Prelimin ary results of the study on the reason s of the Hai Hau erosion phenomenon

Vietnam Journal of Mechanics, VAST, Vol.29, No.3 (2007), pp. 415 - 426 Special Issue Dedicated to the Memory of Prof. Nguyen Van Dao PRELIMIN ARY RESULTS OF THE STUDY ON THE REASON S OF THE HAI HAU EROSION PHENOMENON I PHAM VAN NINH 1, PHAN NGOC VINH, NGUYEN MANH HUNG, DINH VAN MANH Institute of Mechanics, VAST Abstract. Overall the evolution process of the Red Jliver Delta based on the maps and historical data resulted in a fact that before the 20ll' century all the Nam Dinh coast line was attributed to accumulation. Then started the erosion process at Xuan Thuy district and from the period of 1935 - 1965 the most severe erosion was contributed in the stretch from Ha Lan to Hai Trieu, 1965 - 1990 in llai Chinh -- llai Hoa, 1990 - 2005 in the m iddle part of Hai Chinh - Hai Thinh (Ilai Hau district). The adjoining stretches were suffered from not severe erosion. At the same time, the Ba Lat mouth is advanced to the sea and to the North and South direction by the time with a very high rate. The first task of the mathematical modeling of coastal line evolution of l lai lJau is to evaluate this important historical marked periods e. g. to model the coastal line at the periods before 1900, 1935 - 1965; 1965 -- 1990; 1990 - 2005. The tasks is very complicated and time and working labors consuming. In the paper, the primari ly results of the above mentioned simulations (as waves, cur- rents, sediments transports and bottom - coastal lines evolution) has been shown. Based on the obtained results, there is a strong correlation between the protrusion magnitude and the southward moving of the erosion areas. 1. COASTLINE EROSION AT HAI HAU BEACH The length of the erosion stretch at Hai Hau district ranks the second after the Ganl1 Hao stretch which is located in Ca Mau province, north east of Ca Mau cnp. The 30 km of Hai Hau beach has suffered from erosion at least from the beginning of the last century. The coast has been eroded at a rate of 10 -- 15 m/year during the last half century. At present two sea dyke system have been constructed to protect the coast. Some descriptions of the eroded stretch arc below: +The shore line has NE-SvV orientation (coincided with the main predominant north- east monsoon wave direction) . Theoretically the N, NNE, NE wind direction (from land to sea) can not generate waves. But in reality there are quite high wave fields with NE wind direction. This situation can not be simulated by the monochromati c wave theory, but by the spectral method. + The erosion rate for Hai Hau beach is not unique. At present the most severe erosion takes place at Hai Ly - Hai Trieu communes, +The bottom slope is rather gentle ( <0.008) as a results of repeated shore line retreats (more gentler than classical equilibrium profiles). +A small Red river tributary named Vop - Ha-Lan mouth had been closed since 1955. 416 I Pham Van Ninh I, Phan Ngoc \linh, Nguyen Manh flung, Dinh Van Manh + Present eroded rate is approximately lOm/year (with the present of sea dyke sys- tem). It takes 20-30 year for one sea-dyke system retreat. + There is a strong correlation between the seaward advance of Ba Lat mouth and the posit ion of the severest erosion. With the Ba Lat mouth seawards development and formation of underwater sand bars the strongest erosion area moves soutlnvard. Dased on the historical data, four periods of coastline evolution in Hai Hau district have been divided as: * 1912-1935: the coast began to be eroded from the nearest south of Ba Lat mouth to Hai Dong commune; * 1935-1965: the strongest erosion area occurred at the coast from Hai Dong to Hai Trieu communes; Fig. 1. RRD's coastline evolution during 1965-1989 (a) and 1989-2001 (b) Fig. 2. The southward moving of the severest erosion areas in Hai Hau beach * 1965-1990: the strongest erosion area occurred at the coast from Hai Chinh to Ilai Hoa communes; Preliminary results of the study on the reasons of the /lai I/ au ... '117 * 1990-2005: the strongest erosion area occurred at the coast at the coast from Hai Chinh commune to Thinh Long town (Hai Thinh commune) There arc some hypothesizes of the main rneclianism causing the shore line retreat in the area but all of them are designated to wave activities. One of the important theoretical point of the authors is the correlation between the seaward advance of the Ba Lat mouth and the position of the severest erosion in Hai Hau beach. The southward moving of the severest erosion areas arc mapped in the figure 2. It was found out that , with the Bn Lat mouth seawards development and formation of underwater sand bars the strongest erosion area moves southward. This hypothesis will serve as the most important criteria for the explanation of reason of the Hai Hau present state of erosion and for the proposed coastline protection measures at the beach. 2. PRIMARILY EXPLAINATION OF THE COASTLINE EVOLUTION Sediment transport plays an important role in many aspects of coastal, estuarine en- gineering. vVaves, tidal currents and wave-induced currents arc the main forciug of the sediment transport. The sediment transport is responsible for tl1e morphological changes and the shoreline evolution in the coastal zone. However , the understanding of the 111cc11- anism responsible for these phenomena is still not clear. N umcrical rnocleling is one of useful tools for the evaluation of the hydrodynarnic parameters such as the wave height , the currents, the sediment transport , the bottom level change and shoreline evolution for the design of shore protection and environment problems. In order to simulate the evolution of Nam Dinh coast, in the present study, a 2-D numerical model has been developed to simulate the sediment transport, the bottom level change and the coastline evolution under the action of waves. The model includes a total load module which is based on the approach of Watanabe (1988), the advcction-diffusion model which is used to simulate the sediment transport from the river mouths and the morphology and shoreline change modules. The tidal current, the wind-induced current, the wave-induced currents and the waves parameters are supplied as inputs. Tidal and wind drift currents arc obtained by using a 2-D hydrodynamic model based on resolving the full shallow-water equations by a finite differences method developed by Manh and Yanagi (2003). The wave parameters, the most important sea parameters, influencing coastal processes, are obtained by using the wave model STWAVE ( Hung et al. 2006). These matters are not presented here . 2.1. Modeling of the Sediment Transport Rate due to Current (in the presence of wave) The formula of sediment transport rate qc ( 1112 .s- 1) d uc to current (in the presence of wave) proposed by Watanabe (1988) is used: (2.1) The sediment transport rate in x and y direction are as follows: - A ((Tb,cw - Tcr) u). q= - c • pg ( ( Tb, cw - T er) V ) qcy = A c pg (2.2) 418 ~ Van Nfilij, Phan Ngoc Vinh, Nguyen Manh Hung, Dinh Van Manh where Ac is a non dimensional empirical coefficient; lvl is the absolute velocity; p is the fluid density; g is the gravity acceleration; U, V are the components of flow velocity in the x and y directions respectively; Tb,cw is the bed shear stress due to waves and currents, and assumed to be form as linear superposition of the respective wave and current components (Van Rijn, 1989): Tb,cw = Tb,5 + Tb,w, (2.3) Tb,5 is the bed shear stress due to current (modified by wave motion): (2.4) where a is a factor related to the influence of the waves on the bed shear stress. Tb ,c is the bed shear stress due to current alone.Tb,w is the bed shear stress due to waves defined by Van Rijn (1989). Tcr is the critical bed shear stress for incipient motion, determined from the Shield curve for oscillatory flow: Tcr = (s - l)pgd·l/Jcn (2 .5) where d is the diameter of sediment particle; s is the specific gravity; ·l/Jcr is the critical Shields parameter. 2.2. Modeling of the Sediment Transport Rate due to Wave The sediment transport rate due to wave qw (m2 /s) in the direction of wave propaga- tion: F Aw (Tb ,cw - Tcr) Ub qw = D , pg A _ BwWs f1w w - (1 - >.)(s - 1) J(s - l)gd V 2 . The sediment transport rate in x and y direction arc as follows: F Aw ( Tb ,cw - T er) Ub COS() qwx = D ; pg Aw ( Tb, cw - Tcr) Ub sin 0 qwy = FD---------- pg (2.G) (2.7) (2.8) where Bw is a non dimensional coefficient; >. is the void ratio of sediment; W 8 b the fall velocity, estimated by approximate formula of Van Rijn (1993): W = lOv [(O.Ol(s- l)gd 3 ) 0 · 5 - ] s d v2 + 1 1 ' (2.9) where () is the wave angle with respect to the x direction; v is the kinematic viscosity coef- ficie~1t of water (lo- G m2.s-1); FD is the transport direction function defined by Watanabe (1988). 2.3. Sediment Transport Rate due to combined Waves and Currents Sediment transport rate due to combined waves and currents is assumed to be the su111 of the transport rate vectors due to waves and the transport rate vectors due to currents: (2.10) Preliminary results of the study on the reasons of the Hai Hau ... 419 2.4. Advection-Diffusion Modeling The advection-diffusion model is used to simulate the suspended sediment transport from the Ba Lat river mouth taking into account the sediment exchange between the water and the bed. The governing equation is as follows: a(CH) a ( ac) a ( ac) . at +ax CQx - KxH ax + ay CQy - I<yH ay = Fbcd, (2.11) where C is the turbidity, I<x, Ky are the coefficients of the horizontal diffusion. In the general case, Fbed = Fe - Fd is the sediment fluxes exchanged between the water and the bed. Fbed is either erosion or deposit fluxes, depending on the relationship between bed shear stress and bed materials. There does not exist a lot of deposit formulations for fine-sediment transport. The most frequently used one is proposed by Einstein-Krone (Krone, 1962): Fd = WsC (1 - Tb,e), (2.12) Ted where Fd is the deposition rate, Ted is the critical stress for deposition. Ted depends on the turbidity and the currents near the bed. The bed erosion is already included in the formulat ion of Watanabe (1988), so when modeling the advection, the diffusion of suspended sediment, only the deposition of the sediment from river mouths is taken into account, i. e. the erosion flux Fe is set to zero. 2.5. Bed Level Change The conservation of the sediment mass is expressed by: ~~ = l~A [:x (qx-Eslqx l~~) + :Y (qy -Eslqy l~~) +Fbed], (2.13) where Es is an empirical coefficient, dealing with the effect of the bed slop; h is the water depth. 2.6. Shoreline Change Based on the one-line theory, the equation governing shoreline change is as follows (Hanson and Kraus, 1989) : ay + 1 ( aq - Q) = 0 at D l3 + De ax ' (2.14) where y is the shoreline position; DB is the berm elevation; De is the depth of closure; q is the longshore transport rate, Q is the line source or sink of sediment. 2.7. Adaptation of the model to the Nam Dinh coast zone For this application, the study area is the nearshore zone of Nam Dinh province. It covers the lower part of the Ba Lat river mouth, extends alongshore over 60km from the Diem Dien mouth southwards to the Lach Giang mouth and as far as 25km offshore (figure 1). Bathymetric data of the area were digitalised from maps 1/25000 and 1/100000 made by Vietnam People's Navy, published in 2000 and 1984, respectively. Simulation Conditions For the advection-diffusion model, a concentration of 10 mg.1- 1 is imposed for the whole area as the initial condit ion. At the open boundaries, the suspended sediment at 420 I Pham Van Ninh j, Phan Ngoc Vinh, Nguyen Manh /Jung, Dinh Van i\!lanh the river mouth in ebb tide is imposed at l.Og.1- 1 corresponding to a mean river discharge of 4500 m3 .s- 1 for the wet season a:nd 0.2g.1- 1 corresponding to a mean river discharge' of 1500 m3.s- 1 for the dry season (Van Maren et Hoekstra, 2004). Following the approach of Van Maren et Hoekstra (2004), the yearly morphological changes are calculated by simulating half a spring-neap cycle for each wave conditions. and multiply this with a morphological factor which is defined from the frequency of occurrence of each wave class. Summation of the resulting morphological computations yields the yearly morphological change. In this work, calculations arc done with two wave classes with the highest frequencies of occurrence, corresponding to NE and SE winds. The duration of simulation for each wave class and the other statistic data in table 1 arc given by Hung et al (2006). Table 1. Different Cases of Simulations* Computation 'Nave Wave Wave Mean river Duration of Wind heigh Period Dir. p [%] discharge, simulation, velocity / Dir , scenarios t, m d, s deg. m3 .s-1 days m.s - 1 / deg. Case 1 0.81 3.95 180.02 31.36 4500 57.23 4.5/135 (wet season) (SE) Case 2 1.04 3.87 80 .12 37.57 1500 68 .56 G.0 / 45 (dry season) (NE) *Directions of waves and winds are referred to lVIeteorological (Zero from North) Sedimentological Parameters adopted for Simulations The principle sedimcntological parameters are taken as follows: - The median grain diameter D5o= 90µm; - The void ratio >. = 0.4; - The sediment density p 8 =2650 kg.m- 3; - Non-dimensional coefficient for wave Dw = 3; - Non-dimensional coefficient for current Ac= 1; - Critical value for direction of transport rate fl = l; - Effective bed slop coefficient E8 = l. - The horizontal eddy diffusivity for suspended sediment I<x = I<y = lO m2 .s- 1. The critical bed stress for deposition T cd depends on the suspended scdi111C'11t co11- ccutrntion and flows near the bottom. It is obtained from experiments i11 chm111cl. i11 laboratories with sediments of some estuaries, but there is no results co11c:crning the fl11id mud of 13aLat. For reference, some values of Ted arc shown hereafter. Krone ( 1962) 11scd Ted = O.OG N.m- 2 for mud beds in the 13ay of San-Francisco with an initinJ conce11tratio11 of sediments less than 0.3 gl- 1. Mehta (1986) obtained the T crJ is equal to 0.15 N.111 - 2 for kaolin beds , to 0.10 N.m - 2 for mud beds in the 13ay of San-Francisco and to 0.08 N.rn 2 for mud beds in Maraca ibo (Venezuela). Dase on those values and the critical stress vnluc obtained from calculations, the crit ical stress for deposition T ed is chosen at 0.1 N.m - 2 . 2.8. Results and discussions Calculated Net S edimen t Transpor t Preliminary results of the study on the reasons of the Hai lfau .. . 421 The long-shore and cross-shore sediment transports are taken into account in the model. In general, the numerical results (fig. 3) show that the long-shore sed iment trans- port is dominant and considerable within as far as some hundreds of meters offshore. Further offshore, the net sediment transport rates decrease sharply. . In the NE wind the net sediment is mainly transported southwards along the shore. Also, in the SE wind, the net sediment is mainly transported northwards along the shore. The maximum net sediment transport rates reach as much as O.lkg.m- 1 .s- 1 . Wind NE 0.1 ~/ms ~ JI\''''"..---· \\~'\"' \\\''' ~'\- to~U I _)\{km ) Van Ly Hai Thinh 10 20 X(km) 'Wind SE 0 .1 lrtl/ms ~ 0 0.5 1 1.5 2 2.5 3 X(km) Fig. 3. Calculated net sediment transport rates in the NE (left) and SE (right) winds Calculated Bed Level Change - Accretion Caused by the Suspended Sediment from the River Mouth Fig. 4 shows the calculated bottom change caused by the sediment deposition from the Ba Lat river mouth in the NE and SE winds. The sediment from the river mouth is transported by the currents and causes the bed accretion. The numerical results show that the accretion mainly takes place at the north and south areas of t he river mouth, where the sand spits are formed. At the very main course of the river mouth, larger river flovv does not support the bed accretion. Those are in agreement with the observations. Because of the longer duration and the more intense of the NE wind speed compared to the SE wind, in the NE wind the sediment from the river mouth is transported further alongshore, as far as Quat Lam area. The accretion rate in the NE wind is minor. In the NE wind, the maximum rate of bed accretion is 0.1-0.2 m whereas in the SE wind, this value is up to lm. That is caused by the larger suspended concentration, from the Ba Lat mouth in the wet season, supplied to the bed. - Bed Level Change Caused by the Wave Action Figure 5 shows the calculated bottom change in the NE and SE wind seasons. In t be figure, the area filled with dot symbols is in accretion and the area filled with cross :::ymbols is in erosion. In general, we observe that the area in accretion are larger than the area in erosion and the erosion tends to take place closely to the shore. However, the rate of accretion is minor, under 0.1-0.2m. The rate of erosion in the NE wind is more intense 422 I Pham Van Ninh I, Phan Ngoc Vinh, Nguyen Manh Hung, Dinh Van Manh X(km) \ 0·1 \ '" ~ \_ L ,,,,_._,,.-.• . '_•,.'.•,.' '.·.-·Ba'''''···.Lat·,_ •''•'-•.'-'·mou1·.·.••.·.•·_·_,,_, '''h''-•.·.>~- ~""'-._',~-•·.· __ ·.• l:\:/ y[:; ; t5 ·1' /'~~} V.: T ;" ";/ ~ j ,.. -< I c/ . ]'. ~ i /:;:::; i /' !;{ II, l~Lam ~ ! \ ~ I } i M ~{i!,,. 1[~'.~ .,. t.:J0.05 f :oro ~-~ 0.01 II I :r -10.00 ' I -- ,----- , --- - :t . .J. Land t5 . i Dl · CUatlam ~Vanly t'l · X(km) M i:: ~'.}' 0.40 :\t( o.20 -~ .:·.: .,..,_.,0.10 :i;~!oos : . : :- ·. ·: o.ro :0.01 ,o.oo Land Fig. 4. Calculated bottom elevation differences caused by the sediment deposition from the river mouth in the NE wind (left) and SE wind seasons (right) than that in the SE wind. Close to the shore, the rate of erosion can reach up to lm m the NE winds and as much as 0.4m in the SE wind. X(km) X(km) "' "' "' ~~~~ ,• , .. .L:-- ~~"~' c~-~=\~ .. >- '- ~If' ~:$..~:- -:> ' __, 1 '• ~~~ 1 ~~ 1~ ' r ~%r /.~~J 1 _,'.o,<$' i I M ~ ~~ffe ' .I 1!lt&1'' .o ·I\ ;r} \\ ::·:'::' ·.1 [''- io.2 .I I ~ fil:%\,W'' ' ;_~, o.o ~".~::~-=::~- .• .. '. p 0 I ~w 0 0 I B fu.~"t~~ : ~-: . -02 .() -< ®'\f . ' ' - ]'. .. ~~ ?) -0.4 ]'. --~11\~·: I·--. ft~/;<. ~ ~-6 A i@\1I·, . ,·· - %~% o , _;. c::,(;,) ·~o ;~- " .. 9~1 -- ' i -1.0 '. ,.,,~, ··c' _r:::,~ "'~"'&, + fu~~~~: ~1 -:- ,. -w~? ~ ~%--{·-,: '. :t@i ,.,, ' - qi "'Ii·' ,.,~ :;~·-1 /'i: ~~ LnnJ ()'~ \ -~-,, :Irr "q, .~~':~ ;.··r ·· ~''(if\ We --e~·"-"'" !*.~\Tu'@ . ... ~:_;; J ~~:§:::-::.~' . ~ :-: • . • _!. ~. ,. Fig. 5. Calculated bottom elevation differences caused by the wave action and current in the NE wind (left) and SE wind (right) seasons in Van Ly area For a clearer showing of the tendency of bed erosion/ accret ion , figure 6 is made. This figure shows the calculated bottom elevation differences of three profiles in the NE wind PreliminanJ results of the study on the reasons of the Hai Hau . .. 423 season in Van Ly area. In this figure, the minus value of the bottom elevation difference indicates the erosion and the positive value shows the accretion. Due to the sediment transport is considerable within some hundreds of meters offshore, the bed change mainly takes place in this area. In the NE wind, near the shore bed erosion rate can reach up to lm, further offshore there is a tendency of bed accretion or erosion with a minor rate. Over 2 km offshore, the wave action nearly does not cause any considerable effects on the bed level. Elevation differences at 3 profiles in Vanly area 0.4 ' ,_ I .S ! :; I 1' ! ~ i :; i ~ o~-- - - - - - - ~:.....::.--:- - - ~·:, .'> -. ...::::;:::-;:,, 0 I I t -0 4 ' - - - - - - - -0 8 l - - - . - - . --- . --- - I W -1.2 j ---- .. ----·- I o soo 1000 1soo i Distance from the shore (m) 2000 l ______ ·---·- - - -~- --··- ·------- .. Fig. 6. Calculated bottom elevation differences of some profiles in the NE wind season in Van Ly area Calculated Shoreline Evolution Figures 6 shows the calculated shore line evolution with the effects of waves, tide currents, wind-induced currents in 2001 and 1912 in the NE and SE wind seasons. In this figure, in order to see more clearly the shoreline evolution after modeling, the coordinates of the shoreline are reduced to 50 times, and the coordinates differences between the shorelines before and after simulation are kept the same. For the calibration step, the calculations are done in order to obtain that the shore line erosion rate of the year 2001 is as much as 8-lOm per year as observation. Then, the calculations are done with the shore line of the year 1912 for the SE wind season. The calculated results show that the shoreline of the whole interest area does not change much in the SE wind season compared to that in the NE wind (figure 7). In the NE wind, the shore line of the year 2001 is eroded at the south part, from Quat Lam to Hai Thinh, whereas there is a clear tendency of shoreline erosion at the Quat Lam area in 1912. What responsible for this phenomenon should be the convex shape of the shoreline. The land area at the Ba Lat mouth is convex and it plays a role as a groin. It is worth to mention that this shoreline evolution has been calculated for NE and SE monsoons 2001 only (not for other wind directions also) and for NE monsoon 1912 only. The effect of calculated accumulation from the Ba Lat mouth is not added over there and various matters still waiting for the further study. So in comparison with the figure 2 the part from Hai Chinh to Ha Lan river mouth is not the severe erosion part. 3. CONCLUSIONS - A 2-D numerical model of the sediment transport, the bottom level change and the coastline evolution under the action of waves and currents has been developed to simulate the coastal erosion in Nam Dinh province, of which the total load model is based on the approach of Watanabe (1988), the advection-diffusion model is used to simulate the 424 I Pham Van Ninh I, Phan Ngoc Vinh, Ng'Uyen Manh 11'Ung, Dinh Van Manh Shoreline evolution in 2001, Shoreline evolution in 2001, 1400 Wind NE 1400 Wind SE 1200 . 1200 1000 1 1000 800 800 I > 600 i 600 400 400 Van Ly 200 200 Hal Thinh 0 • 0 0 100 200 300 400 0 100 200 300 400 I Fig. 7. Calculated shoreline evolutions in the NE wind (left) and in the SE wind (right) seasons in 2001 sediment transport from the I3a Lat river mouth taking into account the sediment exdw.11g<' bet.ween the water and the bed, and the shoreline evolution is based 011 the one-line theory. - The long-shore and cross-shore sediment transports arc taken into account i11 the model. The dominance of the long-shore sediment transport is shown with the model. The net sediment transport is considerable within as far as some hundreds of meters offshore. The maximum net sediment transport rates reach as much as 0.1 kg.m- 1.s - 1 . - The NE wind season contribution in sediment transport and coast evolution is mucl1 great.er than the SE wind season contribution. - In reality the Ba Lat mouth plays a role of a groin. The more seaward and southward its development this role is more clearly. Preliminarily results of coastal evolution in 2001 and 1912 show a good qualitative agreement with the observation. - The obtained preliminarily results have weaks points they could be caused by: * Not considering the other wind directions, * Not including the water and sediment. discharge of the other river mouths - Ha Lan, Ninh Co and Day. * Not considering the effect of Hoa Binh electric plant lake in the upper basin of the Red River. * Not considering the effects of man activities dykes buicling. Acknowledgements. This work was mainly supported by the project VS /HDE/03 ··The evol'Ulion and sustainable rna.nagernent in the coaslal areas of Vietnmn'' which is provided at the Center for Marine Environment Survey, Research, and Consultation, Institute of Mechanics (2003-2007). Preliminary results of the study on the reasons of the Hai Hau ... 425 REFERENCES 1. H. Hanson , and N. C. Kraus, GENESIS: Generalized model for simulating shoreline chauge, Report 1, Department of the Army US Army Corps of Engineers, Washington, DC 20314-1000 , 1989. 2. N. M. Hung, M. Larson , and D. C. Dien, Modelii1g wave transformation, nearshore currents, and morphology change in the nearshore zone of the Nam Dinh province by C.t\IS-2D, Inter- national vVorkshop on coastal line evolution, Vietnam-Sweden research cooperation program, Project VS/ RDE-30 " The evolution and sustainable management in the coastal areas of Viet- nam", 2006. 3. R. 13. Krone, Flume studies of the transport of sediment in estuarial processes, Final Report, Hydraulic Engineering Laboratory and Sanitary Enginering Research Laboratory, University of California, Berkeley, 1962 . 4. D. V. Manh, T. and Yanagi, Seasonal variation of residual flow in the East sea, Vietnam Journal of Mechanics 25 (3) (2003) 153-169. 5. A . J. Mehta, Characterization of cohesive sediment properties and transport processes in estuaries, Estuarine Cohesive Sediment Dynamics, edited by Mehta, Springer-Verlag, New York, 1986, pp. 290-325. 6. P. Nielsen, Coastal Bottom Boundary Layers and Sediment Transport. World Scientific, 199·1, p. 7. P. V. Ninh, N. M. Hung, L. X. Hong, T . D. Tan , N. T. K. Nga, Overview of the coast line evolution in Viet Nam and especially in the Red River Delta by maps and historica l data, International ivorkshop on coastal line evolution, Vietnam-Sweden research cooperation program, Project VS/ RDE-30 "The evolution and sustainable management in the coastal a reas of Vietnam", 2006. 8. E. Parthenaides, A study of erosion and deposition of cohesive soils in salt water. Ph. D Thesis. University of Ca lifornia, Berkeley, 1962, 182p, 9. D . S. Van Maren, and P. Hoekstra, Seasonal variation of hydrodynamics and sediment dynam- ics in a shallow subtropical estuary: the I3a Lat River , Vietnam , Estuarine, Coastal and Shelf Science 60 (2004) 529-540. 10. L. C. Van Rijn , Handbook Sediment Transport by Currents and Waves , Delft Hydraulics , 1989. 11. L. C. Van Rijn , Principle of Sediment Transport in Rivers, Estuaries and Coastal Seas, Aqua publications, 1993. 12. A. Watanabe, Numerical Model of beach topography changc-Nearshore Dynamics and Coastal Processes, Horikawa Kiyoshi (ed.) University of Tokyo Press, 1988, 522p. Received June 20. 2007. 426 I Pham Van Ninh j, Phan Ngoc Vinh, Nguyen Man.h Hung, Dinh Van Manh c.Ac KET QUA NGHIEN cuu sa B6 vit NGUYEN NH.AN HIEN TUONG x61 LO Ba BiEN HAI H~ u · · T6ng quan ve S\f tien trien duang ba chau th6 song Hong dva vao cac ban do va so li~u lich su da chi ra rang: Tnr&c the ky 20 toan b9 ba bien Nam Dinh chi co boi Jang. Sau do bat dau thai ky ~oi la khu V\l'C huy~n Xuan Thuy va t\.r 1935 - 1965 xoi la t;%p trung m~nh nhat a do~n cua Ha L~n - Hai Trieu, 1965 - 1990 a Hai Chlnh Hai Hoa, 1990 - 2005 a giua xa Hai Chinh - Hai Thjnh. Dong thai ta cling thay ro rang cua Ba L~t ngay cang Joi ra (ve phia bien va phia bac - nam) v&i cuang d9 rat m~nh. , Nhi~m Vl,J. clia mo hlnh boa toan h9c qua trlnh bien d6i ducmg ba Hai H~u tnr&c bet Ia ph<ii mo phong l~i dm;:rc cac moc quan tr9ng do ti'.rc la duang ba xoi v&i cac moc tnr&c 1900, 1935 - 1965; 1965 - 1990; 1990 - 2005 . Day la nhi~m Vl,J. phi'.rc t~p doi hoi nhieu cong SlrC, thai gian. Bai nay trlnh bay mi;it so ket qua ban dau cua vi~c mo phong do (song, dong chay, v<Itn chuyen bun cat va bien d(>ng ba day bien). Ket qua nghien ClrU cho tqay moi tuang quan rat m~nh gitra vung dat Joi ra a cua Ba L~t va chuyen dich ve phia nam cua vung xoi la.