, flooding boundary depends on variations of flood water level

There are 3 approaches are being applied in creating flood hazard maps:

- Traditional approach: Creating flood hazard maps based on geomorphological survey

- Creating flood hazard maps based on big flood events which occurred in history

- Creating flood hazard maps based on hydraulic and hydrologic models

Each approach has its advantages and disadvantages in generating and estimating flooding areas Flood hazard maps which are built by traditional approach, only reproduce flooding status It’s unpredictable but it still has great significance in commanding flood prevention as well as being the basis in evaluating and comparing further studies However, this approach takes a long time, has low predictability and does not meet the actual needs

Generating flood hazard maps based on survey data and the data of flood events which occurred in the past is the most reliable However, the data for major flood events is inadequate and unpredictable in the future Thus, this approach limits many advantages and applicability of flood hazard maps in practice

Applying hydrologic and hydraulic models is an effective way in simulating the flooding Besides, this is also a modern approach and being used widely in the world

as well as in Vietnam Otherwise, with the fast-paced development of computer, information system and database, an increasing number of applications developed based on geographic information system (GIS), in which, creating flood hazard maps

is one of important application, which brings practical benefits in flood prevention and disaster mitigation

Therefore, this thesis will focus on calculating and analyzing inundation process by applying hydrodynamic model The outputs of the model then will be used for creating flood hazard map in GIS environment

3.3.1 Application of 2D hydrodynamic model - MIKE 21 Flow FM in flood simulation

3.3.1.1 Hydrodynamic model – MIKE 21FM

MIKE 21 Flow Model FM is a two-dimensional hydraulic model of the MIKE software The model has been built and developed by the Danish Hydraulic Institute (DHI) since the late 90s Mike 21FM model was presented in Vietnam in November

2005 by technology transfer between DHI and Irrigation Planning Institute MIKE 21

is dedicated software which is used to simulate the 2 dimensional variations of water level and flow in lake, estuary, bays, coastal and offshore areas

MIKE 21 Flow Model FM was built and incorporated with new model techniques and uses unstructured mesh approach (triangular mesh) This technique has been developed for applications related to environment in estuary, coastal areas, oceans and inland flood overflow

MIKE 21 Flow Model FM is composed of following modules:

- Hydrodynamic module

- Transport module

- Eco Lab/Oil spill module

- Particle tracking module

- Mud Transport module

- Sand Transport module

Hydrodynamic module is the basic computational component of the entire MIKE 21 Flow model FM modeling system providing the hydro-dynamic basis for the Transport module, Eco Lab/Oil spill module, Particle tracking module, Mud Transport module and Sand Transport module

With the advantages in creating a flexible mesh, the scientific basis of MIKE 21 FM showed the applicability with following researches:

- Studying overall hydraulic regime across the river and details at each location including characteristics of water level, discharge, flow velocity and their horizontal distribution Especially, the model has a great performance in calculating the flow in the rivers which have many changes about direction of flow – an important component

in the study of erosion and accretion

- Calculating the changes of river bed and erosion river banks in its natural status as well as plans of exploitation in the river in the future

3.3.1.2 Study area

The study area is the Huong river section – Hue city with the length of 10 km

Figure 3.3: Study area

3.3.1.3 2D MIKE 21 FM calibration and verification

The model calibration is primarily conducted by modifying the value of Bed resistance The method is used for model calibration in the study is trial-error method

Figure 3.4: Steps for model calibration and verification The concrete steps are described as below:

- Select an observed data series which is adequate and reliable for model calibration

- Set up parameters for model calibration

- Identify the criteria to evaluate the error for calibration process

- Use trial-error method to find out the parameters satisfying the give criteria

- After the suitable parameters for calibration is found out, the parameters are used for model verification with a different observed data series

The result of the model is evaluated by criteria for error assessment:

- The difference between the peak of observed flood data and calculated flood data:

∆H = Hpcal – Hpobs

Whereas: Hpcal : Value of calculated peak of water level

Hpobs: Value of observed peak of water level

- The Nash – Sutcliffe model efficiency coefficient:

Assume the

parameters

Run the model

Select assumed parameters Good

Modify the parameters

Bad

Compare the calculated result with observed data

Stop

Whereas: Ho,i: Value of observed water level

Hs,i: Value of water level which is calculation in Mike 21 FM

3.3.2 Flood Hazard mapping

3.3.2.1 Overview of ArcGIS

ArcGIS (ESRI Inc - http://www.esri.com): is the state of the art GIS system, provides

a comprehensive solution from collecting, revising, analyzing data and distributing information on the Internet to different levels as personal geographic database and business database In term of technology, GIS professionals consider ESRI technology

as an open and completed solution which has capacity in exploiting all functions of GIS in various applications such as: desktop (ArcGIS Desktop), Server (ArcGIS server) and Web applications (ArcIMS , ArcGIS online), or mobile system (ArcPAD)…

ArcGIS Desktop (with the newest version as ArcGIS 10) consists of powerful tools in managing, updating, analyzing information and creating a complete geographic information system, allows:

- Creating and revising integrated data (spatial data integrated with attribute data) – allows using different data formats, even the data which is downloaded from the Internet

- Querying spatial data and attribute data from many sources and in many different ways

- Displaying, querying and analyzing spatial data in combination with attribute data

- Establishing thematic maps and professional-quality presentation

In case of associating with hydrologic and hydraulic models, ArcGIS is an indispensable component The role of GIS tools is presented in:

,,1

Ho i Ho

i Hs i Ho

- Synthesizing and selecting documents as important inputs in hydrologic and hydraulic models

- Analyzing, visualizing and assessing the area and the levels of inundation using the outputs of above models

3.3.2.2 Flood hazard mapping

Flood hazard map can be assessed through basic indicators such as flood depth, flood duration, flood velocity, debris in the flood flow (sediments, salts, chemical substances, waste water and soil),etc…In the elements mentioned above, flood depth, flood velocity and flood duration play an important role in identifying flood damage The integration between flood depth and flood velocity demonstrates the ability to destroy objects in affected areas It has a direct impact on the objects such as houses, buildings, people lives… Flood duration has an indirect effect on socio-economic activities, environmental pollution, diseases, community‘s health… Based on the map overlaying method to overlay flood depth, flood velocity and flood duration maps (the outputs of MIKE 21), the study will build the flood hazard map for flood event in

1999

3.4 Flood hazard index identification

3.4.1 Analytical Hierarchy Process (AHP)

AHP method is a pairwise comparison method which has the added advantages of providing an organized structure for group discussion, and helping the decision making group focus on areas of agreement and disagreement when setting criterion weights (Drobne & Lisec, 2009)

The technique of pairwise comparisons has been developed by Saaty in 1977 in the context of a decision making process known as the Analytical Hierarchy Process (AHP) In Saaty's technique, weights of this nature can be derived by taking the principal eigenvector of a square reciprocal matrix of pair-wise comparisons between the criteria The comparisons deal with the relative importance of the two criteria involved in determining suitability for the stated objective Ratings are provided on a nine-point continuous scale (Table 3.1):

Table 3.1: Saaty Rating Scale

Source:(Haryanto, 2014)

In developing weights, an individual or group compares every possible pairing and enters the ratings into a pairwise comparison matrix or ratio matrix Since the matrix is symmetrical, only the lower triangle actually needs to be filled in The remaining cells are then simply the reciprocals of the lower triangle (Drobne & Lisec, 2009)

The procedure then requires that the principal eigenvector of the pairwise comparison matrix must be computed to produce the best fit set of weights A good approximation

to this result can be achieved by following the operations below: (Drobne & Lisec, 2009)

- Sum the values in each column of the pairwise comparison matrix;

- Divide each element in the matrix by its column total (the resulting matrix is referred

to as the normalized pairwise comparison matrix); and

- Compute the average of the elements in each row of the normalized matrix, that is, divide the sum of normalized scores for each row by the number of criteria

These averages provide an estimate of the relative weights of the relevant criteria Here, the weights are interpreted as the average of all possible ways of comparing the criteria

In practice, it is not always possible to build transitive relation during pair-wise comparisons For example, the plan A may be better than B, B may be better than C,

but A may not be better than C This called inconsistency The inconsistency is real but inconsistency level should not be too high because it shows the incorrect assessment then AHP provides mathematical measures to determine the inconsistency level of judgments through the consistency ratio (CR) If the value is less than or equal

to 0.10, it is acceptable Otherwise, if this value is greater than 0.10, it is necessary to re-assess the previous steps

Estimation of the consistency ratio involves the following operations: (Drobne & Lisec, 2009)

- Determination of the weighted sum vector by multiplying the weight for the first criterion times the first column of the original pairwise comparison matrix, then multiplying the second weight times the second column, the third criterion times the third column of the original pairwise matrix, and so on to the last weight, and finally summing these values over the rows; and

- Determination of the consistency vector by dividing the weighted sum vector by the criterion weights determined previously

The consistency ratio is defined as:

Table 3.2: Random inconsistency indices (RI) for different number of criteria

Steps to solve AHP can be summarized in the following diagram:

Figure 3.5: Diagram for converting qualitative indexes to quantitative value

Source: (Dang et al, 2011)