Sediment Resuspension in the Yellow River Subaqueous Delta During Gale Events

Abstract

During winter, strong winds and waves significantly enhance sediment resuspension in the Yellow River Delta. Based on the continuous and high-resolution data on water levels, wave heights, current velocities, and echo intensities collected by the Acoustic Doppler Current Profiler at different depths (5 m and 12 m) in the northern Yellow River Delta simultaneously, this study investigated the sediment resuspension during gale events and tranquil conditions. In deeper waters (12 m), the suspended sediment volume concentration (SSVC) showed a strong correlation with current speed (r = 0.74), while in shallower waters (5 m), the SSVC correlated more closely with wave height (r = 0.72). The thorough analysis of gale events revealed that the maximum wave heights during northwest gales were 23.80% and 34.59% lower than that during northeast gales at deep and shallow stations, respectively, primarily due to the longer wind fetch associated with northeast gales. Conversely, the maximum current velocities during northwest gales were 10.34% and 37.31% higher than that during northeast gales at deep and shallow stations. In deeper waters, the maximum wave–current induced shear stress $({\tau }_{cw}$) and SSVC during northwest gales were 30.38% and 3.70% higher than those during northeast gales, highlighting current-driven resuspension. In contrast, in shallower waters, the maximum ${\tau }_{cw}$ and SSVC during northeast gales were 47.35% and 4.94% higher than those during northwest gales, underscoring the dominance of wave-induced resuspension.

1. Introduction

Suspended particulate matter serves as a vital component of seawater, and its transportation, sedimentation, and resuspension impact the surrounding seabed topography, consequently affecting the port channel construction and coastal projects [1]. Additionally, suspended particulate matter, as a carrier of nutrients and pollutants, plays a crucial role in the marine ecological environment [2]. Previous studies have indicated that wave action, tidal currents, and wind conditions significantly impact the distribution of suspended particles [3], highlighting the importance of studying sediment resuspension during gale events.

The Yellow River delivers approximately 1.1 billion tons of sediment annually into the Bohai Sea [4]. Active sediment movement occurs in the Yellow River subaqueous delta under complex ocean dynamics, including waves, tides, and currents [5]. Previous studies have investigated sediment resuspension and its influencing factors in the Yellow River Delta [6] Some studies suggested that sediment resuspension in the Yellow River Delta was primarily wave-driven, with strong waves disrupting the vertical structure of suspended sediments in shallow waters [7]. The suspended sediment concentration (SSC) exhibited a positive correlation with wave height [8]. Other studies noted a positive correlation between wind speed and the SSC but observed that the peak SSC in the upper-middle water layer lagged behind high wind speeds by approximately 8–10 h [9,10]. However, the tidal currents exerted an overarching control on the SSC [11]. Specifically, based on data collected during a 5-month field observation of the Yellow River Delta, previous study discovered that the formation of approximately 2–12 cm thick fluid mud during winter wind events was mainly attributed to seabed liquefaction induced by waves [12]. After the formation of fluid mud, strong currents would transport the fluid mud away 10 h after storm arrival, whereas currents during calm weather would hardly cause suspension [13]. However, due to the limitations of observational technical difficulties and extreme weather conditions, the differences in the impact of storms on sediment resuspension at different water depths remain unclear. The lack of high-precision and long-term systematic field observation data hinders the understanding of sedimentary dynamics [14].

Observations of the SSC could be classified into traditional and modern methods [15]. Traditional methods involve synchronous water sample collection by multiple vessels along cross-sections, stratification by water depth, or single-point measurements with a turbidimeter [15]. Modern methods include optical techniques, acoustic techniques, and satellite remote sensing. Acoustic observations offer advantages such as non-intrusive data collection, simultaneous profiling, and high spatial resolution [16,17,18]. Optical backscatter sensors (OBS) correlate turbidity with the SSC but require site-specific calibration [19]. Satellite remote sensing retrieves the SSC from water, leaving radiance in visible and near-infrared bands, suitable for large-scale monitoring but limited by water transparency and atmospheric conditions [20]. The application of acoustic techniques in shallow seas and estuarine areas with complex hydrodynamics enhances the development of sedimentation dynamics [21,22,23].

The data acquired by the Acoustic Doppler Current Profiler (ADCP) offers long-term suspended material data across varying weather scenarios, with measurement accuracy comparable to specialized suspended sediment instruments [24,25]. By analyzing data from two stations at different water depths in the northern Yellow River Delta, this study investigated the varied responses of sediment resuspension during gale events and evaluated the impacts of wind direction. The findings advance sedimentation dynamics in shallow estuaries, offering significant guidance for marine ecological protection, marine infrastructure maintenance, and disaster risk management.

2. Data and Methods

2.1. Study Area

The study area is located in the underwater slope zone of the northern Yellow River Delta, with a shallow water depth and gentle topographic slope (Figure 1). One M₂ amphidromic point is located in the northeastern Yellow River Delta [26] and shifts continuously, resulting in complex tidal current dynamics in the adjacent region [27]. The current velocities are notably high, reaching up to 90 cm/s within the 5 m isobath, with the strongest currents occurring at depths of 5–15 m. The waves are primarily wind-generated waves, with minimal occurrence of swell waves. Both normal and strong wave directions are oriented northeast, with a maximum wave height recorded at 5.2 m [28]. The Yellow River annually delivers approximately 1.1 billion tons of sediment into the Bohai Sea [4]. Approximately 70% of the sediment from the Yellow River accumulates within 15 km of the estuary [29]. The sediment in the study area, primarily sourced from the Yellow River, comprises loose sandy silt, which is easily resuspended and transported by waves and currents [10,30,31].

2.2. Data Source

Seabed observation stations CBD01 (118.9304° E, 38.2058° N, water depth of 12 m) and CBS01 (118.8722° E, 38.1457° N, water depth of 5 m) in the northern Yellow River Delta are illustrated in Figure 1a. The Acoustic Wave and Current Profilers (AWAC-600 kHz) (Nortek AS, Rud, Norway) deployed at these stations collect continuous, high-resolution (1 h) data on water level, wave height, current velocity, and echo intensity simultaneously. The instrument employs three synchronized transducers positioned approximately 0.7 m above the seabed, observing upward with a vertical resolution of 1 m. Throughout the observation period from 3 January 2023 to 14 May 2023, three gale events exceeding fresh gale conditions (13.8 m/s) were recorded. Wind speed and direction data for both stations were interpolated from the European Centre for Medium-Range Weather Forecasts (ECMWF)—ERA5 wind field data (1 h time resolution, 0.25° spatial resolution).

2.3. ADCP Data Processing

By considering and modifying the geometry and absorption attenuation of sound waves, the echo intensity recorded by ADCP could be converted to the suspended sediment volume concentration (SSVC) according to Formula (1) [32,33,34].

$$Sv=Kc\ast \left(EI-Nc\right)+20\mathit{lg}\left(R\right)+2\alpha R+10lg(273.16+T)$$

(1)

where $Sv$ is the backscattering intensity (dB); $EI$ denotes the average echo intensity received by the ADCP transducer (count); $Nc$ is the noise background (count); $Kc$ is the proportional coefficient of $EI$ (dB/count), typically around 0.45 [34,35]; $R$ is the vertical distance from the suspended matter to the transducer plane (m); $\alpha$ is the absorption coefficient (dB·m⁻¹); and $T$ is the transducer temperature (°C). The absorption coefficient ($\alpha$) included the absorption of sound waves by seawater (${\alpha }_w$) and the absorption by suspended matter (${\alpha }_s$), that is, $\alpha$ = ${\alpha }_w+{\alpha }_s$. ${\alpha }_w$ can be expressed according to Formulas (2)–(5) [36,37]:

$${\alpha }_w=A_3{P}_3{f}^2$$

(2)

$$P_3=1-3.83\times {10}^{-5}D+4.9\times {10}^{-10}{D}^2$$

(3)

When $t$ ≤ 20 °C,

$$A_3={10}^{-3}\times (4.937\times {10}^{-4}-2.59\times {10}^{-5}t+9.11\times {10}^{-7}{t}^2-1.5\times {10}^{-8}{t}^3)$$

(4)

When $t$ > 20 °C,

$$A_3={10}^{-3}\times (3.964\times {10}^{-4}-1.146\times {10}^{-5}t+1.45\times {10}^{-7}{t}^2-6.5\times {10}^{-10}{t}^3)$$

(5)

where $D$ is the water depth (m), $t$ is the water temperature (°C), and $f$ is the frequency of the sound wave (kHz). Since the water depth in the study area was less than 40 m, the $P_3$ value was approximately 1 [34]. The absorption by suspended matter ${\alpha }_s$ at each layer was about 0.001 dB·m⁻¹, significantly different from ${\alpha }_w$. Therefore, the impact of ${\alpha }_s$ was ignored in this study, and ${\alpha }_w$ was considered as $\alpha$ [38]. At the CBD01 station, simultaneously, 223 groups of SSVC data derived from ADCP and turbidity data collected by OBS were compared and analyzed (Figure 2). The calculation using the Matlab fitting function showed that there was a significant logarithmic correlation between the SSVC data and the turbidity data, with a correlation coefficient of 0.79. Therefore, analyzing the backscattering intensity in the study area can provide a comprehensive understanding of the local suspended matter distribution.

2.4. Calculation of Bottom Shear Stress

The bottom shear stress was crucial for assessing the force of waves and currents on the seafloor [39]. The average current velocity in the near-shore bottom boundary layer usually meets the logarithmic distribution [40]. Based on the accurate current velocity profile observed by ADCP, the log-profile (LP) method was used to calculate the current-induced shear stress (${\tau }_{\mathrm{c}}$) in this study area [41]. The LP method could be expressed by the von Karman–Prandtl equation using Equation (6) [42], and ${\tau }_{\mathrm{c}}$ was calculated based upon Formulas (6) and (7):

$$U\left(z\right) = ({\mathrm{u}}_{\ast }/\mathsf{\kappa })\mathrm{In}(z/z_0)$$

(6)

$${\tau }_{\mathrm{c}}{= \rho {\mathrm{u}}_{\ast }}^2$$

(7)

U(z) is the average velocity (m/s) of the sampling period above the seabed $z$ (m), ${\mathrm{u}}_{\ast }$ is the friction velocity (m/s), κ is the Carmen constant (0.4), $\rho$ is the seawater density (1025 kg/m³), and $z_0$ is the roughness length of the seabed ($z_0$ = k_s/30, where k_s = 2.5 D₅₀ is the Nikuradse particle roughness, and D₅₀ is the median grain size of the sediment). The median grain sizes (D₅₀) at the CBD01 and CBS01 stations were 0.01 mm and 0.015 mm, respectively [43].

According to the linear wave theory, the amplitude of the wave orbit velocity $U_w$ and the major axis radius of the near-bottom wave orbit A_w were calculated according to Formulas (8) and (9) [44]:

$$U_w={\displaystyle \frac{\mathsf{\pi }{H}_{rms}}{T\mathrm{s}\mathrm{i}\mathrm{n}\mathrm{h}(2\mathsf{\pi }d/L)}}$$

(8)

$$A_w = U_{w}T/2\mathsf{\pi }$$

(9)

H_rms is the root mean square wave height (m), $T$ is the wave period (s), d is the water depth (m), H_rms = Hs/$\sqrt{2}$, Hs is the significant wave height (m), and L is the wavelength (m), calculated according to Formula (10) [45,46]:

$$L = g T^2 \mathrm{t}\mathrm{a}\mathrm{n}\mathrm{h}(2\mathsf{\pi }\mathrm{d}/L)/2\mathsf{\pi }$$

(10)

The orbital motion of waves increased the bed shear stress. Herein, wave-induced shear stress $({\tau }_w$) was calculated according to Formula (11) [47]:

$${\tau }_w={\displaystyle \frac{1}{2}}\rho f_w{U}_{w}2$$

(11)

$\rho$ is the seawater density (1025 kg/m³), and $f_w$ is the wave friction coefficient related to the wave Reynolds number Re_w (Re_w = A_w$U_w$/ν, where ν is the seawater kinematic viscosity coefficient (10⁻⁶ m²/s)).

$${\mathit{Re}}_w \le {10}^5, f_w=2 {{Re}_w}^{-0.5};$$

$${\mathit{Re}}_w> {10}^5, f_w=0.0521 {{Re}_w}^{-0.187};$$

The wave–current-induced shear stress (${\tau }_{cw}$) was calculated according to Formulas (12) and (13) [47]:

$${\tau }_{cw}=[({\tau }_m+{\tau }_w|\mathit{cos}\phi |{)}^2+({\tau }_w{\left|\mathit{sin}\phi \right|)}^2{]}^{1/2}$$

(12)

$${\tau }_m= {\tau }_{\mathrm{c}}[1+1.2 ({\tau }_w/({\tau }_{\mathrm{c}}+{\tau }_w){)}^{3.2}]$$

(13)

${\tau }_m$ is the average shear stress under the action of waves and currents, and $\phi$ is the propagation angle of the waves and currents.

2.5. Calculation of Critical Shear Stress

For fine-grained sediment, the critical shear stress (${\tau }_{cr}$) was estimated using the Shields’ method by Formulas (14)–(17) [42,48,49]:

$${\tau }_{cr}=g{\theta }_{c\mathrm{r}}\left({\rho }_s-\rho \right)D_{50}$$

(14)

$${\theta }_{\mathrm{cr}} = 0.24/D_{\ast } +0.055(1-\exp (-0.02 D_{\ast }\left)\right), D_{\ast }> 5$$

(15)

$${\theta }_{\mathrm{cr}} = 0.3/(1+1.2 D_{\ast }) +0.055(1-\exp (-0.02 D_{\ast }\left)\right), D_{\ast }\le 5$$

(16)

$$D_{\ast }=[g(s-1)/v^2 {]}^{1/3} D_{50}$$

(17)

$g$ is the gravity acceleration, θ_cr is the critical Shields number, $\rho$_s is the sediment density (2650 kg/m³), D₅₀ is the sediment median grain size, $s$ is the density ratio of sediment to seawater, $D_{\ast }$ is the dimensionless grain size number, and ν is the seawater kinematic viscosity coefficient (10⁻⁶ m²/s). The critical shear stress ${\tau }_{cr}$ was 0.037 Pa at the CBD01 station and 0.05 Pa at the CBS01 station.

3. Results

3.1. Holistic Analysis

During the observation period, average wind speeds reached 5.47 m/s and 5.19 m/s at the CBD01 (12 m) and CBS01 stations (5 m), respectively (Figure 3a and Figure 4a). Three gale events exceeding fresh gale conditions (13.8 m/s) were highlighted in the green boxes (Figure 3a and Figure 4a). The average significant wave heights (Hs) were 0.62 m and 0.38 m at the CBD01 and CBS01 stations, respectively, with maximum Hs of 4.45 m and 3.89 m (Figure 3a and Figure 4a), indicating a strong correlation with wind speed. Water levels at the CBD01 station fluctuated between −2.09 m and 1.27 m, exhibiting significant variations during gale events (Figure 3b). Correspondingly, water levels ranged from −1.99 m to 1.53 m at the CBS01 station (Figure 4b).

The average bottom and surface current velocities were 0.33 m/s and 0.62 m/s at the CBD01 station (Figure 3c), while average bottom and surface current velocities were 0.24 m/s and 0.38 m/s at the CBS01 station (Figure 4c), showing lower current velocities. The layer-average and bottom SSVC were 110.36 μL/L and 133.26 μL/L at the CBD01 station (Figure 3d). Occasionally, an abnormally high SSVC was observed on the surface with lower values beneath, possibly influenced by the errors of light and large plant debris and the increased surface acoustic impedance due to the surface air–sea exchange [50,51]. In contrast, the SSVC at the CBS01 station displayed pronounced responses to strong wave events, recording a layer-average SSVC of 99.87 μL/L and a bottom SSVC of 107.49 μL/L, both lower than those at the deeper water station (Figure 4d).

The average and maximum ${\tau }_{cw}$ at the CBD01 station were 0.14 Pa and 1.83 Pa, significantly surpassing the ${\tau }_{cr}$ of 0.037 Pa (Figure 3e). The correlation coefficient between the ${\tau }_{cw}$ and Hs was 0.73, whereas the correlation with current velocity was 0.74 (Figure 3e). At the CBD01 station, the correlation coefficient between the SSVC and current velocity was also 0.74. Additionally, the average and maximum ${\tau }_{cw}$ at the CBS01 station were 0.16 Pa and 1.92 Pa, significantly above the ${\tau }_{cr}$ of 0.05 Pa (Figure 4e). The correlation between the ${\tau }_{cw}$ and Hs was 0.92, although the correlation with current velocity was only 0.35 (Figure 4e). In contrast, the correlation coefficient between the SSVC and Hs was 0.72 at the CBS01 station.

3.2. Northeast Gale Process

Three strong wind events exceeding the fresh gale were highlighted in the green boxes (Figure 3a and Figure 4a), with the highest wind speed in gale 3, which was used as an example to study the northeastward gale process. Gale 3 occurred from 2 April to 6 April 2023 (Figure 3 and Figure 4). At the CBD01 station, wind speeds surpassed the strong breeze threshold (10.8 m/s) for 27 h, peaking at 15.87 m/s (Figure 5a). The maximum Hs was 4.45 m, slightly lagging behind the peak wind speed (Figure 5a). The highest water level was 1.19 m (Figure 5c), while the maximum layer-average current velocity was 1.09 m/s (Figure 5e). The maximum bottom SSVC was 140.14 μL/L, 5.16% above the observation period average (Figure 5g). The maximum ${\tau }_{cw}$ at the CBD01 station was 1.17 Pa, significantly exceeding the ${\tau }_{cr}$ of 0.037 Pa (Figure 5i).

At the CBS01 station, the wind speed surpassed the strong breeze threshold (10.8 m/s) for 26 h, peaking at 15.25 m/s (Figure 5b). The maximum Hs was 3.49 m (0.96 m lower than at the CBD01 station; Figure 5b). This abrupt decrease in wave height during a gale event was not caused by a shift in wind direction, which consistently remained north. Combined with the maximum wave height and water depth, the wave deformation and breaking in the shallow waters resulted in a sudden drop in wave height (Figure 5b). The highest water level was 1.15 m, and the maximum layer-average current velocity was 0.7 m/s (Figure 5d,f). Notably, the shallow-water station exhibited significant sediment resuspension, indicating heightened sensitivity to waves. The maximum bottom SSVC reached 128.93 μL/L, approximately 19.95% higher compared to the observation period average (Figure 5h). The average ${\tau }_{cw}$ at the CBS01 station was 1.92 Pa, significantly exceeding the ${\tau }_{cr}$ of 0.05 Pa (Figure 5j).

3.3. Gale Event with Changing Wind Directions

During the gale event from 21 January to 25 January 2023, wind directions shifted from northeast in the early period to northwest later (Figure 6a,b). Wind speeds were relatively similar throughout both strong wind processes. At the CBD01 station, the peak wind speed of 11.64 m/s occurred during the northeast gale, with sustained winds exceeding the strong breeze (10.8 m/s) for 8 h, while the maximum wind speed was 12.86 m/s during the northwest gale, with winds exceeding the strong breeze (10.8 m/s) for 17 h (Figure 6a). Northeast winds raised the water level to 0.92 m, whereas northwest winds led to a water level decrease of −2.09 m (Figure 6c). The maximum Hs during the northeast and northwest gales were 3.53 m and 2.69 m, respectively, indicating a 23.80% reduction compared to the northeast gale (Figure 6c). Although northeast gales had shorter strong-breeze durations and lower peak wind speeds, their longer fetch generated larger waves. The peak layer-average current velocity during the northwest gale was 1.28 m/s, 10.34% higher than that during the northeast gale (1.16 m/s) (Figure 6e). Additionally, the bottom SSVC during the northwest gale was 134.81 μL/L, 3.70% higher than that during the northeast gale (Figure 6g). The maximum ${\tau }_{cw}$ during the northwest gale was 1.03 Pa, 30.38% higher than that during the northeast gale (Figure 6i). The higher SSVC during the northwest gale was associated with strong currents, confirming current-dominated resuspension in deep waters.

At the CBS01 station, the peak wind speed reached 11.06 m/s during northeast gales, with sustained winds surpassing a strong breeze (10.8 m/s) for 6 h, while the maximum wind speed was 12.24 m/s during the northwest gale, with winds exceeding the strong breeze (10.8 m/s) for 11 h (Figure 6b). The highest water level was 1.16 m during the northeast gale, 5 cm higher compared to the CBD01 station, attributed to coastal flooding caused by the northeast gales. In contrast, the lowest water level during the northwest gale was −1.99 m (Figure 6d). The maximum Hs during the northwest gale was 2.08 m, indicating a 34.59% decrease compared to the northeast gale (3.18 m), mainly due to the larger wind fetch during the northeast gale (Figure 6d). Additionally, the maximum layer-average current velocity was 0.67 m/s during the northeast gale, whereas it increased to 0.92 m/s during the northwest gale, showing a 37.31% rise compared to the northeast gale (Figure 6f). Furthermore, the bottom SSVC during the northeast gale was 127.44 μL/L, 4.94% higher than that during the northwest gale (Figure 6h). The maximum ${\tau }_{cw}$ during the northeast gale was 1.71 Pa, 47.35% higher than that during the northwest gale (Figure 6j). The elevated SSVC during the northeast gale coincided with strong waves, supporting wave-dominated resuspension in shallow areas.

4. Discussion

4.1. Wind Speed

According to the wind speed magnitude, the Hs, maximum wave height (Hmax), bottom and surface current speeds, and bottom and vertical average SSVCs at the CBD01 and CBS01 stations were statistically analyzed, as shown in Figure 7.

At the CBD01 station, under high wind conditions (>13.9 m/s), the average Hs and Hmax were 3.47 ± 0.76 m and 5.16 ± 1.16 m, respectively. Concurrently, current speeds increased substantially to 0.55 ± 0.34 m/s (bottom) and 0.72 ± 0.4 m/s (surface), while the SSVC showed elevated concentrations of 133.56 ± 3.77 μL/L (bottom) and 118.25 ± 4.66 μL/L (vertical average)—representing a marked increase compared to calm wind conditions. Similar trends were observed at the CBS01 station, where extreme winds produced Hs = 2.28 ± 0.81 m and Hmax = 3.62 ± 1.38 m, with current velocities of 0.37 ± 0.09 m/s (bottom) and 0.51 ± 0.22 m/s (surface), accompanied by high SSVC values of 127.94 ± 6.22 μL/L (bottom) and 118.08 ± 3.03 μL/L (vertical average).

The strong wind effect significantly enhanced sediment transport through a dual mechanism: (1) increased wave energy amplified bed shear stress, inducing large-scale sediment resuspension; and (2) wind-enhanced tidal currents substantially elevated the advective transport capacity of suspended sediment. This aligns with the well-established “wave resuspension–tidal advection” coupling mechanism documented in the Yellow River Delta [5].

4.2. Wind Direction

The prevailing wind direction was from the south, while the strong wind direction was from the northeast, north, and northwest in the Yellow River Delta (Figure 8a,b). Both normal and powerful wave directions were east–northeast and northeast (Figure 8c,d). Since wave generation depended on wind speed, fetch, and duration, the westward and southward waves hardly developed (Figure 8c,d). The maximum wave height in shallower waters was significantly lower than that in deeper waters. As waves propagated to the shallow regions where water depth was less than half the wavelength, waves underwent deformation due to bottom friction, leading to a reduction in wavelength, an increase in wave steepness, and ultimately wave breaking, which resulted in diminished wave heights [52]. Surface current velocities surpassed bottom current velocities at both stations, attributed to substantial seabed friction and wind impacts on surface currents (Figure 8e,h). Previous studies have analyzed the characteristics of strong winds (>17 m/s) around the Yellow River Delta for several years, revealing that northwest winds predominantly occur in the winter and northeast winds are more prevalent in the spring, summer, and autumn [53]. The intensity, direction, and duration of the wind play crucial roles in determining the patterns of circulation and sediment transport [54,55,56,57].

4.3. Wave Action

The time ratio of ${\tau }_w$ exceeding ${\tau }_{cr}$ accounted for 12% and 33% at the CBD01 and CBS01 stations, respectively (Figure 3e and Figure 4e). This indicated that waves were capable of impacting the seabed during significant wave events at the CBS01 station (5 m), whereas the impact at the CBD01 station (12 m) was minimal. Nonetheless, the time ratios of ${\tau }_{cw}$ exceeding ${\tau }_{cr}$ accounted for 78% and 60% at the CBD01 and CBS01 stations (Figure 3e and Figure 4e), with the CBD01 station exhibiting a notably longer duration, indicating the substantial impacts of currents at the CBD01 station. Consequently, sediment resuspension at the deeper water was predominantly controlled by currents, while wave activity had a pronounced effect on resuspension in shallower water. However, previous studies believed that at shallow depths (<20 m), waves could directly act on the seabed and induce sediment resuspension [8].

4.4. Bottom Residual Current

The bottom current velocities at the CBD01 and CBS01 stations underwent harmonic analysis, obtaining the bottom residual current velocity and direction (Figure 9). At the CBD01 station (12 m), the residual current velocity was relatively large, reaching a maximum of 0.97 m/s, with the residual current direction predominantly oriented towards the northwest, indicating that the SSC would be transported northwest (Figure 9a). In contrast, the residual current velocity at the CBS01 station (5 m) was comparatively lower, with a maximum of 0.53 m/s. The residual current exhibited a scattered directional distribution, frequently occurring in the northwest, west, south, and southeast directions, potentially influenced by nearby artificial structures [58].

5. Conclusions

High-resolution observations at two stations (CBD01: 12 m; CBS01: 5 m) in the northern Yellow River Delta revealed distinct sediment resuspension responses to gale events.

During the observation period, the correlation coefficient between ${\tau }_{cw}$ and Hs was 0.73, while the correlation with current velocity was 0.74 at the CBD01 station (12 m). The correlation coefficient between the SSVC and current velocity was also 0.74 at the CBD01 station. Conversely, at the CBS01 station (5 m), the correlation between the ${\tau }_{cw}$ and wave height reached 0.92, whereas the correlation with current velocity was only 0.35. The correlation coefficient between the SSVC and Hs was 0.72 at the CBS01 station. During the northeast gale from 2 April to 6 April 2023, the maximum bottom SSVCs during gale events were 19.95% and 5.16% higher compared to the observation period at the CBS01 and CBD01 stations, respectively, reflecting more severe sediment resuspension at the CBS01 station.

The gale event from 21 January to 25 January 2023 initially developed northeastward winds, later transitioning to northwest. The maximum wind speed and the duration of winds exceeding 10.8 m/s during the northwest gale surpassed those during the northeast gale at both stations. At the CBD01 station (12 m), the maximum Hs during the northwest gale was 23.80% lower than that during the northeast gale, attributed to the shorter wind fetch. Furthermore, the maximum layer-average current velocity during the northwest gale was 10.34% higher than during the northeast gale. The maximum ${\tau }_{cw}$ and SSVC during the northwest gale were 30.38% and 3.70% higher than those during the northeast gale, indicating that sediment resuspension in deeper waters was primarily driven by currents. Conversely, at the CBS01 station (5 m), the maximum Hs during the northwest gale was 34.59% lower than that during the northeast gale, while the maximum layer-average current velocity during the northwest gale was 37.31% higher than during the northeast gale. The maximum ${\tau }_{cw}$ and SSVC during the northeast gale were 47.35% and 4.94% higher than those during the northwest gale, indicating that sediment resuspension in shallow waters was predominantly driven by wave dynamics. Strong winds in the Yellow River Delta originated from the northeast, north, and northwest directions, with their intensity, direction, and duration collectively controlling sediment transport.

Future work will integrate numerical modeling to quantify wave–current effects on sediment resuspension across water depths, identify critical wave-dominated thresholds, and extend the framework to other deltas.