The study area has a temperate continental to semi-arid climate, with average annual rainfall ranging from 500-600 mm and evaporation ranging from 1700-1900 mm. Rainfall is mainly concentrated in the July-August period. The topography of the study area is higher in the southern part and lower in the northern part, resulting in lower topography in the coastal area compared to the southern part of Laizhou Bay. The piedmont alluvial-proluvial and marine plains serve as the regional groundwater recharge and discharge zones, respectively. The central part of the piedmont alluvial-proluvial plain is the groundwater transition zone (Chen et al., 2004; Xu and Ding, 2008; Feng, 2016; Guo, 2016). The WF80 borehole is located in the groundwater discharge area, where the sources of groundwater recharge are complex. These sources include brine in the palaeosedimentary environment, lateral water flow from the Yu River, subsurface runoff from the southern region, atmospheric precipitation, and seawater from Laizhou Bay ( Figure 2 ).

The southern coastal area of Laizhou Bay is affected by changes in relative sea level, resulting in the development of marine and continental strata in alternating sequences. Previous studies have identified three high paleo sea-level periods in the southern coastal area of Laizhou Bay in the southern Bohai Sea since the late Pleistocene: Cangzhou (Qp₃ ¹), Xian (Qp₃ ³), and Huanghua (Q₄) transgressions ( Figure 2A ). These transgressions have led to the formation of three sets of vertical marine strata, which have resulted in the presence of abundant underground brine mineral deposits in the southern coastal area of Laizhou Bay (Zhao et al., 1978; Han and Wu, 1982; Xu and Wang, 1990; Gao et al., 2015; Zhang et al., 1996; Wang et al., 2003; Yang, 2016; Yang et al., 2016).

Previous research has shown that the total dissolved solids (TDS) in the three marine brine strata exhibit a zonal distribution pattern from the southern (land area) to northern (sea area) parts of the study area (Zheng et al., 2014; Gao et al., 2015). The coastal area and the distant coastal area, spanning a width of 4-8 km and 5-10 km respectively, have low TDS contents (<100 g/L). In contrast, the central part of the south coast of Laizhou Bay displays high TDS contents (>100 g/L) over a width range of 10-15 km (Zhang and Peng, 1998; Leng et al., 2009; Gao et al., 2015; Yang, 2016). Specifically, the brine water within each marine stratum exhibits varying TDS contents: 50 to 140 g/L for the uppermost layer, 50 to 165 g/L for the intermediate layer, and 50 to 130 g/L for the lowermost layer ( Figures 2B–D ). Additionally, the saline aquifer in the southern coastal area extends seaward, resulting in higher TDS contents in the submarine groundwater compared to the surrounding seawater ( Figure 2E ) (Bi et al., 2012; Zheng et al., 2014; Gao et al., 2015; Liu, 2018).

The WF80 borehole (119.26741 E°; 37.20253 N°) is located in the Yu River Delta of the southern coastal area of Laizhou Bay ( Figure 2A ). The Lz908 borehole, on the other hand, is situated near the south coast of the Bohai Sea, China (118.97 E°; 37.15 N°) ( Figure 2A ). The Lz908 borehole was drilled to a depth of 101.3 m in the summer of 2007 by the First Institute of Oceanography, State Oceanic Administration, China. The WF80 borehole is part of the delta cluster formed by the Weihe and Mihe rivers (Xue and Ding, 2008) between the eastern and western branches of the Yishu Fault in the Shandong section of the Tancheng-Lujiang Fault Zone. In August 2020, core drilling was conducted in the WF80 borehole, with a core length of 78.2 m. For this study, the upper 36.6 m layer, which consists of fluvial and coastal sediments, was selected. Below 36.6 m, there is a dense semi-water-resisting layer with a thickness of 5-6 m and significant lithologic changes. Prior to 2017, the WF80 drilling site was located in the intertidal zone with a seawater depth of approximately 1 m. However, land reclamation in 2017 raised the altitude to 4.3 m. The lithology of the upper 36.6 m layer, including 5.3 m of artificial backfill, primarily consists of yellow-gray clay, silty clay, clay silt, and silty sand. The North China Plain, including the study area, has experienced stable sedimentary conditions with consistent strata and no major unconformities or topographic reliefs (Wu et al., 2006). This provides a relatively complete record of environmental changes in both marine and terrestrial settings. Wang et al. (1981) reported that the southern coastal area of Laizhou Bay has undergone three cycles of seawater transgression and regression since the late Pleistocene. These cycles have led to the formation of various sediment facies, including neritic, delta, tidal flat, littoral neritic, and continental facies. During the seawater transgression period, littoral neritic sediment facies developed and retrograded to delta/tidal flat-fluvial sediments. The contribution of fluvial sediments from the Yu River decreased, while sediments from the Yellow River increased. Conversely, during the seawater regression period, there was an increase in terrigenous input, resulting in the development of fluvial facies and delta progradation (Gao et al., 2018).

In this study, we collected a total of 319 sediment samples from the study area to analyze their particle sizes. The samples were collected at intervals of 1-2 cm. We used a MASTERSIZER-2000 laser particle size meter to determine the sediment particle sizes ( Appendix 1 ). To remove organic matter from the sediment samples, we first mixed them evenly and then mixed 1g of sediment with 10-15 ml of 30% hydrogen peroxide (H₂O₂) at room temperature. We added 5-10 ml of 10% hydrochloric acid (HCl) to remove calcium carbonate. The sediment samples were then mixed with distilled water and stored for 24 hours before being centrifuged. In addition, we collected 70 porewater samples from the WF80 borehole to analyze their hydrochemical characteristics. The porewater samples were collected at intervals of 50 cm. We used ion chromatography (ICS-3000, American DIONEX) to determine the concentrations of major anions (Cl^-, Br^-, and SO₄ ^2-) and inductively coupled plasma-atomic emission spectrometer (ICP-MS, American Thermo Fisher) to determine the concentrations of major cations (Ca²⁺, Na⁺, and Mg²⁺) ( Table 1 ). Before analyzing the cation concentrations, we acidified the collected porewater samples to a pH of approximately 2 using a nitric acid (HNO₃) solution. The charge balance errors of the porewater samples were less than 8%.

In this study, we used an LGR liquid water isotope laser mass spectrometry at Nanjing Hydraulic Research Institute to analyze the stable isotopic compositions (¹⁸O and ²H). The δ¹⁸O and δ²H values were calculated using Equations (1) and based on the Vienna Standard Mean Ocean Water (VSMOW). The analytical precision values for the long-term standard measurement of δ¹⁸O and δ²H were ±0.2‰ and ±0.6‰, respectively ( Table 1 ).

In formula (1), δ¹⁸O and δ²H represent the ratios of oxygen (¹⁸O/¹⁶O) and hydrogen (²H/¹H) isotopes, respectively. R_sample represents the ratio of ¹⁸O or ²H in the porewater samples, while R_VSMOW represents the ratio of ¹⁸O or ²H in the VSMOW.

In this study, we conducted sedimentary unit (SU) and sedimentary facies analyses of the Lz908 borehole using optically stimulated luminescence (OSL) and K-feldspar/polymineralplRIR₂₉₀ age (Unpublished data from Yan Li, China University of Geosciences) ( Figure 3 ). The division of sedimentary units was based on the lithology, color, particle size, shell contents, porewater content, and other characteristics of the WF80 borehole sediment samples. The WF80 borehole can be classified into four SUs:

Sedimentary unit A (SUA) (5.3-12.8m): The OSL dating value at 12.8m from the bottom was 12.66 ± 1.05 ka B.P, indicating that Layer A is a marine sedimentary layer formed during the Holocene seawater transgression.

Sedimentary unit B (SUB) (12.8-16.45m): It is important to note that no pore water was observed in the SUB layer. The OSL dating value at 14.8m was older than 48.52 ka B.P, suggesting that SUB is a continental confining bed formed during the late Pleistocene.

Sedimentary unit C (SUC) (16.45-30.8m): In this layer, the average particle size showed significant variation below 27m, with a lack of sand particles. The K-feldspar dating value at 17.6m of the borehole was 91.5 ± 7.7 ka B.P. Additionally, there were three peaks of foraminifera counts in this layer. Based on the relationship between global sea-level changes and foraminifera counts ( Figure 4 ), it can be concluded that SUC is a marine sedimentary layer formed during the late Pleistocene seawater transgression (Liang et al., 2012; Yi et al., 2012a).

Sedimentary unit D (SUD) (30.8-36.6m) had a wide range of particle sizes, but was notably lacking in sand particles. This suggests that SUD is a fluvial sedimentary layer that formed during the middle Pleistocene.

To summarize, the formation of SUA and SUC can be attributed to a period of high sea levels with abundant brine, while SUB and SUD were formed during a period of low sea levels.

The porewater samples from SUA, SUC, and SUD showed different average ion concentrations but had similar orders ( Figure 5 ). The main cation and anion concentrations followed the order of Na⁺ > Mg²⁺ > Ca²⁺ and Cl^- > SO₄ ²- > Br^-, respectively. Na⁺ and Cl^- accounted for more than 75% of the milligram equivalent percentages, while HCO₃ ^- concentrations were negligible. The hydrochemical characteristics of the porewater in SUA, SUC, and SUD were consistent with local seawater, indicating a Na-Cl facies type according to the Shukarev classification method.

These characteristics differed significantly from those observed in the southern brine water area with high total dissolved solids (TDS>50g/L)(Zhang, 1993; Zhang and Peng, 1998; Leng et al., 2009; Gao et al., 2015; Yang, 2016). However, the hydrochemical characteristics of the porewater in SUA, SUC, and SUD were relatively similar to those of the local coastal seawater. The porewater in SUA had higher average Na⁺, Mg²⁺, Cl^-, SO₄ ^2-, and Br^- concentrations compared to SUC and SUD, while the average Ca²⁺ concentration in SUA was lower. The concentrations of Na⁺, Mg²+, Cl^-, SO₄ ^2-, and Br^- in SUA showed significant changes with depth compared to SUC and SUD. These concentrations were generally higher than those in seawater, while Ca²⁺ concentrations were generally lower. The hydrochemical characteristics of the porewater in SUC showed slight changes and were relatively similar to seawater. In SUD, the concentrations of hydrochemical parameters gradually decreased with depth near the water-resisting layer. The porewater in SUA and SUC had similar characteristics to seawater due to the presence of a potential high-salinity brine layer. However, the water-resisting layer in the 12-16 m sediment layer caused slightly different ion distributions in the porewater of SUA and SUC. This could be attributed to the interaction between evaporation, seawater, and aerosol deposition in the southern salt-pan area. The concentrations of Na⁺, Mg²⁺, Cl^-, SO₄ ^2-, and Br^- in the porewater of SUA might increase near the surface due to rainwater infiltration and storm surges, while the Ca²⁺ concentration decreased due to the formation of gypsum through the bonding of Ca²⁺ and SO₄ ^2-. The concentrations of hydrochemical parameters in SUD were lower than those in SUA and SUC.

In this study, we analyzed the δ¹⁸O and δD values of porewater samples to investigate the sources of porewater in the study area. The δ¹⁸O and δD stable isotopes can be used to identify the sources and mixing history of water molecules in different waters, based on the global meteoric water line (GMWL) equation (δD=8δ¹⁸O + 10) (Chen et al., 2021). The local meteoric water line (LMWL) equation for Laizhou Bay was determined as δD=7.4δ¹⁸O+1.1, using the isotopic compositions of monthly precipitation recorded at the Changyi weather station in 2006. The average annual δ¹⁸O and δD values of infiltrated rainwater in Laizhou Plain were -7.7 and -52.2‰, respectively. In the southern mountains of Shouguang, the average annual δ¹⁸O and δD values of infiltrated rainwater were -8.3 and -59.3‰, respectively (Yang et al., 2016).

Based on Figure 6 , the δD and δ¹⁸O values of SUA ranged from -21.0 to -10.9‰ and -3.6‰ to -1.7‰, respectively. The δD and δ¹⁸O values of SUC ranged from -20.4 to -10.3‰ and -3.5 to -2.2‰, respectively. For the five continuous data points of SUD, the δD and δ¹⁸O values ranged from -27.7 to -22.6‰ and -4.3 to -3.8‰, respectively. Most of the stable isotope values of porewater samples from SUA and SUC were plotted between the GMWL and LMWL, with only a few values plotted below the LMWL. The lines representing the δD and δ¹⁸O values of SUA and SUC were similar and can be expressed by the equation: δD = 5.2δ¹⁸O-3.5. These lines intersect with the GMWL and LMWL, while the extension line passes through the stable isotope data points of local seawater.

The ions profile ( Figure 5 ) suggests that the sedimentation funnel caused by subsurface brine mining activities in the southern part of Laizhou Bay (Liu, 2018) may have influenced the groundwater dynamics. This resulted in a hydraulic gradient close to 0 in the horizontal direction on both sides of SUC, leading to a dynamic equilibrium with seawater. The horizontal direction of water flow is generally stable. However, brine mining reduced the porewater pressure in SUC, causing freshwater from the lower fluvial facies to flow through different sediment layers. This dilution process resulted in lower ion concentrations in the bottom SUD layers compared to the average ion concentrations of the SUC porewater.

Groundwater can acquire salt from various sources, including seawater intrusion, mobilization of salt from groundwater formation, and dissolution of evaporites in marine deposits (Zhou and Li, 1995; Han et al., 2011). The Cl/Br ratio can be used to identify the sources of groundwater salinity, especially when the total dissolved solids (TDS) content is high (above 2-3 g/L) (Cartwright et al., 2006; Du et al., 2015). In our study, the TDS content in the porewater samples from the WF80 borehole was significantly higher than 2-3 g/L, indicating the effectiveness of the Cl/Br method in determining the sources of porewater salinity.

The concentrations of Cl^- and Br^- in the world’s oceans are relatively constant, with an average Cl/Br molar ratio of approximately 655 (Cl^- = 550 mmol/L and Br^- = 0.84 mmol/L) (Siemann, 2003). However, the concentrations of Cl^- and Br^- can vary in marginal seas due to factors such as terrigenous inputs and brine chemical wastewater discharge. In our study, the Cl/Br molar ratio of the local seawater was approximately 710 (Cl^- = 417.18 mmol/L and Br^- = 0.59 mmol/L), which is higher than the average value of the ocean.

The Cl/Br ratio values of the porewater samples from SUA, SUC, and SUD were similar to that of the local seawater and lower than those found in dissolved evaporated salt and diluted brine ( Figure 7 ). Notably, there were significant differences between the variations in Cl/Br ratio and Cl^- concentrations in SUA and SUC. The Cl/Br ratios in SUA ranged from 748.8 to 883.2, with an average of 801.3, which was higher than that of the local seawater. These data points were mainly associated with high Cl^- concentrations. The average Cl/Br ratios in SUC and SUD were lower than that of SUA, ranging from 686.8 to 771.7 and 708.7 to 812.4, respectively. However, it is important to mention that six points in SUC had lower Cl/Br ratios than the average Cl/Br ratio of the local seawater, indicating lower Cl concentrations ranging from 14318.5-15582.8 mg/L.

The observed Cl/Br ratios suggest that marine salt is the primary source of salinity in the porewaters of SUA, SUC, and SUD. High Cl/Br ratios may be attributed to halite dissolution, while low values may indicate brine dilution. However, it is challenging to determine whether the porewater was diluted by seawater. Overall, SUA exhibited high Cl^- concentrations and Cl/Br ratio values, with substantial variations in Cl^- concentrations at different depths. This suggests that the porewater in SUA is influenced by evaporation, precipitation, dissolution, and infiltration processes of salts in the southern salt pans. These salts may be transported from the southern salt pans by the south wind. On the other hand, the observed Cl/Br ratio values in SUC were lower than the average seawater value, indicating the presence of small amounts of brine in this layer.

The δD and δ¹⁸O values of SUA and SUC were significantly different from the average annual stable isotope values of rainwater in the Laizhou Plain and southern mountain areas ( Figure 6 ). SUC data points were concentrated between the Global Meteoric Water Line (GMWL) and Local Meteoric Water Line (LMWL), while the five data points of SUD were notably different from those of SUC, with negative δD and δ¹⁸O values. These observations suggest that the porewaters of SUA and SUC mainly originated from local seawater, showing a slight correlation with the rainwater in the Laizhou Plain and southern mountain areas. Compared to SUC, the porewater samples from SUA had δD and δ¹⁸O values closer to local seawater. Additionally, SUC exhibited greater stability and lower water exchange capacity, as indicated by the more concentrated δD and δ¹⁸O values in Figure 6 . The porewater samples of SUD were more influenced by the water-bearing layers below the water-resisting layers, and they were mixed with water characterized by ion concentrations much lower than that of seawater in the water-bearing layers below the water-resisting layers ( Figures 5 , 6 ).

In summary, as mentioned in Sections 4.2 and 4.3, marine salt was identified as the primary source of salt in the porewater of the WF80 borehole. The porewater mainly originated from local seawater, as indicated by the relatively consistent Na⁺, Mg²⁺, Ca²⁺, Cl^-, SO₄ ^2-, and Br^- concentrations with those in local seawater. Therefore, the saline water occupying the original brine layer is likely derived from seawater through modern seawater intrusion.

The stable properties of δ¹⁸O and Cl^- have been studied by Kelln et al. (2001) and Nefzaoui et al. (2023). The δ¹⁸O values of groundwater are closely related to the mixing of water sources. Therefore, the deviation of observed δ¹⁸O and Cl^- values from the mixing lines can be used to identify the main sources of water mixing (Chen et al., 2021). In the Yu River delta, there are four water resources that affect groundwater: seawater, rainwater, Yu River water, and brine. This study investigates the relationship between δ¹⁸O values and Cl^- concentrations to assess porewater mixing in the WF80 borehole ( Figure 8 ).

The results show that the δ¹⁸O values of SUA, SUC, and SUD increase with increasing Cl^- concentrations in the porewater. These values fall between the mixing line of local seawater with rainwater and that of brine with rainwater. The observed values are closer to the mixing line of seawater with rainwater. The data points are classified into three regions. The data points of SUA are close to the values of local seawater, exhibiting high δ¹⁸O values and Cl^- concentrations. The data points of SUC are plotted in the middle, parallel to the data points of SUA, with lower δ¹⁸O values and Cl^- concentrations. The data points of SUD are located in the lower part, deviating from seawater values, and exhibiting the lowest δ¹⁸O values and Cl^- concentrations. These findings support the previous results, suggesting that SUA and SUC represent brine layers, while SUD experienced infiltration of upper brine. Although the brine was displaced by seawater, a small amount of residual brine mixed with the intruded seawater. The porewaters of SUA and SUC have the same source, with SUA being more affected by salt dissolution and infiltration from salt pans carried by the south wind. In contrast, the porewater of SUC is relatively stable, while that of SUD is more influenced by water inputs from the lower layer, characterized by low salinity.