ABSTRACT

The bed change in the Vietnamese Mekong Delta has been increasingly altered caused by natural processes and anthropogenic activities, and it becomes even more complicated under the influence of rising sea levels. At the Tan Chau and Chau Doc stations, the water volume entering the Mekong Delta did not change between 2008 and 2017, but the sediment load decreased by one-third, which caused a significant bed change in the river. This study evaluates riverbed evolution in the Mekong Delta under sediment deficiency and forecasts erosion dynamics until 2030 due to sea-level rise. Results indicate that increased riverbed erosion in 2017 is linked to a 30% drop in sediment supply compared to 2008. Simulations for 2017 indicate a 0.15% decrease in accretion rate—measured as the change in bed elevation—in the upper Tien River compared to 2008, and a 0.5% decrease in the lower reaches. Erosion rates nearly doubled in the upper reaches from Tan Chau to My Thuan, while the lower reaches showed minimal change (0.33%). By 2030, erosion will intensify, especially along the Tien River from Tan Chau to Hong Ngu, reaching 1.6 m/year. Accretion will decrease sharply, with the highest rate at 0.1 m/year near Long Khanh islet.

1 INTRODUCTION

The Mekong Delta, the final stretch of the Mekong River, is a region of immense agricultural potential, characterized by an intricate network of canals and dikes. The pre-dam sediment flow from the Mekong River to the East Sea is estimated at approximately 160 Mt/year (Kondolf et al. 2014; Meade 1996). In recent years, due to the influence of climate change/sea-level rise and human development activities, the construction of upstream dams has exacerbated the situation of sediment deficiency and erosion in rivers (Bussi et al. 2021; Phung et al. 2021; Van Binh et al. 2020a; 2020b).

Regarding dam construction, according to statistics from the Mekong River Commission, seven hydropower dams have either been completed or are currently under construction along the upper reaches of the Mekong River, also known as the Lancang River. In addition, a total of 133 hydropower projects have been constructed, are under construction, or are planned throughout the basin. Several studies have assessed the potential impacts of hydropower development on the Lancang River, particularly with respect to sediment transport (Kameyama et al. 2013; Liu et al. 2013; Walling 2008; Kummu and Varis 2007; Lu and Siew 2006). Notably, Bussi et al. (2021) projects that if all planned dams are built, the suspended sediment load could decrease by up to 50% compared to current levels—currently estimated at 99 million tons per year at the Vietnam-Cambodia border. Such a reduction would have profound consequences for local ecosystems and livelihoods (Bussi et al. 2021; Bravard et al. 2013; Liu et al. 2013). Reservoirs trap both bedload and a substantial portion of suspended sediments, which significantly reduces sediment supply to downstream areas. This disruption in sediment continuity ultimately leads to reduced sediment delivery to the lower Mekong, with implications for river morphology and delta sustainability (Kondolf et al. 2014).

Kummu et al. (2010) estimated that sediment retention could reach up to 85-90%, and Kondolf et al. (2014) estimated up to 96%. The deficiency in sediment, including sand and silt, leads to increased flow velocity during the flood season, causing erosion of the riverbed and heightened risk of bank erosion (Hoai et al. 2019; Kondolf et al. 2014; Bussi et al. 2021). This affects agriculture and the local ecosystem (Baran and Myschowoda 2009). The development of hydropower projects will also alter the flow regime downstream, specifically: dry season flow could increase, while wet season flow is expected to decrease (Piman et al. 2013; Räsänen et al. 2012; Keskinen et al. 2012).

The study by Van Binh et al. (2020b) indicates that sediment has decreased by 74.1% in the Mekong Delta (VMD), with dam construction along the Lancang River being responsible for 40.2% of the decline. On the Tien River, the channel erosion rate increased to around -0.5 m/year in 2014–2017, compared to only -0.16 m/year in the previous period (1998–2014) (Van Binh 2021). Kummu et al. (2008) estimated the bank erosion rate on the main channel of the Mekong River to be between 0.8 and 1.0 m/year (Kummu et al. 2008). In addition to the increasing erosion rate, erosion pits have also formed due to sediment shortages, which is a serious consequence. The study by Van Binh et al. (2022) also emphasized that up to six additional erosion pits are likely to form by the end of 2026 if the suspended sediment supply is reduced by 85% due to dams.

According to data from the Southern Region Hydro-Meteorological Center, Vietnam at the Tan Chau and Chau Doc stations, the water volume entering the Mekong Delta did not change between 2008 and 2017, but the sediment load decreased by one-third. If we analyze the temporal distribution of erosion points and correlate it with water volume and sediment load from the Mekong main stem into the Mekong Delta, a coincidence in the timing of increased erosion points (after 2010) becomes evident. This demonstrates the significant erosion caused by the sediment deficiency in the Mekong Delta river system.

Furthermore, the influence of rising sea levels is also altering the hydrodynamic regime and affecting sediment transport in the region (Mehta 2022; Van Manh 2015). The impacts of sea-level rise on the Mekong Delta have been studied by Smajgl et al. (2015). However, hydraulic and hydrological issues are still not deeply investigated (Smajgl et al. 2015). Wassmann et al. (2004) assessed the impact of sea-level rise on each agricultural region in the Mekong Delta and demonstrated its influence on the hydrodynamic regime in the area. Dang et al. (2018) demonstrated the effects of water infrastructure, sea-level rise, and land subsidence on the Mekong Delta, depending on the hydrological characteristics of each region using a one-dimensional model. The limitation of this study is that it didn't evaluate the impact of these effects on the hydraulic regime of each area (Dang et al. 2018).

Numerous studies have highlighted the compound impacts of dam construction, climate change, and sea-level rise on the hydrological regime and agricultural productivity in the Mekong Delta (Pokhrel et al. 2018; Dang et al. 2018; Lauri et al. 2012; Mekong River Commission 2010). These works have significantly advanced our understanding of the large-scale environmental stressors affecting the Delta. However, there remains a critical knowledge gap concerning how sediment reduction—driven primarily by upstream hydropower development—directly affects riverbed dynamics in the Mekong Delta’s main river channels.

This study seeks to address that gap by focusing specifically on the evolution of riverbed morphology in the Tien and Hau Rivers under conditions of sediment deficiency, with a forward-looking analysis that forecasts erosion trends through to the year 2030 in the context of projected sea-level rise. While sediment dynamics are often mentioned in broader hydrological models, few studies have quantitatively assessed how sediment scarcity alters fluvial processes at the scale. The novelty of this study lies in its integrated approach, combining hydrodynamic modeling, sediment transport analysis, and long-term scenario forecasting to evaluate the impact of reduced sediment input on riverbed evolution. The findings are expected to inform upstream infrastructure planning and provide valuable insights for sediment management strategies and erosion mitigation in downstream areas.

Specifically, this study pursues the following objectives:

• Quantify changes in the flow velocity field between 2008 (baseline scenario) and 2017 (sediment-deficient scenario with sediment input reduced to one-third of the 2008 level), to understand the effect of sediment scarcity on hydrodynamic behaviour;
• Analyze patterns of sediment deposition and erosion in the Tien and Hau Rivers during the same period, identifying key factors such as river morphology, flow regime, and bed material properties that influence sediment transport processes;
• Forecast riverbed changes through 2030 by simulating a scenario where sediment input from the upper Mekong is further reduced to 60% of the 2008 level, in conjunction with a projected sea-level rise of 12 cm. This projection will help anticipate future erosion hotspots and inform risk management efforts.

Through this focused investigation, the study provides a scientific foundation for developing more adaptive and resilient water infrastructure policies in the Mekong Basin, addressing both current and future sediment management challenges.

2 MATERIALS AND METHODS

2.1 Materials

The topographical data in the study were collected from two areas: the riverine terrain data within the research area, and the offshore terrain data. The riverine terrain data within the research area encompass the years from 2008 to 2011 and were gathered from research projects (Bay 2017–2021) and the Departments of Natural Resources & Environment in Mekong Delta provinces. This data comprises 302,775 depth measurement points and elevation data ranging from 0 m to -43 m. The offshore terrain data, extending from the river mouths to the sea, was collected from global terrain datasets Earth Topography (ETOPO) and General Bathymetric Chart of the Oceans (GEBCO), encompassing 894 678 depth measurement points ranging from -1 m to -127 m.

Boundary data, including water level, river discharge, and Total Suspended Solids (TSS) data, were extracted from the MIKE21 FM model (Thuy et al. 2020). The offshore water level boundary data were derived from global water level data using the MIKE toolbox.

Discharge, water level, and TSS data for calibration and model validation were collected from the National Hydro-Meteorological Center and local hydro-meteorological centers. Hourly data from January 1, 2012 to December 31, 2017, were collected from 5 flow stations: Chau Doc, Tan Chau, Can Tho, My Tho, and My Thuan, and 15 water level stations: Chau Doc, Tan Chau, Long Xuyen, Vam Nao, Cao Lanh, My Thuan, Cho Lach, My Tho, My Hoa, Can Tho, Dai Ngai, Vam Kenh, Binh Dai, An Thuan, Ben Trai, Tra Vinh, and Tran De.

The discharge and water level data from the measurement stations were used for calibrating and validating the hydraulic model. Daily TSS data from the 5 stations, Chau Doc, Tan Chau, Can Tho, My Tho, My Thuan, were employed for calibrating and validating the sediment transport model. The locations of these measurement stations are presented in Figure 1.

Figure 1 Hydrologic stations.

2.2 Study area

The research area in the main river system of the Mekong Delta, stretching from the upper reaches of the Tien River at Tan Chau (An Giang province) and the upper reaches of the Hau River at Chau Doc (An Giang province) to the sea. The coastal area is delimited as shown in Figure 2.

Figure 2 Study area.

The study area is divided into two regions: the Upper Reach area, influenced by floodwaters, and the Lower Reach area, influenced by tides. The division point between the two regions is marked by the My Thuan hydrological station on the Tien River and the Can Tho hydrological station on the Hau River.

Due to the significant size of the research area (extending over 100 km in each upper and lower sub-region), for clarity of presentation, the calculation area is further divided into five zones:

• Two zones within the Upper Reach area: U1, U2 (black border). U1 stands for Upper Reach 1, and U2 represents Upper Reach 2.
• Three zones within the Lower Reach area: L3, L4, L5 (yellow border). L3 is Lower Reach 3, L4 is Lower Reach 4, and L5 is Lower Reach 5.

2.3 Methodology

MIKE21 model

The MIKE21 model, developed by DHI Software, is a world-leading software package for simulating flow, waves, sediment transport, morphology, and environmental processes. The user-friendly interface, fast computation, and reliable simulation capabilities have established DHI Software as a significant player in modeling terrestrial, coastal, and offshore processes. MIKE21 is a two-dimensional (2D) model based on a flexible mesh, suitable for high-resolution simulations in areas requiring detailed resolution. It includes modules for advection-dispersion, water quality, heavy metals, eutrophication, sediment transport, and bed morphology changes in river, estuarine, coastal, and marine areas. It has also been expanded to include wave modeling with short wave periods. The modeling system has been extensively applied and has demonstrated high reliability in simulation results across various international studies, including applications in the Mekong Delta (An et al. 2025; Kim et al. 2025; Vu et al. 2024; Le Xuan et al. 2022; Tran Anh et al. 2018).

Morphological Index SI

The Sinuosity Index (SI) is a ratio used to measure the curvature of a river's course. It is calculated as the ratio of the actual length of a river (L) to the straight-line distance between its start and endpoint (L’). SI = L/L’.

2.4 Model set up

Computational mesh

The research group attempted to simulate a 2D model for the entire main river system of the Mekong Delta from the Cambodian border to the sea. However, they encountered certain limitations in the results as follows:

Due to a large number of grid cells (106,428 elements, 66,787 nodes), the computation time was slow.

The smallest grid cell size of 50 m near the shoreline was too large to accurately calculate sediment deposition, resulting in high errors.

The computational area extended over 250 km, and errors accumulated as one moved farther from the border. This was due to the inherent limitations of the 2D model, which does not account for bank overflow. As a result, the accuracy of calculations for flood season in areas like Can Tho and My Thuan within the central part of the computational domain showed significant errors.

In response, the research group decided to divide the Mekong Delta into two sub-regions, delineated by the My Thuan and Can Tho hydrological stations. This division considerably reduced the error in the results due to the mentioned factors. Consequently, they chose to compute the two sub-regions with grid sizes: 48,230 elements, 30,970 nodes for the Upper Reach area, and 74,274 elements, 46,792 nodes for the Lower Reach area.

The riverbed topography and the boundaries for calculations in the Upper Reach and Lower Reach areas are presented in Figure 3 and Figure 4.

Figure 3 Bathymetry and boundaries in Upper Reach area.

Figure 4 Bathymetry and boundaries in Lower Reach area.

BOUNDARY CONDITIONS

Hydrological boundaries

Hourly discharge boundaries (Q1, Q2, Q3, Q4, Q5) and water level boundaries (Z1, Z2) were extracted from the MIKE21 model (Thuy et al. 2020). Data for water levels at offshore boundaries (Z3, Z4, Z5) were collected from January 1, 2008, to December 31, 2008, and from January 1, 2017, to December 31, 2017.

Sediment boundaries

Sediment boundaries at the computation boundaries were extracted from the MIKE21 model (Thuy et al. 2020) for the period from January 1, 2008, to December 31, 2008, and from January 1, 2017, to December 31, 2017.

Geological structure characteristics and grain size distribution

Geological structure characteristics in the research area are described in Table 1 as follows:

Table 1 Parameters for sediment transport model.

Parameter	Symbol	Unit		Value
Bottom layer thickness		m	Layer 1	1
			Layer 2	5
			Layer 3	10
Settling velocity	ws	m/s	ws1	0.05
			ws2	0.05
			ws3	0.09056
Critical stresses of erosion	τce	N/m2		0.25
Critical stresses of deposition	τcd	N/m2		0.005
Erosion speed	E0	kg/m2/s	Layer 1	2×10-6
			Layer 2	5×10-6
			Layer 3	5×10-5
Bottom roughness	kn	m		0.0005

2.5 Modeling scenarios

The calculation scenarios are established for two specific time periods, 2008 and 2017, with changes in various influencing factors including bathymetry, hydrological data, and sediment data. Specifically:

Scenario 1 (SC1) represents the current state, using the bathymetry of the main rivers in the Mekong Delta as of 2017, along with hydrological and sediment boundary conditions from 2017. Analyzing the results of SC1 (2017) will evaluate the evolution of the riverbed under the combined effects of hydraulic, morphological factors, and sediment reduction from the upstream.

Scenario 2 (SC2) represents the past condition, using the bathymetry of the main rivers in the Mekong Delta as of 2008, along with hydrological and sediment boundary conditions from 2008. Analyzing the results of SC2 (2008) will assess the evolution of the riverbed under the influence of hydraulic and morphological factors.

Scenario 3 (SC3) is an assumption that sediment input from the upstream continues to decrease. This represents a worst-case scenario with sediment input to the computation area reduced by 60% compared to 2008 (this value is consistent with historical sediment reduction trends and matches the projected scenario for 2030 by the Mekong River Commission (2010). Hydrological conditions are chosen for a typical year, 2011 (with a 5% frequency). This scenario is also combined with sea level rise in 2030 (a 12 cm increase).

2.6 Model evaluation

Two parameters are adjusted during the calculation process: the Strickler coefficient "M" (which is inversely proportional to the depth) and the phases of water level oscillations at the boundary locations (adjusted to lead or lag compared to the measured positions near the boundary).

Adjustments and validation of model parameters are carried out to ensure that the computational results match the observed data at measurement stations (refer to Figure 1).

Both graphical and numerical methods are employed in the calibration and validation process. Graphs and charts are used to compare simulation results with observed data.

To assess model performance, the Nash-Sutcliffe Efficiency (NSE) and Coefficient of Determination (R²) were employed, following recommendations by Moriasi et al. (2015). NSE is widely used for evaluating continuous simulations, capturing the variance between observed and simulated data. R² complements NSE by quantifying the linear relationship and serving as a benchmark for model fit. Both indicators are effective across different temporal and spatial scales, particularly for streamflow and sediment modeling. Their selection aligns with the study’s focus on hydrodynamic and morphodynamic processes in riverine environments.

The mathematical expressions for NSE and R² are given below:

(1)

(2)

Where:

	=	observed value of the i-th measurement,
	=	simulated value of the i-th measurement, and
	=	the mean of value.

The values of NSE and R² are described in Table 2.

Table 2 Evaluate the accuracy of the model NSE and R² (Moriasi et al. 2015).

Simulation performance	R2	NSE
Very good	0.85 < R2≤1.0	0.8 < NSE≤1.0
Good	0.7 < R2 ≤0.85	0.7 < NSE≤0.8
Moderate	0.5 < R2 ≤ 0.7	0.5 < NSE≤0.7
Below moderate	R2 ≤ 0.5	NSE ≤ 0.5

3 RESULTS

3.1 Modeling verification

The Strickler Coefficient (M) is a parameter primarily used for model calibration. After calibrating the Strickler Coefficient in the Upper Reach, it is obtained linearly with depth, ranging from 25 m^1/3/s to 80 m^1/3/s. Maps of the Strickler Coefficient in the Upper Reach and Lower Reach areas are depicted in Figure 5 and Figure 6.

Figure 5  Strickler Coefficient M in Upper Reach area.

Figure 6  Strickler Coefficient M in Lower Reach area.

The set of parameters for calculating the bed evolution and sediment transport in the Upper Reach and downstream are is presented in detail in Table 2.

3.2 Velocity field 2017 (SC1) and 2008 (SC2)

The specific flow velocities in the five designated regions (Figure 2) within the study area for SC1 and SC2 are presented in Table 3.

Table 3 Flow velocity comparison between SC1 and SC2.

Area	Description	Velocity (m/s)
		SC1	SC2
U1	At the constricted reach on the Tien River passing through Hong Ngu (Dong Thap) and Tan Chau (An Giang)	1.9	1.8
	At the meandering section on the Hau River passing through Phu Tan District (An Giang)	1.4	0.9
	At Vàm Nao, the confluence between the Tien River and the Hau River	1.6	1.2
U2	On the Tien River	1.7	0.1–1.4
	On the Hau River	1.6	0.2–1.0
L3	The junction between the Tien River and the Ham Luong River	0.75	0.7
	On the Co Chien River, at the constricted section passing through Mang Thit District (Vinh Long)	1.8	1.1
	On the river segment passing through Vung Liem District (Vinh Long)	1.0	0.5
L4	On the Tien River segment passing through Ben Tre Province up to Dai estuary	0.55	0.45
	On the meandering section of the Ham Luong River passing through Ben Tre City and Mo Cay District (Ben Tre)	1.4	0.9
L5	The stretch passing through Cau Ke District (Tra Vinh) and Ke Sach District (Soc Trang)	2.6	1.6
	The stretch passing through Long Phu District (Soc Trang) up to Tran De estuary	0.8	0.7
	The stretch passing through Tieu Can District (Tra Vinh) up to Dinh An estuary	1.6	1.0

The comparison results from Table 3 indicate that in SC1, the maximum flow velocities on the rivers reach up to 2.6 m/s in the main channel, while the flow velocities in the channel only reach 1.8 m/s in SC2. Overall, the flow velocities on the Tien River and Hau River are higher in SC1 compared to SC2.

The distribution of velocities in both scenarios share a common characteristic: when the flow passes through the two branches of any given river islet, the flow velocities are not evenly distributed (Table 4). Instead, one branch tends to have higher velocities. For example, the flow velocity on the right branch (viewed from upstream) of Long Khanh islet, Ong Ho islet, Tan Thuan islet (>1.2 m/s) is greater than the left branch (~0.8 m/s), while the flow velocity on the left branch (above 1 m/s) is greater than the right branch (above 0.5 m/s) for Tan Loc islet, Gieng islet, and Tan Phong islet.

Table 4 Flow velocity and discharge in islet in SC1.

No.	Islet	Velocity (m/s)		Discharge (m3/s)
		Right branch	Left branch	Right branch	Left branch
I	Tien River
1	Long Khanh	1.14	0.92	11817	6363
2	Ma	1.07	0.93	8809	12985
3	Gieng	0.64	1.07	1066	13199
4	Tan Thuan	1.34	0.78	15487	400
5	Tan Phong	0.88	1.67	7459	5570
6	Thoi Son	0.51	0.73	3216	4916
7	Tan Phu Đong	0.66	0.69	8602	3777
II	Co Chien River
8	Dai	0.84	0.67	5685	4354
9	Hoa Minh	0.47	0.92	7994	8453
10	Long Tri	0.82	0.91	4300	9570
III	Hau River
11	Vinh Truong	1.33	1.42	1334	3213
12	Ong Ho	1.24	0.84	10157	2248
13	Tan Loc	0.41	1.29	735	17550
14	Phu Thanh (May)	1.38	0.48	16912	1338
15	Nai	1.33	1.56	6943	8703
16	Tan Quy	0.31	0.31	843	230
17	Dung	0.88	1.00	6997	14797

3.3 Accretion and erosion

The simulation results for accretion and erosion after 1 year of computation are presented using a color scale in all scenarios. The color scale ranges from light to dark green (positive values) to indicate increasing sedimentation, and from dark to light red (negative values) to represent decreasing erosion.

Due to the significant impact of sedimentation and erosion on bank erosion and accretion, the simulation results for sedimentation and erosion in the present condition of 2017 (SC1) are compared with the results of computed shoreline changes using remote sensing imagery (right figure). The shoreline changes are represented using a color scale, where shades of green to blue indicate increasing accretion, and colors from yellow, orange, to red indicate increasing erosion.

UPPER REACH

The accretion and erosion processes in the study area after 1 year in SC1, which represents the present condition of 2017, are shown in Figure 7 to Figure 11. The results are presented for the five sub-regions (U1, U2, L3, L4 and L5) within the computational area, as shown in Figure 2.

Accretion and erosion in U1

In the U1 area (Figure 7), it can be observed that erosion occurs more frequently on the Tien River than on the Hau River. The trend of sedimentation and erosion is in good agreement with the analysis of flow velocity patterns in both scenarios.

Figure 7 Erosion and accretion in U1 (A – Accretion, and E – Erosion) with (a) SC1 – 2017, (b) Remote sensing results in marked area (Khoi et al. 2020), and (c) SC2 – 2008.

In the current condition (2017), at the Ong Ho islet area on the Hau River in An Giang province (highlighted in red), the left branch of the islet (viewed from upstream to downstream) tends to experience accretion with a sedimentation level of about 0.1 m/year of computation. On the right branch of the islet, which passes through Long Xuyen city, there is a trend of erosion with a maximum velocity of 0.7 m/year. These results are cross-referenced with the analysis of shoreline changes using remote sensing imagery (Khoi et al. 2020). Locations with sedimentation near the bank cause the riverbank to become steeper, leading to unstable banks prone to erosion and potential bank collapse. These locations correspond well with sections of riverbanks that have experienced erosion based on field surveys and remote sensing analysis. The computed results of the morphological changes in the Ong Ho islet area align well with the results from project KC.08.21/11-15.

The distribution of accretion and erosion in both scenarios is quite similar, with erosion rates on the Tien River higher than on the Hau River. This pattern is observed in narrow sections and curved stretches of the rivers where erosion is more likely to occur. Comparing the two scenarios, the erosion rate in SC1 is higher than in SC2. Specifically, the maximum erosion depth after 1 year of computation is approximately 1 m/year in SC1 and 0.75 m/year in SC2. In narrow sections of the Tien River through Hong Ngu district (Dong Thap province) and Tan Chau town (An Giang province), as well as on the narrow section of the Hau River through Chau Phu and Phu Tan districts (An Giang province), the maximum erosion depth in SC1 is around 1 m/year, greater than SC2 (approximately 0.75 m/year). The sedimentation depth in both SC1 and SC2 is approximately 0.1 m/year.

Accretion and erosion in U2

In the U2 area (Figure 8), due to the decreased flow velocity on the Tien River compared to the U1 area, the extent of erosion in the U2 area on the Tien River is lower than in the U1 area in both scenarios. On the Tien River, the maximum erosion depth after 1 year is approximately 0.7 m/year in SC1 and around 0.5 m/year in SC2. On the Hau River, there is a tendency for more erosion compared to the U1 area, with an erosion rate of 0.5 m/year. This could be attributed to the influence of flow dynamics in this area, as the Vam Nao River serves as a channel that redirects flow from the Tien River. As a result, the distribution of flow velocities in the downstream area of the Vam Nao River is more balanced, leading to higher flow velocities in the downstream section of the Hau River compared to the upstream section.

Figure 8 Erosion and accretion in U2 (A – Accretion, and E – Erosion) with (a) SC1 – 2017, (b) Remote sensing results in marked area (Khoi et al. 2020), and (c) SC2 – 2008.

Accretion and settling processes occur in straight river segments and at the downstream end of the islets. The accretion level in these areas is higher in SC2 compared to SC1. Specifically, accretion fluctuates around 0.5 m/year in SC1, while accretion fluctuates around 0.7 m/year in SC2. In the highlighted red area on the Tien River as it passes through Cao Lanh district (Dong Thap province), it can be observed that the simulated results and computed bank changes are quite consistent, compared to the study of (Khoi et al. 2020). In this section of the river, there is an erosion trend on the left bank, while a slight accretion process occurs at the upstream and downstream ends of the point bar (Cao Lanh).

LOWER REACH

Accretion and erosion in L3

In the L3 area (Figure 9), due to the significant difference in flow velocities between the two scenarios, there is also a difference in the extent of accretion and erosion. In the vicinity of the My Thuan Bridge, the narrow section in Cai Be district (Tien Giang province) on the Tien River, the upstream end of the Tan Phong islet (Tien Giang province), the narrow section of the Co Chien River passing through Vinh Long city, and the right bank of the Cai Cao islet and the upstream end of the Quoi Thien islet in Vinh Long province, there is an erosion trend with a relatively high depth of around 1.6 m/year in SC1 and around 1 m/year in SC2.

Figure 9 Erosion and accretion in L3 (A – Accretion, and E – Erosion) with (a) SC1 – 2017, (b) Remote sensing results in marked area (Khoi et al. 2020), and (c) SC2 – 2008.

It is noticeable that sections of the river with sudden and narrow constriction experience erosion. This is primarily due to the relatively high flow velocities in these areas, reaching about 1.8 m/s, which is significantly higher than the sediment initiation velocity. Additionally, the geological structure in these regions mainly consists of weak soil layers, making erosion occurrences quite reasonable. This observation is well-aligned with the results obtained from the analysis of bank change using satellite imagery, compared to the study of (Khoi et al. 2020).

In straight river segments or at certain locations such as the left bank of the Cai Cao islet, the confluence of the Tien and Ham Luong rivers, and the downstream end of the Thanh Binh islet (Vinh Long), accretion occurs with a slight depth of around 0.1 m/year in both scenarios.

Accretion and erosion in L4

In the L4 area (Figure 10), most areas along the Tien River, Ba Lai River, and Co Chien River in Tien Giang and Ben Tre provinces exhibit a slight accretion trend. The accretion depth is approximately 0.15 m/year in SC1 and around 0.16 m in SC2. These rivers experience relatively low average flow velocities of about 0.2 m/s, which is less than the sediment initiation velocity of the entire region (0.3 m/s).

Figure 10 Erosion and accretion in L4 (A – Accretion, and E – Erosion) with (a) SC1 – 2017, (b) Remote sensing results in marked area (Khoi et al. 2020), and (c) SC2 – 2008.

On the Co Chien River, as it passes through Long Tri islet in Tra Vinh city, the Co islet, and the upstream and left side of the Hoa Minh islet in Chau Thanh district, Tra Vinh province, erosion is observed. The depth of erosion is about 1.2 m/year in SC1, higher than the depth of around 0.9 m/year in SC2.

The results depicted in Figure 10 show that the trends of accretion and erosion in this area align well with the analysis conducted using satellite imagery (as shown in the image on the right) (Khoi et al. 2020).

Accretion and erosion in L5

The simulation results for accretion and erosion processes in the L5 area (Figure 11) after one year reveal a dominant erosion trend in areas with numerous islets on the Hau River, passing through Cau Ke, Tra On (Tra Vinh), Chau Thanh (Hau Giang), and Ke Sach (Soc Trang) districts. Analyzing the simulation results and comparing them with the changes in bank, it can be observed that the segment between the Phu Thanh and Nai islets, the upstream and downstream ends of the Nai islet, experiences erosion with a fluctuation depth of around 1.2 m/year in SC1 and around 0.7 m/year in SC2. Furthermore, erosion is observed at the upstream end of the Dung islet (Soc Trang), primarily affected by tidal forces and sea waves during the windy season. The depth of erosion here is smaller compared to the upstream areas. Accretion predominates in the river segment from Long Phu (Soc Trang) to the Tran De estuary and from Tra Cu (Tra Vinh) to the Dinh An estuary, with accretion depths of around 0.08 m/year in both scenarios.

Figure 11 Erosion and accretion in L5 (A – Accretion, and E – Erosion) with (a) SC1 – 2017, (b) Remote sensing results in marked area (Khoi et al. 2020), and (c) SC2 – 2008.

From the results, it is evident that the extent of erosion in SC1 is greater than in SC2 across most rivers within the research area. It is noticeable that changes in riverbed topography, hydrology, and sediment deposition influence the accretion and erosion conditions in the changing area.

Comparing the erosion patterns between the two computational scenarios, it can be observed that the erosion depth in the research area is higher in SC1 compared to SC2.

The increased occurrence of bed erosion in SC1 is attributed to a reduction of over 30% in the sediment load from the upstream region in 2017 as compared to 2008. Specifically, the suspended sediment flux was 66.54 Mt/year in 2008 and 44.12 Mt/year in 2017 (based on collected data). Furthermore, this area features weak geological formations and lies within a geologically faulted region, which contributes to the severity of the erosion process when the fault boundaries shift.

Currently, alongside socioeconomic development, there is a growing demand for sand for construction purposes, such as residential buildings, industrial zones, tourism, and resorts. This excessive sand mining disrupts riverbed stability and increases the risk of erosion in the Mekong Delta region. In recent years, sand mining activities have been intensifying across all countries within the Mekong River basin, particularly in the downstream area, where sand mining volumes are equivalent to the natural sediment supply. According to a report by WWF (2013), the total average sand mining volume in the entire Mekong River system was approximately 35 mm³ in 2011–2012, with Vietnam accounting for 8 mm³, the riverbed topography has undergone significant changes before and after 2010.

The riverbed topography, hydrology, and sediment conditions within the research area are influenced by both natural and anthropogenic factors. This means that both natural and anthropogenic factors contribute to increased erosion levels in the area, particularly when the combined impact of both factors comes into play. Factors such as climate change, variations in upstream flow rates, fluctuating water levels, tidal regimes, geological formations, hydropower dam construction, sand mining, dredging, infrastructure development, and waterway transportation all exacerbate erosion conditions in the region.

3.4 Evolution of accretion and erosion in SC1 and SC2

Table 5 provides a summary of the accretion and erosion rates obtained from the computational model for both SC1 and SC2 scenarios.

Table 5 Comparison of accretion and erosion rates between SC1 and SC2.

Typical Area	Description	Accretion – Erosion Rate (m/year)
		SC1 (2017)	SC2 (2008)
Meandering river, incised channel, constricted reach	Tien River: Tan Chau, Hong Ngu River (T01): Accretion: Thuong Phuoc 1 Commune, Dong Thap (between 2 islands) Erosion: Tan Chau Town, An Giang Province (incised channel in front of Tan Chau embankment)	+0.8 -1.81	+0.72 -1.31
	Sa Dec, Tien River (T05): Accretion: Binh Thanh Commune, Sa Dec City, Dong Thap Province Erosion: An Hiep Commune, Chau Thanh District, Dong Thap Province	+0.54 -2.14	+0.5 -0.75
	Hau River, constricted reach at National Highway 91 (H3, before the junction of Vam Nao River): Accretion: Binh Thanh Dong Commune, Phu Tan Commune, An Giang Province (downstream of erosion area) Erosion: Binh Thanh Dong Commune, Phu Tan Commune, An Giang Province (constricted erosion area)	+0.04 -1.86	+0.43 -1.84
River confluence	Vam Nao, Hau River (H3): Accretion: Tan Trung, Phu Tan, An Giang Province Erosion: Binh Thuy, Chau Phu, An Giang Province	+0.57 -1.5	+0.21 -1.44
Islet	Long Khanh islet (T01): Accretion: Long Khanh A, Hong Ngu, Dong Thap Province Erosion: Phu Thuan B, Hong Ngu Town, Dong Thap Province	+0.018 -1.46	+0.23 -1.28
	En Island (T3): Accretion: Binh Thanh Commune, Thanh Binh District, Dong Thap Province Erosion: My Hiep, Cao Lanh City, An Giang Province (Tien River)	+0.04 -0.63	+0.3 -0.54
	Cao Lanh Islet (T04): Accretion: Ward 6, Cao Lanh City, Dong Thap Province Erosion: Tan Thuan Dong Island, Tan Thuan Dong, Cao Lanh City, Dong Thap Province (Tien River)	+0.35 -1	+0.34 -0.54
	Ong Ho Islet (H4): Accretion: My Hoa Hung Island, Long Xuyen City, An Giang Province Erosion: Binh Duc Commune, BinhDuc Ward, Long Xuyen City, An Giang Province (Hau River)	+0.81 -1.93	+0.48 -1.85
	Tan Loc Islet (Thot Not, Can Tho City), H05: Accretion: Thuan Hung Ward, Thot Not District, Can Tho City Erosion: Tan Loc Ward, Thot Not District, Can Tho City (Hau River)	+0.68 -1.15	+0.63 -0.84
Lower Reach influenced by tidal effects	T6: Accretion: Phu Phung Commune, Cho Lach District, Ben Tre Province Erosion: Hoa Khanh, Hoa Khanh, Cai Be, Tien Giang Province	+0.3 -1.7	+0.61 -1.11
	T7: Accretion: Ngu Hiep Ward, Cai Lay District, Tien Giang Province Erosion: Thanh Tan Ward, Ben Tre City, Ben Tre Province	+0.3 -0.27	+0.8 -0.13
	Dai and Tieu estuaries T08: Accretion: Phu Tan Commune, Tan Phu Dong Ward, Tien Giang Province Erosion: Binh Thang Commune, Binh Dai Ward, Ben Tre Province	+0.4 -0.04	+0.53 -0.0009
	Tra Vinh, T09: Accretion: My Long Bac Commune, Cau Ngang District, Tra Vinh Province Erosion: Hung My, Chau Thanh, Tra Vinh Province	+0.24 -0.0005	+0.71 -0.0005
	Nai islet (Soc Trang), H7: Accretion: Dai An 1, Dung islet, Soc Trang Province Erosion: Bo De, An Thanh Tay, Dung islet, Soc Trang Province	+0.1 -0.1	+0.22 -0.002
	Tran De and Dinh An estuaries, H7: Accretion: Long Vinh, Duyen Hai, Tra Vinh Province Erosion: An Thanh 3, Cu Lao Dung, Soc Trang Province	+0.04 -0.001	+0.3 -0.0005
Channel and canal region	Ong Chuong canal: Accretion: Long Giang, Cho Moi District, An Giang Province Erosion: DH1, Long Dien B, Cho Moi District, An Giang Province	+0.2 -0.4	+0.2 -0.33
	Long Xuyen – Rach Gia canal	+0.015 -0.06
	Ca Mau Canal (Luong The Tran)	+0.005 -0.02

NOTE: (-) Erosion; (+) Accretion

From the results, it is evident that the erosion level in SC1 is greater than that in SC2 for most rivers within the research area. It can be observed that alterations in riverbed topography, hydrology, and sediment distribution contribute to the phenomenon of accretion and erosion within the changing area. The calculations and analyses of causes and mechanisms mentioned above are consolidated for each region and presented in Figure 13 and Figure 14. The left axis represents the accretion (+) and erosion (-) volume ratios within each area (from T1 to T9 along the Tien River and from H1 to H7 along the Hau River (Figure 12). This is based on the statistical principle of dividing the total accretion or erosion volume by the volume of the riverbed within each area. The solid line represents the accretion and erosion ratios within a one-year simulation period of 2008, a year in which the results were influenced solely by hydraulic and morphological factors of the river (denoted as A_H,M and E_H,M). The dashed line represents the accretion and erosion ratios within a one-year simulation period of 2017, a year in which, in addition to the aforementioned factors, the entire region was influenced by the dynamic hydraulic regime, incorporating the effect of reduced sediment supply by 30% compared to 2008 (denoted as A_Sum and E_Sum). The column charts depict the maximum velocities for each area during 2017 when the water flows downstream (bottom right, positive axis) and when the water flows upstream (top right, negative axis). The river morphology index SI is represented by the shaded yellow background area, with the corresponding axis on the right.

Figure 12 Regions along the Tien and Hau Rivers.

Figure 13 Synthesize the effects of each factor causing accretion and erosion on each region of the Tien River.

Figure 14 Synthesize the effects of each factor causing accretion and erosion on each region of the Hau River.

When analyzing the results from the representations in the figures, several observations can be made:

Morphology of the river (SI = 1 for a straight river, SI > 1 for a more meandering river)

On the Tien River, from Tan Chau to My Thuan Bridge, SI > 1, fluctuating between 1.35 and 1. This area is significantly influenced by the topography, especially at T1 (Tan Chau) and T5 (Sa Đec), where SI > 1.3. Downstream from My Thuan to Vinh Long, the river becomes relatively straight (SI = 1), with only localized bends such as at Cai Be, Vinh Long, and Ben Tre. Bends with SI > 1 often create deeper channels along concave banks and sediment deposition along convex banks.

Compared to the Tien River, the Hau River is relatively straight, with an average SI of about 1.1. However, in the H2 and H3 regions (from Chau Doc to Vam Nao), the SI is generally higher (around 1.2).

Hydraulic regime

The maximum flow velocity during the year (flood season) gradually decreases from 1.8 m/s to 1.2 m/s from Tan Chau to My Thuan (T1 – T5). In the T4 area (Cao Lanh islet), where water is distributed among islands and sandbanks, the flow velocity decreases further to around 0.9 m/s. Conversely, during the dry season, the maximum upstream flow velocity (T1 – T5) is generally less than 0.4 m/s.

In contrast, the maximum downstream flow velocity increases from the upper reaches to the Vam Nao area (from H1 to H3), reaching its peak at around 1.5 m/s. As the river progresses downstream, the maximum flow velocity gradually decreases. Near the river mouth (H7), the maximum downstream and upstream flow velocities are approximately equal, around 0.65 m/s. The upstream flow has a relatively low maximum velocity in the upper reaches (H2 – H4), reaching about 0.2 m/s.

Bed material composition

In the stretch from Tan Chau to Hong Ngu (T1 and part of T2), the predominant bed material is gravelly sand, with a threshold erosion velocity of around 0.36 m/s. From Hong Ngu to Cao Lanh (T2, T3, T4), the bed material consists of fine sand with a threshold erosion velocity of V = 0.18 m/s – 0.2 m/s. In the Sa Dec area, the bed material is sandy mud with a threshold erosion velocity of V = 0.17 m/s – 0.18 m/s. Downstream from Sa Dec to the river mouth, the bed material mainly consists of mud and sand brought in from the sea, forming a mixture of sandy and gravelly mud.

In the Hau River, from the upper reaches to the H5 region (Can Tho area), the bed material consists of gravelly sand combined with mud, with a threshold erosion velocity of around 0.17 m/s – 0.18 m/s. In the lower reaches of the Hau River, the bed material composition is similar to the Tien River, comprised of a mixture of sandy and gravelly mud.

Accretion and erosion ratios

In 2017, under the influence of a 30% reduction in sediment supply compared to 2008. On the Tien River, the upper reaches, affected by floods, experienced a decrease in accretion ratio of about 0.15% compared to 2008. In the lower reaches, the accretion ratio decreased even more (0.5%). The reduction in sediment supply more significantly affected sedimentation in the lower reaches than the upper reaches. Meanwhile, the erosion ratio increased noticeably (almost doubled) in the upper reaches from T1 to T6, while the erosion ratio in the lower reaches (0.33%) did not change significantly. This indicates that the reduction in sediment supply has a substantial impact on erosion in the upper reaches but has less effect on the lower reaches.

On the Hau River, influenced by a sediment deficit, the accretion ratio decreased more significantly from the upper reaches to the river mouth, particularly in the H3 (Vam Nao area) and H7 (river mouth) regions, where the reduction was more than half compared to 2008. When compared to 2008, the erosion ratio increased by over 0.55% in the H1 and H2 areas. Particularly in the H3 and H4 areas, the erosion ratio increased significantly (over 1%). As we move downstream, the impact of sediment deficit on erosion gradually decreases. In the H6 and H7 areas (after Can Tho), the effect of sediment deficit on erosion is negligible (an additional increase of nearly 0.1%).

From the analyses above, a synchronous trend in erosion along the Tien and Hau Rivers can be observed: erosion ratio decreases, and accretion ratio increases. The reduction in sediment supply significantly affects erosion in the upper reaches but has less impact on the lower reaches. Conversely, the accretion ratio is more affected by sediment deficit in the lower reaches than in the upper reaches.

According to Van Binh et al (2020b), the river thalweg in the Tien River, from Tan Chau to My Thuan, has experienced significant incision. The incision rates ranged from -0.03 m/year to -0.93 m/year, with a mean value of -0.33 m/year. The average river thalweg incision was -3.73 m, varying between -0.45 m and -11.37 m, which is equivalent to -1.24 m/year from 2014 to 2017 (Van Binh 2020b). Similarly, this study calculates erosion rates in Tan Chau as 1.81 m/year and up to -2.14 m/year in Sa Dec in 2007, which is significantly higher than the SC2 rates (which were -1.31 m/year in Tan Chau and -0.75 m/year in 2008). These results are consistent with Van Binh et al. (2020a), showing no significant differences. Furthermore, Van Binh (2020b) demonstrated that morphological changes in the Tien River from Tan Chau to My Thuan and the Vam Nao channel between 2014 and 2017 revealed severe riverbed incision in meandering areas. In this study, meandering rivers, incised channels, constricted reaches, and river confluences such as Tien River: Tan Chau – Hong Ngu River, Sa Dec – Tien River, Vam Nao – Hau River also exhibited much higher erosion rates compared to sediment deposition.

3.5 Forecast of erosion and accretion until 2030 (SC3)

Assuming a continuation of the trend observed over the past 10 years, from 2008 (SC2) to 2017 (SC1), with a 30% reduction in sediment supply, it is postulated that sediment supply from the upper sources to the Mekong Delta will continue to decline. In the worst-case scenario, sediment supply to the region is assumed to decrease by 60% compared to 2008 (this aligns with historical sediment reduction trends and is consistent with the 2030 projection of the Mekong River Commission (2010)). Hydrological conditions are based on a typical year in 2011 (with a 5% frequency). Combined with an expected sea level rise of 12 cm by 2030, the calculated results show that erosion will become increasingly severe.

In the SC3 (Figure 15), erosion rates across the entire region increase, ranging from 0.82 m to 1.6 m. Erosion typically occurs in narrow stretches of the river, such as the winding section of the Tien River through Hong Ngu district (Dong Thap province) and Tan Chau town (An Giang province). Areas around My Thuan Bridge, the narrow stretch of Cai Be district (Tien Giang province) on the Tien River, the head of Tan Phong islet (Tien Giang province), the Co Chien River stretch passing through Vinh Long city, and the right bank of Cai Cao islet and the head of Quoi Thien islet in Vinh Long province exhibit high levels of erosion, ranging from 1.1 m to 2.45 m. In straight sections of the river and in some areas like the left bank of Cai Cao islet, the confluence of the Tien and Ham Luong rivers, and the downstream end of Thanh Binh islet (Vinh Long province), sedimentation occurs with a minor degree, approximately 0.05 m.

Figure 15 Erosion and accretion in SC3.

Most areas along the Tien, Ba Lai, and Co Chien rivers in Tien Giang and Ben Tre provinces show a mild sedimentation trend, with sedimentation rates of about 0.08 m. Along the Co Chien River, as it passes through Long Tri islet in Tra Vinh city, Co islet, and the head and left bank of Hoa Minh islet in Chau Thanh district, Tra Vinh province, erosion occurs with levels ranging from 0.56 m to 0.8 m.

On the Hau River, as it passes through Cau Ke and Tra On districts (Tra Vinh province), Chau Thanh district (Hau Giang province), and Ke Sach district (Soc Trang province), the erosion trend is dominant. Simulation results indicate that between Phu Thanh islet (Vinh Long province) and Nai islet (Soc Trang province), as well as at the head and tail of Nai islet, erosion processes occur with fluctuations between 0.23 m and 0.42 m under the hydropower scenario. Additionally, erosion processes occur at the head of Dung islet (Soc Trang province), where impacts are primarily driven by tides and ocean waves during the monsoon season. The erosion level here is lower than in the upper reaches. Sedimentation processes are observed along the river stretch from Long Phu district (Soc Trang province) to the Tran De river mouth, as well as from Tra Cu district (Tra Vinh province) to the Dinh An river mouth, with mild accretion levels in both scenarios, around 0.01 m – 0.03 m.

The analysis results of accretion and erosion levels for unfavorable scenarios in several typical areas within different regions are summarized in Table 6 below.

Table 6 Accretion and erosion levels in SC3.

Description	2030
Erosion occurs in narrow and winding river sections, such as the narrow stretch of the Tien River through Hong Ngu district (Dong Thap), Tan Chau town (An Giang), and the stretch of the Tien River through Long Khanh and Tay islets	-1.6
Sedimentation occurs in straight river sections on the Tien and Hau Rivers	+0.02
Erosion level on the Tien River	-1.1
Concentrated sedimentation in straight river sections and at the downstream ends of islets	+0.06
Along the Hau River, from the right bank of Tan Loc islet, Thot Not district, to Can Tho city	-1
Erosion trend around My Thuan Bridge and the narrow stretch of Cai Be district (Tien Giang) on the Tien River, exhibiting high levels of erosion	-2.45
In straight river sections and specific areas like the left bank of Con Cai Cao, the confluence of the Tien and Ham Luong rivers, and the downstream end of Thanh Binh islet (Vinh Long), sedimentation occurs with a mild degree	+0.05
Areas along the Tien, Ba Lai, and Co Chien rivers in Tien Giang and Ben Tre provinces show a tendency for mild sedimentation	+0.08
Along the Co Chien River, as it passes through Long Tri islet in Tra Vinh city, Co islet, and the head and left bank of Hoa Minh islet (Chau Thanh district, Tra Vinh province), erosion occurs	-0.8
Between Phu Thanh islet (Vinh Long province) and Nai islet (Soc Trang province), at the head and tail of Nai islet, erosion processes occur	-0.42
Sedimentation processes occur along the river stretch from Long Phu district (Soc Trang province) to the Tran De river mouth and in the vicinity of the Dinh An river mouth	+0.03

NOTE: The values indicate the change in sedimentation or erosion levels compared to a baseline scenario in 2030. Positive values represent sedimentation, while negative values represent erosion.

In SC3, the erosion rate across the entire region increases, with the most severe erosion occurring along the narrow stretch of the Tien River from Tan Chau to Hong Ngu, reaching up to 1.6 m/year. Around My Thuan Bridge, the narrow stretch of Cai Be district (Tien Giang province), the head of Tan Phong islet (Tien Giang province), the winding section of the Co Chien River, the right bank of Con Cai Cao, and the head of Con Quoi Thien (Vinh Long province), erosion reaches high levels, up to 2.45 m/year. Along the Co Chien River, as it passes through the Long Tri islet area (Tra Vinh province), Co islet, and the head and left bank of Hoa Minh islet (Chau Thanh district, Tra Vinh province), erosion occurs with levels ranging from 0.56 m/year to 0.8 m/year.

Across the entire region, accretion rates decrease significantly compared to 2017, with the highest sedimentation rate reaching only 0.1 m/year (Long Khanh islet, the downstream end of Cao Lãnh islet). Most areas near the river mouths of the Tien, Ba Lai, and Co Chien rivers in Tien Giang and Ben Tre provinces show a mild sedimentation trend, with sedimentation rates of around 0.08 m/year. Sedimentation processes are observed along the river stretch from Long Phu district (Soc Trang province) to the Tran De river mouth, as well as from Tra Cu district (Tra Vinh province) to the Dinh An river mouth, with mild sedimentation levels ranging from 0.01 m/year to 0.03 m/year.

4 DISCUSSION

The simulation results indicate that in 2017, erosion prevailed, accounting for 71% of the total along the Hau River, while accretion accounted for 21%. This finding is consistent with the study by Brunier et al. (2014), which utilized bathymetric surveys by the Mekong River Commission (MRC) to document channel changes in the Song Hau deep channel between 1998 and 2008 surveys. The survey comparison from Vam Nao to the ocean calculated thalweg erosion in 70% of the channel, against only 12% accretion and 18% (Brunier et al. 2014).

Regarding the Tien River, the simulation results show that erosion accounted for 69%, while accretion was relatively low at about 22%. This outcome aligns with the study by Gruel et al. (2022), where 60% of the surveyed area experienced erosion, while 27% had accretion across the entire Tien River system in 2017 (Gruel et al. 2022). Therefore, the overall erosion and deposition results in this study are in good agreement with previous research.

Furthermore, there has been an increase in erosion in both the Tien and Hau Rivers when comparing data from 2008 and 2017. In 2017, erosion had decreased by more than 30% compared to 2008. The total sediment load in 2008 was 66.54 Mt/year, while in 2017, it was 44.12 Mt/year. This reduction in sediment load is one of the factors contributing to increased erosion in the study area.

Additionally, the impact of climate change, water level fluctuations, tidal regimes, SLR has significantly affected flow regimes and sediment dynamics in the Mekong Delta. According to the Climate Change and Sea-Level Rise Scenarios for Vietnam in 2020, if sea levels rise by 100 cm, the Mekong Delta region, particularly the Cuu Long Delta, faces a high risk of inundation, with Ca Mau and Kien Giang being the provinces most vulnerable (79.62% and 75.68% of their respective areas) (Ministry of Natural Resources and Environment 2020). This could increase the likelihood of coastal erosion and river mouth areas. Climate change scenarios also predict an increase in rainfall during the wet season and a decrease during the dry season, creating water level disparities between the flood season and the dry season. This disparity can increase the risk of erosion in the region. Moreover, increased rainfall during the wet season can lead to higher flow rates, further exacerbating erosion. These factors, combined with the young alluvial plain nature of the Mekong Delta, its soft and weak geological foundation, low resistance to forces, and sensitive coastal geology, contribute to increased erosion susceptibility. Nazneen Aktar's study also demonstrates the complex relationship between climate change and river morphological changes (Aktar 2013). This study predicts that riverbank erosion will increase by 13% and 18% by 2050 and 2100, respectively, due to the assumed 15% and 20% increment in flood discharge for those years.

However, it's important to note that this study has primarily focused on the impact of sea-level rise, and other factors have not been thoroughly investigated. This represents an element of uncertainty in this research.

5 CONCLUSION

It can be observed that the erosion level in SC1 is greater than in SC2 for most of the rivers in the study area. It is evident that changes in riverbed topography, hydrology, and sediment deposition contribute to the sedimentation and erosion dynamics in the study area. The increased riverbed erosion in SC1 can be attributed to a significant decrease of over 30% in sediment and sand supply from the upper source in 2017 compared to 2008.

Simulation results for 2017 indicate that in the upper reaches of the Tien River, there was a decrease in the accretion ratio of approximately 0.15% compared to 2008. In the lower reaches, the accretion ratio decreased even further, by 0.5%. Meanwhile, the erosion ratio increased significantly, nearly doubling, in the upper reaches from Tan Chau to My Thuan. However, the erosion ratio in the lower reaches (0.33%) did not experience significant changes, compared to 2008. This suggests that the reduction in sediment supply has a substantial impact on erosion in the upper reaches but has a lesser effect on the lower reaches of the river.

In 2030, the erosion rate across the entire region will increase, with the most severe erosion occurring along the narrow stretch of the Tien River from Tan Chau to Hong Ngu, reaching up to 1.6 m/year. Around My Thuan Bridge, the narrow stretch of Cai Be district (Tien Giang province), the head of Tan Phong islet (Tien Giang province), the winding section of the Co Chien River, the right bank of Cai Cao islet, and the head of Quoi Thien islet (Vinh Long province), erosion will reach high levels, up to 2.45 m/year. Along the Co Chien River, as it passes through the Long Tri islet area (Tra Vinh province), Co islet, and the head and left bank of Hoa Minh islet (Chau Thanh district, Tra Vinh province), erosion will occur with levels ranging from 0.56 m/year to 0.8 m/year. There is a notable decrease in accretion rates compared to 2017, with the highest accretion rate recorded at only 0.1 m/year (located at Long Khanh islet, the downstream end of Cao Lanh islet). Most areas near the river mouths of the Tien, Ba Lai, and Co Chien rivers in Tien Giang and Ben Tre provinces exhibit a mild accretion trend.

ACKNOWLEDGMENTS

We acknowledge Ho Chi Minh City University of Technology (HCMUT), VNU-HCM, and Nguyen Tat Thanh University, Ho Chi Minh City, Vietnam for supporting this study.

References

1. Aktar, N. 2013. "Impact of climate change on riverbank erosion." International Journal of Sciences: Basic Applied Research 7 (1): 36–42. https://www.gssrr.org/index.php/JournalOfBasicAndApplied/article/view/1255
2. An, N.N., P.V. Hong, N.A. Binh, G.T.P. Thao, L. Van Tinh, N.C. Hanh, and T.T. Tran. 2025. "Spatiotemporal dynamics of suspended sediment in coastal Mekong Delta: A hydrodynamic modelling approach under tropical monsoon climate." Scientific Reports 15 (1): 5851. https://doi.org/10.1038/s41598-025-89111-z
3. Baran, E. and C. Myschowoda. 2009. "Dams and fisheries in the Mekong Basin." Aquatic Ecosystem Health Management 12 (3): 227–234. https://doi.org/10.1080/14634980903149902
4. Bay, N.T. 2017–2021. Research to identify causes, mechanisms and propose feasible technical and economic solutions to reduce erosion, sedimentation for the Mekong river system (2017 – 2020), KHCN-TNB.ĐT/14-19/C10.
5. Bravard, J.-P., M. Goichot, and S. Gaillot. 2013. "Geography of sand and gravel mining in the Lower Mekong River. First survey and impact assessment." EchoGéo 26. https://doi.org/10.4000/echogeo.13659
6. Brunier, G., E.J. Anthony, M. Goichot, M. Provansal, and P. Dussouillez. 2014. "Recent morphological changes in the Mekong and Bassac river channels, Mekong Delta: The marked impact of river-bed mining and implications for delta destabilisation." Geomorphology 224, 177–191. https://doi.org/10.1016/j.geomorph.2014.07.009
7. Bussi, G., S.E. Darby, P.G. Whitehead, L. Jin, S.J. Dadson, H.E. Voepel, G. Vasilopoulos, et al. 2021. "Impact of dams and climate change on suspended sediment flux to the Mekong Delta."  Science of the Total Environment 755, Part 1, 142468. https://doi.org/10.1016/j.scitotenv.2020.142468
8. Dang, T.D., T.A. Cochrane, and M.E. Arias. 2018. "Future hydrological alterations in the Mekong Delta under the impact of water resources development, land subsidence and sea level rise." Journal of Hydrology: Regional Studies 15: 119–133. https://doi.org/10.1016/j.ejrh.2017.12.002
9. Gruel, C.-R., E. Park, A.D. Switzer, S. Kumar, H.L. Ho, S. Kantoush, D. Van Binh, and L. Feng. 2022. "New systematically measured sand mining budget for the Mekong Delta reveals rising trends and significant volume underestimations." International Journal of Applied Earth Observation Geoinformation 108, 102736. https://doi.org/10.1016/j.jag.2022.102736
10. Hoai, H.C., N.T. Bay, D.N. Khoi, and T.N.Q. Nga. 2019. "Analyzing the causes producing the rapidity of river bank erosion in Mekong Delta." Journal of Hydrometeorology 703: 42–50 (in Vietnamese). https://doi.org/10.36335/VNJHM.2019(703).42-50 
11. Kameyama, S., H. Shimazaki, S. Nohara, T. Sato, Y. Fujii, and K. Kudo. 2013. "Hydrological and sediment transport simulation to assess the impact of dam construction in the Mekong River main channel." American Journal of Environmental Sciences 9 (3): 247. http://doi.org/10.3844/ajessp.2013.247.258
12. Keskinen, M., M. Kummu, M. Käkönen, and O. Varis. 2012. "Mekong at the Crossroads: Next Steps for Impact Assessment of Large Dams." Ambio 41: 319–324. https://doi.org/10.1007/s13280-012-0261-x
13. Khoi, D.N., T.D. Dang, L.T. Pham, P.T. Loi, N.T.D. Thuy, N.K. Phung, and N.T. Bay. 2020. "Morphological change assessment from intertidal to river-dominated zones using multiple-satellite imagery: A case study of the Vietnamese Mekong Delta." Regional Studies in Marine Science 34, 101087. https://doi.org/10.1016/j.rsma.2020.101087
14. Kim, T.T., T.N.Q. Nga, N.D.Q. Huy, N.K. Phung, H.C. Hoai, N.T. and Bay. 2025. "The impact of sand mining on the bed morphology of the Tien River, Mekong Delta, Vietnam." Environmental Earth Sciences 84 (4): 1–18. https://doi.org/10.1007/s12665-024-12079-y
15. Kondolf, G.M., Z.K. Rubin, and J. Minear. 2014. "Dams on the Mekong: Cumulative sediment starvation." Water Resources Research 50 (6): 5158–5169. https://doi.org/10.1002/2013WR014651
16. Kummu, M., X.X. Lu, A. Rasphone, J. Sarkkula, and J. Koponen. 2008. "Riverbank changes along the Mekong River: Remote sensing detection in the Vientiane-Nong Khai area." Quaternary International 186 (1): 100–112. https://doi.org/10.1016/j.quaint.2007.10.015
17. Kummu, M., X.X. Lu, J.-J. Wang, and O. Varis. 2010. "Basin-wide sediment trapping efficiency of emerging reservoirs along the Mekong." Geomorphology 119 (3–4): 181–197. https://doi.org/10.1016/j.geomorph.2010.03.018
18. Kummu, M. and O. Varis. 2007. "Sediment-related impacts due to upstream reservoir trapping, the Lower Mekong River." Geomorphology 85 (3–4): 275–293. https://doi.org/10.1016/j.geomorph.2006.03.024
19. Lauri, H., H. de Moel, P.J. Ward, T.A. Räsänen, M. Keskinen, and M. Kummu. 2012. "Future changes in Mekong River hydrology: impact of climate change and reservoir operation on discharge." Hydrology Earth System Sciences 16 (12): 4603–4619. https://doi.org/10.5194/hess-16-4603-2012
20. Le Xuan, T., P. Nguyen Cong, T. Vo Quoc, Q.Q. Tran, D.P. Wright, and D. Tran Anh. 2022. "Multi-scale modelling for hydrodynamic and morphological changes of breakwater in coastal Mekong Delta in Vietnam." Journal of Coastal Conservation 26 (3): 18. https://doi.org/10.1007/s11852-022-00866-3
21. Liu, C., Y. He, E. Des Walling, and J. Wang. 2013. "Changes in the sediment load of the Lancang-Mekong River over the period 1965–2003." Science China Technological Sciences 56, 843–852. https://doi.org/10.1007/s11431-013-5162-0
22. Lu, X.X. and R.Y.L. Siew. 2006. "Water discharge and sediment flux changes over the past decades in the Lower Mekong River: possible impacts of the Chinese dams." Hydrology Earth System Sciences 10 (2): 181–195. https://doi.org/10.5194/hess-10-181-2006
23. Meade, R.H. 1996. "River-sediment inputs to major deltas." In Milliman, J.D., Haq, B.U. (eds) Sea-Level Rise and Coastal Subsidence. Coastal Systems and Continental Margins, vol 2. Springer, Dordrecht. https://doi.org/10.1007/978-94-015-8719-8_4
24. Mehta, A.J. 2022. An Introduction To Hydraulics of Fine Sediment Transport. Advanced Series on Ocean Engineering. World Scientific Europe, United Kingdom. https://doi.org/10.1142/12873
25. Mekong River Commission. 2010. Mekong River Commission: State of the Basin Report 2010. MRC: Vientiane, Laos. https://reliefweb.int/report/world/state-basin-report-2010
26. Ministry of Natural Resources and Environment. 2020. "Climate Change and Sea-Level Rise Scenarios for Vietnam in 2020." Department of Climate Change, Vietnam. https://pilot.dcc.gov.vn/en/publications/scenario-of-climate-change-sea-level-1508
27. Moriasi, D.N., M.W. Gitau, N. Pai, and P. Daggupati. 2015. "Hydrologic and water quality models: Performance measures and evaluation criteria." Transactions of the ASABE 58, 1763–1785. http://doi.org/10.13031/trans.58.10715
28. Phung, D., T. Nguyen-Huy, N.N. Tran, D.N. Tran, S. Nghiem, N.H. Nguyen, T.H. Nguyen, and T. Bennett. 2021. "Hydropower dams, river drought and health effects: A detection and attribution study in the lower Mekong Delta Region." Climate Risk Management 32: 100280. https://doi.org/10.1016/j.crm.2021.100280
29. Piman, T., T. Lennaerts, and P. Southalack. 2013. "Assessment of hydrological changes in the lower Mekong Basin from Basin‐Wide development scenarios." Hydrological Processes 27 (15): 2115–2125. https://doi.org/10.1002/hyp.9764
30. Pokhrel, Y., M. Burbano, J. Roush, H. Kang, V. Sridhar, and D.W. Hyndman. 2018. "A review of the integrated effects of changing climate, land use, and dams on Mekong river hydrology."  Water 10 (3): 266. https://doi.org/10.3390/w10030266
31. Räsänen, T.A., J. Koponen, H. Lauri, and M. Kummu. 2012. "Downstream hydrological impacts of hydropower development in the Upper Mekong Basin." Water Resources Management 26, 3495–3513. https://doi.org/10.1007/s11269-012-0087-0
32. Smajgl, A., T.Q. Toan, D.K. Nhan, J. Ward, N. Trung, L. Tri, V. Tri, and P. Vu. 2015. "Responding to rising sea levels in the Mekong Delta." Nature Climate Change 5 (2): 167–174. https://doi.org/10.1038/nclimate2469
33. Thuy, N.T D., D.N. Khoi, D.T. Nhan, T.N.Q. Nga, N.T. Bay, and N.K. Phung. 2020. Modelling Accresion and Erosion Processes in the Bassac and Mekong Rivers of the Vietnamese Mekong Delta. APAC 2019: Proceedings of the 10th International Conference on Asian and Pacific Coasts, 2019, Hanoi, Vietnam.
34. Tran Anh, D., L.P. Hoang, M.D. Bui, and P. Rutschmann. 2018. "Simulating future flows and salinity intrusion using combined one-and two-dimensional hydrodynamic modelling—The case of Hau River, Vietnamese Mekong Delta." Water 10 (7): 897. https://doi.org/10.3390/w10070897
35. Van Binh, D., S.A. Kantoush, M. Saber, N.P. Mai, S. Maskey, D.T. Phong, and T. Sumi. 2020a. "Long-term alterations of flow regimes of the Mekong River and adaptation strategies for the Vietnamese Mekong Delta." Journal of Hydrology: Regional Studies 32, 100742. https://doi.org/10.1016/j.ejrh.2020.100742
36. Van Binh, D., S. Kantoush, and T. Sumi. 2020b. "Changes to long-term discharge and sediment loads in the Vietnamese Mekong Delta caused by upstream dams." Geomorphology 353, 107011. https://doi.org/10.1016/j.geomorph.2019.107011
37. Van Binh, D., S.A. Kantoush, T. Sumi, N.P. Mai, T.A. Ngoc, L.V. Trung, and T.D. An. 2021. "Effects of riverbed incision on the hydrology of the Vietnamese Mekong Delta." Hydrological Processes 35 (2): e14030.  https://doi.org/10.1002/hyp.14030
38. Van Binh, D., S.A. Kantoush, R. Ata, P. Tassi, T.V. Nguyen, J. Lepesqueur, K.E.K. Abderrezzak, et al. 2022. "Hydrodynamics, sediment transport, and morphodynamics in the Vietnamese Mekong Delta: Field study and numerical modelling". Geomorphology 413, 108368. https://doi.org/10.1016/j.geomorph.2022.108368
39. Van Manh, N., N.V. Dung, N.N. Hung, M. Kummu, B. Merz, and H. Apel. 2015. "Future sediment dynamics in the Mekong Delta floodplains: Impacts of hydropower development, climate change and sea level rise." Global Planetary Change 127, 22–33. https://doi.org/10.1016/j.gloplacha.2015.01.001
40. Walling, D.E. 2008. "The changing sediment load of the Mekong River." Ambio 37 (3): 150–157. https://www.jstor.org/stable/i25547872
41. Wassmann, R., N.X. Hien, C.T. Hoanh, and T.P. Tuong. 2004. "Sea level rise affecting the Vietnamese Mekong Delta: water elevation in the flood season and implications for rice production." Climatic Change 66 (1–2): 89–107. https://doi.org/10.1023/B:CLIM.0000043144.69736.b7
42. Vu, M.T., C. Luu, D.Q. Bui, Q.H. Vu, and M.Q. Pham. 2024. "Simulation of hydrodynamic changes and salinity intrusion in the lower Vietnamese Mekong Delta under climate change-induced sea level rise and upstream river discharge." Regional Studies in Marine Science 78, 103749. https://doi.org/10.1016/j.rsma.2024.103749

APPENDICES

A1 Calibration and validation of hydraulic model

The hydraulic model was calibrated over a period of 30 days from 0:01 am on April 1, 2012, to 11:00 pm on April 30, 2012, and validated over a period of 30 days from 0:01 am on October 1, 2017, to 11:00 pm on October 30, 2017, at 5 flow measurement stations and 15 water level measurement stations. The calibration process primarily involved adjusting the Manning's coefficient (M) to ensure that the simulated flow rates and water levels matched the observed data. Following calibration, the Manning's coefficient was linearly interpolated based on water depth, ranging from 25 m^1/3/s to 80 m^1/3/s in the Upper Reach and from 20 m^1/3/s to 70 m^1/3/s in the Lower Reach.

The comparison results between observed flow rates and simulated values during the calibration and validation periods are presented in Figure 16. Similarly, the comparison results between observed water levels and simulated values at the measurement stations are shown in Figure 17.

Figure A1 Results of discharge calibration and validation.

Figure A2 Results of water level calibration and validation.

Using NSE and R2 indices to evaluate model performance, simulation results of the indices at measurement stations presented in Table 7.

Table A1 Calibration and validation results.

Station		Calibration (01 – 30/04/2012)		Validation (01 - 30/10/2017)
		NSE	R2	NSE	R2
Discharge	Tan Chau	0.84	0.85	0.84	0.85
	Chau Doc	0.87	0.87	0.81	0.83
	Vam Nao	0.83	0.86	0.80	0.82
	Can Tho	0.83	0.85	0.83	0.84
	My Thuan	0.85	0.86	0.84	0.86
Water level	Tan Chau	0.87	0.88	0.88	0.88
	Chau Doc	0.90	0.90	0.89	0.90
	Vam Nao	0.88	0.90	0.86	0.87
	Can Tho	0.91	0.94	0.89	0.90
	My Thuan	0.91	0.92	0.86	0.87
	Long Xuyen	0.91	0.92	0.92	0.92
	Cao Lanh	0.91	0.92	0.86	0.88
	My Tho	0.91	0.91	0.92	0.91
	Dai Ngai	0.94	0.93	0.92	0.93
	Vam Kenh	0.91	0.93	0.92	0.93
	Binh Dai	0.94	0.94	0.92	0.93
	An Thuan	0.93	0.93	0.92	0.93
	Ben Trai	0.90	0.91	0.92	0.91
	Tra Vinh	0.91	0.92	0.90	0.92
	Tran De	0.91	0.92	0.92	0.92

The evaluation results of the correlation between observed and simulated values using the NSE and R2 indices show that discharge observed and simulated at the Chau Doc, Tan Chau, Vam Nao, Can Tho, and My Thuan stations exhibit strong correlation, with NSE > 0.8 and R2 > 0.8 during both calibration and validation periods.

The calibration and validation results of water levels at the 15 measurement stations are also quite satisfactory, particularly in terms of water level fluctuations. The correlation between observed and simulated water levels is high, with NSE > 0.85 and R² > 0.85 at all measurement stations (Moriasi et al. 2015). The achieved calibration and validation results demonstrate the capability of the MIKE 21FM HD model to accurately simulate flow in the study area.

A2 Calibration and validation of sediment transport model

Data on sediment concentration at measurement stations were collected twice daily during both rising and falling tide conditions. The calculated results were then compared with the average sediment concentration values per day from April 1, 2012, to April 30, 2012, for the calibration period, and from October 1, 2017, to October 30, 2017, for the validation period. The calibration and validation results of sediment concentration at the Tan Chau, Chau Doc, Vam Nao, Can Tho, and My Thuan stations are depicted in Figure 18, respectively.

Figure A3 Results of sediment concentration calibration and validation.

The study employs the Nash-Sutcliffe Efficiency (NSE) and coefficient of determination (R2) to assess the effectiveness of the model. The calculated indices for sediment concentration at measurement stations are presented in Table 8.

Table A2 Results of calibration and verification of the sediment transport model.

Station	Calibration (01 – 30/04/2012)		Validation (01 - 30/10/2017)
	NSE	R2	NSE	R2
Tan Chau	0.8	0.83	0.75	0.81
Chau Doc	0.76	0.82	0.79	0.82
Vam Nao	0.8	0.84	0.79	0.8
Can Tho	0.84	0.86	0.8	0.82
My Thuan	0.74	0.81	0.82	0.84

The computation results reveal that the observed and simulated sediment concentrations at the Tan Chau, Vam Nao, Can Tho, and My Thuan stations achieve a relatively high level of agreement, with NSE values above 0.74 and R2 values above 0.81.

This demonstrates that the parameter set utilized in the sediment transport model is suitable, ensuring reliability for simulating variations in sediment and bed conditions within the study area.

The calibration and validation results of the hydraulic model (MIKE 21FM HD) and the sediment transport model (MIKE 21FM MT) during the dry season (April 2012) and the rainy season (October 2017) demonstrate that the hydraulic and sediment transport parameter sets have achieved a level of reliability suitable for conducting hydraulic simulations, sediment transport modeling, and analyzing sediment dynamics within the study area over an extended period. The models were assessed under various computational scenarios for different time periods to evaluate the influence of natural factors and human activities on the hydraulic regime, sediment transport processes, as well as sedimentation and erosion dynamics along the main river system in the Mekong Delta region.

This comprehensive analysis contributes to a better understanding of the complex interactions between natural processes and human interventions in the study area. The reliable models provide valuable insights into hydraulic behavior, sediment transport patterns, and morphological changes, facilitating informed decision-making for water resources management, flood risk assessment, and sustainable development in the Delta region.