Modelling storm surge for the north central coast of vietnam

Conner and the other researchers 1957 has developed a simple model for calculating approximate peak coastal surges, then draw up the largest water-based monitoring pressure data observed

Sustainable Hydraulic Structures

Supervisor: A/Prof Dr Nghiem Tien Lam, Faculty of Marine and Coastal Enginering Thuy Loi University, Ha Noi, Viet Nam Co-supervisor: Prof Jean-Michel Hiver Université de Liège & Université Libre de Bruxelles

Ha Noi- 2016

Declaration

I hereby certify the work which is being presented in this thesis entitled, “Modeling storm surge for the North central coast of Vietnam” in partial fulfillment of the requirement for the award of the Master of Coastal Engineering Management and Sustainable Hydraulic Structures is an authentic record of my own work carried out under supervision of Ass Prof PhD Nghiem Tien Lam and Co-supervisor: Prof Jean-Michel Hiver The matter embodied in this thesis has not been submitted by me for the award of any other degree or diploma

Date: Aug 15, 2016

Tran Manh Linh

Acknowledgements

I would like to express my sincere gratitude to my advisor Ass Prof PhD Nghiem Tien Lam and Co-supervisor Prof Jean-Michel Hiver for their guidance, suggestion and inspiration

I would like to express my sincere thanks to University Board of Management, professors and lectures at Department of Marine and Coastal Engineering of Thuy Loi University and professors and lecturers of the Niche programme and Belgian-Vietnamese Master Programme for supporting me throughout my study progress

Finally, I would like to express my special appreciation to my friends and colleagues for their support, encourage and advices The deepest thanks are expressed to my family member for their unconditional loves

CONTENTS

CONTENTS iii

LIST OF FIGURES vi

LIST OF TABLES ix

ABSTRACT 1

1 Problem statement 1

2 Literature review 2

3 Study objectives 7

4 Approach and methodology 7

5 Thesis structure 9

CHAPTER 1 PHYSICAL SETTINGS 10

1.1 Geographic conditions 10

1.2 Topographic conditions 11

1.3 Climatic conditions 11

1.4 Observation stations and hydrologic conditions 12

1.5 Historical disasters of typhoons and storm surges 16

1.6 Conclusions 19

CHAPTER 2 MODELLING OF NON-TYPHOON CONDITION 20

2.1 Introduction 20

2.1.1 Introduction model MIKE 21 20

2.1.2 Module Mike21 FM 21

2.2 Model setup and boundary conditions 24

2.2.1 Basic impute data 25

2.2.2 The digitalized area 26

2.3 Model calibration 28

2.3.1 Boundary condition and initial condition 29

2.3.2 The model parameters 31

2.3.3 Result of model calibration 31

2.4 Model verification 33

2.4.1 The parameter for verification 34

2.4.2 The result of verification model 35

2.5 Model results 36

2.6 Conclusions 37

CHAPTER 3 MODELING OF STORM SURGE 39

3.1 Introduction 39

3.1.1 Typhoon model 39

3.1.2 General description 45

3.2 Model setup and boundary conditions 47

3.3 Model calibration 48

3.3.1 The input data 48

3.3.2 Result of calibration 51

3.4 Model verification 53

3.4.1 The result of verification 56

3.5 Model results 58

3.5.1 The result calculation of storm surge 58

3.5.2 Sensitivity analysis model 58

3.6 Conclusions 66

Conclusion 67

Recommendation 68

REFERENCES 69

APPENDIXES 71

LIST OF FIGURES

Figure 1-1 Administrative map North Central Coast of Vietnam 10

Figure 2-1 Introduction Mike 21 20

Figure 2-2 Digital elevation model of North central coast of Vietnam in Mike21 25

Figure 2-3 Simulated area 26

Figure 2-4 Location of simulated area 27

Figure 2-5 Hon Dau, Sam Son location on model 28

Figure 2-6 Diagram of calibration process 29

Figure 2-7 Water level in East condition 3/2007 30

Figure 2-8 Water level in North East condition 3/2007 30

Figure 2-9 Water level in South East condition 3/2007 30

Figure 2-10 Comparison of water level at Hon Dau observation station 32

Figure 2-11 Comparison of water level at Sam Son observation station 32

Figure 2-12 Water level in East condition 11/2007 33

Figure 2-13 Water level in North East condition 11/2007 33

Figure 2-14 Water level in South East condition 11/2007 34

Figure 2-15 The detail of Manning’s number 34

Figure 2-16 Comparison of water level in Hon Dau station 35

Figure 2-17 Comparison of water level in Sam Son station 36

Figure 2-18 The detail of Manning’s number 37

Figure 3-1 The track of the storm Damrey (2005) 45

Figure 3-2 Flood at coastal zone Nam Dinh in the Damrey typhoon (09/2005) 46

Figure 3-3 The track of the LEKIMA 2007 46

Figure 3-4 Water level in East boundary condition 9/2005 47

Figure 3-5 Water level in North East condition 9/2005 47

Figure 3-6 Water level in South East condition 9/2005 48

Figure 3-7 Pressure in Sep 26 2005 05:00 PM 49

Figure 3-8 Wind speed in Sep 26 2005 05:00 PM 50

Figure 3-9 Wind direction in Sep 26 2005 05:00 PM 50

Figure 3-10 Comparison of water level in Hon Dau station in 9/2005 52

Figure 3-11 Comparison of water level in Sam Son station in 9/2005 52

Figure 3-12 Pressure in Oct 03 2007 07:00 AM 54

Figure 3-13 Wind speed in Oct 03 2007 07:00 AM 54

Figure 3-14 Wind direction in Oct 03 2007 07:00 AM 55

Figure 3-15 Water level in East boundary condition 10/2007 55

Figure 3-16 Water level in North East boundary condition 10/2007 56

Figure 3-17 Water level in South East boundary condition 10/2007 56

Figure 3-18 Comparison of water level in Hon Dau station in 10/2007 57

Figure 3-19 Comparison of water level in Sam Son station 10/2007 57

Figure 3-20 The results of calculation storm surge in typhoon Damrey in 2005 58

Figure 3-21 The track of the typhoon Damrey (2005) 59

Figure 3-22 Storm surge at Hon Dau when change typhoon track 60

Figure 3-23 Storm surge at Sam Son when change typhoon track 60

Figure 3-24 Storm surge at Hon Dau station 62

Figure 3-25 Storm surge at Sam Son station 62

Figure 3-26 Water level at Hon Dau when change wind speed 63

Figure 3-27 Water level at Sam Son when change wind speed 63

Figure 3-28 Models in mike 21 tool box 64

Figure 3-29 Necessary parameters from the formulas which calculate wind field 64

Student: Tran Manh Linh viii Class: Figure 3-30 Water level at Hon Dau Station 65Figure 3-31 Water level at Sam Son station 65

LIST OF TABLES

Table 1.1 The water level corresponding to the frequency 12

Table 1.2 The water level corresponding to the cumulative frequency at Hon Dau 13

Table 1.3 Height, length, speed and largest wave period (19561985) 14

Table 1.4 Statistics storm in North Central Coast province from 1946 to 2015 17

Table 2.1 Boundary condition of model 27

Table 2.2 The coordinate of observation stations 28

Table 2.3 The result of error in observation station 33

Table 2.4 The result of error in observation station 36

Table 3.1 The parameter of storm Damrey 2005 48

Table 3.2 The result of error in observation stations in Storm Damrey 51

Table 3.3 The parameter of storm LEKIMA (2007) 53

Table 3.4 The result of error in observation station in Storm 56

ABSTRACT

1 Problem statement

Vietnam is located in the region where the world's largest storms in the Pacific North West, occur frequently This is one of the main types of disasters and dangers in Vietnam Annually, about 6 to 10 hurricanes or tropical storms hit the country, Storms and tide accompanied by heavy rain that causes coastal flooding In recent times, along with the effects of climate change, there are many strong storms and super typhoons in the world causing great damages to people and properties, such as Hurricane Katrina in the United States in 2005, Cyclone Nargis in Myanmar in 2008, Typhoon Bopha in the Philippines in 2012, In particular, the super typhoon Haiyan in 2013 was the most powerful storm which hit Philippines with strong winds up to more than Beaufort scale

17, caused storm surge up to 7m It killed 6000 people and seriously destroyed local infrastructure

About 80-90% of the population lives in the coastal plains in Vietnam which are directly affected by the storms In the trend of climate change and global warming, the flooding problem is becoming more serious in terms of quantity, frequency and intensity Moreover, the occurrence of the typhoons is unpredictable

Every year, hundreds of people are dead or missing, and thousands of people are affected

by storms and floods Damages to houses, crops and socio-economic are worth up to ten billions Coastal areas which are densely populated with developed economics, and tourism are very vulnerable and are of very high risk These areas are often threatened

by coastal erosion, dike breaching, flooding or salinity intrusion caused by storms

In the context of global climate change, many researchers have warned that the storms will occur more often, and more intensely with stronger winds and heavier rains and storm surge is greater than by sea level rise Among the 13 largest super typhoons in the world from 1935 to now, 6 new storm occurred since 1998, including 3 hurricanes occurring from 2010 to now in the East Sea including Haiyan (2013), Bopha (2012) and Magi (2010) When the air pressure in the center of storm reduces, large wind often causes of storm surges which can lead to inundating coastal regions In addition, storm

surges can result in big wave when the storm transfer into the land, destroying sea dikes and coastal works This then increases the seriousness of the storm

2 Literature review

a) International studies on storm surge

Globally, modeling sea level rise by storms is still being improved in the 21st century That the storms caused storm surges phenomenon has been studied in areas of the world with different names such as: storms in North America, hurricanes in the Gulf of Mexico and the East Coast of the United States; tropical storms in Europe; and typhoons in Asia and Oceania In the first half of the 20th century, scientists have studied storm surges with existing tools over time: from the simple empirical method based on sparse observation (Conner et al, 1957; Harris, 1959) to the derivatives which are more complicated analysis tools for storms and sea basin, but the value is limited (Proudman, 1954; Doodson, 1956; Heaps, 1965) In the US, Congress has directed the research institution of the army and the Weather Bureau conducted in-depth research on hurricanes and storm forecasting methods after the immense damage to the east coast in

1954 This is considered as the beginning of a systematic study of storm surges in North America (Murty, 1984) Before the developed model, the monogram was designed to predict surges whenever a typhoon landed on a beach Conner and the other researchers (1957) has developed a simple model for calculating approximate peak coastal surges, then draw up the largest water-based monitoring pressure data observed at the center of the storm to determine the best surges envelope

There are many methods and numerical models which have been used to quantify storm surges One of the earliest guidelines developed for this purpose is the hurricane scale Saffir / Simpson Hurricane scale Saffir / Simpson is oriented like a comprehensive guide for the emergency rescue agencies use in case of a storm However, this approach does not reflect the impact of localized factors such as bottom topography changes, shoreline shape, obstructions or other factors to high surges in other locations together when the storm enter the land

Harris (1959) have found a systematic variation of surges with two parameters: the minimum pressure at the center of the storm and 91.44 m distance from the center of the

storm to the shore, but concluded that the slope of the continental shelf does not greatly affect the water level rise (Harris, 1959; Harris, 1963)

Harris and Jelesnianski (1964) gave a two-dimensional hydrodynamic equations linearized to calculate hydronamic tide when fluid uniform pressure is calculated by the hydrostatic equation without the surface waves Jelesnianski (1966) has used a system

of linear motion equations without bottom friction to calculate and concluded that the effect of wind which is 4 times higher than air pressure to water level rise The largest peak water levels appeared to speed about 37 mph and the angle between the direction

of the storm moving and perpendicular to the shore line is 650

Jelesnianski (1972) has used the size of the storm as a parameter in his monogram The monogram uses the monitoring results and the model number of SPLASH (Special Program to List the Amplitude of surges from Hurricanes) Two versions of this model are SPLASH1, SPLASH 2 which are developed for designated areas in the Gulf and Atlantic coasts Although Splash models offer the reliable high water level, but the limitations of this model is only water level for coastal passages straight shoreline and topography with little variation Jelesnianski has concluded that the slope of the continental shelf is the main factor affecting water level rise He remains the wrong conclusion by Harris (1959) into the parameter "distance of 91.44 m of the line from the shore", so this factor does not resolve the variability of bottom topography Jelesnianski has mistakenly concluded like other studies from 1950 to 1970 that the impact of storm size to water level is relatively small However, recently analysis of historical data of the storm along with model simulations demonstrated that the size of the storm has played an important role in the generation of storm surges in coastal areas, particularly especially the super strong typhoon moved on very shallow waters (Irish et al, 2008)

In the years from 1950 to 1960, the water level calculation applies only to areas with a straight coastline with little terrain change They are the basis for the design of infrastructure to avoid the risk of being destroyed by storm surge until the 1970s and many are still in use today Since the late 1960s, high-speed computers allow differentiation of basic equations on the basis of calculation using structured mesh (usually 12 grid squares) in the 2D model (vertical integration) and then the 3D model

This capability eliminates the assumption of stable force balance in model (not suitable for the rapid transformation of the wind in the storm), while the simulation is a balance between water flux and the changes in water level in the basin In the 1980s, the model SLOSH (the Sea, Lake, and Overland surges from Hurricanes) and NOAA were set and Jeleśniański (1992) completed to calculate and provide maps assessing insurance by flooding FEMA (Federal Emergency Management agency) and the US weather forecasters still use to predict the water level rise This is a two-dimensional model which was developed to predict the instantaneous water level rise by the storms operating in the coastal waters of the Atlantic, but it is also applied in many parts of the world.b) Studies on storm surge in in Vietnam

In Vietnam, storm surge phenomenon has been studied since the 1970s of the last century Some authors and works featured include: Le Phuoc Trinh and Tran Ky (1969

- 1970), Nguyen Van Cu (1979), Pham Van Ninh (1982), Do Ngoc Quynh (1982), Vu Nhu Hoan (1988), Nguyen Ngoc Thuy (1989), Ta Dang Minh (1989), Le Trong Dao (1989), Bui Xuan Thong…

Do Ngoc Quynh and Pham Van Ninh (1995) in Project KT.03.06 had used many numerical models with different numerical algorithms and developed up 7 finite difference schemes including explicit schemes on staggered grids, semi-implicit central difference scheme with smoothing, implicit schemes on staggered grids, characteristic method, semi-implicit schemes on triangular grid and curvilinear grid These models have been investigated for their stability, convergency, and accuracy compared to some analytic solutions Their simulation speed and efficiency have been also investigated Finally the implicit finite difference scheme with ADI algorithem has been selected The result has provided an overall picture about the possibility of surges has and can happen for each latitude along the coast

Recently, some studies (Bui Hong Long, Nguyen Ky Phung Dinh Van Manh, Nguyen Tho Sao, Le Trong Dao .) have calculated storm surges using 3D models using 16 software either self-developed or commercial ones which numerically solve the basic system of equations in spherical coordinates horizontally and sigma stretch coordinate vertically

The distribution of wind and air pressure fields in the storm are very important factors

in study of storm surge The accuracy of the forecast wind and barometric fields play an important role in the precision of storm surge prediction Most previous studies have used empirical formulas to calculate wind and air pressure fields in storms However, the wind and pressure fields calculated by these formulas cannot reflect the influences

of the terrain and the effects of air pressure patterns outside the scope of the storm On the other hand the value of the largest wind speed and minimum pressure at the center

of the storm are often not achieved close to the actual value of the storm

Currently in Vietnam, the calculation of surges and flooding due to storm surges have been made through the mathematical model or software such as MIKE 21 Flood (Denmark), Sobek2D (Netherlands), etc…Some examples are the reseach project

“Assessment of risk due to sea level rise in Thua Thien-Hue province and development

of decision support system”, preformed by Institute of Mechanics in 2010, the Project KC09.23/06-10 “Assessment the variation of extreme sea water level due to climate change support marine development strategy” by University of Science (VNU), or the project “Estimating coastal flooding risk caused by surge in super typhoons” by Tran Thuc and collaborators carried out in 2014 These are usefull research results but still lacking the level of details as well as low map resolution which are not met the requirements on specific and details for disaster prevention when storms occur Therefore, more detail and more specific results of storm surge are nessesary as well as the specific recommendations for natural disaster prevention The research on storm surges and the development flood risk maps due to super typhoons are especially urgent

in the coastal province of Vietnam

 Method to calculate storm surge

- Statistical and field survey methods

The method is based on the statistical data of measured water level in the research area and data on the storm occured at the same time From the statistical data, the correlation relationships between storm data and water levels are determined for the study area In this method, field survey data is essential It is required to calibrate and verify the

accuracy of other methods However, this method has just been applied to each area of research and it also takes a lot of time to collect data

This is the traditional method to provide relatively positive results The result of this approach is very important It builds up a database catering for the integrated assessment

of water level in the research areas and to serve other methods

- Empirical formula methods

In this method, the height of storm surge above the average water level is calculated according to the wind speed and the average water depth determined based on long-term average water level with or without the tide The formula has the form:

∆𝑆 = 𝑘𝑊

2 𝑋 𝑐𝑜𝑠𝜑𝑔ℎ

where: ∆𝑆 (m): Storm surge;

k = 2.10-6: A coefficient, value can be adjusted;

This formula for the biggest surges is highly probable because the parameters are defined for stable conditions Therefore, this formula is used to design because the result

is often greater to ensure the safety of the work

3 Study objectives

The main objective of this research is to develop a numerical model and to simulate storm surges for North Central Coast of Vietnam The model will be used for severing the predicting storm surge when typhoon occur in the future

4 Approach and methodology

The study objectives can be achieved by using the numerical modeling approach with following steps:

 Collecting basic data including topographic data, bathymetric data, water level, wind speed, wind direction, atmospheric pressure and typhoon tracks The wind data, atmospheric pressure and water level data during typhoon periods can be collected from Vietnam Hydro-Meteorological Services (HMS) which are necessary for model calibration and validation The typhoon tracks can be downloaded from Japan Meteorological Agency (JMA), Typhoon Warning Center (JTWC) and the National Climatic Data Center (NCDC) These data include information on time, location of typhoon eye, maximum wind speed, minimum central pressure, and the moving speed

of typhoon centre

 Setting up a 2D hydrodynamic model to simulate water level for non-typhoon condition The model will be calibrated and validated for tidal-only conditions using observed water level

 Setting up a 2D hydrodynamic model to simulate storm surge The model will be calibrated and validated for typhoon conditions using observed water level

Collect data (Winds, Tide, Bathymetry,…)

Setup Model (Winds, Tide, Bathymetry,…)

Simulation storm by model (Calibrate and validate)

Results of the model

Analysis and Discussion

Conclusions

CHAPTER 1 PHYSICAL SETTINGS

1.1 Geographic conditions

The North Central coasl is located in the northern part of Central Vietnam, starting at south Tam Diep Mountains to northern Hai Van Pass North Central has an area of approximately 51 552 km² including 6 provinces: Thanh Hoa, Nghe An, Ha Tinh, Quang Binh, Quang Tri and Thua Thien-Hue.The northem North Central borders with northem moutains and the Red River Delta, the west borders with Laos, to the south by the South Central Coast, and the east is the East Sea

Figure 1-1 Administrative map North Central Coast of Vietnam

North Central Region is the transitional zone between the northern economic region and the southern economic region The narrowest region is Quang Binh (50km), located on the trans-Vietnam's traffic which has facilitated economic exchanges between the North and the South To the west is the East Truong Son, bordering with Laos with 1,294 km long border with the gate of Quan Hoa, Lang Chanh (Thanh Hoa), Ky Son (Nghe An),

Huong Son (Ha Tinh province), Lao Bao (Quang Tri), faciliating economic exchanges with Laos and other Southeast Asian countries on the continent Towards the East Sea

is the coastal route about 700 km, with lots of seafood and many deep-water port that can form the seaport North Central is a region with its strategic location that is very important in fighting, national defense, and protection of the territorial integrity of Vietnam

a narrow plain, and the western final is the midland mountains of northern Truong Son mountain sierra Overall the terrain of North Central is complex because the majority of the territory is mountains and hills, overlooking the sea, slope, water rapids, which often cause sudden floods, leading to difficulties for production and people's life

1.3 Climatic conditions

North Central is located in the tropical monsoon area, so the climate is influenced by the transition of the North and South During the year there are two seasons: The rainy with high temperature from April to October, and the cold dry seasons from November to April of the next year In the rainy season there are many typhoons, high rainfall (over

200 mm), and the summer is influenced by the southwest wind blowing from Laos which lead to large evaporation and severe droughts

North Central is the region where the climate is the harshest in Vietnam Annually, there are more frequent natural disasters such as storms, floods, wind Laos, drought, but the underlying cause is due to the location and the terrain structure North Truong Son mountain range created walls that stop the winds entering to the North Central:

- Blocking the southwest monsoon, causing southwest wind that is hot and dry, blowing down the coastal plain in the early rainy season (around May to July)

- Blocking the Northeasterly wind, the humid air masses from the ocean (due to storm activity tropical convergence zone), causing heavy rain in many places

The complexity of the climate leads to a large difference between the temperature of the winter and summer The average temperature is about 24.25oC Climatic conditions of the region make it difficult for business activities, especially agricultural production

1.4 Observation stations and hydrologic conditions

a) Hon Dau Station

- Tidal

The water level at Hon Dau is diurnal tide There are about 25 days per month in which

1 time the water level is high and 1 time is low tide The tidal range at Hon Dau Station

is big, the great tidal range about 3m to 4m At high tide the water level which is highest measured reaches 421cm (22/10/1985), the lowest level is -3cm (02/01/1991)

Base on the highest water level data from 1974 to 2004, calculating the highest water level corresponding to the frequency as the Table 1.1

Table 1.1 The water level corresponding to the frequency

(altitude chart system)

Depending on the water level, crest of tidal, bottom of tidal, average cumulative frequency calculation gives result of the water level frequency (Table 1.2)

Table 1.2 The water level corresponding to the cumulative frequency at Hon Dau

(altitude chart system)

The current of coastal areas and estuaries Cat Hai have complex mode that shown interactions river - bottom topography - waves - tides

The influence of wind and wave regime in the winter to Cat Hai coastal areas is not large The current regime results from topography factor and fluctuating water levels (by tides) In the winter, when waves and wind is relatively quiet, the current appears mainly due to the tidal current and the deference of the water level due to water block

of tidal wave is cornered when entering the shore However, in the high tide days when waves and wind have east or southeast direction, coastal currents caused by the waves associated with different types of flows to increase (or decrease) in aggregate flow velocity of coastal regions

- Waves

Wave mode at Hon Dau specific trait of the waves in the Gulf regimes Hai Phong Every year, high altitude wave forces on May to September, the highest wave in July and September In winter, the wave height is not high because it is covered by Cat Ba Island The highest waves which are observed in this season appear only in the direction South, Southeast

Factors extreme waves are observed on the July 03 1964:

Year 1964 1983 1974 1959 1957 1961 1964 1962 1962 1917 1983 1971 1964 Period (s) 9,1 7,6 7,5 9,3 9,3 8,2 11 8 7,7 68 6,7 7,1 11

there are 2 tide cycles, each of which lasts for 14 days) Every month there is a tidal cycle up and down for about 25 days and one day also for a top tide and a bottom tide Average tidal height in high tide period is about 1.2 to 1.5m Tidal current speed around Sam Son Station is quite large

- Wave

Thanh Hoa coast is relatively straight towards the northeast to the southwest / south Without, external shielding island, the coast has low slope, which leads to waves with long momentum entering the coastal without any obstruction causing strong impact

north-on the coastal or shore protectinorth-on works

In the winter, from September until March next year, the waves are very rough with waves with a height from 0.8m to 1m and maximum up to 2.0m to 2.5m Wave period

is from 7s to 10s Prevailing wave direction is northeast to the shoreline which creates a variable angle from 300 - 450, the frequency is 40% In October and November wave combine high tide threatens beaches and shore protection works

In summer, from May to October, there are very few days high waves But in the summer, severe storms often occur causing significant damage to the coast and shore protection works The average wave height is from 0.6 - 0.7m, maximum up to 3.0 - 3.5m From June to August the prevailing wave southwest direction and wave height 0.6 - 0.7m Most of the summer storms hitting the coast of northern and central provinces are affecting the coast in North Central coast, accompanying typhoon is storm surges and swells phenomena When a big storm hit the wave height may reach about 6m

- Currents

In the Gulf of Tonkin, the cold water runs to the east after that combining with warm water flow back to the north They make a reverse cycle clockwise Due to the circulation of the Gulf of Tonkin the seas area around Sam Son station is affected by the cold water flow towards southwest and south

In cold season: Northeast monsoon and the transitional period the general circulation of the Gulf of Tonkin in general and in particular of North Central coastal, coastal currents

direction north - south (along the North Central coast to the south) Ocean intensity in this season is enhanced by control of the northeast monsoon, depending on the strength

or weakness, continuous or interrupted, which leads to the intensity of ocean rise or fall

In hot season: Due to the influence of the southwest wind, coastal currents have opposite direction compared to the cold season, but the intensity is lower February to March, North Central often have whirlpool and are concentrated in the north, but in July this phenomenon going to down the south

1.5 Historical disasters of typhoons and storm surges

In recent times, under the effects of climate change, there have been many strong storms such as Hurricane Katrina hit the United States in 2005, Cyclone Nargis hit Myanmar

in 2008, typhoon Bopha hit the Philippines in 2012, hurricanes, causing great losses of lives and property In particular, the super typhoon Haiyan in 2013 was the most powerful storm which hit the Philippines with strong winds on level 17 and the water height up to 7m The storm killed more than 6,000 people and seriously destroyed infrastructure

Vietnam has more than 3600km coastal line, and the country is located in the zone which

is directly affected by typhoons that are formed in the Western Pacific and East Sea According to the data of Disaster Management Unit of project UNDP VIE/97/002 at Vietnam, storms begin on early June on the North and are going down to the South in December The number of typhoons also decreases from the North to the South

From 1954 to now, there are about 30 storms occurring in the Western Pacific each year, including about 10 storms arising in the East Sea and 4 or 6 typhoons directly affecting Vietnam There are some years when the hurricane hits Vietnam with more than 10 storms such as: 1964 (18 storms), 1973 (12 storms), 1978 (12 storms), 1989 (10 storms) and 1996 (10 storms) Typically, the storm hit northern and central Vietnam, the typhoon hit the southern is less

Hurricane in coastal estuary area often causes high waves, strong winds, and surges such as breaking embankments, homes, bed… In the past, the Damrey storm with storm surge more than 2m in Nam Dinh caused significant damage to the dike system and the

lives and livelihoods in the region In the area study and the next area landfall density is quite high compared to other coastal provinces

According to the Table 1.4, from 1946 to 2015, there are 69 typhoons hitting to North Central coastal It is the region with greater frequency of storms with the rate on 30% of the storm, focusing primarily on the provinces of Thanh Hoa, Nghe An, Ha Tinh Table 1.4 Statistics storm in North Central Coast province from 1946 to 2015

No Name of storm Type Date

1.6 Conclusions

With the tremendous harm to human and material caused by storm surge, forecasting surges become an important task of the meteorological sector, particularly to areas which have greater frequency of storms like North Central Vietnam The research

“Modeling storm surge for the North Central Coast of Vietnam” will be carried out to make a numerical modeling to calculate the height of storm surge at the central coast of Vietnam which is often affected negatively by storms The results will be used to serve the expectance in cases of response to disaster The question is how to prevent to minimize adverse impacts by storm surge Moreover, the issue of enhancing the accuracy of forecast models has always been of interest and research

In many methods and mathematical models, Mike 21 Flow Model FM model is a model

of modern, dedicated, result in rising water features stability and accuracy This is the pattern of trade and have been used in many research projects in Vietnam and the world With the help of the project "Building up flood map due to storm surge in super typhoon-2016" this research have used this software

CHAPTER 2 MODELLING OF NON-TYPHOON CONDITION

2.1 Introduction

2.1.1 Introduction model MIKE 21

Vietnam has about 1.000.000 km2 of sea surface, 3260 km of coastline; it is an opportunity for developing economy in coastal zones According to stategy to 2020, Vietnam is a sea country, and coastal activities will account for 54 per sent of GDP Beside the opportunities, Vietnam has to withstand several typhoons which lead to breaking sea dykes, floor and economic damage To prevent and reduce hazard due to typhoon, studying and choosing solution to protect for each area that become more important and essential

From wind field transferring on the sea, we can know which areas are affected from storm surge the most; hence, there will be solution to protect or strengthen the area So calculating storm surge in the area by 2D model is a opimun method

Figure 2-1 Introduction Mike 21 Mike software is software package of DHI Water & Enviroment applyied hydraulic, sources and enviromental simulation, including river, river mount, sea and coastal zone The software package has been utilized rather effectively in different countries

In Mike 21 model package, there are modules, such as Hydrodynamic (HD); Transport (TR); ECO Lab (EL); Mud Transport (MT); Sand Transport (ST , and the master research mainly applies the module: Module Flow Model FM

In all module of Mike 21, Hydrodynamic HD is basic module It provides the regime of hydrodynamic that is foundation for calculation process of other modules

2.1.2 Module Mike21 FM

Mike 21 FM which is new model system in untilizing flexible gird node, is developed

by DHI Water & Enviroment The model system is developed for studying about sea and envirironment of coastal zone The model consists of continunity equation, moment equation, concentration equation, salt equation

The Mike 21 consists of all modules as:

 Hydrodynamic Module

 Sediment transport module

 Ecological Module

 Observation point module

Hydrodynamic Module is the basic ingredient of mike 21FM system model, providing basic hydraulic mode for area calculation

General description

2D Shallow water equation of Basic hydrodanamic module is a combination of the average of Navier- Stoke and Renold coefficient It consists of continuity, temperature, satlt, concentration and momentum equation

Storm surge can be presented by the 2DH hydrodynamic equations for long waves in shallow water It includes the continuity equation presenting the mass conservation law and momentum equations for the x- and y-directions presenting the momentum conservation law

The continuity equation:

The momentum equations

x, y Cartesian co-ordinates in the horizontal directions (m)

η water surface elevation above the reference level (datum) (m)

h total water depth, h = d + η (m)

d water depth below the reference level (datum) (m)

u depth-averaged flow velocity in x direction (m/s)

v depth-averaged flow velocity in y direction (m/s)

q rate of inflow (m³/s/m²)

f Coriolis parameter (inertial frequency) (1/s)

g acceleration due to gravity (m/s²)

ρa air density (kg/m³)

ρ water density (kg/m³)

Px,Py gradient of air pressure in x- and y-directions (kg/m²/s²)

Fx,Fy friction components in x- and y-directions (m/s²)

τwx,τwy wind shear stresses in x- and y-directions (kg/m/s²)

Swx,Swy radian stresses in x- and y-directions (m/s²)

τix,τiy internal stress components (m/s²)

The momentum equations include terms presenting the inertial term (1), convective terms (2,3), Coriolis term (4), gravity term (5), pressure term (6), friction term (7), free surface stress term (8), short wave influence (9), and internal stress term (10)

The pressure term (6) accounts for the influence of atmospheric pressure

a x

a y

p

p P

Where pa is the atmospheric pressure (N/m²) The distribution of atmospheric pressure field in a typhoon can be presented by the typhoon air pressure models of Bierknes (1921), Takahashi (1939), Fujita (1952), Mayers (1954) and Jelesnianski (1965)

The friction term accounts for the bottom friction (7)

2

x y

Cd wind drag coefficient (dimensionless)

V the wind speed at 10 m above the surface, 2 2

V  V V (m/s)

Vx,Vy wind speed components in x- and y-directions (m/s)

The variation of wind field in a typhoon can be presented by the typhoon wind models such as the Modified Rankine vortex model (1947), Holland (1980), DeMaria (1992), Fujita (Tan, 1992), and SLOSH (Jelesnianski et al., 1992; Houston and Powell, 1994)

The short wave influence (9)

1

xy xx

wx

S S

where ε is the horizontal eddy diffusion (m²/s)

The system of equations (1) to (3) can be solved numerically by a finite difference method (FDM), a finite element method (FEM) or a finite volume method (FVM) Boundary conditions are required for the model Model boundary locations should be selected far enough from the interested area at the places where there is almost no storm surge influence Therefore, only tidal levels can be used as model boundary conditions The tidal level can be computed based on tidal harmonic constants Together with parameters of the models presenting typhoon pressure and wind fields, the above coefficients of Chézy roughness coefficient (C), wind drag coefficient (Cd), eddy diffusion coefficient (ε) are model parameters and can be changed during model calibration process

There are many modeling software available that can be used to solve these above equations including sophisticated commercial models like Delft3D, MIKE 21, SMS,

…The advantages of these models are stable, reliable, and easy to use because they are equipped with graphical user interface (GUI) and pre- and post-processing software The disadvantages of the models are that they are not always available, expensive and are not allowed to change the computational code Beside of these, there are also many open source modeling codes such as POM, ECOMSED, HAMSOM These models are free and allowed to modify computational code if necessary

2.2 Model setup and boundary conditions

In regard with rule of application, each model has to follow 3steps:

- Setup model: in this step, the model is set up, simulate real condition with boundary in a specific time to find out the parameters for the model

- Verification: assess the realization of the model

- Using the model to simulate different scenarios for the system

The basic data for calculation

2.2.1 Basic impute data

a Topography

The topography used in the research is surveyed under the project name ”Building up flood map due to storm surge in super typhoon-2016” with map scale 1/10000 It is displayed in the study area as Figure 2-2

Figure 2-2 Digital elevation model of North central coast of Vietnam in Mike21

In the North central coast of Vietnam, the topography has relatively shallow continental shelf The coastline has complex geographic features such as estuaries, lowland and mountains In general, the shoreline is rather flat and the slope is gentle The depth contour line of 10 m and 50 m is far from the shore and these characteristics support the development of high storm surge The topographic depth increases from north to south

b The oceanographic data

- Boundary condition

- Observation station is the station that is collected in the project “Building up

flood map due to storm surge in super typhoon-2016”

2.2.2 The digitalized area

a The simulated area.

Based on tophography and other documents as the simulated area of North central coast

of Vienam is selected as Figure 2.2

Figure 2-3 Simulated area

Domain properties, grid computing: Is set for North central coast, in the model coordinates are converted on the WGS 1984 UTM Zone 48N Grid computing in the near shore was subdivided small to increase accuracy when calculating and gradually increase from shore to offshore to reduce computing time which does not influence muchon the accuracy of the model

b Boundary conditions

The boundary of the model include 3 tidal boundaries are tide Northeast, East, Southeast and 1 shoreline that is shown on Figure 2-4 Figure 2-4 The water level data of the boundaries are generated from the module tool box Tidal mike 21- Tide prediction of Heights

Table 2.1 Boundary condition of model

of points Longitude Latitude Longitude Latitude

c Observation station

Due to the real data collection, the study selects 2 stations for the calibration and verification model The water data of 2 stations are got from the project “Building up flood map due to storm surge in super typhoon-2016” The coordinate of station shows

on Table 2.2 and station’s location on the map shows on Figure 2-5

Table 2.2 The coordinate of observation stations

calibration process that is refered to collecting data, and assessing realization over data

of station by gradual testing method

Figure 2-6 Diagram of calibration process The calibration process follows 3 steps as:

Step 1: Assmussing parameters, initial condition

Step 2: Running the model with the relevant parameters

step 3: Comparing the result of model with real data If not come back step 1

Comparing is excuted by visualation( Comparing mode data and read date),simultaneously combine with Nash coefficient to check

In which: X0,I: is observed data

Xs,i: is simulated model data

0

X : is the average of real data

2.3.1 Boundary condition and initial condition

The boundary conditions of the hydrodynamic model are the agent which determines the motion of the sea In land boundary, the velocity components obtained zero The model allows the use of different boundary conditions at the liquid boundary including water level fluctuations or velocity and river traffic from flowing into the study area On

the surface is distributed in space and time of the wind, the atmospheric pressure There are 3 liquid boundaries in model: East, Northeast, and Southeast Tide oscillations in the liquid boundary are created with module Mike 21 tool box and put on the model by file

*.dfs1 Flow speed and water level were equal to 0 at the beginning

Figure 2-7 Water level in East condition 3/2007

Figure 2-8 Water level in North East condition 3/2007

Figure 2-9 Water level in South East condition 3/2007

Time Time Time