Dingzi Bay is located in the southern marine area of Shandong Peninsula, China ( Figure 1 ), connecting upstream to Wulong River estuary and downstream to Yellow Sea (Zhan et al., 2017). This bay is the largest and most representative ria-type tidal channel bay on Shandong Peninsula (Tian et al., 2017). It has unique geomorphological properties, with contours resembling “small mouth and big belly” shape. The northwestern area is the estuarine section of Wulong River, while the central to southeastern areas extend into coastal regions outside Wulong River estuary (Wang et al., 2023). During high tide in Dingzi Bay, the area of the bay reaches 143.75 km², with an intertidal zone covering about 119.01 km² and bay mouth width of about 6.0 km. During low tide, seawater mainly covers tidal channel areas, with tidal flats exposed on both sides.

Geomorphological pattern of Dingzi Bay is characterized as “intertidal flats with a tidal channel.” A large ebb-tide delta, with bay mouth as its apex, develops outside the bay entrance. Tidal channel is less than 1000 m wide and about 15 km long above the bottom of the channel and channel sidewall slopes range from 0.9 to 5.2% ( Figure 2 ). The area around the brine discharge outlet, located in tidal channel, features a seabed geomorphological pattern of “two channels with a sandbank in between.” In the middle section of cross-section DM11, central sandbank top rises about 3.4 m above the western ebb-tide channel and 6.0 m above the eastern flood-tide channel ( Figure 2 ). In cross-section DM16, central sandbank top is 3.4 m above the bottom of the northwest ebb-tide channel and 2.5 m above the bottom of the southeast flood-tide channel.

The tidal characteristics of Dingzi Bay are typical of regular semidiurnal tides, with a maximum tidal range of 4.7 meters. The tidal currents of the bay are primarily due to the narrow, elongated shoreline and the topographic effects of tidal channel, predominantly presenting a reciprocating flow from southeast to northwest. Regarding the waves outside the bay, the primary wave direction is from southeast. In summer, typhoons often cause high waves outside the bay entrance, with wave heights typically exceeding 1.5 m and reaching up to 3.5 m. However, these Most of the waves have minimal impacts inside the bay; therefore, this research model did not consider wave effects (Zhan et al., 2024).

Wulong River at the head of Dingzi Bay is a monsoon rain-fed river, with an average annual runoff of 92.7×10⁸ m³, accounting for 70-80% of the total annual runoff during flood season ( Figure 3 ). Based on the data observed at Tuanwang station (hydrological station) from 2000 to 2022, the average monthly discharge of Wulong River into the sea was 10.71 m³/s, with multi-year average daily flow in winter being 3.21 m³/s, which was significantly increased to 24.28 m³/s in summer. On August 12th, 2006, the highest daily flow of 1870 m³/s was recorded.

Research area was located between latitudes 36.1°N and 36.8°N and longitudes 120.6°E and 121.5°E, including Wulong River Estuary, Dingzi Bay, and parts of Yellow River marine area. Bathymetric data for Wulong River Estuary and Dingzi Bay area were collected by onsite measurements with precision of 1:5000. Bathymetric data for areas outside Dingzi Bay were obtained from ETOPO global bathymetric dataset.

Considering the complex topography of Dingzi Bay coastline, to more accurately simulate natural shoreline, the developed model employed an unstructured triangular mesh (Fowler et al., 2016) to delineate the computational area and utilized dynamic boundaries to handle wet-dry grid transformations (Guo et al., 2021). Computational area included 41,866 grids and 23,393 nodes( Figure 4 ), divided vertically into 6 Sigma layers. To precisely calculate salinity variations near discharge outlets, the grids near these outlets were refined (Zhang et al., 2023), with the smallest grid scale in the refined area being 2 m. Using a bilinear interpolation method (Minh et al., 2024), bathymetric scatter data were applied to grid points to obtain distributions of water depths, as illustrated in Figure 5 . The model has a minimum time step of 0.05 seconds and a maximum time step of 30 seconds. The boundary conditions for the Wulong River use runoff input, while the ocean boundary conditions are driven by water levels. The water levels are forecasted based on harmonic constants of various tidal constituents assimilated from satellite altimetry data using the Tide Model Driver (TMD) (Ma et al., 2024).

Seawater salinity data for Dingzi Bay included measurements from two tidal stations (S1 and S2 as shown in Figure 1 ) collected between December 25th, 2022, and January 1st, 2023. Remote sensing imagery from the same period was used to obtain the spatial distribution of surface salinity in Dingzi Bay.

Brine discharge outlet was located within the main tidal channel of Dingzi Bay, significantly affected by the upstream flow of Wulong River. The salinity of the estuarine area varied greatly in different seasons. During the dry seasons of winter and spring, salinity in the estuary was higher while In the wet seasons of summer and autumn, increased river flow led to a significant decrease in salinity due to seawater dilution by freshwater. Considering the impact of brine discharge on ecological environments, marine biomass in Dingzi Bay reached its peak in autumn, when salinity variations have the most significant impact. Furthermore, the average river flow in September was at its highest. After considering seasonal variations in Wulong River flow into the sea and biomass in Dingzi Bay, winter and autumn were selected as the representative seasons for 3D numerical simulations. The average river flows in December and September served as runoff input conditions for the model. The water intake during autumn and winter is 15.52 m³/s and 11.12 m³/s, respectively, while the discharge rates are 9.99 m³/s and 7.16 m³/s, respectively.

To verify the feasibility and accuracy of the developed model, comprehensive validation was performed considering various parameters such as tide level, flow speed, flow direction, and salinity. Figure 6 compares simulated and observed tide levels at Nvdao Station ( Figure 1 , Station T) from July 16th to July 23rd, 2022, with results indicating good consistency in tide amplitude and phase. The speed and direction of flow at measurement points P1, P2, P3, and P4 from December 29th to December 30th, 2022, are also compared ( Figure 7 ). It was found that simulation results matched well with the experimental values and flow directions were almost identical, proving that the model could accurately reflect the hydrodynamic characteristics of Dingzi Bay. Figure 8 compares the observed and simulated salinity results at measurement points S1 and S2, from December 29th to December 30th, 2022. The calculated values strong agreement with the observed data, ensuring the accuracy of subsequent model calculations.

Dingzi Bay is a narrow estuarine tidal channel bay with seabed geomorphological pattern of “two channels with a sandbank in between.” Flood tide flows propagate upstream along the eastern side of the main tidal channel, while ebb tide flows move downstream along the western side of the channel ( Figure 9 ). Figure 10 shows a depth average flow speed map of Dingzi Bay. In this area, the highest flow speeds are typically concentrated near the bay mouth and within the internal tidal channels of the bay. Maximum surface flow speed during flood tide is about 75-85 cm/s and during ebb tide, it is about 70-80 cm/s. Ebb tide duration (6 h 25 min) is longer than that of flood tide (5 h 58 min). Flow speeds in other bay areas are relatively lower, with flood tide flows generally stronger and of shorter duration compared to ebb tide flows.

After entering from the open sea through Dingzi Bay mouth, the tidal flow progresses from southeast to northwest, turns west at the west side of Magu Island, and then north past the west side of Xiang Island into narrow Wulong River. Because of the narrowing of the tidal channels upstream, flood tidal currents downstream of Xiang Island are strong, presenting distinct reverse flow properties. On tidal flats on both sides of the channels, flow speeds are relatively lower and flow direction includes components moving towards the shore.

As shown in Figure 11 , high salinity variation zones are primarily concentrated near discharge outlets, which are affected by the tidal channels of Dingzi Bay and dynamic geomorphological conditions near these outlets, giving rise to the formation of a narrow strip extending in NW-SE direction. The magnitude of salinity variation after brine discharge was changed with season, tidal pattern, tidal phase, water layer, and specific location. Numerical simulations on Dingzi Bay indicated that in both winter and autumn, areas within the bay and near the bay mouth experienced varying degrees of salinity increase. By analyzing maximum salinity changes in vertical envelopes, it was observed that in winter, most areas had salinity increase values of less than 3psu, particularly in the narrow regions of tidal channel at discharge outlets where increases exceeded 3psu, the statistics for the area of salinity increase are shown in Table 1 . In autumn, increase in most areas did not exceed 2psu, with a large area southwest of the discharge outlet experiencing 2psu increase with no instances exceeding 3psu. Seasonal variations showed that high salinity areas near the discharge outlets were more concentrated in winter, possibly due to stronger tidal effects and hydrodynamics during this season. However, areas with increased salinity in autumn were more dispersed. Although some marine communities can adapt to salinity variation exceeding 3psu, long-term and widespread ecological impacts still require thorough assessment, particularly in winter, where the ecological impacts of high salinity in tidal channel areas should be carefully considered.

By comparing the distribution maps of increased salinity after freshwater discharge in winter ( Figure 12 ) and autumn ( Figure 13 ), distribution characteristics of high-concentration saline water under the effect of Wulong River runoff and tidal action were further investigated. Salinity increase distributions during the four typical tidal phases of flood rapid, flood slack, ebb rapid, and ebb slack in winter and autumn were evaluated. The areas affected by salinity increases were significantly changed with tides, especially in regions where salinity was increased by more than 2psu. During flood rapid, areas with salinity increase of more than 2psu were located at the bend of tidal channel near the discharge outlet, which roughly corresponded to the central marine area of the bay. During flood slack, they were located in the tidal channel near the discharge outlet and at bay top. During ebb rapid, they were upstream of discharge outlet in the tidal channel and at bay top while during ebb slack, they were near tidal channel close to discharge outlet.

Statistical analysis of cumulative durations for salinity increases of over 2psu and 3psu in winter and that over 2psu in autumn showed the temporal patterns of salinity changes ( Figures 14 , 15 ). In winter, most areas with salinity increase of 2psu had cumulative duration of less than 150 h, but within 300 m range from discharge outlet, there were areas where cumulative duration exceeded 180 h. Areas with salinity increases of over 3psu and cumulative durations exceeding 60 h were mainly concentrated near the discharge outlet. Areas with 3psu increase in salinity lasting over 30 h were almost entirely within 200 m of the discharge outlet, while most areas had cumulative durations of less than 120 h, with only the area around the discharge outlet presenting durations exceeding 120 h. In autumn, due to higher Wulong River runoff and the effect of water dilution due to runoff, no high-concentration salinity increase areas occurred and overall salinity increases within the bay were lower. Areas with salinity increases of over 2psu and cumulative durations exceeding 30 h showed elliptical spatial distributions, primarily around the discharge outlet. Most areas with over 2psu increase in autumn had cumulative durations of less than 30 h, with those exceeding 30 h mainly concentrating within tidal channels. In summary, regardless of the season, the areas near the discharge outlets were key regions for the accumulation of increased salinity durations. Especially in winter, due to decreased runoff, areas with prolonged salinity increases might have more significant impacts on marine communities. In contrast, salinity increases in autumn were more gradual with fewer long-duration areas, mainly distributed in tidal channels with stronger water dynamics.

Figure 16 illustrates the cross-section ( Figure 5 ) of depth -tide-salinity increase relationship. The figure revealed that flow speeds at surface exceeded those at mid-depths and bottom layers. Under the effect of topography, in 0-500 m depth range, due to deeper water and gentle slope, the flow was slow and steady. In 500-1000 m depth range, where the topography was more variable and water was shallower, flow speed was increased. Near discharge outlet at 1000 m depth, flow speeds were smooth but fluctuated significantly near the outlet due to discharged water and density currents. Furthermore, as the terrain flattens and water body widens, flow speeds were gradually stabilized. The vertical distribution of salinity increase was similar to those found by Tsiourtis (2001), who investigated the impacts of brine discharge from desalination plants near Garin Islands. The results revealed that while brine dispersed well at surface layer, its higher density caused it to accumulate in deep water sections, resulting in the formation of high-salinity zones, which in this research were also affected by the terrain near discharge outlets. Research has shown that since natural seawater salinity could fluctuate by ±10psu of the average annual salinity of the area, marine organisms had natural adaptability to these fluctuations. A 10psu increase in seawater salinity could be applied as a conservative estimate of the tolerance of marine organisms to salinity increase (Gao and Ruan, 2004). Monitoring results of brine discharge into the sea indicated that most marine communities could tolerate salinity increases of 2-3psu, some even up to 10psu; therefore, a 2-3psu increase had no impact on the ecosystem (Jenkins et al., 2012). Therefore, we identified areas with 3psu increase as high-concentration salinity zones. The dispersion of high-concentration saline water was found to mainly occur near incoming tide peak moments, indicating a close relationship between brine dispersion and flow speed. At low speeds, high-density brine gathered together under the effect of density gradients, dispersing to other areas near the discharge outlet when flow speed was increased. During ebb tide, the area affected by high-concentration saline water was smaller because, as the tide fell, freshwater from Wulong River diluted the water near the intake, decreasing the salinity of the brine discharged from the outlet and preventing the formation of large high-salinity zones near the outlet.

After brine discharge, it rapidly diffused. To investigate salinity diffusion after brine discharge, a statistical analysis was performed on the attenuation process of salinity increases along four directions (E, W, S and N) within a 500 m range from the discharge point during winter ( Figure 17 ). Following discharge, there was a rapid attenuation of salinity within a 50-100 m range near the discharge outlet, with attenuation rates reaching 42-59% within 100 m along all four directions, with maximum attenuation rate being 0.03799 psu/m ( Table 2 ). Within 100-500 m range, the attenuation of salinity increases was more gradual, due to the brine being sufficiently diluted and oscillating within the bay under tidal effects.

Field simulation results were generally consistent with the findings of most previous researchers (Aljohani et al., 2022), indicating that brine discharge led to significant increases in local salinity, which could potentially impact marine communities (Flavin et al., 2023). This research showed that the amount of salinity changes was related to other factors such as runoff and tides. Under the effect of tidal dynamics, the diffusion characteristics and impact ranges of brine varied between winter and autumn; therefore, when planning and managing brine discharge into the sea, it was necessary to consider the effect mechanisms of these natural conditions on discharge to adopt more effective management strategies to minimize environmental impacts.

The water in estuarine bays is supplied by rivers and the open sea and salinity variations are jointly affected by runoff and tidal flows. Due to the offsetting effects of tidal actions, the effect of tides on salt intrusion was not significant (Uncles and Stephens, 2011). He et al. (2018) analyzed historical data and numerical simulation results and found that an increase in runoff decreased salinity in estuarine areas. When freshwater flow was increased, more freshwater entered the estuary, diluting saline water and thus reducing salinity. By integrating observed data and numerical model results, Chang et al. (2024) found that with the increase of runoff, mixing of salinity intensified, leading to a significant decrease in salinity in estuary areas. In Dingzi Bay, there was a significant seasonal variation in runoff. With high runoff in autumn, salinity within the bay was decreased, resulting in a lower degree of salinity increase from brine discharge into the sea during autumn compared to winter.

Saltwater intrusion is a common phenomenon in estuarine areas (Liu et al., 2011; Jiang et al., 2024), with river runoff being the main factor affecting saltwater intrusion changes (Tian, 2019). In this research, due to internal water extraction and drainage within the bay, where extraction volume exceeded drainage volume, concentrated brine discharge decreased total water volume inside the bay, thereby increasing tidal volume within the bay and intensifying saltwater intrusion. Seawater entered freshwater areas through tidal forces and since seawater salinity outside the bay far exceeded that of freshwater, this led to significant variations in the salinity and its spatial-temporal distribution in Dingzi Bay and nearby maritime areas. The average salinity of seawater outside the bay was 28psu in winter and 27psu in autumn. Reductions of freshwater volume within the bay due to neap tides in winter and autumn were 5132160 and 7039872 m³, respectively. Based on these figures, the areas where salinity was increased by more than 0.1psu (0.1psu salinity increase zones) and salinity changes due to additional seawater intrusion within the bay were determined ( Table 3 ).

Additional seawater intrusion resulted in average salinity increases of 0.36psu in winter and 0.28psu in autumn for 0.1psu salinity increase zones during neap tides. Average increases in bay salinity due to additional seawater intrusion during neap tides were 0.80psu in winter and 1.05psu in autumn. Therefore, brine extraction and discharge did not alter the total amount of salt in the original seawater of the bay and only reduced the amount of freshwater within the bay resulting in the addition of higher-salinity seawater from outside the bay, thus increasing total salinity within the bay. Redistribution of original salt due to concentrated brine discharge only had a certain impact on the existing salinity distribution patterns and their dynamics within the bay.

Paparella et al. (2022) showed that the Arabian Gulf was a natural extreme marine system where marine organisms had adapted to high and variable natural salinity. Minor salinity changes due to concentrated brine discharge from desalination plants did not impact the marine environment of the gulf. Even under most dire climate predictions, increase of seawater salinity and tendency toward more desalination over the coming decades are likely to have minimal impacts on gulf salinity. Furthermore, an increase in salinity could lead to greater flow through the Strait of Hormuz, thereby accelerating the renewal of water in the gulf. This accelerated water exchange process might increase high-concentration saline water influx from outside Dingzi Bay, potentially leading to further salinity increase in the bay. This issue requires continued research in future works.