modeling storm surge and flood mapping for kien thuy and do son district, hai phong city

Scope of the study: This study only determined total water level and storm surge caused by wind field, atmospheric field, tide but not water level rise in the rivers and effect of wave..

THUY LOI UNIVERSITY

PHAN TUAN ANH

MODELING STORM SURGE AND FLOOD MAPPING

FOR KIEN THUY AND DO SON DISTRICT,

HAI PHONG CITY

THESIS OF MASTER DEGREE

Ha Noi, 2016

MINISTRY OF EDUCATION

AND TRAINING

MINISTRY OF ARGRICULTURE AND RURAL DEVELOPMENT

THUY LOI UNIVERSITY

PHAN TUAN ANH

MODELING STORM SURGE AND FLOOD MAPPING

FOR KIEN THUY AND DO SON DISTRICT,

HAI PHONG CITY

MINISTRY OF EDUCATION

AND TRAINING

MINISTRY OF ARGRICULTURE AND RURAL DEVELOPMENT

Major: Coastal Engineering and Management Majors code: 62580203

SUPERVISORS: 1 Assoc.Prof.Dr Tran Thanh Tung

2 Dr Le Tuan Hai

Declaration

I declare that I have developed and written the enclosed Master Thesis completely by myself, and have not used sources or means without declaration in the text Any thoughts from others or literal quotations are clearly marked The Master Thesis was not used in the same or in a similar version to achieve an academic grading or is being published elsewhere

Ha Noi, Aug 17, 2016

Author

Phan Tuan Anh

Preface

I first thank my parents and my brother for their support throughout not only the past two years, but throughout my entire life I am truly grateful for all the opportunities and support provided by my family at every juncture of my life

After 20 weeks with my efforts and help of my teachers, colleagues, I have completed

my thesis with the subject “MODELING STORM SURGE AND FLOOD MAPPING FOR KIEN THUY AND DO SON DISTRICT, HAI PHONG CITY”

I am thankful for my advisors Assoc Prof Dr Tran Thanh Tung and Dr Le Tuan Hai This thesis would never have reached completion without their guidance and assistance Assoc Prof Dr Tran Thanh Tung provided insightful edits and innovative approaches that helped shaped parts of this thesis Finally, Dr Le Tuan Hai provided guidance, especially in Mike model, that ensured the result of calculation was comprehensive and accurate

I am forever grateful to the Faculty of Marine and Coastal Engineering and Thuy loi University gave me a meaning course I had a lot of knowledge and reality experiences

TABLE OF CONTENT

INTRODUCTION 1

1 Necessity of research 1

2 Research objects and Scope 3

3 Research content 3

4 Approach and methodology 3

CHAPTER 1: OVERVIEWS OF STUDY OF STORM SURGE 5

1.1 Overview of study in the world 5

1.2 Overview of study in Viet Nam and Hai Phong 7

1.3 Overview of research methodology and tools 10

1.4.1 Geographical location, administrative boundaries 12

1.4.2 Topographic characteristics 15

1.4.3 Geological and soil characteristics 16

1.4.4 Meteorology and climate characteristics 17

1.4.4.1 Thermal mode 17

1.4.4.2 Relative Humidity 17

1.4.4.3 Piche Evaporation 18

1.4.4.4 Wind 18

1.4.4.5 Storm and tropical depression 19

1.4.5 Economical and social characteristics 20

CHAPTER 2: SETUP NUMERICAL MODEL TO SIMULATE STORM SURGE FOR STUDY AREA 22

2.1 Data collection and Analysis 22

2.1.1 Topographical data 22

2.1.2 Hydrodynamic data 23

2.1.3 Storm data 24

2.2 Introduction of Mike Model 31

2.2.1 General 31

2.2.2 Theory of Mike 21 model 32

2.2.3 Theoretical basis of MIKE 21 Toolbox to calculate tide 33

2.2.4 Theoretical basis of MIKE 21 Toolbox to calculate atmospheric field and

wind forcing 35

2.3 Setup Model 36

2.3.1 Domain 36

2.3.2 Mesh 38

2.3.3 Bathymetry data 39

2.3.4 Boundary 40

2.3.5 Setting parameters of big model 41

2.4 Calibration and verification of simulation model 43

2.4.2 Verification of storm surge model 45

CHAPTER 3: SIMULATING STORM SURGE 50

3.1 Develop scenarios 50

3.4 Develop calculated station 54

3.5 Result of storm surge simulation 56

3.5.1 Scenario 1 56

3.5.2 Scenario 2 57

3.5.3 Scenario 3 61

3.6 Analysis results of simulation model 64

CHAPTER 4: FLOODING SIMULATION AND DEVELOP FLOOD MAPPING DUE TO STORM SURGE OF KIEN THUY AND DO SON DISTRICT 65

4.1 Setup flood model 65

4.1.1 Domain and mesh 65

4.1.2 Bathymetry data 66

4.1.3 Input boundaries 67

4.1.4 Infrastructure data 69

4.1.5 Setup parameters of flood model 70

4.2 Results of flood model 71

4.3 Building flood mapping 73

4.3.1 Appying GIS software to build flood mapping 73

4.3.2 Input data 74

4.3.2.1 Flood depth 74

4.3.2.2 Administrative map 77

4.4 Results of flood mapping 78

CONCLUSION AND RECOMMENDATION 82

REFERENCES 85

APPENDIX 89

LIST OF FIGURES

Fig 1.1 Administrative map of Hai Phong city 13

Fig 1.2 Sea dikes 2 is protecting populated areas 15

Fig 1.3 Do Son beach 21

Fig 1.4 Exploiting seafood 21

Fig 2 1 Hai Phong city map scale 1:50000 22

Fig 2.2 Digital elevation model DEM (30x30m) of study area 23

Fig 2.3 Distribution of storm’s direction landfall in Hai Phong coast 30

Fig 2 4 Module Mike 21FM of Mike Zero 32

Fig 2.5 Domain of big model cover small model 37

Fig 2.6 Domain of small model 37

Fig 2.7 Mesh of big model 38

Fig 2.8 Mesh of small model 39

Fig 2.9 Bathymetry of big model 39

Fig 2.10 Bathymetry of small model 40

Fig 2.11 Boundary conditions of big model (03 water boundaries, 01 land boundary) 40 Fig 2.12 Water boundary in North East – Code 2 of big model 41

Fig 2.13 Comparision of water level in calibration of tidal model 43

Fig 2.14 Comparision of water level in verification of tidal model 44

Fig 2.15 Tracking of the typhoon Kalmaegi on 09/2014 46

Fig 2.16 Wind forcing data of storm surge model 48

Fig 2.17 Comparision of water level in verification of storm surge model 48

Fig 3.1 Track of the typhoon Sarah in 1977 51

Fig 3.2 Setup storm surge model (big model) of scenario 1 53

Fig 3.3 Wind forcing of scenario 1 54

Fig 3.4 Location map of calculated stations 55

Fig 3.5 Water level map of scenario 1 56

Fig 3.6 Comparison of total water level of scenario 1 56

Fig 3.7 Water level of tidal model 57

Fig 3.8 Wind forcing data of storm surge model in scenario 2 59

Fig 3.9 Water level map of scenario 2 60

Fig 3.10 Comparision of total water level (storm surge + tide) of scenario 2 60

Fig 3.11 Wind forcing data of storm surge model in scenario 3 62

Fig 3.12 Water level map of scenario 3 63

Fig 3.13 Comparison of total water level (storm surge + tide) of scenario 3 63

Fig 3.14 Compare highest water level of storm surge model in 03 scenarios 64

Fig 4.1 Domain of flood model (small model) 65

Fig 4.2 Mesh Generator in Flood model (small model) 66

Fig 4.3 Bathymetry data of flood model 67

Fig 4.4 Tool for extracting results of big model to boundaries of small model 67

Fig 4.5 Boundaries of flood model 68

Fig 4.6 One of boundaries (code 2) of flood model 68

Fig 4.7 Import characteristic of infrastructure to flood model 69

Fig 4.8 After import infrastructures (sea dike1, sea dike 2, express ways…) 69

Fig 4.9 Result of flood simulation in scenario 1 71

Fig 4.10 Result of flood simulation in scenario 2 72

Fig 4.11 Result of flood model in scenario 3 73

Fig 4.12 Methodology for building flood mapping 74

Fig 4.13 Total water depth of flood model 75

Fig 4.14 After eliminate unsubmerged area of study area 75

Fig 4.15 Classification of flood depth 76

Fig 4.16 Administrative map of study area 77

Fig 4.17 Flood mapping of study area in scenario 1 78

Fig 4.18 Flood mapping of study area in scenario 2 79

Fig 4.19 Flood mapping of study area in scenario 3 80

LIST OF TABLES

Table 1.1 Typical monthly and annual average temperature (unit oC) 17

Table 1.2 Typical monthly and annual average relative humidity (Unit: %) 17

Table 1 3 Typical monthly and annual average evaporation (Unit: %) 18

Table 1 4 Average monthly and annual wind speed 19

Table 2.1 List of typhoons affected to Hai Phong coast from 1952 to 2015 24

Table 2 2 Details of some strong typhoon landfall on Hai Phong coast 26

Table 2.3 Tidal components 33

Table 2.4 Parameters of big model 41

Table 2.5 Track of the typhoon Kalmaegi 45

Table 2.6 Parameters of the typhoon Kalmaegi 47

Table 2.7 Result of calculation of the Nash Sutcliffe index 49

Table 3.1 Classification of flooding zone of Ministry of Natural Resources & Environment 50

Table 3.2 Scenarios of storm surge simulation 50

Table 3.3 Tracking information of the typhoon Sarah 51

Table 3.4 Parameters of the typhoon Sarah 52

Table 3.5 Location of 06 calculated stations 55

Table 3.6 Highest water level of 06 stations in scenario 1 57

Table 3.7 Higher high water of tide 58

Table 3.8 Parameters of the supposal typhoon in scenario 2 58

Table 3.9 Result of storm surge model of scenario 2 61

Table 3.10 Parameters of the supposal typhoon in scenario 3 61

Table 3.11 Result of storm surge model of scenario 3 64

Table 4.1 Parameters of flood 70

Table 4.2 Flood depth is presented by color 76

Table 4.3 Statistics of flooded area in scenario 1 81

Table 4.4 Statistics of flooded area in scenario 2 81

Table 4.5 Statistics of flooded area in scenario 3 81

1 Necessity of research

Along with increase of greenhouse effect, pollution due to climate change cause of global warming, melting ice and rising sea levels are increasingly more and more complex developments in the world According to the fifth Assessment Report of the IPCC on the status of current climate change, sea level rise will not be uniform across regions By the end of 21st century, it is very likely that sea level will rise more approximately 95% of the ocean Sea level rise depends on quantity of CO2 not only total cumulative; emissions total accumulated together and sooner rather than later due

to a larger mitigation of sea level rise About 70% of worldwide coast are projected to experience sea level change within ± 20% the global average [1]

Scientific studies show that increased storms due to climate change include two key elements The first one is the sea surges due to thermal expansion and ice melting constantly A recent study showed that sea level rise could reach 1 meter or more in this century The second, a warmer ocean increases the likelihood of a tornado caused

of increasing high storm surge Because of climate change rise, storm surge flooding will facilitate more damage in adjacent coastal areas and low-lying areas Large storm surges greater threat of destruction in the future because they will move inland, the threat level larger area than before In addition, major storms can cause burst dikes, severely affecting coastal works, with coastal cities increased, people's lives will be affected more World history has witnessed the typhoon Sidr which attacks Bangladesh coast in November 2007, killed more than 3,000 people, injured more than 50,000 people, damaged or destroyed more than 1.5 million households, and affected

to life of more than 7 million people [2]

Vietnam is a country located on the west coast seaside of the East China Sea, with geopolitical and geoeconomic very important not only to Viet Nam but also other countries There are over 3,260 km coastline stretching from North to South in Viet Nam, so Viet Nam have ranks 27th among 157 coastal countries, island nations and territories of the world Coastal line of Viet Nam have about 3,000 islands and two

offshore Hoang Sa and Truong Sa islands equally distributed along the country's coast, some special positions are similare to important line for defending outpost on the west

of the country; some coastal islands also have an important position to be used as national landmarks in the sea to establish baselines for the coastal Vietnam Three-quarters of Viet Nam is mountainous with elevations ranging from 100 m to 3400 m, while delta area includes two large river deltas; the Red River Delta in the north and the Mekong Delta in the south Lowlands in Viet Nam are extremely fertile and populous and most of the agricultural sector and industry of Vietnam are concentrated there In 63 provinces and cities of the country, 28 provinces and cities have coastal line and nearly half of population is living in the coastal province and coastal cities

Storm surge is rise of the water level with short dephan pressure and stress field storm

on the sea [3] Storm surge occurs with high tide is the cause of the loss of human life and property in hurricane region and surrounding areas During the setdown, the high cross-shore velocity may give rise of shore erosion Storm surge/setdown is very dangerous in natural hazards, but one pays much attention to the surge because it affects on the hydrostructures [4] According to the report of UNDP (2008) which predicts Viet Nam to be one of the most disaster-prone country special in the climate change status [5] The report of Vietnam Central Committee for Flood and Storm Control (CCFSC, 2005) showed that in the period of 1990-2010, Viet Nam experienced 74 floods [6] Typhoons and floods always cause serious consequences For example, the typhoon Damrey had landfall in Viet Nam include Hai Phong city on

27 September 2015, it is considered the most severe storm to hit Viet Nam in the last

50 years Damrey affected all coastal provinces of the Red River delta region with level 14 of wind forces in the eye of the storm High storm surges coincident with high tides led to extensive overtopping ofsea dikes in the area Storm surges from Damrey reached a height of three to four metersand the seawater penetrated inland three to four kilometers The typhoon Damrey destroyed at least 1194 houses and damaged another 11,576 More than 130,000 hectares of rice fields were submerged and damaged, most

of which had not been harvested prior to the typhoon Damrey [7] Although it is difficult to associate singular events like the damaging typhoons of 2005 with climate

change, events like these clearly demonstrate the extreme vulnerability of the region to exaggerated climate variability

Climate change and storm surge are changing complex and their consequences caused increasingly unpredictable results Researching and forecasting storm surge as well as flooding scenarios in the future to get suitable migration plans which should reduce losses caused by natural disasters and make adaptation with climate change

2 Research objects and Scope

Subjects research: Simulated storm surge and develop inundated map in different scenarios for Kien Thuy and Do Son District, Hai Phong city

Scope of the study: This study only determined total water level and storm surge caused by wind field, atmospheric field, tide but not water level rise in the rivers and effect of wave

3 Research content

With the climate change and sea level rise, all hydro-meteorological parameters are dramatically changed, specially for typhoons As the results announced by Ministry of Natural and Environment (MoNRE), typhoon’s intensity can be risen up to 16 order (Beaufort scale) with wind velocity reaches 180-200 km/hr Also from announcement

of MoNRE, the storm surge in super typhoons at the coast are up to 5 to 6m above mean sea level With very high storm surges in super typhoons, sea dikes may be broken and hinter land will be inundated The simulation can assess the inundated level and propose the solutions to mitigate damages caused by these phenomena

With above reasons, the study is very important and really having scientific and practical significances It is also the requirement of Viet Nam Government

4 Approach and methodology

- Research scientific literatures in the world and Viet Nam about storm surge and related literatures;

- Data collection: all required data such as topography, mangrove forest, water level, waves and collected and extra investigation for the study;

- The MIKE21FM is applied to simulate totally water level and storm surges is subtracted for the south central coasts of Viet Nam with different scenarios Results

will be a basis for making strategy to mitigate effect of storm surge for coastal areas in Viet Nam It is absolutely good and modernized tools for the study

- GIS is coupled with simulated results to make inundated maps for different scenarios

METHODOLOGY

ANALYSE DATA

DEVELOP FLOOD MAPPING

CONLUSIONS AND RECOMMENDATIONS

(Winds, waves, tide, water depth, storm…)

SETUP TIDAL AND STORM SURGE MODELING FOR BIG AREA

SETUP STORM SURGE (BIG MODEL)

CHAPTER 1: OVERVIEWS OF STUDY OF STORM SURGE

1.1 Overview of study in the world

A researcher once said "model the phenomenon of water surges is an art" [8] Storm surge modeling is still being improved in the 21st century and worldwide The storm caused water surges were studied in these regions of the world with different names such as: in North America, storms appear in the Mexican Gulf and East Coast are called Hurricanes; in Europe, storm named Tropical Storms; in Asia and Oceania storms called Typhoons In the first half of the 20th century, scientists have studied storm surge by existed over time tools: from the simple experiment method based on sparse sampling of Conner [9] và Harris [10] to derivatives are a more complex analysis for the Sea-basin and storm with simple but limited practical value as the study of the Proudman [11], Doodson [12] In the US, Congress has directed the research agency of army and Weather Bureau to conduct intensive research on storm and storm forecasting methods after tremendous damage of the eastern coastal region

in 1954 This is considered as the beginning of a systematic study of storm surges in North America [13] Before digital model is developed, storm map has been designed

to forecast water levels rose whenever a typhoon landing on a shore Author Conner and others have built a simple experience models to calculate approximately crest of water level rise then draw up the largest monitoring water levels based on pressure monitoring data are at the center of the storm to determine how cover of water level [9]

In the world there are many studies using numerical models to simulate storm surge through that reviews potentially flooding due to water level rise by storms on coastal area Many storm surge models were used to simulate storm surge and provided high accuracy Research of group R.Christina used ADCIRC model to calculate the crest of wave in storm surges and ability of coastal flooding in West Bengal and Bangladesh caused by ultra strong tropical storm Alia landed near the Sagar Islands in West Bengal border and Bangladesh coast on 25/5/2009 This is the storm has caused huge damage to persons and property along the coastal belt of West Bengal and Bangladesh Results calculated by ADCIRC model showed the highest wave in the storm is up to 4

m in the Indian Sunderban area then spread to all the major river systems of the region,

submerge all the rivers as well as the inland area in water The extent of flooding are calculated via simulation models showed the storm caused flooding level inside area

is 350m average and up to a maximum 600m in different areas in the coastal region of West Bengal and Bangladesh city [14]

Using numerical models to simulate the water surges in storms and flooding simulations by coastal storms are increasingly applied in recent global studies of simulation capabilities and highly accurate forecasts In a study entitled "Numerical Model of Storm Surge and Inundation in Bohai Bay" by the author Daming and other co-authors refers to the application of models to simulate storm surge and flooding due

to the super storm caused in the Gulf of Bohai The authors used the ADI method to support 2D image simulation of storm surge in the FVM methods and models to support the level of flooding simulations in 2 cases: case1 with jetties high 2m and case 2 wihout jetties The results have identified flood map of region with two cases include jetty and no jetty Simulation results by model is verified accurate offers comparison with observed data in history and they were consistent with observed data [15]

The area North-East of the Arabian Sea include Gujarat coast of India and the western coast of Pakistan also is the area prone to the effects of the storm and suffered damage

of life and property by storm surges To simulate sea level and storm surge warnings contributed to the risk of flooding can occur, Indu Jain and other authors used the experimental model to calculate the water level in the big storm struck the coast Pakistan in 1999 and 2001 According to the results calculated by the model, sea-level rise during a storm in 1999 was greater than 3.8 m compared with the observed results

of the organization Unisys Weather, it was similar to hurricanes in 2001 and higher than 1,0m However, the results of study not mentioned dimensional mode especially the case storm tide coincides with the highest but predicting hurricane surge could completely done by the model based on real time [16]

Along with the development of computer and information technology at present, research using models to forecast the impacts of sea level rise due to storm surge more popular than empirical statistical methods The authors LIU Juan, JIANG Wensheng

and other authors use Matlab model to simulate storm surge and ArcGIS model to determine the extent of flooding in the reseach name "A high resolution forecast model

of storm surge inundation" The storm was chosen to simulate including 02 typhoons are 6903 (Viola) and 0104 (Utor) that has caused huge damage to Shantou coast Comparing the simulation results with observed data of 2 typhoons on the ability showed high accuracy between the model and reality so that we can use the model to predict risk of flooding caused by storm surge [17]

The authors Jun Wang, Xu Shiyuan, Mingwu Ye, Jing Huang and etc used Mike 21 model to assess the risk of spills from the sea dyke sea dike and the effects of sea level rise, storm tide combined land subsidence in a study entitled "The MIKE Overtopping Application to Risk Assessment Model of Seawalls and Levees in Shanghai" The study results showed that the risk of overflowing the sea dike and sea dykes from disasters combined effects of three natural hazards which are highly anticipated capability and comprehensive assessment of risks in the future to build preventable plans [18]

1.2 Overview of study in Viet Nam and Hai Phong

In Viet Nam, the research on sea mostly focused on the tide change and storm surge phenomenon Particularly storm surge phenomenon has been studied since the 1970s

of the last century, including a number of authors and their featured works such as: Programs and Tran Le Phuoc States (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 (1995), Nguyen Thi Vietnam Union (1996), Nguyen Vu Thang (1999), Bui Hong Long (2005), Nguyen Ky Phung (2006), etc

The first study of storm surges using statistical methods to calculate and chart surges

in the position to be calculated Recently, the research approach surges value method

to simulate and calculate storm surges more widely used It can be said, the review study of storm surges in Vietnam is showed most clearly from 1984 to the present, within the framework of the state-level 3 topics Research of author Pham Van Ninh

on sea level rise due storm is divided into 2 parts in 2 research topics of state level

The project 48.06.15 of the author Pham Van Ninh has developed a standard method optimal standard use "Tidal table" with high precision in order to clearly identify the movements and distribution time spatial phenomena of the storm surge This study has confirmed the phenomenon caused storm surges in Vietnam can be studied to give better results into practical applications [19] The project 48B.02.02 (1986-1990) studied the distribution characteristics of the phenomenon of rising water in time and space, location and time of the maximum current and surges the relationship between the time of the tidal surges to the challenge Practical significance of the research topic

to determine the elevation of the marine construction to serve the planning for economic areas of coastal [20] Research forecasts storm surges for specific (1991-

socio-1995, subject KT.03.06, chaired by Do Ngoc Quynh) In addition, from the year

1996-2000 state level project "Scientific basis and technical characteristics of coastal, " by Pham Van Ninh and Do Ngoc Quynh have reviewed calculation of characterized in southern storm mode by additional data [21] In addition to state projects, there are many other research topics such as the research of author Le Trong Dao presented idea

in research storm surges by finite element method to calculate tides and storm surges for the south china sea [22] Scientific research projects of Nguyen Vu Thang using computational models of coastal storm surges in Hai Phong which gave storm surge chart in forecast for the region [23] Overall studies of storm surge in the beginning have achieved certain accuracy, however because limitations of actual data about topography, tides and storms, especial limitations of speed of computer so mesh detail level and expansion of calculated boundaries during encounter many restrictions To overcome limitations of computer speed in simulating and calculating storm surge for small area, author Bui Xuan Thong used nested grid method in value model application to simulate storm Vietnam coastal waters has brought results quite detailed and accurate Along with studies of storm surge by numerical models, recent studies tend to use the commercial model and open source model has already been built in countries to apply the calculated storm surge of Viet Nam coast and provide high accuracy of results [24]

At present, the popular commercial model in the world include: MIKE model of the Danish hydraulic Institute (DHI), SMS model of the U.S Navy, the Delft 3D model of

Delft, Netherlands, and a group of open source models including the POM model of Princeton University, ROMS model of Rutgers University and University of California, USA, model GHER of the University of Liege, Belgium, In this direction, a number of typical projects as research of Le Trong Dao using Delft3D models of Dutch to set up and simulation, forecasting storm surges to coastal areas of Viet Nam The research of Nguyen The Tuong, Tran Hong Lam and others in coordination between Vietnam - China on research forecasting ocean waves, storm surge by using different models such as Delft 3D Netherlands, JMA (Japan Meteorological Agency storm surge model) of Japan and CTS (China typhoon Surge)

of China to calculate and provide forecasting processes of storm surge [25] In another research used the open source model such as Vu Thanh Ca and et al who applied and developed using POM model of the United States to apply storm surge calculations taking into account the influence of tide [26] From results of Vietnam National project KC.09.04/01-05 (2001-2005) “Short time prediction of hydrodynamic processes in EVS” then meteorological fields forecast models in 72h, in which, storm surge forecasts for entire coast of Viet Nam continue to be developed by Professor Dr Tran Tan Tien in project KC08-05 [27] Also may include study of Nguyen Tho Sao in the study “Storm surge predictions for Vietnam coast by Delft3D model using results from RAMS model” who used RAMS model to build an associated procedure for prediction of storm surges using Delft3D-FLOW model [28] The study of Bui Xuan Thong used a method for determining possible maximum storm surge at a sea dike [29]

Studies of storm surge in Hai Phong city currently attracting interest of many experts expressed through many researchs of expects include: Nguyen Xuan Hien and et al have used the ADCIRC model to calculate storm surge due to the typhoon Damrey on coastal estuary area of Hai Phong city in 2005 Simulation results using ADCIRC model were compared with measured data and produce results quite similar Base on results of these study that showed phenomenon of storm surges not only effect to sea dike, but also directly damage the estuary dikes, river dikes [30] Continue towards storm surge research by using models and empirical formula in coastal areas of Hai Phong city inlcuding research of the authors Nguyen Xuan Hien and et al (2012) In the framework of this study, total storm surges is calculated by sum of sea level rise in

storm and tide plus sea level rise by wave surges The results showed that the magnitude of storm surge in the storm happened constituted about 16% to 18% of the mean wave height offshore and contribute significantly to total sea level rise in storm Storm surges in the largest total time period of 1000 years can be approximately 500cm and potentially flooded area after dikes [31] In framework of the project

"Assessment fluctuations extreme sea levels due to climate change serves marine economic strategy" belong to the program KC09.23/06-10 "Marine science and technology serve sustainable development of social economy" was developed by Prof Dinh Van Uu, software ADCIRC of SMS suite (using triangular mesh) was applied to calculate storm surge for North Bay area and details for Hai Phong coast [32]

1.3 Overview of research methodology and tools

In general, the research methods of storm surge at the moment includes following methods:

Statistical measurement methods: This method is based on the statistical data to

measure the sea level in the study area and data of the storm affect at the same time From the statistics that found the rules or build relationships correlation between data and storm surge in the study area Survey data in the field is essential which is used to calibrate and test the accuracy of the method However, this method can only be applied to each area of research, because every region has different natural conditions This is the traditional method for relatively positive results The result of this approach

is very important, it built up a database of catering for the integrated assessment of water level in the sea areas of research and service to other methods

Experimental Methods: To determine the height of storm surges, often use the

empirical formula is summarized from the data measured for each area In our country,

on the basis of the survey the relationship between high-speed storm surge from 1959

to 1970 in coastal area gave formula as follow:

Inwhich Wmax – Average of wind speed (m/s); ∆h- Sea level rise (m)

- Formula of Karausev A.V., Labzovski N.A in standard 06.04.82 of Russia: = (1.2)

+ Standard 22TCN222-95 of Ministry of Transport:

=

+ Standard TLC-1-78 of Ministry of Agriculture & Rural Development:

In which: W – wind speed (m/s); X- wind velocity (km); H- water high (m);

kw – corrective ratio; – angle between coast and wind route (Degree)

Numerical method: Numerical modeling is one of the optimal approach in the study of

hydrodynamic processes The first characteristic of the model is the subject of research

as geographic regions, time scale, the process should simulate The second feature is the size of the space and state variables The third characteristic of the model is resolution and final characteristic is the precision

The dynamic processes caused storm surge mentioned above are described in the shallow water equations by two-dimensional numerical models:

H – water depth

Pa – air pressure

,, : wind stress on surface route axis x, y

, = + (1.9) , = + (1.10) : bottom friction coefficient

= 2 – Coriolis ratio

: geographical latitude

: 0.729.10-4 s-1 : earth rotation angular velocity

Currently in the world there are many application models using this method for calculating have results effectively due to the speed of the computer include: model MIKE, Delft 3D, POM, ROMS

Each calculation methods mentioned above have their advantages and disadvantages, method uses empirical formula and selling experience has advantage is easy to apply and fast calculation, however calculation results are too large and maybe differences compare with measured results Theoretical calculation methods using modern computational model can calculate and forecast storm surge on a large scale (85,000 points can be calculated), but fixing of the input parameters such as angle created by perpendicular axis and wind direction , coefficient Kw, wind momentum W, coefficient surface friction , water depth of area that make difference between the calculated results and actual measurement

1.4 Overview of study area

1.4.1 Geographical location, administrative boundaries

Hai Phong is a coastal city, located on the east of the Northern coastal area, 102 km far from Ha Noi capital, with Quang Ninh province on the North, Hai Duong province on the West, Thai Binh province on the South and East Sea on the East The total natural area of 152,318.89 hectares (2001 statistics) accounts for 0.45% of the natural area of the country With Bach Long Vi island district in Tonkin Gulf, with coordinates of 20°07'35'' - 20° 08'36'' North latitude and 107°42'20'' - 107°44'15” East longitude, Hai Phong has advantages of sea, rail, road and air, resulting in favorable exchange conditions with other provinces in the country and the nations of the world Due to the port, Hai Phong city plays tremendous role in import and export of North Vietnam, having quick access to scientific and technology achievements from abroad then spreading them all over the country Hai Phong seaport along with Cai Lan port (Quang Ninh city) with a capacity of several millions of tons create increasingly-large-scale port clusters, contributing to the transportation of Northern goods to other regions of the country, as well as participating in transportation of transiting goods for

Southwest China region Along with the economical and commercial development social and cultural development of the region is oriented to development of Hai Phong City in 2025 Currently the region is a bridge between the other economic centers and

Do Son tourist center In the future, there will be more strategic projects for the economical and political development of the region Land area is divided by the rivers into five separate irrigation systems: Vinh Bao; Tien Lang; Red stones; An Duong and Thuy Nguyen The whole area is located in the downstream of Thai Binh river, an important part of North Delta and is one of the main drainage direction of Red River and Thai Binh river Upstream flooding, heavy rain in the midlands and plain; tides, waves and storm surges from the sea passing through the estuary are the factors causing flooding in areas, severely affected agricultural production and people's life In recent years, together with the strong development of people's livelihood, infrastructure, economy, shoreline areas were leveled out to pave for the development

of infrastructure, tourism, aquaculture , resulting in increasing threat of natural disasters to people's safety in coastal areas

Fig 1.1 Administrative map of Hai Phong city

To deal with storms and prevent flood, Hai Phong has built dyke system consists of 24 dikes with a total length of 420.824km In which there are 18 river dikes with 314.789km long and 6 sea dykes with 106.035km There are also 387 culverts under sea dike and 52 dams to ensure drainage and sea dike protection

The subregion including Kien Thuy District, Duong Kinh and Do Son District is protected by sea dikes I and II, being concentrated residential and economic-developed areas of Hai Phong City In which, Do Son is a district of Hai Phong city, about 20 kilometers far from the city center to the southeast, being a small peninsula created by Rong Mountains spreading to the sea to 5 km, with dozens of hills from 25m to 130m high Do Son district was established on September 12, 2007 on the basis of the entire area of the former Do Son town under Decree 145/2007/ND-CP of the Socialist Republic of Vietnam Government To the west and northwest, Do Son District is next

to Kien Thuy district, the remaining direction adjacent to the East sea Thanks to two estuaries Lach Tray and Van Uc of Thai Binh river system in the north and south of the district flowing into the sea bringing a lot of silt, together with luxury resort on the sea dykes Hon Dau Island, the area is very attractive to the tourists Kien Thuy is a suburban district located on the southeast of the city, having a natural area of 102.56 km², with a population of over 125 thousand people Northern and eastern borders are

Do Son and Duong Kinh district, the southern is Tien Lang district and the western are Kien An and An Lao district Besides the advantage tourism with Do Son beach, the subregion including three districts also has important role in economical development, fishing and aquaculture seafood, domestic and international trade Therefore, the role

of sea dikes 1 and 2 to protect this sub-region are identified by Hai Phong City People's Committee as extremely important for people's lives and social economy of the city

Fig 1.2 Sea dikes 2 is protecting populated areas

(Tien Lang district) and Trung Lap (Vinh Bao district) The low-lying areas often have difficulty in drainage

Hai Phong has a coastline of over 125km including coastlines around the islands The coast is a curve concave towards the Gulf of Tonkin, low and relatively flat, composed mainly of sand, silt from five river mouths Hai Phong sea is a part of the northwestern Gulf of Tonkin The structural and topographic features of the seabed and oceanographic characteristics of Hai Phong sea attach to the general characteristics of the Gulf of Tonkin and the South China Sea Hai Phong sea is not very deep Bottom water level is at 2m around Do Son coast, going down to 5m from offshore On the sea floor where the shed estuaries, the erosion of the flow causes a greater depth The offshore area has many islands scattered across the sea, the largest island is Cat Ba, most far one is Bach Long Vi Island

1.4.3 Geological and soil characteristics

Large parts of the inner city and suburban city of Hai Phong have fourth sediment with thickness ranging from 40-60m Stratigraphic column shows that alluvial sediments arrange from top to bottom as follows: on the surface clay layer has thickness averaging 0,4-2,0m, the next layer consists of layers of mud, mud sand interspersed irregularly, from 5-20m thick, muddy floor is the next floor, from 3-22m in thickness, clay layer with a thickness of 2-26m and finally the small sand floor gradually shift into large one with thickness of 9-30m

Hai Phong has two aquifer floors in the fourth sediments The first floor is located in the clay layer of sediment, sand lenses in the clay layer has thickness of 18m on average The first floor only affect engineering geology The second floor is located between the clay layers and bedrock This floor is salinity and has deepness of 20-40m Underground water is 0.5 - 2.0m deep This water levels lower during the dry season (from November to April) and are added in the rainy season (from May to October) Additional underground water resource is from agricultural irrigation systems Tide can affect water levels and groundwater quality of the coastal zone

1.4.4 Meteorology and climate characteristics

Hai Phong is influenced by the coastal climate Annual climate is divided into two seasons: the rainy season from May to October, the weather is hot, humid and rainy The dry season from November to April, cold and little rain

1.4.4.1 Thermal mode

The average annual temperature in Hai Phong is 23,1oC, classified into two quite distinct seasons: hot summer, the average temperature is above 25°C and cold winter, average temperatures is below 20°C The highest average temperatures is in July, reaching 28,40oC in Phu Lien, 29oC in Hon Dau and 28,7oC in Bach Long Vi The average monthly temperature in January reaches the lowest of 16,3oC in Phu Lien, 16,8oC in Hon Dau and Bach Long Vi

Table 1.1 Typical monthly and annual average temperature (unit oC)

Phù Li u 16.3 17.0 19.4 23.0 26.5 28.1 28.4 27.7 26.9 24.7 21.3 17.9 23.1 Hòn D u 16.8 16.8 19.2 22.8 27 28.5 29 28.4 27.6 25.3 22.3 19 23.6

B ch

Long V 16.8 16.6 18.7 22.3 26.2 28 28.7 28.3 27.3 25.3 22.4 18.9 23.3

Source: Hydrology Report [34] 1.4.4.2 Relative Humidity

The average relative humidity in research region is 83-86% High humidity reach peak

in the last months of winter with drizzle in March with 90-92%, in march with drizzle The lowest monthly average humidity is in November and December when dry northeasterly wind blowing on installments

Table 1.2 Typical monthly and annual average relative humidity (Unit: %)

Phù Li u 84 88 91 90 87 87 86 88 86 82 79 78 86 Hòn D u 80 85 90 90 86 85 84 85 82 77 75 77 83

B ch

Long V 84 90 92 92 90 88 85 86 84 84 79 80 86

Source: Hydrology Report [34]

1.4.4.3 Piche Evaporation

Evaporation in mainland is quite small compared to the amount of evaporation in the island due to high wind speeds During the year the average monthly evaporation peaks in July when the air temperature is high due to the hot dry wind and also high in November and December when the northeast monsoon overflow dry and stinging installments Minimal evaporation occurs in March with damp drizzle

Table 1 3 Typical monthly and annual average evaporation (Unit: %)

Phù Li u 52.9 34.5 32.9 39.9 62.4 67.5 70.4 57.8 63.3 77.6 78.8 71.1 709.5 Hòn D u 64.9 42.6 35.7 39.8 70 77.2 89.5 72.9 80.4 98.2 88.6 79.6 839

In Summer (May to September), Hai Phong is influenced by the flow of hot air and moisture from the west and south Prevailing wind direction is mainly East, Southeast and South The average wind speed reaches 3.5 to 4.0 m/s with a maximum of 20-25 m/s In the summer sometimes appear southwest winds, with small speed but bringing hot dry weather (Table 2.6) During the transition period (April and October), the reducing impact of monsoon, the sea wind and breeze wind usually appear with a velocity of about Level 3 - Level 4, the wind blows from the sea to land during the day and blew back from the land to the sea during the night According to space, with

effect of natural conditions, wind speed decreases from offshore to shore The average wind speed in islands is usually 1-4m/s larger than coastal mainland area

Table 1 4 Average monthly and annual wind speed

Station I II III IV V VI VII VIII IX X XI XII Year

Phù Li u 3,4 2,7 2,7 3,1 3,3 3,1 3,2 2,7 2,8 3,0 3,0 2,8 3,0 Hòn D u 4,8 4,6 4,4 4,7 5,6 5,7 6,0 4,7 4,6 5,0 5,0 4,7 5,0

Typically, the typhoon Kate (1955) caused to 158 broken dykes, 669 died people, 1,200 injured people and 13,000 ha of submerged fields; the typhoon Wendy (1968) destroyed 87.000 m3 dikes, 155 died people, 400 injured people, 18.000 ha flooded fields; the typhoon Sarah (1977) caused to 48 people died, 228 injured people, 160.000

m3 eroded dikes, 48.000ha flooded fields Especially in 2005, Hai Phong is influenced directly by three powerful storms, causing severe damage to economic livelihoods as well as the system of irrigation works such as dykes Typhoon No 02 (Washi) hit Hai Phong coinciding with high tide period at 11 o'clock cause huge surges in coastal areas and estuaries, the largest wave heights measured at 13h is 3,60m Typhoon No 6 (Vicente) landed in Hai Phong together with large waves, tides and storm surges were maintained for long periods of time from 10h to 22h corresponding to the tide level at Hon Dau from 2.2 to 3.0m Typhoon No 7 (Damrey) cause strong winds of level 9,

10, shock on level 10, combined with high tides (at Hon Dau is 3.2m) cause seriously large surges in estuaries, threatening levee systemand estuary dykes

1.4.5 Economical and social characteristics

Do Son district has 4237.29 hectares of natural area and a population of 51417 people The district has seven ward: Bang La, Hop Duc, Minh Duc, Ngoc Hai, Ngoc Xuyen, Van Huong, Van Son Do Son district has favourable conditions to develop tourism, restaurants, hotels In some other areas people mainly earn their living by fishing, sailing, marine aquaculture Tourism service plays important role, accounting for nearly 70% of the economic structure of the area This area has the potential to develop and expand the tourism industry, attracting both domestic and international visitors Do Son beach is divided into three main zones: Zone 1 is located in the beginning of Do Son Town, Zone 2 has many modern hotels, Zone 3 is quiet and discreet

Fishing industry and agriculture account for 23% of the area’s economic structure However, the fishing and marine aquaculture must be linked together with environmental protection, as well as avoiding pollution and water salinization In recent years, the agricultural economy shifted towards industrialization and modernization and ensure food security There are also gradually changing pattern of agricultural production towards agriculture serving urban (with fresh vegetables, flowers, plants, fruits, ), the introduction of high technology, advanced techniques in production and important achievement Hai Phong has built the first and most modern high-technology agricultural zone in the North The model of biotechnology applications are replicated Until now, the city has about 2,000 hectares of arable land cultivation technology applied with a canopy, agricultural mulch, 10,000 hectares of arable land can produce 3-4 crops/year, with revenue of more than 50 million VND/ha/year

Along with the development of agricultural production, the city has focused on infrastructure investment and technical service of aquaculture in all three areas: intensive farming, semi-intensive farming in marine and freshwater; improve the fleet with high technology and new equipment

CHAPTER 2: SETUP NUMERICAL MODEL TO SIMULATE STORM

SURGE FOR STUDY AREA

2.1 Data collection and Analysis

2.1.1 Topographical data

To localize risk of flooding to area of Do Son, Kien Thuy District, Hai Phong City requires more time and quite complex, input data has a very important role because it determines accuracy of a model The data used in a model include: topographic maps scale 1/50,000; digital elevation map (DEM) with resolution (30x30) m; tidal data at Hon Dau station; meteorological data; information about current infrastructure, situation of flooding (flood marks, risk, size, flood influence, )

Topographic maps scale 1/50,000 with seven layers of information include topography, administrative boundaries, transportation, vegetation cover, river systems, premises, residential This map is used as a base map to calculate and show the results flood simulating for study area

Fig 2 1 Hai Phong city map scale 1:50000 Digital elevation model (DEM) with a resolution (30x30 m): SRTM (Shuttle Radar Topography Mission) is a coordinate project NASA and NGA mix (ground Defense

Agency, US) to draw terrain earth’s surface in 3-D large area at a level of detail not seen before Base on combining SRTM terrain data with resolution of 30x30m and administrative map Hai Phong City scale of 1:50,000 which start to conduct assign attributes of contour and properties of special points then build digital elevation models for input as topographic base for study area in the model MIKE Zero

Fig 2.2 Digital elevation model DEM (resolution 30x30m) of study area

2.1.2 Hydrodynamic data

The data is tidal boundary conditions needed for setup Mike Zero modeling in calculating as well as to calibration and verification the model before applying the model to simulate the storm surge

The tidal data in this research is tide of one year at the Hon Dau station which measured by Institute of Coastal and Offshore Engineering Viet Nam

2.1.3 Storm data

Data on storms and tropical depressions in the recent period from 2010 to 2015 were collected from the annual reports of storms (Tropical Cyclone Annual Report) of US JTWC Center often give more information than storm data of other agencies

Data of storms and tropical depressions are considered in this study included storms beachhead from Quang Ninh City to Thanh Hoa City in period from year 1952 to

2015 The storm could cause fluctuations in sea level, waves high in large area include Hai Phong coast Base on that, the list of storm which impacted and influenced Hai Phong coast are presented in Table 2.1

Table 2.1List of typhoons affected to Hai Phong coast from 1952 to 2015

No Storm Time City Category

No Storm Time City Category

Details some of strong storm landed in the sea of Hai Phong is listed in table 2.2

Table 2 2 Details of some strong typhoon landfall on Hai Phong coast

Based on storm data, statistical characteristics including trajectory (direction and speed

of movement) and storm intensity (maximum wind speed and barometric pressure of the storm) are analyzed Input data was taken from a database of characteristics by JTWC including locations and maximum wind speed observations under 6 hours apart Analysis results showed that storms affecting coastal areas in Hai Phong move in the direction of focus from West - Southwest to Northwest, in which the highest is the west - northwest (about 44% of the total number of hurricanes affecting the region)

Source: Nguyen Xuan Hien (2013) [35]

Fig 2.3 Distribution of storm’s direction landfall in Hai Phong coast