Fluid Mud Measurement and Siltation Analysis in A Trial Excavated Channel in the Approach Channel of the Xiangshan Port

In order to clarify the distribution and variation of silt and fluid mud in the Waiganmen shallow section of the 50000-ton intake channel of the Xiangshan Port, and to understand the influence of the channel excavation on the surrounding flow conditions and the strength of the backsilting, especially the impact of typhoon on the sudden silting of the channel, so as to demonstrate the feasibility and stability of the channel excavation. The fluid mud, hydraulic, sediment and topographic measurements were carried out in the study area, and the thickness of the fluid mud layers, tidal current, sediment and topographic data were obtained. Dual-frequency sounder, gamma-ray densitometer and SILAS navigational fluid mud measurement system were used to monitor the fluid mud, and the results were compared and verified. The adaptability and accuracy of the three methods were analyzed. The SILAS navigational continuous density measurement system and gamma-ray fixed-point fluid mud measurement are used to detect the density, thickness and variation of the fluid mud accurately. Based on the hydrological observation data, the process of erosion and deposition in excavation channel and its influence mechanism are analyzed, and the distribution characteristics and evolution law of siltation in engineering area are given in the form of empirical formula. The research shows that the super typhoon can produce large siltation, which results in sudden siltation of the channel. The tidal current is the main dynamic factor of the change of erosion and siltation of the excavation trench. Under the influence of reciprocating tidal current and excavation topography, the trial excavation trench is silted on the whole. There is fluid mud in the monitoring area of the trench, and the distribution of fluid mud is different in space. The thickness of the fluid mud at the bottom of the trench is generally larger than that outside the trench and the slope of the trench, and the siltation of the trench tends to be slow. The research results can provide scientific evaluation for channel excavation and maintenance, and support for the implementation of the project.


Introduction
Zhejiang Guohua Ninghai Power Plant is located at the top of the Xiangshan Port (Xu and Wang, 2009). The installed capacity of the first phase of the project is 4×600 MW, and the annual coal consumption is 6 million tons. In the first phase of the wharf, two 35000-ton berths have been built for coal unloading and a 3000-ton comprehensive wharf was also built. The first stage channel project is designed as 35000-ton channel, and the cargo ship enters the port by taking advantage of the natural water depth and tide. The approach channel is basically not dredged. In order to adapt to the situation of large-scale bulk cargo ships, reduce the cost of coal transportation, reduce the operating cost of the power plant, give full play to the economic benefits of the project investment, it is necessary to develop a 50000ton waterway project. According to the research of Zhejiang Transportation Design Institute, it is necessary to dredge the shallow section of the main gate outside the approach channel to ensure the navigation of 50000-ton ships.
In order to ensure the smooth development of the channel, accumulate relevant experience, provide technical data of channel excavation and demonstrate the feasibility of channel excavation, a trial excavation test is carried out for the shallow section outside the Xiangshan Bay gate. According to the characteristics of sediment movement and channel siltation in the vicinity of this area, this section is about 3 km away from the northeast of Dongyu Island. Because it is located in the water area outside the mouth, it is easy to be affected by waves, and the bottom sediment is suspended, settled, sorted and exchanged actively. After the channel is excavated, it is relatively easy to be silted up. Through the observation and analysis of the trial excavation section, we can understand the silting situation of the channel after the implementation of the project, comprehensively evaluate the excavation width, navigable depth, channel slope and channel maintenance volume of the channel, accumulate the experience and data for further channel excavation, and ensure that the channel excavation is carried out stably, which is of great practical significance for the implementation of the project.

Status of survey areas
The Xiangshan Port is located on the central coast of Zhejiang Province, China. It lies to the west of Liuheng Island, Zhoushan archipelago, east of Damu Ocean. The climate is mid-subtropical monsoon. The harbor area has asuperior natural environment, a vast water area, beaches, fertile water quality, various aquatic products, and strong natural productivity in the waters. The trial trench (the main area of field monitoring and research) is located in the shallow section of the Waiganmen Island outside the entrance of the Xiangshan Bay, approximately 3 km up the northeast side of Dongyu Mountain. Fig. 1 shows the geographical location of the study site. Based on relevant researches Zhou and Tian, 2001) the hydrological conditions in this area are summarized as follows.

Wind conditions
The northwest wind prevails from October to next February every year. When the strong cold waves come, the maximum wind power can reach up to 8−9 grades, and the south-southeast wind prevails from June to August. Except for typhoon season, the general wind power is only 2−3 grades. Typhoons have an average of 2.2 times per year, mainly from July to September. The impact duration of each typhoon averages 3 days, with the maximum of 5−6 days.

Tidal pattern
The Xiangshan Port waters are characterized by typical irregular half-day shallow water tides (Wu et al., 2015). The shallow water effect is more significant. The phenomena of diurnal and nighttime tides are obvious. The duration of flood tide is longer than that of ebb tide, the velocity of ebb tide is larger than that of flood tide, and the velocity of tidal current along the way is different.

Sediment
The sediment movement in the Xiangshan Port is often controlled by its hydrodynamic conditions and material sources. According to the analysis of the characteristics of the natural geographical environment of the Xiangshan Port, its hydrodynamic environment is mainly affected by land runoff, coastal current of the East China Sea, tidal current and wave. According to the measurement data, research results (Zhou et al., 2014) show that the suspended sediment concentration in winter is significantly higher than that in summer, and the magnitude is larger during spring tide than that during neap tide in the same season. Spatially, the suspended sediment concentration decreases constantly from the mouth to the top of the bay. The distribution of suspended sediment is mainly influenced by the alternative effect of coastal current in Fujian and Zhejiang with high suspended sediment concentration in winter and Taiwan warm current with low suspended sediment concentration in summer, and by the tidal dynamic changes in semimonthly tidal cycle.

On-site monitoring
In order to effectively obtain the thickness, distribution and variation of mud and fluid mud in the study area, the dual-frequency sounder, gamma-ray densitometer and SILAS cruising mud observation system are used to monitor the fluid mud.

Test section
The fixed-section and underwater topographic surveys were conducted in the same configuration. Twenty-four cross sections and four vertical sections were uniformly arranged in the monitoring area of the trial trench. The cross sections and vertical sections were placed perpendicular and parallel to the center line of the test trench, respectively. The distances between the measuring lines and measuring points were 50 m and 4 m, respectively. The system delay of the ship borne equipment was determined by conducting a round-trip survey, and the measuring accuracy of the cross section was checked on the longitudinal sections. When an obvious cross-sectional change was detected, the real underwater-section condition was determined via the artificial infill point method implemented in real-time. Fig. 2 shows the fixed longitudinal and cross section layout of the monitoring area.
As shown in Fig. 2, the base points of both the transverse and longitudinal survey sections are the north end of the trench. In order to facilitate the analysis and discussion below, 24 transverse survey sections are labeled T-#1 to T-#24, and 4 longitudinal survey sections are labeled L-#1 to L-#4. Fig. 3 shows the layout of hydrographic survey, suspended sediment concentration and sediment sampling survey lines and survey points. The specific discussion of the relevant measurements will be given below.

Content and frequency of tests
This project included the following observations: 1) the whole study area, monitored by a control network of Global Positioning Systems (GPSs); 2) long-term tidal-level observations, monitored by a layout of tidal stations; 3) fixed-section monitoring; 4) underwater topographic survey; 5) hydrometry; 6) sampling and analysis of the suspended sediment concentration; 7) sampling and analysis of the seabed sediment; and 8) special observation of fluid mud. Table 1 lists the time and frequency requirements of each monitor-ing study.

Observation of hydrological elements
The hydrological test of this project was carried out three times, namely, before the start of the trial excavation,   DING Jian et al. China Ocean Eng., 2020, Vol. 34, No. 3, P. 421-431 after the completion of the excavation and after the impact of severe cold wave. Among them, the pre-construction test is the tide observation, and after the completion of acceptance and the impact of the severe cold wave, three total tides, i.e. high tide, middle tide and low tide, were measured. The sediment concentration of water body has been observed four times. The measurements are set before dredging, 3 months after acceptance and 6 months after completion. There are 10 stations for suspended sediment concentration measurement. The measurement time were selected for the maximum flood velocity, flood slack, the maximum ebb velocity and ebb slack during the spring tide period. The stratified water samples of 0.2H, 0.6H and 0.8H are collected at each measuring point. The velocity and direction of flow are observed at the same time. The seabed sediment samples were arranged in plum-blossom shape for five times. The specific sampling time was 5 days, 1 month, 2 months, 4 months and 6 months after completion. A towed sampler was used for sediment sampling, and 35 sediment samples were collected each time.

Observation of flow velocity
During the implementation of the test, GPS beacon was used to locate the ship, and the ZSX series digital current meter produced by Chongqing hydrologic instrument company was used to observe the flow velocity and direction. The time interval for the measurement is 1 hour. Based on the statistical analysis of the previous observation data of hydrological elements in the study area, the following hydrological and sediment characteristics of the study area can be obtained. In the monitoring area, the average duration of rising tide is 5 hours and 45 minutes, and the average duration of falling tide is 6 hours and 37 minutes. The flow direction of flood tide is 310°−320° and that of ebb tide is 130°−140°. Compared with the flow pattern before and after excavation, the flow direction of flood tide changes within 5° and that of ebb tide is 4°−9°. The flow direction of rising and falling tide changes slightly. The maximum velocity of rising tide is 1.75 m/s, and the mean velocity in vertical is 1.41 m/s; the maximum velocity of falling tide is 1.41 m/s, and the average velocity in vertical is 1.11 m/s. The average tidal range before excavation is 4.58 m, the average tidal range after excavation is 3.22 m, The corresponding tidal current velocity before excavation is higher than that measured after excavation. Compared with the ratio of ebb tide velocity to flood tide velocity, the ratio decreases after excavation, that is to say, the strength of flood tide increases after excavation, and the variation amplitude in the groove is larger than that in the outer groove.

Water sample collection and analysis
The horizontal sampler is used for water sample collection. Vertical suspended sediment samples and salinity water samples, whose volume is 1000 ml, were collected syn-chronously with velocity in fixed-point test. Suspended sediment particle graded water samples were collected at the first half of each tidal cycle at the flood rush, flood slack, ebb rush and ebb slack moments.
The indoor sediment content was analyzed by baking method. The gradation analysis of the samples was carried out by the MASTER SIZER 2000 laser particle size analyzer made in Britain. The indoor salinity analysis uses a certain volume of clarified water directly extracted from suspended sediment samples. The actual analysis was carried out with SYA2-2 salinity meter produced by the National Institute of Marine Technology.
The results show that, during spring tide, the maximum average sediment concentration of rising tide is 0.502 kg/m 3 , and that of falling tide is 0.619 kg/m 3 . During the middle tide period, the maximum average sediment concentration of rising tide is 0.271 kg/m 3 , and that of falling tide is 0.413 kg/m 3 . During neap tide, the maximum average sediment concentration of rising tide is 0.107 kg/m 3 , and that of falling tide is 0.115 kg/m 3 . On the measuring line in the trough, the spring tide sediment concentration is about 5−6 times of that in the neap tide; on the two measuring lines outside the trough, the spring tide sediment concentration is about 7−10 times of that in the neap tide. Most of the median particle size of suspended sediment is smaller than 0.008 mm, which belongs to clayey silt. Sand, silt and clay account for smaller than 10%, 60%−65% and 30%−35%, respectively. The salinity is between 29.7 and 30.4.
The sediment concentration before excavation is larger than that after excavation, and the average sediment concentration before excavation is 1.78−2.21 times of that after excavation; the average sediment concentration before excavation is 1.59−1.89 times of that after excavation. After excavation, the salinity is higher than that before excavation. The increase range of salinity in flood tide is 6%−7%, and the change range of salinity in ebb tide is 5%−6%, which belongs to weak mixing type. The fluctuation amplitude of high tide is slightly larger than that of low tide.

Seabed sediment sample collection and analysis
Sampling is carried out by a towed sampler, and the weight of each sample is no lighter than 500 g. In the laboratory analysis of sediment particle size distribution, the laser particle size analyzer, pipette method and sieving method were used, depending on the size of the sample. The distribution of the median grain size of the bottom material near the trial excavation channel were measured in May, August and December in 2007. The measurement results shows the bed sediments are mainly clayey silt, in which silt accounts for 65.9%−73.5% and sand accounts for 21.7%−30.1%. The size difference between suspended load and bed sand is small, and they exchange frequently. After trenching, the grain size of the local bottom sediment in the navigation channel is coarser. Except for the slope area, the grain size of the bottom sediment is about 0.01 mm, slightly coarser than before the project, which is mainly related to the change of the distribution of the bottom sediment after trenching. At the same time, it can be seen that the particle size of the seabed sediment is coarser due to the collapse caused by the instability of the slope and the convex part of the trenching. Due to the adjustment of sediment back silting and tidal current, in addition to the collapse of local areas, the particle size of sediment bottom in the navigation channel is smaller than that in May as a whole, which indicates that the sediment deposition in the navigation channel is mainly from fine-grained material, which is similar to the particle size of the bottom material before the project and close to the particle size of suspended sediment. It also indicates that in addition to the collapse caused by local excavation, the particle size of the main sediment in the channel is the same as that of the sea bed on both sides before and after the project. The particle size of the suspended sand is close to each other, which means that the siltation of the sediment in the excavated navigation channel is mainly the deposition of the suspended sand, and the unstable collapse caused by the excavation of the channel is only part of it.

Siltation monitoring and fluid mud observation
The measurement of fluid mud and its related influencing factors are the research basis of sediment problems in estuary and waterway engineering (McAnally et al., 2007a(McAnally et al., , 2007b. Systematic field observation with high-resolution observation methods is an important prerequisite for discussing estuary velocity and sediment distribution structure and studying its formation mechanism (Hoitink and Hoekstra, 2005;Wu et al., 2003). Because the fluid mud is easy to be disturbed, the detection and sampling of the fluid mud is difficult (Guo, 2010). These difficulties prompt hydrologists to invent new instruments and improve the measurement method (McAnally et al., 2007b). With modern advanced optical and acoustic instruments, some new mud measurement systems were developed for the observation of water and sediment in the bottom boundary layer (Liu et al., 2006;Xu et al., 2009;Ling et al., 1997;Yang and Pan, 2012).

Fluid mud observation method
Fluid-mud observation data mainly include dual-frequency measurement data, SILAS system data, gamma-ray densitometer data, and fluid-mud distribution maps drawn at different densities. Because the dual-frequency measurements are performed in non-tidal depth mode, low-frequency data are processed with the same method as those used for single-beam echo soundings. Then, the silt thickness at the measuring point is calculated from the processing results. Further, the SILAS system edits the sediment column information of the simulated reflection signals of the line seabed strata. First, the SILAS system determ-ines the accuracy of the fixed-point densities measured by the tuning-fork vibration densitometer in each route, and then establishes the relationship between the densitometermeasured densities and the intensities of the reflected signals. After quantitative analysis of the reflected signals, the densities in different layers of the column profiles are determined at all measuring points. The existing conditions and thicknesses of the fluid mud are subsequently determined by classifying the densities. To process the data of the gamma-ray density meter, standard samples of known densities are prepared from the number of the samples obtained in the field. The densities of the standard samples are measured with the gamma-ray density meter, and the corresponding count rates of the gamma-ray decay are recorded. The density data are regressed against the count rates. Finally, the mud and silt thicknesses are calculated with the regression equation and the variation curve of the gamma-ray density meter in water.

Observation accuracy of fluid mud
The mud in the study area was observed with the SILAS system. To ensure the quality of the mud observations, the tuning-fork density meter was deployed twice in the field. Fig. 4 and Fig. 5 show the corresponding calibration and comparison results of six fluid mud samples numbered 1 to 6 for the two stages. As confirmed in Figs. 4 and 5, the accuracies of the calibrated densitometer results are better than the instrument's nominal precision. Along the SILAS aerial survey line, sampling was repeated at three sections, and five test points were checked with the tuning-fork density meter. The silt thicknesses D d calibrated by the densitometer and D tf measured by the tuning-fork density meter are compared in Fig. 6.
As shown in Fig. 6, the silt thicknesses were accurately observed with tuning fork vibration densitometer at the five points. To verify the reliability of the observation results of the SILAS system, the same points numbered from 1 to 28 were observed by the gamma-ray fixed-point silt-density profiler and the SILAS cruise mud measurement system. Based on the comparison of the measurement data of the previous dual frequency sounder, SALIS system and Gamma ray densitometer, it is found that the thickness of the muddy layer measured by the dual frequency sounder and SALIS system is basically the same, the depth range measured by the dual frequency sounder is larger, which can clearly reflect the profile of the soft layer on the seabed, and is suitable for the navigable bathymetry of the channel.
The disadvantage is that it is difficult to define the argillaceous characteristics of mud and fluid mud measured by the dual frequency sounder. In terms of fluid mud measurement, the thickness of fluid mud measured by SALIS system and gamma ray densitometer at the same position is basically the same, and they can accurately measure the density of fluid mud and other characteristics. However, Gamma ray densitometer adopts the operation mode of measuring line hanging, which is easy to be affected by wind and waves, water flow, etc., so it is difficult to be accurately located, and only the density vertical gradient of a single point can be obtained. Therefore, in this study, the observation results of the SALIS system are used as the basic data for the analysis of the characteristics of the fluid mud, and the measurement data of the dual frequency sounder and gamma ray densitometer are used for the rationality check and evidence.

Statistics of fluid mud reserves of various densities
After calculating the density-gradient mud reserves (see Fig. 8 to Fig. 13), the density of the fluid mud was found to vary from 1.1 to 1.3 g/cm 3 . Moreover, the average thickness of the fluid mud was of decimeter order in the trench area.     In the above figures, D Smin , D Smax , and D Save respectively represent the statistical minimum, maximum and average values of the fluid mud thickness corresponding to each density measured in September, similarly, D Dmin , D Dmax , and D Dave represent the corresponding measurement statistics for December, respectively. S S , S D , V S , and V D represent the area and volume of the fluid mud layer corresponding to each density measured in September and December. It can be seen that the variation trend of each quantity with density is obvious, and the quantity related to thickness is approximately positively correlated with density. From September to December, the area of the mud layer corresponding to each density is increasing, while the thickness and volume of the mud layer are decreasing, which indicates that the mud layer gradually expands and diffuses with time.

Variability
Owing to the strong fluidity of the fluid mud, the amount of mud significantly changed between August-September 2007 and September-December 2007 observation periods (called the first and second stages, respectively, in Fig. 14). The volume difference of the fluid mud layer between different regions in different stages is shown in Fig. 15. V tf −V bf represent the volume difference between the mud layer of the trench area and mud layer of the bottom      DING Jian et al. China Ocean Eng., 2020, Vol. 34, No. 3, P. 421-431 427 area in the first measurement stage, while, V sf −V tf represent the volume difference between the mud layer of the survey area and mud layer of the trench area in the first measurement stage. V ts −V bs , V ss −V ts correspond to the same definition of the second measurement result, respectively. As seen in Fig. 14 and Fig. 15, the volume of the fluid mud is proportional to the density of the fluid mud, which indicates that the sedimentation and compaction evolution of the fluid mud are uniform, and the dynamic environment of the study area is relatively stable. From September to December, the erosion volume (thickness) of each density of the fluid mud layer is basically the same (Winterwerp et al., 2012). The fitting relationship between the volume of the fluid mud in each density layer and its density shown in Fig. 14 is as follows: (1) The mud reserves were obviously lower in the second stage than those in the first stage. The measured data confirmed larger thickness of mud inside than that outside the trench, and higher thickness in the northwest than that in the southwest part of the trench. As a whole, the fluid mud moves from the outside of the trench to the inside of the trench, and transports from dredging southeast to northwest. From September to December, the reserves of fluid mud decreased at each density. In the plane distribution, main scouring appeared at the northwestern side of the trench, where the fluid mud reserves decreased. However, the southeastern side was predominantly silted, and the reserves of fluid mud increased at that side. Therefore, the trench was predominantly scoured and silted along the directions of ebb and flood tide current, respectively. The mud reserves (with densities ranging from 1.20 to 1.30 g/m 3 ) varied more widely as their density gradients increased. The fluid mud reserves in this density range are greatly variable. Under topographic siltation conditions, compaction slowly consolidates the silt layer into a dense silty clay layer. However, the low-density mud reserves (1.10 g/m 3 ) changed only slightly, and the 1.10 g/m 3 mud layer was strongly fluid and existed in dynamic equilibrium.

Calculation and analysis of the erosion and deposition
To study the silting distribution and characteristics of the trial dredging engineering area, the scouring and siltation amounts in the trial trench were analyzed from three-dimensional data of the underwater topography of the fixed section obtained via high-precision modification. The observation data revealed the existing conditions, thickness variations, and distribution characteristics of the fluid mud.

Calculation and analysis of scour and deposition
Erosion and siltation at the study site were calculated from the underwater topographic monitoring data collected from March to December 2007. The corresponding representative results are shown in Fig. 16 and Fig. 17.
Comparing the monitoring data of each fixed section revealed the change of erosion and siltation to be small in the shoal of the trench, and the bottom of the trench was found to be silted as a whole (Schettini et al., 2010). After the excavation, the bottom of the trench was scoured for a short period. Subsequently, the siltation began to accumulate and the deposition speed was gradually accelerated. From May to June after excavation, the deep channel siltation process shows obvious adjustment process in the first two months. At the initial stage of excavation, the slope and bottom of the newly excavated channel are not smooth, there are many concave and convex areas of the section, and the sediment content of the bottom layer is large. The water flow near the navigation channel is affected by the excavation, the crosssection of the water flow increases, and the flow velocity decreases, which causes the local adjustment of the trial excavation channel to be faster, and is reflected in the crosssection diagram that the convex section of the unstable terrain collapses or landslides under the action of the water flow, and the sediment in the groove of the concave area  quickly silts up. The back siltation after June is different from that from May to June. In addition to the normal erosion and siltation of the slope, the back siltation in the channel is strengthened, especially from July to August. The back siltation in the channel is the strongest since the trial excavation of the navigation channel. It can be seen from the water depth distribution map of the above section that there is siltation in some slopes and most of the grooves, and the protruding part of the grooves is scoured, which is mainly the collapse recovery of the local uneven part; while there is siltation in other areas and both sides of the grooves, except that the maximum siltation in the local deep grooves is larger than 1.00 m, the siltation in other areas is smaller than 1.00 m. From July to August 2007, affected by the super typhoon, there was a large area and thick deposition at the bottom of the trench. However, in the trenching area (including the bottom area and the trenching slope), the main part is siltation, but the amount of erosion increases from August to September. This process reflects the change of erosion and deposition caused by the change of tide dynamics due to storm surge. The sediment deposition at the bottom of the dredger decreased from November, 2007to March, 2008 and was generally stable. Overall, the slotting slope was well maintained with a silty bottom.
As confirmed in the above graphics, siltation dominated the main area from May to December. Based on the calculated erosion and deposition in different areas (survey, trench, bottom, and beach-edge areas), the changes in the erosions and depositions in the trench are both spatially and temporally dependent. Specifically, the amount of erosion gradually increased in the beach-edge area, and even more gradually in the bottom area. In contrast, the sediment was mainly deposited in the trench area (including the bottom area and trench slope). However, the amount of erosion increased from August to September, reflecting the change of tidal current dynamics after a storm surge. In the plane distribution, the beach erosion mainly accumulated in the middle part of the survey area. By analyzing the scouring and silting changes in the trench area, significant increases on the slotting slope are found in the middle part of the trench area during December. In the plane distribution of the trench bottom, the siltation was larger at both ends than that in the middle. According to the hydrological survey results (Han et al., 2014), the trench lies approximately parallel to the direction of the flood and ebb tide; furthermore, the siltation at both ends of the trench bottom was larger in the west and north directions than that in the southeast. Thus, the deposition is larger in the direction of ebb tides than that in the direction of flood tides. The siltation was always continuous and developing toward the middle part. Observing the erosion on the marginal bank and on the slope in the middle of the survey area, one finds continuous scouring in the middle part of the beach and the trench slope, whereas the slope part was constantly collapsing. Consequently, scouring sediments continuously flowed into the mid-bottom of the trench, and accumulated there.

Influence of gale weather on siltation
In 2007, a total of 8 Tropical Cyclones Landed in China, among which there were three super typhoons, namely No. 0708 super typhoon "Septa", No.0713 "Vipha" and No.0716 super typhoon "Krosa", which had significant impacts on the study area. The characteristics and impact process of these three typhoons are as follows.
On August 12, 2007, the super typhoon "Septa" was generated. At 20 o'clock on August 16, 2007, it was located in the sea about 630 km southeast of Taitung, Taiwan Province. It landed in Chongwu town, Hui'an, Fujian Province in the early morning of August 19. Xiangshan Port Area in Zhejiang Province was mainly affected by typhoon "Septa". Longgang Town, Cangnan County, Wenzhou City was hit by a tornado at 11:30 p.m. Tuesday. Fig. 18 shows the time variation process of the significant wave height during the super typhoon in the study area of trial trenching. It can be seen that from 0:00 on August 10 to 0:00 on August 11, 2007, the significant wave height is larger than 1.31 m, the maximum value is 3.2 m, and the average value is 1.04 m. From 1:00 on August 18 to 3:00 on August 20, 2007, the significant wave height is larger than 0.8 m, the maximum value is 1.45 m, and the average value is 1.09 m. According to the analysis of historical experience, when a typhoon crosses the Taiwan Strait, the coastal area of Fujian is prone to double water increase peaks. The first peak appears after the typhoon leaves Taiwan Island and enters the Taiwan Strait; the second peak appears before and after the typhoon lands on the coastal area of Fujian. The time course of significant wave height in Fig. 18 shows this characteristic.
At 08:00 on September 16, 2007, tropical storm "Vipha" was formed in the east of the Philippines, and then moved to the northwest. Its intensity increased rapidly. At 05:00 on September 18, it became a super typhoon. It landed in Xia- Fig. 18. Significant wave height distribution in the trenching area during the super typhoon 0708 and its adjacent period.
DING Jian et al. China Ocean Eng., 2020, Vol. 34, No. 3, P. 421-431 guan Town, Cangnan County, Zhejiang Province at 02:30 on September 19, with a central air pressure of 950 HPA and a central wind force of 14 levels (45 m/s). After landing, the typhoon weakened and entered Lishui, Zhejiang Province, and then gradually turned to the north direction. The duration of the great wind effect of "Vipha" was up to 60 hours. Fig. 19 shows the time variation process of the significant wave height during the super typhoon "Vipha" in the study area of trial trenching. It can be seen that from 17:00 on September 17 to 18:00 on September 19, 2007, the significant wave height is larger than 0.8 m, and the maximum value reaches 3.2 m, with an average value of 2.0 m.
At 08:00 on October 2, 2007, tropical storm No. 0716 "Krosa" was generated in the ocean east of Luzon Island, Philippines. At 20:00 on October 2, 2007, it was intensified into a strong tropical storm. At 02:00 on October 3, it was intensified into a typhoon, and at 02:00 on October 5, it was intensified into a super typhoon. At 15:30 on October 7, it landed at the junction of Zhejiang and Fujian near Xiaguan Town, Cangnan County, Zhejiang Province. At the time of landing, the central air pressure was 975 hpa, and the near central wind force was 12 levels (33 m/s). During the influence period of "Krosa", there was a 7-level gust at 17:00 on October 4, and with the typhoon center approaching, the wind increased rapidly. Before it landed and turned to the sea again, the wind in the coastal sea and inland areas of Wenzhou maintained above 12 and 9 levels, respectively, and the duration of the wind above 7 levels reached 115 hours in the whole process. Fig. 20 shows the time variation process of the significant wave height during the super typhoon "Krosa" in the trial trenching area. It can be seen that from 2:00 on October 5 to 6:00 on October 9, 2007, the significant wave height H s is larger than 0.8 m, and the maximum value is 2.8 m, with an average value of 1.65 m.
After the above three super typhoons, the thickness of the fluid mud in the trial trench was measured in order to understand the impact of the typhoon on the sudden siltation of the navigation channel. Fig. 21 is the measurement results of the fluid mud in representative transverse section after the three super typhoons.
It can be seen from the cross-section summary that the larger sudden in the excavation occurred after the impact of super typhoon "Septa" in August, that is, the deposition was slow and uniform before the impact of super typhoon; the two super typhoons in August and September also caused the overall scouring of the beach and the excavation slope, with an average scouring range of 6−7 cm. There is generally about 0.05−0.15 m scouring on the surface of the channel topography after "Vipha". The maximum siltation in the channel under the action of typhoon "Vipha" is smaller than 0.10 m, which is partially scoured by the action of wind and waves. The siltation characteristics in the channel under the action of typhoon "Krosa" are similar to those under the ac-tion of "Vipha", but its siltation intensity is larger than that under the action of typhoon "Vipha". The effect of "Krosa" on the seabed is relatively small, and the scouring depth of the seabed is generally smaller than 0.05 m. The typhoon wave in this area has a certain effect on the sea bed of −8.0 m, but the extent of wind wave caused erosion on the sea bed is small, generally smaller than 0.10 m.

Conclusions
In view of the characteristics of complex coastline and significant tidal changes in the survey area, using Silas navigation type continuous density measurement system of bottom mud, combined with gamma fixed-point fluid mud measurement, the accurate detection of fluid mud density, thickness and its change is realized.
As a whole, the test trench is silted up, and the super typhoon can produce a large amount of siltation. About one year after excavation, the average deposition thickness at the bottom of the excavation trench has been nearly 1 m, and the deposition tends to be slow at present. The thickness of the fluid mud at the bottom of the trench is generally larger than that outside the trench, and the side slope of the trench. The thickness of the fluid mud outside the trench is 0.15−0.20 m, and the thickness of the fluid mud at the bottom of the trench is generally 0.25−0.35 m, and the sedimentation tends to be slow. The plane distribution of silt thickness in the survey area is mainly affected by the fluctuating tidal current velocity.
The obtained monitoring results show that the special climate such as super typhoon has a significant impact on the erosion and deposition of the trench. It is suggested that the monitoring should be carried out immediately after the impact of storm surge, flood, strong cold wave and super typhoon in the later maintenance of the project.