Assessing River Corridor Stability and Erosion Dynamics in the Mekong Delta: Implications for Sustainable Management

Abstract

This study assessed riverbank erosion and stability along the Mekong and Bassac Rivers to propose safe river corridors and mitigate erosion risks in the Mekong Delta. Using Landsat imagery (2000–2023), field surveys, and numerical simulations, we identified severe erosion hotspots, where erosion rates reach up to 40 m annually, in the meandering sections of the Mekong River,. In contrast, the Bassac River exhibited significant sedimentation, though this trend was diminishing due to upstream sediment deficits caused by hydropower dams. Stability assessments revealed optimal safety corridor distances ranging from 20 to 38 m, influenced by local geotechnical conditions and structural loads. A significant proportion of riverbanks in Dong Thap (88%) and An Giang (48%) do not comply with conservation standards, exacerbating erosion risks and threatening infrastructure. The results of this study highlight the urgent need for enforcing conservation regulations, implementing nature-based solutions like riparian buffers, and adopting sustainable land-use planning. By addressing the interplay between natural processes and anthropogenic pressures, these findings offer actionable insights to enhance riverbank stability, protect ecosystems, and sustain livelihoods in the Mekong Delta amidst growing environmental challenges.

1. Introduction

Rivers are inherently dynamic systems that shape landscapes and sustain both human and ecological communities, yet they also pose hazards such as bank erosion and flooding [1]. Nowhere is this dual character more evident than in the Mekong Delta, where fertile floodplains coexist with rapidly retreating riverbanks. Recent estimates indicate the Vietnamese Mekong Delta (VMD) loses approximately 500 hectares of land to riverbank erosion each year [2], jeopardizing homes, infrastructure, and livelihoods. These losses result from a combination of natural forces and human interventions—such as seasonal flooding, channel migration, sediment trapping by upstream hydropower dams, and extensive sand mining [2,3,4]. This context underscores the need for sustainable river corridor management strategies that balance economic development with the river’s natural dynamics.

Recent advances in remote sensing and numerical modeling have significantly improved our ability to assess river corridor stability and erosion dynamics. Remote sensing has emerged as a powerful tool, allowing monitoring of channel migration and bank line changes over large spatial scales and multidecade periods [5,6]. Multi-temporal satellite imagery analyses have revealed alarming trends in bank retreat, with some Mekong River banks retreating up to 40 m per year [7,8]. Such studies demonstrate that remote sensing, including the use of high-resolution imagery and unmanned aerial vehicles (UAVs), can efficiently quantify erosion dynamics [5,9].

Numerical modeling and field assessments provide a complementary, process-based understanding of riverbank instability. Hydrodynamic models, combined with bank erosion algorithms, simulate flow events and resultant bank retreat [10,11]. Field investigations—including erosion pin studies, bank profiling, and geotechnical testing—further validate these numerical models [12,13]. These integrated methodologies enable detailed assessments of bank stability under varying hydrological conditions and inform strategies for effective river management.

Beyond their ecological and hydrological value, Vietnamese rivers are deeply significant culturally and historically. Rivers such as the Mekong and Bassac are vital to local communities, playing integral roles in traditional customs, festivals, commerce, and daily life. Riverbanks serve as key cultural landscapes, supporting activities ranging from floating markets and traditional fisheries to religious rituals, all of which form an essential aspect of local identity and heritage [14,15]. Historically, river corridors have been central to transportation and communication networks, facilitating social cohesion and economic prosperity in the region [16].

Despite regulatory emphasis on water source protection corridors in Vietnam, exemplified by Ho Chi Minh City’s Decision 22/2017/QD-UBND [17], significant challenges persist. This decision regulates the management and use of land corridors along rivers, streams, canals, ditches, and public lakes within the city, outlining specific buffer widths:

• Class I and II waterways: 50 m from the riverbank edge on each side.
• Class III and IV waterways: 30 m from the riverbank edge on each side.
• Class V and VI waterways: 20 m from the riverbank edge on each side.
• Drainage canals and ditches:
  • Waterways wider than 15 m: 10 m from the riverbank edge on each side.
  • Waterways between 5 and 15 m wide: 7 m from the riverbank edge on each side.
  • Waterways narrower than 5 m: 5 m from the riverbank edge on each side.
  • Waterways narrower than 5 m with embankments: 3 m from the riverbank edge on each side.

However, despite these regulations, there remains a notable lack of comprehensive scientific research guiding the definition and management of these corridors, particularly in the context of local hydrological and geomorphological conditions [18,19]. The current regulatory frameworks are largely reactive rather than proactive, often responding to erosion impacts rather than mitigating them through preventive planning.

Globally, river management increasingly advocates for proactive strategies such as the delineation of erodible river corridors, which allow for natural river processes within predefined spatial boundaries [20,21]. These corridors aim to accommodate the dynamic behaviors of rivers, such as lateral migration and sediment deposition, without significant harm to human settlements or economic activities. Approaches integrating ecological engineering, also known as nature-based solutions (NBS), such as establishing riparian vegetation buffers, bioengineering techniques, and mangrove restoration, have been demonstrated to effect widely mitigate erosion risks while enhancing ecological integrity [22,23]. NBS interventions not only stabilize banks but also enhance biodiversity, carbon sequestration, and the resilience of riverine ecosystems to climate variability and extreme events [24]. In contrast, Vietnamese river management practices have historically relied heavily on conventional structural solutions such as concrete embankments and revetments [25,26]. Although effective in the short term, these rigid structures often exacerbate erosion downstream or across river reaches and disrupt the natural ecological processes essential to maintaining river corridor health [27,28]. Recognizing these limitations, there is a growing urgency for Vietnam to transition toward incorporating nature-based and adaptive management strategies into official policy frameworks [29].

Given these complexities, robust scientific studies are imperative to inform practical corridor delineation and management approaches. Such studies must encompass comprehensive spatial and temporal analyses of river dynamics, sediment transport processes, and geomorphological changes, integrated with socio-economic assessments to ensure the sustainability and effectiveness of the proposed interventions [30,31,32].

The present study aimed to address these critical gaps by providing an integrated assessment of river corridor stability and erosion dynamics in the VMD. By combining remote sensing analysis, numerical simulations, and extensive field surveys, we identified erosion hotspots, assessed current safety standards for river corridors, and proposed scientifically grounded delineations of river corridor widths. It is expected that these insights will significantly enhance sustainable land use planning, inform policy decisions, and facilitate the implementation of proactive, adaptive management strategies and nature-based solutions. Ultimately, this study contributes not only to the sustainable management of river corridors in the Mekong Delta but also provides valuable insights applicable to similar deltaic regions worldwide, facing analogous environmental pressures.

2. Materials and Methods

2.1. Study Area

This study was conducted along key stretches of the Mekong and Bassac Rivers within the VMD, one of Southeast Asia’s most dynamic and densely populated deltaic regions. The Mekong and Bassac Rivers are the two primary distributaries of the Mekong River system, which supports agriculture, fisheries, transportation, and livelihoods for over 17 million people in southern Vietnam [2,14,16].

Eight representative locations were selected across five provinces—Dong Thap, Vinh Long, An Giang, Can Tho, and Soc Trang—based on observed erosion risks, land use pressures, and geomorphological characteristics. These sites capture both meandering and relatively straight river segments, enabling a comparative assessment of erosion dynamics and corridor stability under varied physical and anthropogenic conditions.

The Mekong River in this region exhibits highly sinuous channel patterns, particularly in Dong Thap and An Giang, where outer-bank erosion and inner-bank deposition are common [7]. In contrast, the Bassac River, flowing through Can Tho and Soc Trang, is more linear and exhibits comparatively lower erosion but higher sediment deposition—though both trends are shifting due to upstream sediment trapping by hydropower dams and increased bank encroachment from urban and infrastructural development [3,4,25]. These differences make the region a valuable case study for developing spatially differentiated river corridor management strategies [2].

This study followed a structured methodological framework (Figure 1), ensuring consistency in data collection, analysis, and interpretation. This stepwise approach provided a systematic and structured process for data collection, analysis, and interpretation, ensuring methodological rigor and consistency throughout the study.

First, remote sensing images were collected and used to extract the riverbank. These riverbank lines, which were downloaded and analyzed, served as input data for the DSAS model to assess the erosion/deposition changes at various locations along the Mekong and Bassac Rivers. Primary data such as water levels and soil properties were collected and cross-sections were surveyed. The current corridor conditions were checked based on the regulations of the Vietnamese Government [33] and provinces [17,34,35]. The stability of the riverbank was then assessed at different locations along the two rivers, taking general safety of the river corridor into account. Finally, the river safety corridor was proposed for the eight representative locations along the two rivers.

2.2. Remote Sensing Image Analysis

This study used Landsat 5, Landsat 8, and Landsat 9 imagery from the United States Geological Survey (USGS). The Landsat scene corresponding to Path 125/Row 053 was used. This scene covers a region in southern Vietnam, including parts of the Mekong Delta. Each Landsat scene typically spanned an area of approximately 170 km in the east–west direction and 183 km in the north–south direction. Selection criteria prioritized image clarity and minimal cloud cover to ensure accuracy and reliability. Initially, the study aimed to use images with less than 10% cloud coverage [5]. However, due to limitations of available data, some images with cloud coverage of up to 30% were used, provided that the clouds did not affect the study area. The data are summarized in

Table 1. Detailed information on satellite imagery collected from USGS.

Date (dd/mm/yyyy)	Satellite	Sensors	Resolution (m × m/Pixel)	Coordinates	Cloud Coverage (%)
19/03/2000	Landsat 5	TM	30	UTM	38.00
28/04/2005	Landsat 5	TM	30	UTM	1.00
18/05/2010	Landsat 5	TM	30	UTM	11.00
24/01/2015	Landsat 8	OLI/TIRS	30	UTM	1.55
06/01/2020	Landsat 8	OLI/TIRS	30	UTM	2.18
16/12/2023	Landsat 9	OLI/TIRS	30	UTM	1.83

.

In this study, the processing of Landsat data involved the application of the Normalized Difference Water Index (NDWI) to facilitate the classification of water bodies and shoreline/riverbank areas. The NDWI is a widely used spectral index that leverages the near-infrared (NIR) and green (GREEN) spectral bands to enhance the detectability of water features. The index values typically range from −1 to +1, with higher values indicating the presence of water bodies and lower values corresponding to terrestrial features such as soil and vegetation. This classification approach enabled the extraction of coastline and riverbank features in vector format, providing a spatially explicit representation of these features for further analysis. The use of NDWI in this context enhances the accuracy of water boundary delineation, which is critical for hydrological assessments and river corridor management. When the NDWI value is 0 or higher, the pixel is considered classified as water. The NDWI was calculated using the following formula [9]:

$$NDWI={\displaystyle \frac{GREEN-NIR}{GREEN+NIR}}$$

(1)

where on images from Landsat 5, channel 2 (B02) corresponds to the green (GREEN) band and channel 4 (B04) corresponds to the near-infrared (NIR) band. For Landsat 8, the near-infrared and green bands are designated as channel 5 (B05) and channel 3 (B03), respectively.

2.3. Assessment of Shoreline and Riverbank Dynamics

One of the most common methods for calculating the shoreline change rate is the Linear Regression Rate (LRR) method. This method is based on statistical principles that use all available data points in the shoreline position time series. The slope of the regression line represents the rate of shoreline change. The formula for the regression line is expressed as follows:

$$y=a.t+b$$

(2)

where y is the shoreline position (m), a is the slope of the regression line and represents the rate of shoreline change velocity (m/year), t is the time (years), and b is the y-intercept of the regression line. At least two shoreline positions from different time points are required to calculate the change rate using DSAS, which provides several statistical calculation methods. The most frequently cited methods are the EPR (End Point Rate) method and the LRR (Linear Regression Rate) method [6]. The LRR method was selected over the EPR method because it incorporates all available shoreline positions across the study period (2000–2023), providing a more statistically robust estimate of average change rates by reducing the influence of short-term variability and measurement errors.

2.4. Determination of River Safety Corridor

Google Earth Pro in combination with Legally-Specified Conservation Regions (LSCR) was used to determine the length that was not following the LSCR along the two rivers. According to the research results of GEF-ICRSL on riverbank stabilization and the erosion measures for the Chau Ma islet area under the Technical Assistance Project for the Integrated Climate Resilience and Sustainable Livelihoods in the Mekong Delta [36], it was found that the LSCR is greater or equal to 15 m when the river depth (h) ≤ 10 m; LSRC ≥ 20 m when 10 m < h ≤ 15 m; and LSCR ≥ 30 m when h ≥ 15 m. In addition, the LSCRs according to Decision No. 16/2013/QĐ-UBND [34] and Decision No. 47/2015/QD-UBND [35], regulations on the scope of protection for irrigation works and dikes in Soc Trang Province in 2015 concerning buried canals and rivers, are listed in

Table 2. Regulatory framework for safety corridors in Hau Giang.

Width of River (m)	LSCR (m) Both Sides	Sources
>100	25	Decision No. 16/2013/QD-UBND [34]
70 < a ≤ 100	25
50 < a ≤ 70	20
30 < a ≤ 50	15
10 < a ≤ 30	10
>100	25
≥25	20	Decision No. 47/2015/QD-UBND [35]
15 < a ≤ 25	15
8 < a ≤ 15	10
Depth of river (m)	LSCR (m) both sides
≤10	≤15	GEF-ICRSL [36]
10 < a ≤ 15	≥20
≥25	≥30

.

2.5. Riverbank Stability Assessment

The survey points located in the provinces of An Giang, Can Tho, and Soc Trang are presented in

Table 3. Survey locations.

Location	Region
1.	Phu Thuan B Commune, Hong Ngu District, Dong Thap Province
2.	Tan Khanh Trung Commune, Lap Vo District, Dong Thap Province
3.	Tan Hoa Ward, Vinh Long City, Vinh Long Province
4.	Binh Thuy Commune, Chau Phu District, An Giang Province
5.	Long Giang Commune, Cho Moi District, An Giang Province
6.	Thoi Thuan Commune, Thot Not District, Can Tho City
7.	Bui Huu Nghia Ward, Binh Thuy District, Can Tho City
8.	Dai Ngai Town, Long Phu District, Soc Trang Province

and Figure 2. At each point, the cross-sections were measured around 30 m from the riverbank (for drawing the slope), and the current river corridors (from riverbank to the houses/roads) were also measured (for estimating the surcharge loads). These data were then used for stability assessment (Equation (3)).

Maximum and minimum water levels corresponding to 10% and 90% exceedance probabilities were estimated using annual peak and low water level time series (2000–2022) from five stations in the study area. The Gumbel distribution was applied to fit the data and calculate the respective design values, representing high-water and low-water scenarios used in slope stability simulations.

The stability analysis was carried out with two cases: maximum water level with a 10% frequency and minimum water level with a 90% frequency (for the period 2000–2022) (Figure 3 and Table 4).

• In Dong Thap (Cao Lanh station): the minimum water level is at an elevation of −0.15 m, and the maximum water level is at an elevation of +2.2 m;
• In Vinh Long (My Thuan station): the minimum water level is at an elevation of −0.6 m, and the maximum water level is at an elevation of +1.3 m;
• In An Giang (Long Xuyen station): the minimum water level is at an elevation of −0.55 m, and the maximum water level is at an elevation of +2.4 m;
• In Can Tho (Can Tho station): the minimum water level is at an elevation of −0.2 m, and the maximum water level is at an elevation of +1.4 m;
• In Soc Trang (Đai Ngai station): the minimum water level is at an elevation of −0.8 m, and the maximum water level is at an elevation of +0.6 m.

Figure 3. Illustration of the simulation cases in An Giang.

Figure 3. Illustration of the simulation cases in An Giang.

Table 4. Summary of the simulation cases for the survey locations with two water level cases.

Case	No Surcharge Load	Surcharge Loads of Houses and Roads	Consideration of LSCR	No LSCR Consideration	Proposed LSCR
1	✓
2		✓	✓
3		✓		✓
4		✓			✓

Table 4. Summary of the simulation cases for the survey locations with two water level cases.

Case	No Surcharge Load	Surcharge Loads of Houses and Roads	Consideration of LSCR	No LSCR Consideration	Proposed LSCR
1	✓
2		✓	✓
3		✓		✓
4		✓			✓

Determination of riverbank safety coefficient (K_at).

The riverbank safety coefficient (K_at) was calculated using the following formula [10]:

$$K_{at}={\displaystyle \frac{{\sum }_{i=1}^{i=n}\left(W_i\times cos{\alpha }_i\times tg{\phi }_i+C_i\times L_i\right)}{{\sum }_{i=1}^{i=n}\left(W_i\times sin{\alpha }_i\right)}}$$

(3)

where C_i is the cohesion of the i-th slice within the sliding arc length; L_i is the length of the i-th sliding arc; W_i is the weight of the i-th slice; α_i is the inclination angle of the i-th slice relative to the sliding plane; φ_i is the internal friction angle of the i-th slice within the sliding arc length; W_ith is the method for general slope stability analysis based on deep sliding with the assumption of a circular sliding surface, according to QCVN 04-05:2022/BNNPTNT [37]; and the required value of the riverbank safety coefficient (K_at) is 1.15.

3. Results

3.1. Erosion/Deposition Dynamics Along the Two Rivers

Remote sensing analysis of the Mekong and Bassac Rivers (Figure 4) revealed significant morphological changes, particularly in meandering sections of the Mekong. These areas exhibit complex flow regimes, resulting in outer-bank erosion and inner-bank deposition. The primary cause of this phenomenon is that at the meanders, the flow is directed toward the outer bank, resulting in higher flow velocity on the outer side compared to the inner side. This observation aligns with the findings of other studies [8,38], which also showed that at cross-sections of river bends, the flow velocity on the outer bank is significantly higher than on the inner bank. In addition, it is noted that for rivers with clay and silt beds and slopes less than 0.02—similar to the Mekong Delta—if the sinuosity ratio, defined as the ratio of the meander length (L) to the straight-line distance between two meander ends (L′), exceeds 1.2, lateral erosion occurs on the concave bank while deposition occurs on the convex bank [39]. In this study, L and L′ were measured using polyline and straight-line tools in GIS from satellite imagery. This approach follows the standard methodology for assessing river planform characteristics, as described by Rosgen (2001) [39]. Based on the field observation about the riverbank conditions at these meandered sections, it confirms that the erosion/deposition phenomenon in the meanders described above accurately reflects ongoing realities. On the other hand, this phenomenon is less frequent in straight river sections, where the flow is relatively stable. This understanding can be applied to determine appropriate river corridors for different locations.

Unlike the Mekong River, the Bassac River has a relatively straight shape and a lower level of erosion. Instead, its overall level of sediment deposition is relatively higher than that of the Mekong River. To better illustrate this, the erosion/sedimentation level was considered and the extent of erosion and deposition along the two banks (left bank and right bank) of the Mekong and Bassac Rivers are also illustrated in Figure 5a and Figure 5b, respectively.

Figure 5a shows that for the Mekong River, the erosion width range of 0 to 0.5 m accounts for the longest total riverbank length affected by erosion (27.5 km on the left bank), whereas the erosion width range of 2 to 3 m affects the shortest total riverbank length (6.6 km on the left bank). However, the widest erosion range is between 3 and 33.4 m, with a total length of 16.7 km. For the right bank, the erosion width range of 0 to 0.5 m also affects the longest total length (13 km), while the 2 to 3 m range affects the shortest length (5.7 km). The widest erosion range on the right bank is also from 3 to 33.4 m, with a total length of 11.6 km. Overall, the distribution of erosion levels on both sides of riverbank is relatively similar; however, the extent of erosion on the left bank is greater across all ranges. Regarding deposition levels, on the left bank, the most significant deposition occurs in the range of 0 to 0.5 m, while the least occurs in the range of 2 to 3 m. However, the widest deposition range is from 3 to 46.6 m, with a total length of 27.7 km. Similarly, for the right bank, the most significant deposition is in the range of 3 to 46.6 m (total length: 21.5 km), while the least significant is in the range of 2 to 3 m (total length: 5 km). It can be observed that the extent of deposition on the left bank is greater than on the right bank across all ranges. The primary deposition ranges are 0 to 0.5 m and 3 to 46.6 m.

On the other hand, sedimentation levels in the Bassac River are relatively high. The section with the greatest sediment deposition length is the one with a sedimentation width of 0 to 0.5 m (with a left bank length of 30.2 km and a right bank length of 42 km). The lowest sedimentation level is observed in the section with a sedimentation width of 2 to 3 m (with a left bank length of 8.2 km and a right bank length of 7.1 km). Meanwhile, the section with the largest sediment deposition width is the one ranging from 3 to 23.8 m (with a total left bank length of 20.4 km and a right bank length of 11.8 km). This demonstrates that sediment deposition is significantly higher than erosion and is influenced by sediment inflow from upstream sources.

In addition, it can be also observed from Figure 5a that the left bank experiences the highest erosion levels across all ranges. However, the left bank also exhibits the highest deposition levels in all ranges. To further analyze this association/relationship, the erosion map has been incorporated into an OXY coordinate system and is presented in Figure 6 for the Mekong and Bassac rivers, respectively.

The relationship between deposition and erosion in the Mekong River is more clearly illustrated in Figure 6. Along the OX axis, from upstream to downstream, it can be seen that areas of erosion typically precede areas of deposition. This pattern occurs because eroded sediment particles often settle at nearby locations with favorable conditions for deposition. This phenomenon arises as sediment particles generated by erosion are not transported far downstream but instead tend to deposit relatively close to the erosion site. However, this behavior depends on the size of the sediment particles—finer particles are carried farther downstream, while coarser particles settle closer to the erosion site. On the other hand, it can be observed that significant erosion locations in Bassac River are relatively few, mainly concentrated in the downstream area. In contrast, sediment deposition points are more numerous and evenly distributed from upstream to downstream. The Bassac River has fewer meandering segments compared to the Mekong River, resulting in fewer instances of complex flows caused by curved river sections.

3.2. Annual Erosion/Deposition Rate and Trend (2000–2023)

To further analyze the data, the annual erosion/deposition rates (from 2000 to 2023) are presented in Figure 7 and the yearly trends of erosion/deposition are illustrated in Figure 8. It is indicated in Figure 7a that the deposition in the Mekong River generally exceeds erosion along the section from km0 to km83, with two erosion points and three deposition points at the rates below 20 m/year. From km80 to km100, there is one significant deposition point with a rate exceeding 40 m/year, extending over approximately 15 km. Similar patterns of rapid, localized accretion in meander belts have also been documented in previous studies. For instance, Anthony et al. (2015) [7] observed dynamic sediment redistribution and channel migration processes leading to rapid morphological changes in the Mekong Delta. These results indicate that large deposition points are typically distributed in the downstream sections of the river.

From Figure 7b in the Bassac River, it is evident that the left bank of the Bassac River has more sedimentation points than erosion points. The section with the highest sedimentation width is from km35 to km45, with a sedimentation width exceeding 20 m/year. Most other sedimentation points have widths ranging from 0 to 5 m/year. In contrast, there is only one significant erosion point (over 20 m/year), which spans approximately 3 km. In the Bassac River, the erosion/sedimentation trends on the left bank from 2005 to 2023 indicate a gradual decrease in sedimentation levels over time. For instance, in 2005, there were locations with sedimentation widths exceeding 60 m, but by 2023, these locations only experienced an additional 20 m of sedimentation. Conversely, some erosion points showed an increase during this period. The erosion/sedimentation rates from 2000 to 2023 on the right bank show that sedimentation is relatively significant, with the highest rate exceeding 15 m/year, while erosion is almost negligible.

Annual erosion and deposition trends on the Mekong River (a) indicate a gradual decline in deposition rates from 2005 to 2023, while erosion rates have increased. The construction of 11 hydropower dams along the lower Mekong mainstream has significantly disrupted sediment dynamics in the delta. The sediment deficit is severe, even during the flood season, which traditionally brings abundant sediment [4]. At the river mouths, sediment concentrations have dropped to very low levels, ranging from 0.02 g/L to 0.06 g/L. The total sediment volume passing through Tan Chau station into the Mekong River has decreased by 80%, from 44.2 million tons to just 8.8 million tons annually. This dramatic reduction in sediment supply may be a key factor contributing to the decline in deposition. The reduced sediment load in downstream areas has led to an increase in erosion due to “sediment-starved” flows. As erosion intensifies over the years, the safety corridors along the river are progressively widening, leading to greater impacts on infrastructure and urban developments along the riverbanks.

In addition, both banks of the Bassac River experience low erosion levels and relatively high sedimentation levels. This disparity suggests that sedimentation is not only influenced by sediment derived from erosion but also significantly supplemented by silt carried from upstream sources. However, the decreasing sedimentation trend indicates that the operation of 11 upstream hydropower dams has reduced the sediment load passing through Chau Doc into the Bassac river by 77%, down to 1.8 million tons (compared to 7.98 million tons under pre-dam conditions) [4]. This highlights that if this situation persists, sedimentation levels will continue to decline, while erosion could increase, potentially impacting safety corridors and the livelihoods of communities along the riverbanks. Moreover, similar to the Mekong River, the Bassac River faces significant challenges regarding bank instability due to encroachment into safety corridors by construction projects and residential developments.

Additionally, another cause of morphological changes and erosion along the river is the encroachment into the river corridor by infrastructure and activities that increase the loadings along the riverbanks. To better understand the extent of river corridor encroachment, data have been analyzed and a summary of the results of the current status of river safety corridors by provinces along the two rivers are shown in Figure 9 and Figure 10.

3.3. Current Status of River Safety Corridor (2000–2023)

From Figure 9 and Figure 10a, for the Mekong River, it can be seen that the length and percentage of unfollowed LSCR in Dong Thap are the most significant, with a total length of approximately 30 km, accounting for 88% of the total length of unfollowed LSCR. This supports the assumption that structures with substantial loads located within the river corridor contribute to increased erosion risk along the riverbanks. This, in turn, exacerbates the encroachment of the safety corridor into the banks, resulting in greater impacts on infrastructure and further narrowing of the corridor. To validate this assumption, different scenarios were developed to evaluate the locations of surcharge loads (placement of structures) inside and outside the river safety corridor under conditions of maximum and minimum water levels. This approach ensures a more accurate reflection of real-world conditions. On the other hand, for the Bassac River (Figure 9 and Figure 10b), it is evident that the highest level of encroachment into the river safety corridor is found in An Giang (48%), followed by Can Tho (21%), with Tra Vinh having the lowest at 1%. This situation indicates a significant risk of bank instability that is difficult to avoid.

A summary of the factor of safety assessment according to four cases of surcharge loads under maximum/minimum water levels in Mekong River and Bassac River are, respectively, shown in Figure 11 and Figure 12.

The results in Mekong River shown in Figure 13 indicate that when the structure is placed near the riverbank during the minimum water level season, the slope safety factor is 1.13. The minimum safety factor for riverbank stability in hydraulic structures is 1.15 [37]. Therefore, in this scenario, with a safety factor of 1.13 < 1.15, the bank slope will lose stability and face a high risk of erosion. When the structure is positioned 10 m away from the riverbank, the safety factor is 1.12. Since this safety factor is still lower than the minimum allowable safety factor of 1.15, the placement of the structure in this scenario is also unsuitable. The results show that when the structure is relocated to a position 37 m away from the riverbank, then the safety factor increases to 1.77 > 1.15 according to regulations. This indicates that the minimum distance for the bank slope to have a safety factor within a stable range is 37 m from the riverbank. This provides a basis for determining the river safety corridor.

On the other hand, to clearly observe the variation in the safety factor when the structure’s position is shifted, the safety factor chart according to the structure’s position has been assessed and is shown in Figure 13 and Figure 14. When the position of the surcharge loads (structure) changes, the safety factor also fluctuates. However, it can be observed that at the 37 m position, the safety factor no longer increases but starts to stabilize. This indicates that, from the 37 m point onward, the load from the structure no longer affects the bank slope’s safety. Therefore, this serves as the basis for determining the minimum river safety corridor at this location, which is 37 m in the case of the minimum water level season.

For the Bassac River, in the scenario of extreme low water levels, it is shown that in all three cases, the positions where the structures are placed have a safety factor of less than one. This indicates that the riverbank itself is not stable, even without considering the load of the structures. What needs to be determined now is the impact of the structures on the stability of the riverbank. Therefore, the detailed results of the variations in the riverbank safety factor are presented in Figure 13. The results show that when the structure is placed at a position approximately 20 m from the riverbank edge, the safety factor of the riverbank no longer increases and starts to level off. This indicates that at this distance, the load from the structure no longer impacts the stability of the riverbank. This finding serves as a basis for determining the safety corridor for this simulation position.

Similarly, the factor of safety vs. different location of surcharge loads under maximum water level at Location 1 has also been developed and assessed, and the results are presented in Figure 14.

In the Mekong River, it can be seen that in all cases, the factor of safety under different surcharge load locations (of the structure) is greater than 1.15. This indicates that in the maximum water level scenario, the bank slope will be more stable than in the minimum water level scenario. This is due to the larger difference in water levels between the bank face and the external water surface, which increases the instability of the bank slope. This also leads to the phenomenon of seepage forces from groundwater inside the soil, which disrupt the bank’s structural integrity and cause bank erosion [12]. Furthermore, to clearly observe the variation in the factor of safety vice versa the surcharge load locations, the results from different scenarios are presented in Figure 13. From Figure 13, it can be concluded that the riverbank slope consistently has a stability coefficient greater than 1.15. Particularly, at the 37 m position, the load from the structure no longer affects the bank slope’s safety, as the factor of safety no longer increases but stabilizes, similar to the minimum water level scenario.

For the Bassac River, the simulation results for the extreme high water level scenario show that in all four cases, the positions of the structures result in safety factors less than 1.15, indicating that the riverbank structure itself is unstable, even without considering the load exerted by the structure. However, similar to the extreme low water scenario, it is necessary to identify the position where the load from the structure no longer impacts the riverbank. The detailed results are presented in Figure 14. The results show that at a position 20 m from the riverbank edge, the safety factor no longer increases and levels off, indicating that this is the position where the load of the structure no longer affects the riverbank. This finding also serves as a basis for determining the safety corridor for this simulation position.

With the results of both scenarios, it is shown that the minimum water level scenario is the most dangerous and both show that the farther the location of the work is from the edge of the bank, the higher the stability coefficient will be. The proposed safety corridors for the Mekong and Bassac rivers in

Table 5. Proposed safety corridors at locations with corresponding stability coefficient values.

Location	1	2	3	4	5	6	7	8
Proposed a safety corridor with the max water level (Kat)	1.35	1.72	0.8	1.14	0.78	0.70	0.48	1.39
Proposed a safety corridor with the max water level (Kat)	1.18	1.34	0.8	0.83	0.45	0.59	0.41	0.92
Proposed distance for the safety corridor with the max water level (m)	37	35	24	20	21	30	14	28
Proposed distance for the safety corridor with the min water level (m)	38	35	23	20	11	28	14	21
Proposed safety corridor distance (m)	38	35	24	20	21	30	14	28

are based on the results of several scenario evaluations.

4. Discussion

The findings of this study reveal substantial encroachment into Legally-Specified Conservation Regions (LSCRs), or safety corridors, with Dong Thap Province being particularly affected, where 88% of the river length is non-compliant. To address this, policymakers should prioritize the revision and enforcement of LSCR standards. Updated guidelines must ensure that setback distances are aligned with site-specific erosion risks, promoting the sustainable management of riverbank areas. Strengthening enforcement mechanisms is essential to deter encroachment, with penalties for non-compliance and incentives for adherence serving as key components of this approach. In addition, to strengthening LSCR standards, zoning regulations should be established to control riverbank development. Designating specific areas where development is prohibited will help to prevent further encroachment into critical conservation zones. Priority areas for protection should be identified based on assessments of erosion vulnerability and riverbank stability. These zoning regulations should be integrated into broader land-use planning frameworks to ensure consistency across local, provincial, and national levels. NBS offers a cost-effective and ecologically sustainable approach to stabilizing riverbanks. Evidence from this study supports the adoption of NBS interventions, such as mangrove reforestation and the establishment of riparian buffer zones. This aligns with global best practices increasingly recognized for riverbank and corridor management. Comparative studies from similar deltaic and riverine environments demonstrate the effectiveness of such measures. For instance, Cohen-Shacham et al. (2019) [21] highlighted successful outcomes in riparian vegetation restoration, noting significant improvements in erosion control and biodiversity enhancement. Palmer et al. (2015) [22] also underscored the efficacy of riparian buffers and bioengineering techniques in stabilizing riverbanks and mitigating sediment transport in river corridors, particularly under conditions of increased hydrological variability.

Furthermore, studies conducted in the VMD provide local evidence of the advantages of vegetative buffers and bioengineered embankments over traditional concrete structures, emphasizing the dual benefits of ecological resilience and structural stability. Similarly, Nelson et al. (2024) [20] demonstrated how proactive delineation of erodible river corridors combined with strategic vegetation planting could reduce the flood risk and manage sediment deposition sustainably.

There is growing worldwide evidence supporting the integration of NBS into flood risk management and river corridor stabilization efforts. Thaler et al. (2023) [27] indicated that the adoption of nature-based flood management strategies on private lands can substantially reduce downstream flood risk while enhancing ecological connectivity. Additionally, Kondolf et al. (2014) [29] discussed sediment management practices involving NBS approaches across different continents, emphasizing the advantages of enhancing natural sediment regimes rather than relying solely on artificial sedimentation control.

Moreover, Fryirs et al. (2018) [28] highlighted successful cases of geomorphic recovery through NBS, demonstrating that strategic vegetation establishment can significantly enhance channel stability and habitat quality. Kiss et al. (2019) [25] provided further support through their study of bioengineered revetments along Hungary’s Tisza River, revealing substantial improvements in bank stability and reductions in erosion rates compared to traditional hard engineering methods.

Socio-economic considerations and community involvement are also crucial to the successful implementation of nature-based solutions. Schmidt and Wilcock (2008) [24] emphasized that effective river management using NBS requires an integrated approach, combining ecological engineering with stakeholder collaboration and adaptive governance.

These comparative insights underline that the application of NBS, as suggested in our study, is not only feasible but also beneficial from ecological, economic, and societal perspectives. Nevertheless, the efficiency of these measures is context specific, influenced by local hydrological, geomorphological, and socio-economic conditions. Therefore, careful site-specific planning, stakeholder involvement, and adaptive management approaches are crucial to ensure the successful implementation and longevity of NBS strategies within river corridors.

5. Conclusions

Remote sensing analysis showed that erosion along the Mekong River is significantly more severe than along the Bassac River, particularly in curved sections where flow dynamics are more complex. The assessment of erosion/deposition change rates along the two rivers indicate that the highest erosion rate on the Mekong River reaches 40 m/year, while on the Bassac River, it is 20 m/year, primarily concentrated in the downstream sections. Conversely, the sedimentation levels of the Mekong River are significantly lower than those of the Bassac River. Although the Mekong River has some areas with relatively high sedimentation (up to 50 m/year), compared to the Bassac River, where the highest is 20 m/year, most sedimentation on the Mekong River is concentrated in sections with sedimentation widths ranging from 0 to 0.5 m (28.3 km on the left bank and 10.3 km on the right bank). Similarly, on the Bassac River, sedimentation is primarily concentrated in the 0 to 0.5 m range (30.2 km on the left bank and 40.2 km on the right bank). The combined results show that the Mekong River has more erosion-prone locations than the Bassac River. This study presents an integrated methodology combining remote sensing, hydrodynamic modeling, and field-based geotechnical assessment tailored to river corridor management in the VMD. Unlike previous regional studies that have focused predominantly on structural erosion control measures, our work emphasizes NBS within a scientifically informed delineation of river corridors, thereby bridging a critical gap in local management strategies. As an innovative contribution, we provide specific, evidence-based guidelines for defining river corridor buffer widths in alignment with local geomorphological and hydrological conditions. Future research should focus on longitudinal monitoring of the proposed NBS implementations, evaluating their long-term ecological and socio-economic impacts, and exploring adaptive management frameworks to optimize these strategies under future climate change and development scenarios.

Factor of safety vs. different location of surcharge loads (scenarios) under min/max water levels was assessed and the results showed that in the extreme low water scenario, the safety factor is lower than in the extreme high water level scenario for both rivers. Various structure placement cases were simulated, and the results indicate that as the structure’s position moves further away from the riverbank edge, the stability coefficient increases until it reaches a certain point where it no longer increases. This demonstrates that at this position, the load from the structure no longer impacts the riverbank. This finding serves as the basis for determining the river safety corridor at that specific location. Future research should continue to explore the complex interactions within these systems and develop innovative approaches to address the challenges posed by urbanization and climate change.

This study provides several innovations. First, it presents a rare integration of long-term remote sensing analysis (2000–2023), hydrodynamic simulation, and field-based geotechnical assessment to delineate river safety corridors tailored to local conditions in the VMD. Second, it offers specific, evidence-based corridor widths derived from stability coefficients under different water levels and surcharge scenarios—something not previously addressed at this scale or spatial granularity in the VMD. These innovations enable a more scientifically robust foundation for regulatory enforcement and land use planning.

Future research should focus on validating the effectiveness of the proposed nature-based solutions over time, including the use of vegetative buffers and managed retreat in erosion-prone areas. There is also a need for participatory approaches involving local stakeholders in monitoring and adaptive management, as well as coupling hydrological models with climate projections to evaluate how safety corridors might shift under sea level rise and altered sediment regimes. In addition, the integration of higher-resolution data sources such as UAV imagery, LiDAR, or InSAR could improve the spatial accuracy of riverbank delineation and morphological assessments. The installation of in situ monitoring equipment for sediment volume measurements would further strengthen the approach by providing ground-truth data for validating remote sensing-based methodologies and enhancing the overall robustness of erosion and deposition analysis.

The river safety corridor was also determined based on the results of several scenarios. For the Mekong River, the safety corridor distances from the riverbank at positions 1, 2, and 3 are 38 m, 35 m, and 24 m, respectively. For the Bassac River, the safety corridor distances from the riverbank at positions 1, 2, 3, 4, and 5 are 20 m, 21 m, 30 m, 14 m, and 28 m, respectively.