Probabilistic Design of Coastal Flood Defences in Vietnam Probabilistic Design of Coastal Flood Defences in Vietnam Proefschrift ter verkrijging van de graad van doctor aan de Technische Universiteit Delft, op gezag van de Rector Magnificus professor ir K.C.A.M Luyben, voorzitter van het College voor Promoties, in het openbaar te verdedigen op dinsdag 21 september 2010 om 10.00 uur door MAI VAN CONG Master of Science in Hydraulic Engineering geboren te Ninh Binh, Vietnam Dit proefschrift is goedgekeurd door de promotor: Prof drs ir J.K Vrijling Copromotor: Dr ir P.H.A.J.M van Gelder Samenstelling promotiecommissie: Rector Magnificus, voorzitter Prof drs ir J.K Vrijling Technical University of Delft, promotor Dr ir P.H.A.J.M van Gelder Technical University of Delft, copromotor Prof dr ir M.J.F Stive Technical University of Delft Prof dr H Hanson Lund University, Sweden Prof dr J.W Hall Newcastle University, UK Ir K.W Pilarczyk Formerly Rijkswaterstaat Dr ir S.N Jonkman Technical University of Delft Prof dr W.S.J Uijttewaal Technical University of Delft, reservelid Copyright © 2010 by Mai Van Cong All rights reserved No part of this book may be reproduced in any form or by any means including photocopy, without written permission from the copyright holder ISBN 978-90-9025648-1 Printed by: Sieca Repro, Delft, the Netherlands Cover layout: Pham Quang Dieu Cover image: Violent ocean wave (source: Ocean Waves Wallpapers) to my family SUMMARY Probabilistic design of coastal flood defences in Vietnam The year 2005 brought hurricane Katrina to New Orleans (U.S), it was also a historical year for Vietnam because of the disastrous impact of typhoons on sea flood defences and coastal regions In total, eight typhoons hit the Vietnamese coast that year resulting in human casualties and enormous economic damage In response to that, both the central and local governments of Vietnam have paid the highest-ever attention to the rehabilitation of sea dike systems and the improvement of the safety of its coastal regions The improvement of the flood defence systems of Vietnam calls for the use of the latest available knowledge at all levels Vietnam has a profound practical experience in the field of flood protection; however, there is a lack of up-to-date knowledge in the field of dike design and flood risk management, especially regarding to coastal flood defences This study focuses on coastal flood defences in Vietnam and its objectives are to evaluate the current level of safety and reliability, to develop safety standards and design practices and to formulate a proposal for the improvement of the flood defences To meet the study objectives, the research presented in this thesis has focused on firstly reviewing, developing and widening the application of probabilistic design and state of the art of risk assessment and risk evaluation in the field of flood defences (theory part) Secondly, as a demonstration of knowledge transfer, the proposed approach has been applied to the coastal flood defence system of Nam Dinh province in Vietnam which is considered as a representative coastal system in the country (application part) The main contents covered in this thesis are summarized below Chapter gives an overview of the coastal area, existing approaches for flood defence in Vietnam and a description of the case study area The coastal zone in the Red River Delta of Vietnam, where the case study was carried out, is under a potential threat of fluvial flooding, coastal flooding and coastal erosion In the region, sea dikes have been used as a predominant countermeasure with two intended functions, i.e protecting lowlying areas from coastal flooding and reducing risk for the hinterland caused by coastal erosion However, as was also experienced during recent typhoons in the last years, the sea dikes not seem to function very well and their strength is insufficient to withstand the hydraulic loads with a sufficiently low frequency of occurrence This is due to the underestimation of the hydraulic boundary conditions and imperfections in the design, construction and maintenance of the dike systems A qualitative multi-criteria analysis has been developed for proposing a suitable coastal protection strategy for the study area It is proposed to use a combination of sea dike systems to provide flood protection, i Summary and beach erosion control measures, such as groins and/or sand nourishments to preserve a shallow foreshore To have a better understanding of the actual hydraulic load conditions in the coastal area and to derive input information for the reliability analysis, methods for the statistical analysis of extreme hydraulic load conditions have been developed and applied (Chapter 3) These include data management techniques (i.e trend analysis, stationarity tests, seasonality analysis and peaks over threshold analysis); and statistical methods to model the occurrence probabilities of extreme values (the value which is exceeded by the random environmental variable with very small probability p) Hydraulic observations for the coast Vietnam are available for a period of 35 years and on that basis an estimate needs to be derived for a design sea load, i.e waves and storm surge levels, corresponding to a 100 to 10,000 years return period The methods of Regional Frequency Analysis (RFA) and Mixed Distribution Fitting (MDF) are used in this study for such a sea load In order to ensure that the statistically predicted values of the interested sea loads will be as close as possible to the true values, recommendations are made to compare these values with physical hindcast models which are based on historical information of typhoon characteristics such as wind speed Inspection of monthly sea water level data at Nam Dinh coast reveals that there exists a quadratic rising trend, with periodicity of months, 12 months and 224 months in the data The 6-month periodicity reflects well the influence of two monsoons per year to the fluctuation of sea water level; the 12-month periodicity is explained by the influence of yearly typhoons in the region The 224-month periodicity, which is approximately 18.6 years, is almost exactly equal to a tidal circle Although sea level rise (SLR) is not a focus of this study, based on a detected linear trend of the observed water level data the rate of SLR can be estimated approximately at 0.68 cm per year at Nam Dinh coast In addition, this study recommends that estimation of SLR should be based on not only mathematical extrapolations, but also the longterm nature of sea level fluctuation The theoretical background of probabilistic design and reliability analysis of flood defences is reviewed in Chapter Models for safety assessment and reliability based design are developed for the specific application to coastal flood defence systems In the safety assessment model, present probabilistic calculation techniques are applied to quantify the failure probability of the system and the system components The effect of the length of the dike system on the system reliability was deployed in the model as well The reliability-based design model is used to search for an optimal geometry of the dike system given an admissible failure probability Chapter demonstrates the application of the models proposed in the previous chapter to Vietnam The case study area is the Nam Dinh coastal flood defence system in the Red River delta of Vietnam It is found that the existing coastal flood defence system is less safe than its required safety standards According to the standards the dikes should be designed to provide a target safety of 1/20 per year However, the total failure probability of the dike system is estimated to be much higher, about 0.15 per year This corresponds to the occurrence of a failure once in years instead of once in 20 years as intended Excessive wave overtopping induced dike failure is the dominant failure mode, ii LIST OF TABLES Table 3.1 Sites characteristics, N: Number of peaks 58 Table 3.2 Heterogeneity measures with 500 simulations 60 Table 3.3 Goodness of fit measures (based on 1000 simulations) 60 Table 4.1 Related input variables and its statistical parameters 94 Table 4.2 Variance and influence factor α of each variable 94 Table 4.3 Example of fragility curve due to an individual failure mode with and without length effect (Total dike length L=10 km) 95 Table 4.4 Correlation distance for load variables of flood defences (Vrouwenvelder et al 1987) 99 Table 4.5 Correlation distance for strength variables of flood defences (Vrouwenvelder et al 1987) 99 Table 4.6 Example result of safety assessment models 100 Table 5.1 Summary of LME of indicated failure mechanisms 119 Table 5.2 Failure probability vs failure mode per dike section 120 Table 5.3 Contribution of failure probability to system failure (new dike system) 121 Table 5.4 Overflow related variables and its statistical parameters 125 Table 5.5 Variance and influence factors 125 Table 5.6 Piping related input variables and its statistical parameters 127 Table 5.7 Variance and influence factor α for Piping 127 Table 5.8 Variance and influence factor α for geotechnical instability of inner slope 128 Table 5.9 Total failure probability of Hai Hau sea dike system 132 Table 5.10 Overall system length effects 132 Table 5.11 Design values of interested parameters per mechanism 136 Table 6.1 Basic dimensions underlying perceived riskiness (adapted from Vlek 1996) 154 Table 6.2 The value of the policy factor βi as a function of voluntariness and benefit (Vrijling, 1998) 155 Table 7.1 Yearly average death rate of Vietnam 168 Table Comparison of actual risk in the Netherlands and Vietnam 172 Table 7.3 Actual policy factor tested for flooding in Vietnam 177 Table 7.4 Information on population for the Nam Dinh coastal districts, including estimates of the number of people remaining in the area and those evacuating (source: Nam Dinh Dike Department) 180 Table 7.5: Estimates of mortality by class of flood depth 182 Table 7.6 General and overall estimates of loss of life in coastal districts of Nam Dinh 183 Table 7.7 Evacuation and number of exposed people during flood in Nam Dinh [Nam Dinh dike department, based on information of Damrey Typhoon in 2005] 183 Table 7.8 Loss of life estimates for coastal districts in Nam Dinh Province 183 Table 8.1 Unit cost factors for Nam Dinh sea dikes 188 Table 8.2 Estimated potential economic loss due to flood at Nam Dinh coastal regions 193 Table 8.3 Optimal safety for Nam Dinh coastal regions 195 231 Table 8.4 Economic risk-based design for base case (polder has no slope) 199 Table 8.5 Summary of the case results: a linear sloping polder 201 Table 8.6 Summary of the Case results: an exponential sloping polder 203 232 APPENDICES APPENDIX LIST OF SCIENTIFIC PAPERS In this section, selected scientific papers which have been written following research results from this PhD project are listed These papers have been published in international journals and proceedings of international conferences in the years from 2005 to 2009 Mai Van, C., Hoffmans, G., van Hoven, A., van Gelder, P.H.A.J.M & Vrijling, J.K 2010 Probabilistic analysis of grass erosion due to wave overtopping Coastal Engineering (accepted & being published in 2010) Mai Van, C., van Gelder, P.H.A.J.M & Vrijling, J.K 2009a Failure Mechanisms of Sea dikes: inventory and sensitivity analysis In L Franco, GR Tomasicchio & A Lamberti (Eds.), Coastal structures 2007 Pp 550-561 ISBN: 978-981-4282-02-4 Singapore: World Scientific Mai Van, C., van Gelder, P.H.A.J.M., Vrijling, J.K & Stive, M.J.F 2009b Reliability - and risk - based design of coastal flood defences In J McKee Smith (Eds) Coastal Engineering Pp 4276-4288 Singapore: World scientific publishing Mai Van, C., van Gelder, P.H.A.J.M., Nguyen, H.Q & Vrijling, J.K 2009c Statistical analysis of extreme sea water level in risk based design of coastal structures In Irtces & H Tanaka (Eds.), Proceedings of the international conference on Estuaries and Coasts, Sendai, Japan Pp 610-617 Japan: Tohoku University Mai Van, C., van Gelder, P.H.A.J.M., Nguyen, H.Q & Vrijling, J.K 2009d Probabilistic design and reliability analysis of coastal structures - a Vietnam case In Tan Soon Keat & H Zhen Hua (Eds.), Proceedings of the 5th Internationa international conference on Asian and pacific coasts Pp 201-210 World Scientific Mai Van, C., van Gelder, P.H.A.J.M & Vrijling, J.K 2009e Risk based design of coastal flood defences - A Vietnam case In R Bris, C Guedes Soares, CG Soares & S Martorell (Eds.), Reliability, Risk and Safety; Theory and Applications Pp 1125-1132 London: Taylor & Francis Group Mai Van, C., van Gelder, P.H.A.J.M., & Stive, M.J.F 2009f Coastal protection strategies for the Red River Delta Journal of Coastal Research, 25(1), pp 105-116 West Palm Beach (Florida) Mai Van, C., van Gelder, P.H.A.J.M., & Vrijling, J.K 2008a Risk based approach for a long-term solution of coastal flood defences - a Vietnam case In S Martorell, C Guedes Soares, & J Barnett (Eds.), Proceedings of the European safety and relibitlity conference, ESREL 2008 and 17th SRA-Europe (pp 2797-2805) Leiden: Taylor & Francis Groups Mai Van, C., Tri, C.M., Quy, N.B., & Vrijling, J.K 2008a Risk based approach for safety standard of 233 coastal flood defences In Dao Xuan Hoc & et al (Eds.), Proceedings of the fifth anniversary workshop of the Marine and Coastal Engineering Faculty (pp 204-216) Hanoi: Hanoi Water Resources University Mai Van, C., Vrijling, J.K., & van Gelder, P.H.A.J.M 2008b Risk based approach for long-term plan of coastal flood defences: a Vietnam case In T.M Kao, E Zio, & V Ho (Eds.), Proceedings of the ninth international conference on probabilistic safety assessment and management PSAM (pp 1-13) Hong Kong, China: Lammar offset printing limited Mai Van, C., van Gelder, P.H.A.J.M., Vrijling J.K & Mai, TC 2008c Risk analysis of coastal flood defences - A Vietnam case In S.P Simonovic, P.G Bourget, & S.F Blanchard (Eds.), Proceedings of the 4th international symposium on flood defence; managing flood risk, reliability & vulnerability (Pp 93-193-8) Toronto: Institute for catastrophic loss reduction Mai Van, C., van Gelder, P.H.A.J.M & J.K Vrijling, 2007a Statistical methods to estimate extreme quantile values of the sea data In: Proceedings of the Fifth International Sysposium on Evironmental Hydraulics (ISEH-V), December 4-7, 2007 Tempe, Arizona, USA Edited by Don L Boyer & Oga Alexandrova, ASU, AZ, USA Mai Van, C., van Gelder, P.H.A.J.M & J.K Vrijling, 2007b Probailistic design and risk based approach in Civil Engineering In Proceedings of The 1st International Conference on Modern Design, Construction and Maintenance of Structures, 10-12 December 2007, Hanoi, Vietnam Construction Publishing House Press, Hanoi Edited by Mandis P.A et al ISBN: 978-0-646-48404-4 Pp 204-211 Mai Van, C., van Gelder, P.H.A.J.M., and Vrijling, J.K., 2007c Inventory and sensitivity analysis of failure mechanisms of coastal flood defences Proceedings of Coastal Structures 2007, July 1st -4th Venice, Italy In press Mai Van, C., van Gelder, P.H.A.J.M & J.K Vrijling, 2007d Reliability of coastal flood defences in Vietnam IAHS Publ 317 Redbook series Wallingford, UK ISBN: 978-1-901502-29-9 Pp 424-431 Mai Van, C., Pilarczyk, K & van Gelder, P.H.A.J.M & Vrijling, J.K., 2006e Foreshore erosion and scour induced failure of sea dikes Proceedings Third international conferences on Scour and Erosion, ICSE 2006 November 1-3, 2006 Amsterdam, CURNET, ISBN-10 90-376-0503-6/ ISBN-13 978-90-3760503-7 The Netherlands Mai Van, C., Stive, M.J.F., and van Gelder, P.H.A.J.M., 2006f Coastline changes and coastal protection within the Red River Delta of Vietnam In: Proceedings of the Third International Conference on Estuaries and Coasts (ICEC2006) Guangzhou, China 28-30 November 2006 Pp 305-317 Mai Van, C., van Gelder, P.H.A.J.M., and Vrijling, J.K., 2006g Safety of coastal defences and flood risk analysis In: Safety and Reliability for Managing Risk, ISBN 13: 978-0-415-41620-7, Taylor & Francis/Balkema, Leiden, The Netherlands, Vol Mai Van, C., and Pilarczyk, K.W., 2005 Safety aspect of sea dikes in Vietnam-A Nam Dinh case study Conference Proceeding of International Symposium on Stochastic Hydraulics, De Vereeniging, Nijmegen, The Netherlands, 23-24 May, 2005 IAHR, Paseo Bajo Virgen del Puerto 3, 28005 Madrid, Spain, 2005 ISBN: 90-805649-9-0 van Gelder, P.H.A.J.M & Mai Van, C 2008 Distribution functions of sea waves and river discharges in flood risk Journal of Hydraulic Research Vol 46 Issue Pp 280-291 van Gelder, P.H.A.J.M., Mai Van, C., Wang, W., Shams, G., Rajabalinejad, M., & Burgmeijer, M.A 2008 Data management of extreme marine and coastal hydro-meteorological events Journal of hydraulic research, Vol 46, Issue Pp 191-210 234 APPENDIX STATISTICAL DESCRIPTION OF THE RANDOM VARI- ABLES USED IN RELIABILITY ANALYSES Not φ ϕ ρ γ γ ρ 8.7 a A b B C c cosα D d H k KΔ K K L m m MHWL N P S S SL.rise Surge tanα T Z Z c sat unsat w B s p w k d m Description Empirical factor -Pilarczyk Friction angle Saturated density of subsoil Soil unit weight (saturated) Soil unit weight (unsaturated) Water density Model factor Safety free board Wave loads on outer slope Power factor Transportation load on dike’s crest Cohesion Bligh constant Cosine of slope angle Required block thickness/rock size Thickness of the top layer Wave height at toe of the dikes (1/20 years) Permeability Reduction factor for slope roughness Transformed coefficient % of wave exceedance Coefficient of wind effect Seepage path length Cotangent of slope angle Model factor High tidal level(+MSL) Number of storms Permeability factor Initial damage level Wave steepness Sea level rise Incr W.level by storm Tangent of slope angle Wave mean period Design water level Inland water level Unit o kG/m kN/m kN/m kG/m kN kN kN/m m m m m/s m m m - 3 3 - m m m s m m Dist Norm Nor Deter Nor Nor Deter norm Norm Nor Norm Nor Nor Deter Norm Deter Norm Lognor Deter Norm deter deter Norm Norm Norm Norm Deter Norm Deter Deter Norm Expon Norm Deter Norm Norm mean value nom 1800 nom nom 1031 8.7 0.5 nom 0.65 100 nom 15 0.97 nom 3.5 2.0 nom Nom 1.65 48 1.67 2.29 7000 0.2 0.02 0.1 1.0 0.25 nom nom std dev 0.5 20 5% 5% 5% 0.05 50 0.15 10 5% 5% 0.2 0.35 0.05 0.15 0.33 0.071 0.05 0.05 0.2 5% 0.2 0.5 nom : normative value; Norm: Normal distributed; Weil: Weibull distributed; Deter: Deterministics; Lognor: lognormal distributed Expon: Exponential distributed bed inland 235 APPENDIX OVERVIEW OF POPULATION DATA OF NAM DINH Appendix 3A- Population data (source: Dike department Nam Dinh) community district total population flood depth remaining / stay evacuated 9729 9902 10121 9935 7075 8469 8683 8290 9995 8671 6853 >3m 2-3m 2-3m 2-3m 1-2m 0-1m 0-1m 2-3m 2-3m 2-3m 2-3m 6510 6450 6800 5900 4530 4430 5950 5100 6745 5875 4530 3219 3452 3321 4035 2545 4039 2733 3190 3250 2796 232 8230 10237 5386 4830 8863 15300 1-2m 1-2m 2-3m 1-2m 0-1m 1-2m 5470 6790 3450 3210 5860 7540 2760 3447 1936 1620 3003 7760 7239 2190 9079 6863 17500 5169 >3m >3m 4850 650 5690 945 11260 3640 2389 1540 3389 5918 6240 1529 Giao Thuy Giao Thiện Giao An Giao Lạc Giao Xuân Giao Hải Giao Long Bạch Long Giao Phong Quất Lâm Giao Hương Giao Thanh Hai Hau Hải Hải Hải Hải Hải TT Đơng Lý Chính Triều Hịa Thịnh Long Nghia Hung Nghĩa Thắng Nghĩa Phúc TT Rạng Đơng Nam Điền Nghĩa Hải Nghĩa Bình >3m 0-1m 1-2m Source: data from Dike Department, Nam Dinh, 2009 236 Appendix 3B- Loss of life estimates due to potential flood village district source Giao Thuy Giao Thiện Giao An Giao Lạc Giao Xuân Giao Hải Giao Long Bạch Long Giao Phong Quất Lâm Giao Hương Giao Thanh Hai Hau Hải Đông Hải Lý Hải Chính Hải Triều Hải Hịa TT Thịnh Long Nghia Hung Nghĩa Thắng Nghĩa Phúc Rạng Đông Nam Điền Nghĩa Hải Nghĩa Bình population Flood depth exposed avg depth mortality exp Nr of fatalities Data DDMFC Flood maps 10% flood maps Mor Func Est 9729 9902 10121 9935 7075 8469 8683 8290 9995 8671 6853 >3m 2-3m 2-3m 2-3m 1-2m 0-1m 0-1m 2-3m 2-3m 2-3m 2-3m 651 645 680 590 453 443 595 510 674.5 587.5 453 3.5 2.5 2.5 2.5 1.5 0.5 0.5 2.5 2.5 2.5 2.5 0.00974 0.00663 0.00663 0.00663 0.00358 0.00082 0.00082 0.00663 0.00663 0.00663 0.00663 6.34 4.28 4.51 3.91 1.62 0.37 0.49 3.38 4.47 3.90 3.00 8230 10237 5386 4830 8863 1-2m 1-2m 2-3m 1-2m 0-1m 547 679 345 321 586 1.5 1.5 2.5 1.5 0.5 0.00358 0.00358 0.00663 0.00358 0.00082 1.96 2.43 2.29 1.15 0.48 15300 1-2m 754 1.5 0.00358 2.70 7239 2190 9079 6863 17500 5169 >3m >3m 485 65 569 94.5 1126 364 3.5 3.5 0.00974 0.00974 4.72 0.63 3.5 0.5 1.5 0.00974 0.00082 0.00358 0.92 0.93 1.30 >3m 0-1m 1-2m 237 LIST OF NOTATIONS The following list covers the global symbols that have been used in this thesis The following types of symbols have not been included: - Symbols of distribution functions and probability density functions which already were defined in the text e.g fN(n) or FN(n) - Expected values and standard deviations of already defined symbols, e.g E(N) - Logical combinations and specific variations in notations of already defined symbols, e.g σR (standard deviation of resistance) Nf (number of fatalities given failure) and IRAV – average individual risk - Abbreviations only used in the text and in graphs, e.g LOL –loss of life - All individual symbols have been defined in the text of this thesis Roman symbols A B Ci CN Ctot D FD FE I IR IRAV IRMAX L N NE NA NPAR PE parameter of the exponential distribution [m] (Chapter and 8) parameter of the exponential distribution [m] (Chapter and 8) constant that determines the vertical position of the FN limit line for an installation [yr-1 fat- ] constant that determines the vertical position of the FN limit line at a national scale [yr-1 fat- ] totals costs [€] economic damage [€] mortality (= number of fatalities / people exposed) [-] effects of evacuation and shelter investment in safety measures [€] (Chapter and 8) individual risk [yr-1] average individual risk [yr-1] maximum individual risk [yr-1] length of object [m] number of fatalities number of people exposed number of installations (section 9) number of inhabitants in the flooded area probability of exposure [-] 239 PV Q R RI S TR Z g h k m m0 n pf pf,0 pf,opt q t u/v present value factor [-] discharge [m3/s] resistance in limit state function [-] Risk integral [fat2/yr2] load in limit state function [-] Total risk [fat/yr] limit state function Acceleration due to gravity [m/s2] water depth [m] risk aversion coefficient [-] population density [person/m2] constant for population density [person/m2] constant in the probit function [-] probability of failure [1/yr] initial probability of failure [1/yr] optimal probability of failure following from an economic optimisation [1/yr] average overtopping discharge [l/m/s] time [s] flow velocity of water [m/s] Greek symbols Φ Cumulative normal distribution α risk aversion coefficient [-] β policy factor used to characterise the severity of an activity [-] θ angle in polar coordinate system [rad] θW wind direction [rad] μ average/mean value μ coefficient of static friction μN average of the normal distribution ρ correlation coefficient [-] ρ density of the water [kgm-3] σ standard deviation σN standard deviation of the normal distribution Abbreviations Name of the models ReOM reliability-based optimal model RFA regional frequency analysis RiOM risk-based optimal model RSAM reliability and safety assessment model 240 Organisations DDMFC department of Dike Management and Flood Control (Vietnam) DWW/RWS rijkswaterstaat, Ministry of transport, public works and water management, the Netherlands MARD ministry of Agricultural and Rural Developments (Vietnam) MONRE ministry of Natural Resources and Environment (Vietnam) PPD provincial dike department (Vietnam) VROM ministry of Housing, Land use Planning and Environments, National Environmental Plan, the Netherlands Others CDF cumulative distribution function CFDS coastal flood defence system DWL design water level (m) LSE limit state equation MHW normative high sea water level (m) MSL mean sea water level (m) PDF probability distribution function RRD Red River delta SLR relative sea level rise (m) 241 ACKNOWLEDGEMENTS This doctoral dissertation is a final product of the work done in several years at the Section of Hydraulic Engineering, Delft University of Technology (TUDelft), where I benefited a lot from the inspiring and friendly atmosphere, for which I am greatly indebted to many people, who have inspired and supported me during my studies My greatest gratitude addresses to prof Han Vrijling and dr Pieter van Gelder, my promoters, for their excellent guidance on the combination of hydraulic engineering and probabilistic design; giving me continuous support and providing me freedom to trying out new ways in doing research I highly appreciate their profound knowledge and precious input into our research works, their encouragement, their constant help and their great sense of humor I am very grateful to prof Marcel Stive, my mentor, for his fruitful discussions, advices on research direction and constant support during my PhD research I would like to express my gratitude to dr Bas Jonkman, from whom I could get both great mentorship and friendship, for sharing his ideas and detailed advices on acceptable risks of flooding We have spent many hours both in Delft and elsewhere, during our international trips discussing on the topic related issues His input has been very helpful to shape my research Many thanks go to the jury members of the Doctoral Examination Committee for reviewing and commenting on my thesis Especially, I would like to give my sincere appreciation to prof Jim Hall from Newcastle University, UK for his critical comments and constructive advices for all parts of my thesis in the finalizing phase It is a great honor for me to have ir Krystian Pilarczyk as a jury member I owe him for his mentorship at all phases of my study in Delft since 2002 Krystian is one of the first persons who introduced the state of the art knowledge on hydraulic engineering and flood defences from the West to Vietnam in the early 1990’s He has been considered as the founding farther of the cooperative activities between Dutch and Vietnamese Institutes in the fields of hydraulic engineering, coastal engineering and flood risk management I would like to express my great gratitude to him for his input through many hours of intense debate, his encouragement, warm welcome, loyal friendship and kindness Krystian, I will never forget the academic-sightseeing trips and family visits that Marian and you offered me The financial support for this PhD project, which is very much appreciated, came from the Section of Hydraulic Engineering, Water Research Center and CICAT of TUDelft 243 and Delft Cluster research programme of the Netherlands and partly from the Water Resources University of Vietnam (WRU) During my PhD study, I have received tremendous supports from colleagues both in the Netherlands and Vietnam My sincery thanks to: drs Paul Althuis, ms Veronique van der Vast, (CICAT); ir Gerrit Jan Schiereck, ir Michiel Tonnijeck (TUDelft-WRU coorporation project); ir Henk Jan Verhagen, ms Chantal van Woggelum, ms Adeeba Ramdjan, ms Inge van Rooij, ir Kees den Heijer, mr Mark Voorendt and Menno Eelkema (Hydraulic Engineering, TUDelft); ir Timo Schweckendiek (TUDelft & Deltares); prof Nguyen Van Mao, prof Dao Xuan Hoc, prof Nguyen Quang Kim, prof Pham Ngoc Quy, dr Le Dinh Thanh and dr Vu Minh Cat (WRU) I am greatly indebted to Wim Kanning, a faithful friend and long term office-mate of mine, for his kindness and unbounded enthusiasm Wim has helped me in many daily-life practical issues and shared with me various aspects of life during my study time in Delft Wim, we made memorable trips together, especially the visits to our parents There is too little space to list all my friends and colleagues in the Netherlands, who created so cozily and joyfully environment in the last years I am thankful to all of my international, Dutch and Vietnamese friends for your kindness and friendliness You have been keeping me away from loneliness and homesickness Especially, I am greatly indebted to thank all colleagues and friends from the department of Hydraulic Engineering of TUDelft, the unit of Geoengineering (formerly GeoDelft) of Deltares and the Vietnamese group in Delft (VCID) for the precious moments of working, chatting and having fun together Most importantly, I am grateful to my family Being abroad for such a long time was an enormous sacrifice of my family My PhD study started since our first son, Trung, was just born He is now still a little boy, but very often questions me: “dad, when the book’s done?”… Yes, it’s ready now! Trung and Nam, I have no excuse anymore for not being home regularly with you boys Mom and dad, I can never express sufficiently my indebtedness for all your love, support and encouragement that enabled me to actively pursue my personal objectives My appreciation to: my sisters Hai Yen and Thuy Linh, my brother Cao Tri, my aunt Di Toan for their unconditional support and taking care of my little family as our boys are still small; my elder sister Tuyet Mai and my brother in law Truong An for taking care of our parents when I am away from home My deeply thanks go to all other members of our big families for their support and inspiring encouragement Definitely, it has been much easier for me to have you all behind Finally, to my wife Kim Thoa, without your faithful love, patient and sacrifice it would have been very difficult for me to reach this day Thanks my love Mai Van Cong 2010 August, 19th 244 ABOUT THE AUTHOR Mai Van Cong was born in Ninh Binh, a province lying in the Red River Delta of Vietnam In 1995, he started his higher education at Water Resources University (WRU) and obtained a Bachelor of Science degree, with an outstanding rank, in hydraulic engineering in 2000 During his study time at WRU he won a number of university and national prizes from the Student Researching Encouragement Program and the National Student Olympic Competition on Mechanics Subjects, an annually prime academic event amongst technical universities in Vietnam He was invited to be a lecturer at the faculty of Hydraulic Engineering, WRU and fully granted for continuing his master study at the School of Postgraduates In 2002, he started studying coastal engineering at IHE Delft (presently named UNESCOIHE) and obtained his Master of Science in 2004 with the first-class honour After graduation, Mai Van went back to Vietnam and continued his academic career at WRU In late 2005, he was awarded a scholarship by Delft University of Technology for his PhD project on probabilistic analysis and risk based approaches for water defence systems of Vietnam While working on the project, he has been involved in a number of research activities at the Dept of Hydraulic Engineering such as the EU project FLOODSite and the Delft Cluster project on Safety against flooding in the Netherlands Since July 2009, while finishing up his PhD thesis, Mai Van has been working at Deltares, Unit Geoengineering (formerly GeoDelft) He has been involved in various activities at the institute include implementing applied research and research consultancy projects Mai Van’s research interests are related to statistics, sensitivity and uncertainty analysis of environmental data; multivariate- extreme value analysis of hydraulic loads on water defence systems; system reliability, flood risk and geotechnical risks 245 ... traffic is involved Based on Vietnamese statistics, baseline mortality has been determined including the risks of traffic and living in a flood- prone area in Vietnam An FN-curve for flooding in Vietnam. .. situation in flood defences, in general, and sea flood defences, in particular 1.5 PROBABILISTIC VS DETERMINISTIC APPROACH The conventional design method in civil engineering design is a deterministic... main branch of the Red River discharging its water into the Gulf of Tonkin, South China Sea 29 Coastal management and flood defences in Vietnam Figure 2.1 Position of the Red River Delta in Vietnam