Effect of Wave, Current, and Lutocline on Sediment Resuspension in Yellow River Delta-Front

Bowen Li 1,2, Yonggang Jia 1,2,3,*, J. Paul Liu 4, Xiaolei Liu 1 and Zhenhao Wang 1,5 1 College of Environmental Science and Engineering, Ocean University of China, Qingdao 266100, China; l1398849394@163.com (B.L.); xiaolei@ouc.edu.cn (X.L.); wzh-ouc@foxmail.com (Z.W.) 2 Shandong Provincial Key Laboratory of Marine Environment and Geological Engineering, Ocean University of China, Qingdao 266100, China 3 Laboratory for Marine Geology, Qingdao National Laboratory for Marine Science and Technology, Qingdao 266100, China 4 Department of Marine, Earth and Atmospheric Sciences, North Carolina State University, Raleigh, NC 27695, USA; jpliu@ncsu.edu 5 First Institute of Oceanography, MNR, Qingdao 266061, China * Correspondence: yonggang@ouc.edu.cn


Introduction
Suspended sediments are an important part of sediment movement in estuarine coastal waters, and their distribution, diffusion, and deposition have a major impact on ports, waterways, and ecological environments. Additionally, marine sediments serve as a key sink for heavy metals that are released into the sea when the seabed sediment is resuspended [1,2]. Therefore, temporal changes in the suspended sediment concentration (SSC) are an important issue in estuarine coastal biogeochemical research. Previous studies have suggested that SSC variations in estuarine waters are significantly affected by flood and ebb tides, spring and neap tides, and seasonal factors [3][4][5][6][7]. A high SSC is the result of the combined effects of sediment transport and resuspension [8].
Previous studies show that the current velocity in the horizontal direction is closely associated with the water depth. The current velocity is the least when the water depth is less than 5 m, and the current velocity reaches the maximum at depths between 10 and 15 m and is reducing in depths from 15 to 20 m [50]. The area selected for in-situ measurement (38 • 10 N, 118 • 54 E) is located in the subaqueous delta front of the Yellow River ( Figure 1). The water depths at the observation site range approximately from 8 to 12 m, and the current velocity is relatively high. The topography of the observation area inclines to the northeast at an average slope of 7-8 • , and the gently sloped terrain is conducive to the formation of catastrophic storm surges [49]. For these reasons, the sedimentary processes are complex and the associated strata are disordered, exhibiting very high occurrences of resuspension in the research area [50,51]. distribution of suspended sediment in the Bohai Sea is mainly dominated by the river input and coastal resuspension [44]. The winter monsoon winds and strong waves are the main cause of the sediment resuspension and offshore transport [45]. The sediment shear strength is uniform under hydrodynamic action [46]. The waves are mainly induced by the wind in the Yellow River Delta and vary seasonally and interannually [47]. As the speed and frequency of the north-west winter monsoon is significantly higher than that in summer monsoon, in the winter the waves are strong [48] and the frequency of storm surges is much higher [49].
Previous studies show that the current velocity in the horizontal direction is closely associated with the water depth. The current velocity is the least when the water depth is less than 5 m, and the current velocity reaches the maximum at depths between 10 and 15 m and is reducing in depths from 15 to 20 m [50]. The area selected for in-situ measurement (38°10' N, 118°54' E) is located in the subaqueous delta front of the Yellow River ( Figure 1). The water depths at the observation site range approximately from 8 to 12 m, and the current velocity is relatively high. The topography of the observation area inclines to the northeast at an average slope of 7-8°, and the gently sloped terrain is conducive to the formation of catastrophic storm surges [49]. For these reasons, the sedimentary processes are complex and the associated strata are disordered, exhibiting very high occurrences of resuspension in the research area [50,51].

Observation Instrumentation
A seabed-based tripod was designed to record the seabed surface hydrodynamic functions and SSCs. The design is optimized to maintain a stable posture on the seabed, and includes features to reduce subsidence and prevent the sideslip caused by storm waves and currents. The threaded holes on the tripod pedestals allow the installation of steel bars for insertion into the seabed. Therefore, the seabed-based tripod can remain stable, as the load is scattered through the pile foundations.
To measure erosion and deposition, current velocities, and turbidity we mounted a pack of sediment and hydrodynamic instruments on a tripod, including autonomous altimeter, current velocity meter, Argus surface meter, turbidity meter and wave tide gauge (see details in Figure 2). The range of the autonomous altimeter (AA400) is 0-3000 mm, with an instrumental error of 2%. The

Observation Instrumentation
A seabed-based tripod was designed to record the seabed surface hydrodynamic functions and SSCs. The design is optimized to maintain a stable posture on the seabed, and includes features to reduce subsidence and prevent the sideslip caused by storm waves and currents. The threaded holes on the tripod pedestals allow the installation of steel bars for insertion into the seabed. Therefore, the seabed-based tripod can remain stable, as the load is scattered through the pile foundations.
To measure erosion and deposition, current velocities, and turbidity we mounted a pack of sediment and hydrodynamic instruments on a tripod, including autonomous altimeter, current velocity meter, Argus surface meter, turbidity meter and wave tide gauge (see details in Figure 2). The range of the autonomous altimeter (AA400) is 0-3000 mm, with an instrumental error of 2%. The sampling frequency is 1.0 Hz, and the sampling interval is 30 min. The range, accuracy, and direction of the ALEC-EM current velocity meter is 0-500 cm/s, 1.0 cm/s, and eastward, respectively. Its sampling frequency and interval is 1.0 Hz and 10 min, respectively. The Argus surface meter (ASM-4) consists of tilt sensors, temperature sensors, microcontrollers, a memory stick, a power supply, and a sensor stick with a length of 96 cm and sensor spacing of 1 cm; the range is 0-2000 NTU, and the sampling interval is 15 min. The range and instrumental error of the XR-420 turbidity meter is 0-4000 NTU and 2%, respectively, while the sampling frequency and interval is 1 Hz and 2 min, respectively. The range and instrumental error of the wave tide gauge (RB16-TWR-2050) is 0-25 m and 0.05%, respectively. This gauge measures the water pressure every 30 min with a sampling frequency of 1 Hz. To assess the sediment strength, a 5 m sediment core was collected from the observation site.
NTU, and the sampling interval is 15 min. The range and instrumental error of the XR-420 turbidity meter is 0-4000 NTU and 2%, respectively, while the sampling frequency and interval is 1 Hz and 2 min, respectively. The range and instrumental error of the wave tide gauge (RB16-TWR-2050) is 0-25 m and 0.05%, respectively. This gauge measures the water pressure every 30 min with a sampling frequency of 1 Hz. To assess the sediment strength, a 5 m sediment core was collected from the observation site.

Calculation
The wave-induced shear stress was calculated by τ wmax = 0.5ρ w f w U max 2 (1)

Calculation
The wave-induced shear stress was calculated by τ wmax = 0.5ρ w f w U 2 max (1) [52], where τ wmax is the maximum wave-induced shear stress (Pa); ρ w is seawater density, assumed to be 1.025 g/cm3; f w is the wave friction coefficient, assumed to be 0.01 [14]; and U max is the maximum wave orbital velocity (m/s).
The current-induced shear stress was calculated as [52] where τ c is the current-induced shear stress (Pa); C d is the traction coefficient, which is assumed to be 3.1 × 10 −3 [14]; and U c is the observed current velocity (cm/s). The bed shear stress due to the current when both currents and waves coexist is determined by Water 2020, 12, 845 5 of 17 where n is Manning's roughness coefficient, and c w is a coefficient, assumed to be 0.65 [53].
In the research region, the direction of waves could be similar to that of currents as they are induced by wind. Therefore, the total shear stress is the sum of the wave-and current-induced shear stresses: [54] where τ is the total stress (Pa), τ wmax is the maximum wave-induced shear stress (Pa), and τ c is the current-induced shear stress (Pa). The critical shear stress (τ cri ) is estimated as: [55] where P is the cohesive force (KPa). We employed measures for decomposing single-wide sediment loads [11,56] to analyse the source of the suspended sediment flux. The per-tidal-cycle current velocity, water depth, and suspended sediment content are decomposed into mean and pulsating values. The single-wide load can be expressed as: [57] where T is the single-wide sediment load (mg/(cm*s)), C is the suspended sediment concentration (g/L), V is the current velocity (cm/s), and H is the water depth (cm). V, C, and H are the average values of the depth-averaged current velocity, water depth, and depth-averaged suspended sediment concentration, respectively, and V , C , H and are the pulsating values of the current velocity, water depth, and suspended sediment concentration, respectively. Over a single tide cycle, A = C = H . The single-wide load during a tide cycle can be expressed as: where C and H are positive constants, and V may be positive or negative. When V is positive, it is the flood tidal current velocity, and when V is negative, it is the ebb current velocity. T 1 and T 3 are the contributions of the advection transport volume and Stokes drift, and the sum of T 1 and T 3 is the transport volume. T 2 , T 4 , and T 5 are components of C , and the sediment concentration in water can be approximated as a constant during a tide cycle. The values of C are affected by the two-way exchange of sediment between the sediment and water. The sum of T 2 , T 4 , and T 5 is the resuspended sediment volume. The suspended sediment content can be expressed: where C i is the suspended sediment concentration at different depths (g/L). From Formulas (7) and (9), the resuspended sediment concentration can be expressed as: C r VH/CVH = (T 2 + T 4 + T 5 )/T Water 2020, 12, 845 where C r is the resuspended sediment concentration (g/L). From the above formulas, the transport sediment concentration can be calculated as: where C a is the advection transport sediment concentration (g/L), C s is the Stokes drift transport sediment concentration (g/L), and C t is the transport sediment concentration (g/L). Using the above formulas, the suspended sediment content and contribution of current transport (tidal current and Stokes drift) and resuspension to the suspended sediment content can be calculated.

Hydrodynamic Parameters
The winter hydrodynamic parameters were measured for 25 days by taking in-situ observations. The observed hydrodynamic state variables are shown in  and 10 January 2017, the significant wave height increased by over 1 m due to wind. The tidal range during the second observation period was 0.12-1.21 m, and the minimum and maximum water depths were 6.62 and 8.93 m, respectively ( Figure 5). The maximum of current velocity reached 49.40 cm/s. Generally, five sea conditions were observed, according to the in-situ measurements, including rough (2.5-4.0 m), moderate (1.25-2.5 m), slight (0.5-1.25 m), smooth-wavelet (0.1-0.5 m), and calm-rippled (0-0.1 m).

Suspended Sediment Concentration
The SSC varies greatly under different sea conditions. The SSC profiles show that during the observation period the SSC was mainly below 2.5 g/L, and its maximum level (>12.5 g/L) occurred during the storm surge ( Figure 6). The high-tide period is normally characterized by increasing SSC, which exceeded 4.5 g/L. During the storm period from 15-17 December 2014, the SSC was well-distributed throughout the water column (Figure 6a), which could reflect the strong vertical

Suspended Sediment Concentration
The SSC varies greatly under different sea conditions. The SSC profiles show that during the observation period the SSC was mainly below 2.5 g/L, and its maximum level (>12.5 g/L) occurred during the storm surge ( Figure 6). The high-tide period is normally characterized by increasing SSC, which exceeded 4.5 g/L. During the storm period from 15-17 December 2014, the SSC was well-distributed throughout the water column (Figure 6a), which could reflect the strong vertical mixing due to the enhanced waves or storm surge [58]. The SSC appeared to be irregular from 26 December 2016 to 4 January 2017 after the significant wave height increased to 2.7 m (Figures 5 and  6b). These irregular SSC (lutoclines) occurred at about 90cm above the seabed (Figure 6b), and these lutoclines resulted in average SSC records between 12/19 and 12/22 (all below 2 g/L), which is less than records between 12/30 and 01/05 (ranging from 2-3 g/L) despite the similar tide/current/wave conditions ( Figure 5). After 5 January 2017, the SSC was also well-distributed in the water as the lutoclines were broken by stronger waves (Figure 6b). Thus, the hydrodynamic parameters under different sea conditions were the main factors affecting the vertical distributions of SSC.

Physical and Mechanical Properties of Seabed Sediment
The 5 m sediment core showed that the sediment is mainly composed of silt (77.3%) and clay (18.5%) ( Table 1). A new cone penetration test system (CPTs) [59], which was developed by the Ocean University of China, was used to test the in-situ sediment shear strength. The results show that there is a comparatively hard crust in the core from 0.8 to 1.6 m. Previous studies found that the sediment strength in the study area is below 150 kPa [60], but our test results show a relatively high strength of 130-290 KPa (

Physical and Mechanical Properties of Seabed Sediment
The 5 m sediment core showed that the sediment is mainly composed of silt (77.3%) and clay (18.5%) ( Table 1). A new cone penetration test system (CPTs) [59], which was developed by the Ocean University of China, was used to test the in-situ sediment shear strength. The results show that there is a comparatively hard crust in the core from 0.8 to 1.6 m. Previous studies found that the sediment strength in the study area is below 150 kPa [60], but our test results show a relatively high strength of 130-290 KPa (Table 2).  a. Data represent the mean values obtained from the empirical formula f k = 0.043P s + 0.06 [61], where f k is the bearing capacity and P s is the cone resistance.

Contribution of Waves and Currents to Resuspension
The suspended sediment is closely related to the sea conditions, which are, therefore, associated with the processes of sediment resuspension and transportation. During the observation period, the transport sediment concentration (TSC) accounted for 69.3%-100.0% of the SSC, and its maximum level was 2.8 g/L. (Figure 7).  a. Data represent the mean values obtained from the empirical formula f k = 0.043P s + 0.06 [61], where f k is the bearing capacity and P s is the cone resistance.

Contribution of Waves and Currents to Resuspension
The suspended sediment is closely related to the sea conditions, which are, therefore, associated with the processes of sediment resuspension and transportation. During the observation period, the transport sediment concentration (TSC) accounted for 69.3%-100.0% of the SSC, and its maximum level was 2.8 g/L. (Figure 7). Equation (5) was used to analyze the contributions of waves and currents, as the correlation between the SSC and total shear stress calculated by Equation (5) (Figure 8) was better than that calculated by Equation (3) (Figure 9). The R, R2 of Figure 8 are larger than in Figure 9, and the mean square and standard errors of Figure 8 are less than Figure 9 ( Table 3). During the first observation period, the study site experienced rough (significant wave height of 2.5-4.0 m) and moderate (significant wave height of 1.25-2.5 m) sea conditions. Under these two conditions, the fraction of the resuspended sediment concentration (RSC) induced by waves was much greater than that induced by currents. The contribution of waves to the total RSC was 85.1%-99.7% during the rough period (Figure 10e), and 71.4-99.6% during the moderate period (Figure 10d). The maximum RSC induced by waves reached 2.2 and 1.2 g/L, respectively.  Equation (5) was used to analyze the contributions of waves and currents, as the correlation between the SSC and total shear stress calculated by Equation (5) (Figure 8) was better than that calculated by Equation (3) (Figure 9). The R, R2 of Figure 8 are larger than in Figure 9, and the mean square and standard errors of Figure 8 are less than Figure 9 ( Table 3). During the first observation period, the study site experienced rough (significant wave height of 2.5-4.0 m) and moderate (significant wave height of 1.25-2.5 m) sea conditions. Under these two conditions, the fraction of the resuspended sediment concentration (RSC) induced by waves was much greater than that induced by currents. The contribution of waves to the total RSC was 85.1%-99.7% during the rough period (Figure 10e), and 71.4-99.6% during the moderate period (Figure 10d). The maximum RSC induced by waves reached 2.2 and 1.2 g/L, respectively.      During the slight (significant wave height of 0.5-1.25 m) period, the RSC induced by currents was similar to that induced by waves (Figure 10c). The contribution of currents to the total RSC was between 3.0% and 87.8%, and the maximum RSC induced by currents and waves reached 0.79 and 1.02 g/L, respectively. Under smooth-wavelet (significant wave height 0.1-0.5 m) and calm-rippled (significant wave height 0-0.1 m) conditions, the RSC induced by currents was much higher than that induced by waves (Figure 10a,b). The fraction of RSC induced by currents during the smooth-wavelet and calm-rippled periods were as low as 30.9% and 77.7% of the total RSC, respectively, and the maximum RSC induced by currents under these two conditions reached 1.26 and 0.78 g/L, respectively.
Generally, the RSC induced by currents exceeded that induced by waves under calm-rippled and smooth-wavelet sea conditions, while the RSC induced by waves under slight, moderate, and rough sea conditions was higher than that induced by currents during the observation period. Although the RSC induced by currents decreased as the significant wave height increased when the sea conditions changed from calm-rippled to rough (Figure 10a-e), the current velocities changed slightly under different sea conditions (Figure 4). The result shows that the stress induced by currents positively affects resuspension when waves are weak. The effect of the current decreased when sediment resuspension increased with a significant increase in wave height.

Relationship between SSC and Resuspension
During the observation period, the SSC did not increase with a significant increase in the wave height beyond 1.25 m ( Figure 11). From 26-27 December 2016, the total shear stress increased to 10 Pa as the significant wave height increased to 2.7 m, and then decreased rapidly (Figures 4b and 12). However, the SSC (including the averaged and near-bed SSCs) decreased slowly. The change in the near-bed SSC lagged behind the average SSC after the total stress reached 10 Pa (Figure 12). This indicates that the sediment does not immediately reach the seabed during flocculent settling. The flocculent settling changed to hindered settling when the SSC exceeded 3 g/L, and the hydrodynamic conditions did not distribute the SSC well in water. Under this condition, the settling velocity of suspended sediment decreased with increasing SSC [62]. This resulted in lutoclines with a thickness of approximately 20 cm (Figure 13). When the total shear stress exceeded the critical shear stress (0.32 Pa), which was calculated by Equation (6), the near bed SSC record between 17 December 2016 and 5 January 2017 was higher, despite the similar shear stress (Figure 14). This shows that the SSC exchange velocity between the lutoclines and lower water was reduced. The absolute value of SSC increased due to the decrease in settling velocity, and the SSC was higher in the lutoclines and lower water. When the significant wave height increased and broke the lutoclines, flocculent settling increased the settling velocity and caused the absolute SSC value to decrease. These effects weakened the resuspension-induced RSC, and the RSC induced by currents decreased when the significant wave height increased while the current velocities changed slightly.

Conclusions
The in-situ monitoring records show that the changes and sources of SSC changes mainly depend on the sea conditions in the Yellow River Delta-front. The SSCs were normally less than 2.5 g/L, and high SSCs (>4.5 g/L) occurred during high-tide periods. During the storm surge, the SSC

Conclusions
The in-situ monitoring records show that the changes and sources of SSC changes mainly depend on the sea conditions in the Yellow River Delta-front. The SSCs were normally less than 2.5 g/L, and high SSCs (>4.5 g/L) occurred during high-tide periods. During the storm surge, the SSC reached its maximum (>12.5 g/L) and was uniform throughout the water column due to the strong vertical mixing caused by the enhanced waves. The main source of suspended sediment during the observation period is sediment transport, accounting for 69.3%-100.0% of the SSC. When the significant wave height is below 1.25 m, the contribution of currents to the total RSC exceeded that of waves. However, the contributions of waves to the total RSC were 85.1%-99.7% and 71.4%-99.6% during the rough and moderate periods, respectively. The 20 cm-thick lutoclines occurred after the significant wave height increased to its peak value and then decreased. When the flocculent settling of suspended sediment hindered in the Yellow River Delta when the SSC exceeded 3 g/L and the hydrodynamic conditions could not break the lutoclines. Under hindered settling, the resuspended sediment stagnated in the lutoclines and lower water as the settling velocity reduced from 27 December 2016 to 11 January 2017. When the significant wave height exceeded 1 m, the lutoclines were broken, the flocculent settling increased the settling velocity, and the SSC was well diffused in the water. This study reveals different controlling factors for the high SSC near a river-influenced delta, and helps us get a better understanding of a delta's resuspension and settling mechanisms. The future studies are comparing winter with summer.