Impacts of human activities on tidal dynamics in a sexta-diurnal tidal resonant bay

Using numerical modelling, we study changes in tidal dynamics in Daya Bay (DYB) between 1989 and 2014. During this period, a total water area of 30 km2 was reclaimed and the average water depth increased by 38 cm. As DYB is a sexta-diurnal tidal resonant bay, the sexta-diurnal tides respond differently to the coastline and bathymetry changes than other tides. Taking K1, M2, M4, and M6 as examples, model results show a decrease in tidal elevation amplitude, tidal current magnitude, and tidal energy flux for K1, M2, and M4 tides. For the M6 tide, however, the model predicted an increase in tidal elevation amplitude, tidal current magnitude in some parts of the bay, and the tidal energy flowing into the bay. Land reclamation leads to the enhancement of sexta-diurnal tidal resonance and thus the magnitude of the M6 tide. Furthermore, due to the magnification of M6, tidal duration asymmetry in DYB changed from ebb-dominance to flood-dominance, and water exchange became much more active. Therefore, owing to the sexta-diurnal tidal resonance, the impact of human activities on tidal dynamics in DYB is different from that in previously reported semi-enclosed bays where large-scale land reclamation has been carried out.


Introduction
Bays and estuaries affected by strong land-sea interactions are known to be indispensable regions with high primary productivity and rich biodiversity. However, coastal areas are suffering from the pressures of urbanization, especially the demand for land. Land reclamation profits the local economy, but it has many negative effects. For example, it changes tidal dynamics (e.g., Kang 1999;Park et al. 2014), shrinks water volume (e.g., Gao et al. 2014;Xiao et al. 2019), alters sediment erosion and deposition (e.g., Flemming and Nyandwi 1994;Gao et al. 2018), and may finally shorten the life of a bay. Large-scale land reclamation can also lead to damaged coastal habitats, rapid reductions in biological resources, and deterioration of marine ecosystems (e.g., Sato and Azuma 2002;Reise 2005;Hodoki and Murakami 2006;Yan et al. 2013;Jickells et al. 2016;Yuan et al. 2016). Over the past three decades, land reclamation has prevailed along China's coastline due to rapid urbanization and massive economic growth (Liu and Diamond 2005;Pelling et al. 2013;Song et al. 2013;Zhang et al. 2013Zhang et al. , 2017Li et al. 2018).
Daya Bay (DYB, Fig. 1) is a large, semi-enclosed bay with a coastline of 92 km and an area of 600 km 2 . It is located in Guangdong Province in southern China (114°29.7′-114°49.7′E, 22°31.2′-22°50.0′N). The maximum water depth is less than 21 m, and the average is 9.7 m (Yang 2001). There are more than 50 islands, but no major rivers discharge into the bay. It is a tidally dominated bay but has a poor water exchange rate . DYB has rich biological resources (Song et al. 2004;Wu et al. 2009), but it is suffering due to intensive human activities in recent years (Li et al. 2015). For example, from 2001 to 2011, more than 16.1 km of coastline was changed due to human activities -17.5% of the total coastline. Roughly 70% of DYB's coastline has been developed in the past decades, and about 8%, mainly in the northwestern end of the bay, has been completely changed from its original natural state ). Furthermore, DYB is the only tropical bay in the world with three nuclear power plants (NPPs); large amounts of cooling water from the NPPs discharge directly into the bay, resulting in heat pollution and declines in biological reproduction (Ke et al. 2010). Harmful algae blooms (Akashiwo sanguinea, Skeletonema costatum, Noctiluca scintillans, Protoperidinium quinquecorne, etc.) in DYB have been recorded since the 1980s, becoming more frequent in recent years (Zhong et al. 2002;Yu et al. 2007;Song et al. 2009;Dou et al. 2015). A comparison of two investigations, one undertaken from 1982 to 1990 and another from 1994 to 2007, shows that 10 genus and 47 species of diatoms, two genus and 14 species of dinoflagellates, and 15 species of zooplankton have disappeared. Furthermore, the individual weights of fish also decreased from 14.8 g/tail in 1985 to 10.8 g/tail in 2004 (Wang et al. , 2006(Wang et al. , 2008(Wang et al. , 2010Qiu et al. 2005;Li et al. 2015). This indicates that the marine ecosystem of DYB is deteriorating. Thus, it is urgent and important to evaluate the impact of human activities on the marine ecological environment of the bay, which requires a strong understanding of tidal dynamics in the bay.
This paper aims to study the changes of tidal dynamics associated with development of the coastline of DYB. The bay is one of two with sexta-diurnal tidal resonance along the South China coast . It is interesting to learn how human activities affect the shallow-water tidal resonance, and it is of great significance to the protection of local marine environment. The rest of this paper is organized as follows: Sect. 2 describes the coastline and bathymetric changes in DYB in detail; Sect. 3 illustrates changes in tidal dynamics between 1989 and 2014 based on numerical modelling; Sect. 4 contains a discussion of the impacts of human activities on tidal resonance, tidal duration asymmetry, and water exchange; and Sect. 5 offers conclusions.

Coastline and bathymetric changes in DYB
From 1973 to 1986, natural siltation in Aotou Harbor, Fanhe Harbor, and Yaling Cove ( Fig. 1) extended the tidal flat seaward by about 100 m on average (650 m maximum). In contrast, the top of the bay from Baishou Cove to Xiachong was eroded by about 1000 m due to strong tides and waves (Xia et al. 2000). Between 1987 and1997, aquaculture, urbanization, and port construction caused anthropogenic coastline evolution in Fanhe Harbor, Aotou Harbor, Baishou Cove, Dapeng Cove, and Mabian Island far beyond the natural changes (Yu et al. 2009). From 2001 to 2011, land reclaimed for NPPs, petrochemical engineering, and other large projects reached 11.63 km 2 , and the coastline was moved offshore by 2.31 km at the head of DYB (Yu et al. 2014). Artificial linkages between islands and the mainland, mainly in Aotou Harbor, Huangmao Island, and Zhima Island, also changed the coastlines. A comparison between the coastlines in 1989 and 2014 ( Fig. 1) shows that a total water area of 30 km 2 was lost in the past two decades.
At the same time, water depth was also altered from 1989 to 2014, as shown in Fig. 2. The average water depth increased from 9.34 m in 1989 to 9.72 m in 2014. This increase is because most of the tidal flat was turned into land. Dredging for ports and navigation channels also deepened the water in some regions in DYB. For example, the Mabian Island Navigation Channel, with a length of 20.2 km, and an inner and outer channel width of 251 and 300 m, respectively, was completed in 2007. For shipping safety, regular dredging maintains the channel depth at 20 m.
Although anthropogenic changes to DYB morphology occurred gradually over the last 30 years, the cumulative effect may have a remarkable impact on the tidal dynamics of DYB and thus the marine ecological environment (Song et al. 2013;Lin et al. 2016).

Model setup
To fit the complex coastline of DYB, an unstructured-grid, primitive-equation, Finite Volume Community Ocean Model (FVCOM) (Chen et al. 2003) is used in this study. Coastlines and water depths with respect to theoretical lowest low tide were obtained from the official marine charts published by the Maritime Safety Administration of P.R. China in 1989 and 2014. The water depth was then converted to the datum with respect to mean sea level by adding 1.6 m in the model. The model is of high resolution (about 50 m) in channels and near the coast, with seven sigma layers in the vertical. It is driven by four diurnal tides (K 1 , O 1 , P 1 , Q 1 ), four semi-diurnal tides (M 2 , S 2 , N 2 , K 2 ), two quarter-diurnal tides (M 4 , MS 4 ), and two sexta-diurnal tides (M 6 ,MS 6 ) at the open boundary. No winds or rivers are given as only tidal dynamics are studied. The model for 2014 (hereinafter R14) has been fully described in Song et al. (2016). The model configurations for 1989 (hereinafter R89) are the same as those in 2014, including the open boundary forcing, bottom friction coefficient, and other model settings (except the coastline and bathymetry). The mesh-grids are consistent in the two cases except where land was reclaimed. By using this method, the impact of coastline and bathymetry changes on tidal dynamics can be better estimated.
The model is run for 60 days, and the last 30 day result is used for analysis. The robust T_TIDE harmonic analysis program (Pawlowicz et al. 2002;Leffler and Jay 2009) is adopted. As in Song et al. (2016), tidal constituents P 1 and K 2 are inferred from K 1 and S 2 , respectively, using the amplitude factors and phase offsets obtained from a 1 year data set with 1 h intervals at the DW station ( Fig. 1) read from tidal tables.

Model validation
R14 has been well validated in Song et al. (2016). Here, tidal elevations measured at DW station ( Fig. 1) from 1 to 31 January 1987 and tidal currents recorded at three stations (C1-C3 in Fig. 1) from 15 January to 14 February 1987 (details can be found in Li et al. 1990) are used to validate R89. The observed and modelled tidal elevation and northward (the main direction) current speed are compared in Fig. 3. Then, the model skill score (SS) is adopted to quantify the model errors. The SS is defined as the ratio of the root-mean-square error normalized by the standard deviation of the observation (Murphy 1988 where X is the variable being evaluated and X is the temporal average. For reference, an evaluation of a hydrodynamic and ecosystem model in the southern North Sea categorized an SS > 0.65 as an excellent simulation, 0.5-0.65 as very good, 0.2-0.5 as good, and <0.2 as poor (Allen et al. 2007). In addition to SS, the correlation coefficient (CC) between the model and in situ observations is also used to quantify the model results The SS and CC for the tidal elevation are 0.942 and 0.971, respectively. The model accurately reproduces the tidal elevation variation and captures the double high water (Fig. 3a). The SS values for the current speed 4 m below the surface are 0.733, 0.583, and 0.468 (Figs. 3b-3d); the CC values are 0.862, 0.789, and 0.812 for C1, C2, and C3 stations, respectively. This indicates that the model reproduces the variations in the tidal current but not exactly the same as observed. This is likely because (i) the observed current includes the baroclinic current, the wind-driven current, and the coastal current (Zhang et al. 2019), but this model only considers the barotropic tidal current; and (ii) the morphology changes between R89 and R14 may affect the bottom friction due to seabed sediment evolution and the open boundary tidal forcing due to far-field effects (Song et al. 2013). Because R89 and R14 are given the same settings in this study, we believe the model is still applicable and valuable for improving our understanding of the tidal dynamic changes in DYB from 1989 to 2014.

Tides
Using T_TIDE, 31 tidal constituents with signal-to-noise ratios (SNR) larger than 1.0 can be decomposed from the tidal elevation datasets. Among them, two diurnal tides (K 1 , O 1 ), two semidiurnal tides (M 2 , S 2 ), two quarter-diurnal tides (M 4 , MS 4 ), and two sexta-diurnal tides (M 6 , 2MS 6 ) are remarkable. The tidal form number F = (a O1 + a K1 )/(a M2 + a S2 ) indicates that DYB is a mixed, mainly semidiurnal tidal regime (1.15 < F < 1.32), a conclusion that is unchanged from 1989 to 2014. The amplitude of M 2 is the largest in all tidal constituents, followed by K 1 . Figure 4 shows the co-tidal charts in 1989 and the difference between R89 and R14 (R14 − R89). The co-tidal charts in R14 can be found in Song et al. (2016). There is little difference between R89 and R14 for the K 1 and M 2 tides outside DYB. However, from the entrance to the end of the bay, the difference in K 1 amplitude gradually increases to about 0.5 cm in Baishou Cove and Fanhe Harbor. The amplitude of M 2 changes more than those of the diurnal tides. It increased by 0.2-0.8 cm, and the largest differences are found in Baishou Cove and Fanhe Harbor. The phase lag of K 1 varied little in DYB, as did M 2 in the central and southern parts of the bay; it slightly advanced in the northern part.
Most of the shallow-water tides are generated inside DYB due to wave deformation by nonlinear effects. A comparison between R14 and R89 (Fig. 5) indicates that the M 4 amplitude decreased by 0.3-1.5 cm in DYB, with the difference increasing from the bay mouth to Fanhe Harbor. In contrast, the M 6 amplitude increased by 1.5-3.5 cm from 1989 to 2014. Compared to other tides, the M 6 amplitude rises more dramatically in R14 than that in R89 when travelling into the bay. Furthermore, the amphidromic point of M 6 outside DYB moved southeastward by about 6 km. Both phases of M 4 and M 6 are advanced, but the latter is much more notable. The modification of sexta-diurnal tides in DYB under different coastline and bathymetry conditions is noteworthy and has never been reported before.

Tidal currents
The ellipticity ε was calculated to show the shape of the tidal ellipse, where |ε| ≤ 0.3 indicates a rectilinear current and 0.3 < |ε| ≤ 1.0 indicates a rotary current. The greater the magnitude of ε, the stronger the current rotation (Yang 2001). As shown in Fig. 6a, K 1 is a rotary current outside DYB, but a rectilinear current inside DYB for 1989. The major axis of K 1 decreased by no more than 4 cm·s −1 from 1989 to 2014 (Figs. 6a and 6b). Notable reductions occurred in Yaling Cove, Baishou Cove, and Fanhe Harbor. At the end of the bay, the maximum |ε| is 0.85 in R14, which is larger than that in R89, indicating that the rotation of the K 1 tidal current was enhanced. Outside DYB, |ε| is 0.3-0.7 in R89 and 0.2-0.5 in R14, indicating that the rotation of the K 1 tidal current was weakened. The M 2 tidal current (Figs. 6c and 6d) shows the same alteration with K 1 , but the major axis of that current decreased by 2-10 cm·s −1 from 1989 to 2014the most remarkable change in all the tidal constituents. Great differences can be found in Huangmao Mt., Yaling Cove, and Baishou Cove. The changes in the shallow water tidal current are obvious (Fig. 7). The major axis of the M 4 tidal current decreased by 2-6 cm·s −1 in central DYB and by 8-14 cm·s −1 between islands and the mainland (Figs. 7a and 7b). Compared to R89, the rotation of M 4 is slightly enhanced in R14. In contrast, the major axis of the M 6 tidal current increased by a maximum 4 cm·s −1 in the central bay but decreased in the mouths of Baishou Cove and Fanhe Harbor (Figs. 7c and 7d). At the end of the bay, the largest |ε| is 0.9 in R89 but only 0.5 in R14, indicating weakening of the rotation of the M 6 tidal current.

Tidal energy flux
As a semi-enclosed bay, the tidal energy in DYB is mostly transported from the deep sea. Based on Song et al. (2013), the tidal energy flux in a tidal cycle can be calculated as follows: where v = (u, v) is the velocity vector; D = H + ζ is the water depth and ζ is the tidal elevation; g is the gravitational acceleration; ρ is the water density; t is the time; and T is the tidal period. Figure 8 shows the energy fluxes of K 1 , M 2 , M 4 , and M 6 tides in R89. Compared to R14 , the distribution of each tidal energy flux varies little. The tidal energy of K 1 enters the bay from the west channel and the middle channel (3000.8 kW per K 1 cycle in R89), and leaves from the east channel (2455.0 kW per K 1 cycle in R89). The energy flux concentrated in the channels is much greater than that in shallow waters (Fig. 8a). The tidal energy dissipated in the bay is 545.8 kW in R89 with a dissipation rate of 18.2%. In R14, the dissipated tidal energy is 500.5 kW per K 1 cycle (Table 1), giving a dissipation rate of 18.7%. The M 2 tidal energy flux has a similar pattern as K 1 (Fig. 8b). Tidal energy flows into DYB from the west and middle channels (4460.5 kW per M 2 cycle in R89) and leaves from the east channel (2645.5 kW per M 2 cycle in R89). About 40.7% of M 2 tidal energy is dissipated in the bay in R89, but the dissipation rate is reduced by 0.8% in R14 due to the decreased inflow caused by the land reclamation.
The M 4 energy flows into the bay from all the three channels (1040.6 kW per M 4 cycle in R89) and out of the bay along the east and west coasts (148.2 kW per M 4 cycle in R89), with about 85.8% of energy dissipated in the bay (Fig. 8c). In both cases, the M 4 energy dissipation rate is higher than 80%, but in R14 it is decreased by 1.7% (Table 1). The energy flux of M 6 has a similar distribution to that of M 4 in DYB, but it flows into the bay from the south-southwest, which is influenced by the amphidromic point. Most of the M 6 energy flux into the bay (316.5 kW per M 6 cycle in R89) is dissipated, and less than 1% (1.7 kW per M 6 cycle in R89) can leave the bay. In one M 2 tidal cycle, the dissipation of M 6 tidal energy is about 52% of that of M 2 (Table 1). However, the ratio is increased to 105% in R14, with an additional 203.9 kW of energy entering the bay per M 4 cycle.

The enhancement of tidal resonance
The reduction of astronomical tidal amplitudes due to land reclamation has also been reported in several other coastal regions, including Ariake Sound (Manda and Matsuoka 2006), Jiaozhou Bay (Gao et al. 2014), and Xiangshan Bay (Li et al. 2018). This study shows a similar consequence of land reclamation on astronomical tides in DYB. The reduction in water surface area lowers the amplitude of the K 1 and M 2 tides and slows the K 1 and M 2 tidal currents. Thus, the astronomical tidal energy flow into the bay is largely reduced by 17% for the M 2 tide and 11% for the K 1 tide. As the first overtide of M 2 , M 4 has a similar variation as M 2 from 1989 to 2014.
Unlike these other regions, DYB is a sexta-diurnal tidal resonant bay. The sexta-diurnal tides represented by M 6 show a different response to anthropogenic alterations of the coastline and bathymetry than the other tides (Figs. 4 and 5). The amplification of sexta-diurnal tides in DYB is dominated by resonance and followed by shoaling effects and nonlinear processes . The resonance-fitting curve (Fig. 9) between stations T1 and T2 (Fig. 1) is obtained by using the same least-square approach employed by Song et al. (2016).
It measures the resonance between these two stations rather than for the full journey. Figure 9 shows that the resonant period between stations T1 and T2 was reduced from 3.40 h in R89 to 3.23 h in R14. It indicates that the phase speed (c = ffiffiffiffiffi gh p ) increased, a finding that is consistent with the increase in the average water depth from 9.34 m in R89 to 9.72 m in R14. It also indicates that the wavelength of M 6 was enlarged due to human activities in the past 30 years. That is why the M 6 amphidromic point moved seaward. According to Cui et al. (2015), in a flat, rectangular, semi-enclosed bay, the resonant tidal elevation can be expressed as ζ = cos kðL − xÞ cos kL cos ωt where k = ω/c is the wave number, ω is the wave frequency, x = 0 is the resonant node, and x = L is the head of the bay. The tidal amplitude can be expressed as A = cos kðL − xÞ cos kL As pointed out by Song et al. (2016), the resonant system is a composite of the DYB and the northern South China Sea continental shelf, and the M 6 amphidromic point is a node preserved within the system. L represents the distance from the amphidromic point to the terminal of M 6 in DYB. At x = L, A = 1/cos(kL), it indicates that as L increased in R14, the amplitude of M 6 also increased. This is also confirmed by the model result (Fig. 5d). Figure 9 also illustrates that the M 6 tidal resonance in DYB was enhanced from 1989 to 2014; this fits with the fact that more M 6 tidal energy flowed into DYB in 2014. Another reason is that the M 6 in DYB is limited more by friction than by distance from resonance ; the reduction of tidal flats due to land reclamation may reduce bottom dissipation and increase tidal amplitude.

The reverse of tidal duration asymmetry
One of the most important characteristics of a coastal tidal system is tidal asymmetry. It has effects on sediment transport and siltation and erosion in tidal estuaries, inlets, and basins (Friedrichs and Aubrey 1988). According to Song et al. (2011), only a combination of two or three tidal constituents satisfying the frequency relationship (i.e., 2ω 1 = ω 2 or ω 1 + ω 2 = ω 3 ) can contribute to tidal duration asymmetry, which can be calculated using the skewness of the tidal elevation time derivative. The falling tidal duration is shorter (stronger ebb tidal current) for negative skewness and rising tidal duration is shorter (stronger flood tidal current) for positive skewness. A larger skewness in magnitude indicates a larger-duration difference between flood and ebb tides (i.e., a stronger tidal distortion). An increase in flood tidal currents (increased skewness) may generate net-onshore sediment transport and result in siltation; in contrast, an increase in the ebb tidal current (decreased skewness) may induce net offshore sediment transport and coastal erosions.
In DYB, the decomposed 31 tidal constituents can produce 65 combinations, among which M 2 /M 4 , K 1 /O 1 /M 2 , M 2 /M 4 /M 6 , S 2 /M 2 /MS 4 , and S 2 /M 4 /2MS 6 contribute most to the tidal duration asymmetry. A comparison between the case in R89 and R14 shows that only the combinations M 2 /M 4 /M 6 and S 2 /M 4 /2MS 6 changed much, increasing the tidal skewness by about 0.10 (Fig. 10a). The rapid growth induced by sexta-diurnal tides dominates the tidal distortion . For example, the skewness of M 2 /M 4 /M 6 is rather weak outside DYB but enlarges quickly within the bay due to the growth of the M 6 tide.
In R89, the total tidal skewness, summarized by all the combinations' skewnesses, is negative in most areas of DYB, indicating that the flood duration is always longer than the ebb duration (Fig. 10b). The total skewness magnifies from the east to the west outside DYB, and from the south to the north within DYB. In R14, the tidal duration asymmetry is increased by 0.2-0.5, making tidal asymmetry positive in the northern DYB, meaning the ebb duration is greater than the flood duration. Based on the contribution to tidal skewness, it can be found that the difference of tidal asymmetry in DYB is induced by the sexta-diurnal tides, the amplitude of which increased by 1-4 cm.

The acceleration of water exchange
Water exchange plays a critical role in coastal regions by promoting purification. Discussions on various definitions of exchange times (e.g., flushing time, turnover time, and half-life time) were given by Prandle (1984) and Luff and Pohlmann (1995). Because of the nature of dilution, pollution in water will never reach a final concentration of 0%. Therefore, a half-life time for the water exchange is defined analogously to the half-life time of radioactive substances (Luff and Pohlmann 1995). In this study, the half-life time is adopted to represent the water exchange time, and the DYE module provided by the FVCOM package (Chen et al. 2003) is employed. For each model run, the model is initialized with a constant dye concentration inside DYB (north of S1 in Fig. 1) but zero elsewhere. The calculation is started at low tide and stopped when the average concentration inside DYB reaches 50% of its initial value. Three scenarios are undertaken with different open boundary forcings; the results are listed in Table 2.
For the sole M 2 tidal forcing from 1989 to 2014, the half-life exchange time increased from 268 to 319 days due to the weakened M 2 tide and tidal current (Figs. 4d and 6c). For the M 2 /M 4 forced case, the half-life exchange time increased by 40 days between 1989 and 2014. However, when the M 6 tide is involved, the water exchange accelerated by 116 days for the triplet of M 2 /M 4 /M 6 forced case and by 63 days for the 12 tides forced case. This effect may be produced by changes in tidal duration asymmetry (Fig. 10).
Another experiment was carried out to separate the impacts of coastline evolution and bathymetry change (e.g., the dredging of Mabian Island Navigation Channel, Fig. 2). The results (Table 2) show that the former accelerated the water exchange by 34 days (54%), and the latter by 29 days (46%). This indicates that, from 1989 to 2014, the alteration of coastline and changes in bathymetry have approximately equal contributions to changes in water exchange in DYB.

Conclusions
Based on numerical models, the impacts of human activities (mainly land reclamation and modification of the water depth) on tidal dynamics are studied in DYB, which is a sexta-diurnal tidal resonant bay. Two scenarios are compared: one using coastline and bathymetry maps published in 1989 and the other using maps published in 2014. In the past 30 years, the coastlines and water depth of DYB changed considerably. About 70% of the coastline has been changed, and about 30 km 2 of water area have been converted to land during this period. Model results show the decrease of tidal elevation amplitude, tidal current magnitude, and the tidal energy flux of astronomical tides and their first overtides and compound tides. This is not surprising, as many similar cases have been reported in other land-reclaimed semi-enclosed bays.
The most impressive part of our results concerns sexta-diurnal tides (represented by M 6 in this study). Changes in tidal resonances caused by morphological alterations enhance M 6 in tidal elevation amplitude, tidal current magnitude in some parts of the bay, and the tidal energy flowing into the bay. Furthermore, the tidal duration asymmetry changed from ebb-dominance to flood dominance due to enhanced M 6 . This indicates that DYB may have experienced a conversion from net-offshore sediment transport to net-onshore sediment transport, which affects ports, harbors, estuaries, and navigation channels, and results in severe siltation. This finding will need to be confirmed by future studies on sediment transport in the region. Water exchange became much more active due to the enhancement of M 6 tide, indicating that the degeneration of the BYD marine ecological environment may be caused by other kinds of human activities, such as the operation of NPPs, increased discharge of wastewater high in nutrients, overfishing, and the rapid expansion of aquaculture, rather than by land reclamation. However, this conclusion is based on a numerical model and will need to be verified by other types of evidence.
The impacts of human activities on coastal waters have been studied worldwide. However, sexta-diurnal tides have not been studied extensively because they tend to be small in magnitude in most coastal regions. However, such tides are significant in DYB The same tidal constituents used as the tidal model. due to its morphological conditions. Thus, the bay provides a good opportunity to extend our knowledge of high-frequency shallow-water tides. The effects of sexta-diurnal tides on coastal hydrodynamics differ from those of quarter-diurnal tides and astronomical tides, a finding that has never been reported before. In conclusion, the impact of human activities on the coastal marine environment will need to be carefully assessed wherever shallowwater, high-frequency tides are notable.