Assessment of the distribution and ecological risks of heavy metals in coastal sediments in Vietnam’s Mong Cai area

Coastal sediments in the Mong Cai area were collected and analyzed for grain size, heavy metals, total organic carbon, and isotopes (210Pb, 226Ra, δ15N, δ13C) to assess sediment quality. The most common sediments were fine sand in surface sediment, very fine sand in core C1, and very coarse and coarse silt in core C2. The total organic carbon was highest in C2 next to the surface and lowest in C1, with content levels of 1.81%, 0.40%, and 0.31%, respectively. The chronology in C1 was 1877–2019 (142 years, 0–5 0 cm), with an average sedimentation rate of 0.71 cm/year. In C2, the chronology was 1944–2019 (75 years, 0–14 cm), with an average sedimentation rate of 0.27 cm/year. These δ13C and δ15N in the sediment reflect the source of the organic matter mix from the marine and terrigenous sediments. All studied heavy metals were lower than the ISQGs, with the exception of As in C1 and C2, which were higher. C1 showed a decline in As over time, while C2 As levels increased between 1996 and 2019. In terms of heavy metal pollution indexes, the geoaccumulation index (Igeo) showed that C1 and C2 were unpolluted to moderately polluted with As, with Li and Pb in C2; the enrichment factor (EF) was moderately enriched with As; the contamination factor (CF) was moderately contaminated (Pb, Cd, Fe, Mo, and Li) in C2 and C1 (Cd, As, Li) and considerably contaminated (As) in C2. The risk factor (ER) of As showed a moderate potential ecological risk in C2. The degree of contamination (CD) ranged from moderate to considerable (C1, C2), and the ecological risk (RI) was low. Although CD ranged from moderate (C1) to considerable (C2), most contamination was concentrated at the bottom of the cores. RI was low. The Mong Cai sediment quality does not currently affect the coastal area’s ecosystem and fauna.


Introduction
Heavy metals are common pollutants in coastal areas. They derive from natural and manmade sources and Abstract Coastal sediments in the Mong Cai area were collected and analyzed for grain size, heavy metals, total organic carbon, and isotopes ( 210 Pb, 226 Ra, δ 15 N, δ 13 C) to assess sediment quality. The most common sediments were fine sand in surface sediment, very fine sand in core C1, and very coarse and coarse silt in core C2. The total organic carbon was highest in C2 next to the surface and lowest in C1, with content levels of 1.81%, 0.40%, and 0.31%, respectively. The chronology in C1 was 1877-2019 (142 years, 0-5 0 cm), with an average sedimentation rate of 0.71 cm/year. In C2, the chronology was 1944-2019 (75 years, 0-14 cm), with an average sedimentation rate of 0.27 cm/year. are transported by rivers into coastal areas, where they are redistributed by waves, tides, and currents. Heavy metals can accumulate in the environment and enter animals through the food chain, resulting in bioaccumulation. The presence of high heavy metal concentrations in organisms induces toxic stress and impacts human health via the food chain (Nour et al., 2022;Tchounwou et al., 2012). Fine sediments, particularly clay minerals that occur in sediments, such as kaolinite, organic matter, and chlorite, can absorb large quantities of metal ions (Uddin, 2017) in the water column. Heavy metals in sediments are influenced by grain size, currents, tidal conditions, and physicochemical factors to rediffuse into the water or accumulate in the sediment.
The Mong Cai area in Quang Ninh province is a border region between Vietnam and China that has been prioritized for economic development. The Ka Long River is 65 km long with a catchment area of 773 km 2 and runs along the Vietnam-China border. Upon entering Vietnam, it splits into two branches: the first flows to the Mui Ngoc area at the end of the Tra Co peninsula, and the second proceeds to the Mui Sa Vi at the peninsula's head. The main Ka Long estuary is located near the border between Vietnam and China (Fig. 1). The water discharge ranges from 3.5 to 4090 m 3 /s, with an average of 55.6 m 3 /s annual discharge into the sea in around 1.7 billion m 3 of water, concentrated during the period between July and September (Thanh et al., 2008). In addition to the Ka Long, several other rivers in the Chinese region-the Qinjang, Nanliu, and Jiuzhou-discharge into the Guangxi province's coast, which borders the Mong Cai area.
The Mong Cai coast has an irregular tidal regime, whereby high tide lasts longer than low tide. The maximum tide level is 4.67 m, and the minimum is 0.39 m (Thuy, 1984). The dominant direction of the current in the Gulf of Tonkin is northeast-southwest during both monsoons (Chung et al., 2015). The Mong Cai area is affected by trans-regional environmental impacts originating from different sources in the Ka Long's catchment area. Under the tide's coastal circulation, the current can transport pollutants from the northeast to the southwest, impacting the Mong Cai coast.
Earlier studies that examined sediment in the Mong Cai area found that heavy metal (Cu, Pb, Zn, As, Cd) concentrations in the sediment core were significantly lower than the interim sediment quality guideline (ISQG) levels; however, they showed an increasing trend over time (Nhon et al., 2019). The sediment distributed abundant sand, and the accumulation of persistent organic pollution (POPs) in the sediment was low (Nhon et al., 2014). Total nitrogen (Nt) was low, while total phosphor (Pt), total organic carbon (TOC), and total sulfur (St) were high, indicating that the marine environment was more severely affected than the terrigenous environment (Nhon et al., 2013). The human impact associated with aquaculture increased phosphorus, nitrogen, and organic carbon in the tidal creek on the Mong Cai coast, and concentrations of Nt, Pt, and TOC in surface sediments near ponds were higher than those outside or further away from ponds (Bui et al., 2013). Trang et al. (2019) analyzed water quality in the Mong Cai area and reported that physical-chemical parameters, such as pH, total suspended solids, dissolved oxygen, biochemical oxygen demand, and chemical oxygen demand, were of good quality; nutrients including N-NO 2 − , N-NO 3 − , N-NH 4 + , and N-PO 4 3− and heavy metals Fe, Zn, Hg, As, Pb, and Cd fell within permissible limits in accordance with Vietnam's coastal seawater quality standards (MORE, 2015;Trang et al., 2019). Previous studies on radioisotopes of 137 Cs and 90 Sr in water and sediments at Mong Cai showed that 137 Cs and 90 Sr in water and sediment were lower than other pollutants worldwide (Nguyen et al., 2020).
Results from the Mong Cai area revealed the impact of human activity on aquaculture ponds, which manifested in the accumulation of nutrients in coastal sediments. The present study aims to contribute to our understanding of the changing status of heavy metals in coastal sediments, both spatially and temporally, as a means of assessing the anthropogenic impact on coastal areas.

Materials and methods
In March 2020, 22 sampling stations were located on the Mong Cai coast, 20 of which collected surface samples, while the remaining two collected sediment cores (Fig. 1). Surface sediment samples were taken using a Petersen Grab (Wildco Company), and sediment cores were collected using a manual piston corer with plexiglass tubes that had an inner diameter of 60 mm and a length of 1 m. Surface sediment was taken at a depth of 0-5 cm, and sediment cores were cut into layers with the following thicknesses: cut 1 cm from tops (0 cm) to 1 cm; cut 2 cm from 1 to 21 cm; cut 3 cm from 21 to 51 cm; and cut 4 cm from 51 cm to the end of the cores. The collected samples were placed in polyethylene (PE) tubes for storage at 4 °C in an icebox until they reached the laboratory (USEPA, 2001). In the laboratory, the samples were dried at 16 °C in air conditioning and analyzed for grain size, 210 Pb, 226 Ra, organic carbon, δ 15 N, δ 13 C, and heavy metal content.
For grain size analysis at the Institute of Marine Environment and Resources, organic matter and salt were removed from the sediments using H 2 O 2 (10%) and distilled water. The sediment was then wet-sieved using a 63-μm sieve. The > 63-µm portion that remained after the water had been evaporated over a warm bath was dried overnight at 105 °C, and the particles were sieved between 2000 and 50 µm. For the < 63-µm portion, once all particles had been deposited, the water was decanted and filtered through filter paper in a vacuum and dried overnight at 105 °C. The < 63-µm fraction (5 g) was added to 1 ml 10% NaOH, placed in an ultrasonic bath for 10 min to allow the particles to separate, diluted with distilled water to 1000 ml, and analyzed using the pipette method (Galehouse, 1971). Particles were calculated as a percentage of each grade. Sediment parameters, including Md, sorting (S 0 ), and skewness (S k ) were calculated according to Folk (1980), and sediment types were classified according to Wentworth (1922). GRADISTAT software (Kenneth Pye Associates Ltd., UK) was used to calculate the sediment parameters.
Radioisotopes 210 Pb and 226 Ra in the sediment samples were analyzed at the Dalat Nuclear Research Institute. The total activity of 210 Pb in the sediments was determined through its daughter radionuclide polonium-210 ( 210 Po) based on an assumption of secular equilibrium. The sediments were digested using nitric acid (HNO 3 ) and concentrated hydrogen fluoride (HF). After digestion of the sample, polonium was extracted from a 5 M HCl solution using 0.1% diethylammonium diethyldithiocarbamate (DDTC) in chloroform. 209 Po was used as a tracer to calculate the yield of the entire chemical procedure. Radionuclide 210 Po was analyzed using an alpha spectrometer (Alpha Analyst, CANBERRA) with passivated implanted planar silicon (PIPS) detectors with an active area of 900 mm 2 . Radionuclide 226 Ra content in the sediments was determined directly through gamma spectrometry using GMX high-purity germanium (HPGe) detectors with 30% relative efficiency (ORTEC). 226 Ra was measured using the 295 keV and 352 keV gamma rays emitted by its daughter isotope 214 Pb and by the 609 keV gamma rays emitted by bismuth-214 ( 214 Bi) (Hai, 1999). Unsupported 210 Pb (or excess 210 Pb) was calculated by subtracting 226 Ra from the total 210 Pb.
The CRS model was first proposed by Krishnaswami (Krishnaswami et al., 1971) and later developed further and corrected by Appleby (Appleby & Oldfield, 1971) and Robbins (Robbins, 1978). It is now widely used to calculate sediment age (1).
where t is time in years; λ is a constant = 0.031; A(0) is the 210 Pb excess in the whole sediment core; and A(x) is 210 Pb excess in the sediment core at depth x.
Accordingly, sedimentation rate (vh) is calculated using formula (2), where vh is sedimentation rate (cm/year); l is slice thickness; and t n and t n-1 are the time calculated in formula (1).
The heavy metals were extracted from the sediment at the Institute of Marine Environment and Resources and analyzed at Ha Noi University of Science. The sediments were ground using a mortar and pestle, and material > 2 mm was removed by sieving. Dry sediment (0.5 g) was added to 10 ml HNO 3 8 N and 3 ml H 2 O 2 under a Vigreux reflux column, heated on a hot plate for 2 h at 120 °C, and then cooled and filtered at 45 µm (Bellucci et al., 2002). Finally, the samples were diluted with 100 ml of deionized water and measured using inductively coupled plasma mass spectrometry (ICP-MS; Elan 9000 Perkin Elmer). All chemicals used were of analytical grade (Merck, (2) vh = l t n − t n−1 Germany). For quality assurance and control (QA/ QC) of the analytical processes, certified reference material samples (PACS2) were used to verify recovery efficiency. The heavy metals recovered from the five PACS2 samples were 106,88,114,118,124,127,123,72,66,and 69% for Fe,Co,Cu,Zn,As,Cd,Pb,Mo,Cr,and Mn,respectively. TOC in the sediment was analyzed by oxidizing TOC with potassium dichromate (K 2 Cr 2 O 7 ) and sulfuric acid (H 2 SO 4 ) with known concentrations. The TOC was completely oxidized by the K 2 Cr 2 O 7 , and the excess K 2 Cr 2 O 7 solution was titrated with Mohr's salt using diphenylamine ((C 6 H 5 ) 2 NH) as an indicator. The TOC in the sediments was then calculated based on the amount of K 2 Cr 2 O 7 consumed for oxidization (Walkley & Black, 1934).
Stable isotopes (δ 13 C, δ 15 N) in sediments were analyzed using an EA-IRMS (UK) isotope ratio mass spectrometer (PDZ Europa 20-20 system by Sercon, Cheshire, UK). This analysis was performed at the Institute of Nuclear Science and Technology.
where X = 13 C or 15 N, R S was the ratio of 13 C/ 12 C or 15 N/ 14 N in the sediment sample, and R Std was the ratio of 13 C/ 12 C or 15 N/ 14 N of the reference sample. Sediment samples for stable carbon isotope (δ 13 C) analysis were weighed in cleaned tin capsules using a Mettler Toledo AT20 balance with 400-500 µg sediment and placed in the EA analyzer. They were then burned in an O 2 atmosphere in a combustion tube containing chromium oxide and silver coated with cobalt oxide. The combustible gas was impelled through the reduction column via a stream of inert He gas and passed into a gas chromatograph, where CO 2 , still in the He stream, was separated from the other gases. Upon warming, the CO 2 was analyzed using a mass spectrometer and compared with IAEA reference gas (the Vienna PeeDee Belemnite [VPDB] limestone standard). Samples for δ 15 N analysis were similarly burned in the EA analyzer, and the generated N 2 was trapped by freezing and introduced to the mass spectrometer input. The analytical processes were subjected to quality assurance/quality control (QA/QC) using certified reference material samples, including IAEA-CO9, IAEA-CO8, IAEA-600, IAEA-603, and IAEA-CH3 for δ 13 C and IAEA-N1, IAEA-N2, and IAEA-600 for δ 15 N with measurement uncertainty of 0.2‰ for δ 13 C and 0.3 ‰ for δ 15 N. Sediment quality assessment: ISQGs and probable effect levels (PELs) (CCME, 1999) were used to assess sediment quality. Other indexes used included the geoaccumulation index (Igeo) (Muller, 1979), enrichment factor (EF) (Ergin et al., 1991), contamination factor (CF), degree of contamination (CD) (Hakanson, 1980), ecological potential risk (ER), and ecological risk (RI) (Hakanson, 1980). The formula indices are calculated, and the values are presented in Table 1.
Data processing and statistical analysis: Heavy metal content, grain size, and TOC were statistically analyzed using Origin Pro. 2021 software to calculate the maximum, minimum, mean, standard deviation, and Pearson correlation coefficient and perform factor analysis (FA) and cluster analysis (CA). Pearson correlation analysis identified a relationship between heavy metals and grain size and TOC. FA determined which parameters control the sedimentary environment and the accumulation of heavy metals. Clustering is a multivariate analysis technique used to combine parameters and stations with similar properties based on correlation coefficients. The purpose of the clustering technique is to identify groups that have the same parameter values and to identify differences between the groups.

Grain size distribution
Six types of sediment were distributed in the Mong Cai coastal area: coarse sand, medium sand, fine sand, very fine sand, very coarse silt, and coarse silt. On the surface, fine sand predominated, followed by very fine sand and then coarse sand and medium sand (Fig. 2a). C1 was dominated by very fine sand (Fig. 2c), while very coarse silt and coarse silt were abundant in C2 (Fig. 2e).
Surface sediments comprised four types: coarse sand, medium sand, fine sand, and very fine sand. Coarse sand was distributed in two samples from Ha Coi Bay, with sand > gravel > silt > clay fractions (Fig. 2a, b). Medium sand was found distributed in the mouth of the Ka Long, Mui Ngoc, and Ha Coi Bay; the content of sand > silt > gravel > clay fraction (Fig. 2a, b). Fine sand was common in the study area, with sand > silt > gravel > clay fraction (Fig. 2a, b). Very fine sand was distributed to the front of the Tra Co peninsula, with sand > silt > clay > gravel fraction (Fig. 2a, b). C1 contained fine sand, very fine sand, and coarse silt, among which very fine sand predominated (Fig. 2c,d). Fine sand was distributed at 62-70 cm, with sand fraction > silt fraction > clay fraction (Fig. 2c, d). Very fine sand was distributed at 0-40 cm and 70-80 cm, sand fraction > the silt fraction > clay fraction (Fig. 2c, d). Very coarse silt was distributed at 40-60 cm, silt fraction > sand fraction > clay fraction (Fig. 2c, d).
The chronology, organic carbon, and stable isotopes in sediments For C1 at the Ka Long estuary, the chronology was 1877-2019 (142 years) from 0 to 50 cm; the sedimentation rate was from 0.08 to 1.62 cm/year with an average of 0.71 cm/year (Fig. 3a). In C2 at Mui Ngoc, the chronology was 1944-2019 (75 years) from 0 to 14 cm; the sedimentation rate was from 0.07 to 0.51 cm/year with an average of 0.27 cm/year (Fig. 3b).

Heavy metals in sediments
In surface sediments, heavy metals were lower than the ISQG levels ( Fig. 6 1944, 1944-1996, and 1996-2019 (Fig. 7b)  Heavy metal concentrations in the surface sediment were lower than in the ISQGs. In C1, the concentration of As was higher than the ISQG (7.2 mg/kg) (Fig. 7a); Cu was higher than ISQGs in some layers at 36-80 cm (18.7 mg/kg). In C2, Cu, Pb, and As were higher than ISQGs in some levels before 1944; between 1944 and 1996, only As was higher than the ISQG level. Meanwhile, only As and Cu in the top layer exceeded the ISQG levels during the 1996-2019 period. Neither the surface nor the core sediments contained heavy metals that exceeded the PEL level (Table 2).

Correlation between sediment parameters
The surface sediment parameters indicated a positive correlation between heavy metals and silt and clay fractions and a negative correlation between heavy metals and gravel and sand fractions (Fig. 8a).
A strong positive correlation (R 0.75-0.95) was observed between silt and clay, Cu, Pb, Zn, Cr, Ni, Fe, Li, and V; clay and Cu, Zn, Cr, Ni, Fe, Li, and V; Cu and Zn, Ni, and Li; Pb and Cr, Ni, Li, and V; Zn and Cr, Ni, Fe, and Mn; Cr and Co, Ni, Fe, Li, and V; Co and Fe; Ni and Fe, Li, and V; and Fe and Li. A moderate correlation (R 0.5-0.75) was observed between silt and Co and Mo; clay and Pb and Co; Cu and Pb, Cr, Fe, and V; Pb and Zn, As, Fe, and TOC; Zn and Co, Li, V, and TOC; As and Ni and V; Cr and TOC; Co and Ni, Li, and TOC; Ni and TOC; Fe and Mn, V, and TOC; Mo and V; Li and TOC; and V and TOC. A weak correlation (R 0.25-0.5) was observed between silt and As and Mn; clay and As, Mn, Mo, and TOC; Cu and Cd, Co, Mn, Mo, and TOC; Pb and Co, Mo, and TOC; Zn and Mn and TOC; Cd and Cr, Ni, V, and TOC; As and Cr, Co, Ni, Fe, Mo, and Li; Cr and Ni, Mn, and Mo; Co and V; Mn and Li; Mo and Mn and Li; and V and TOC.
A strong negative correlation (R − 0.75 to − 0.95) was observed between sand and silt, Mo, Li, and V. A moderate negative correlation (R − 0.5 to − 0.75) was observed between gravel and Mn; sand and silt, Cu, Pb, Zn, Cr, Ni, Fe, and TOC. A weak negative correlation (R − 0.25 to − 0.5) was observed between gravel and sand, Cu, Zn, and Co; sand and Cd, As, and Co.
In C1, the sediment parameters showed positive and negative correlations: a positive correlation was observed between silt and clay and heavy metals, while a negative correlation was observed between sand and heavy metals (Fig. 8b). TOC showed no significant correlation with heavy metals, sand, silt, or clay.
Strong positive correlations (R 0.75-0.95) were observed between silt and Zn, As, Cr, Co, Ni, Fe, Li, and V; clay and Cu, Pb, Zn, As, Cr, Co, Ni, Fe, Mn, Mo, Li, and V; Cu and Pb, Zn, As, Cr, Co, Ni, Fe, Mn, Mo, Li, and V; Pb and Zn, As, Cr, Co, Ni, Fe, Mn, Mo, Li, and V; Zn and As, Cr, Co, Ni, Fe, Mn, Mo, Li, and V; As and Cr, Co, Ni, Fe, Mn, Mo, Li, and V; Cr and Co, Ni, Fe, Mn, Mo, Li, and V; Co and Ni, Fe, Mn, Mo, Li, and V; Ni and Fe, Mn, Mo, Li, and V; Fe and Mn, Mo, Li, and V; Mn and Mo, Li, and V; Mo and Li, and V; and Li and V. A moderate correlation (R 0.5-0.75) was observed between silt and clay, Cu, Pb, Mn, Mo; Zn and Cd; and Cd and Mn, Co, Ni, and Fe. A weak correlation (R 0.25-0.5) was observed between Cd and silt, clay, Cu, Pb, As, Cr, Mo, Li, and V.
A strong negative correlation (R −0.75 to −0.95) was observed between sand and Cu, Pb, Zn, As, Cr, Co, Ni, Fe, Mn, Li, and V. A moderate negative correlation (R − 0.5 to − 0.75) was observed between sand and clay, Mo. A weak negative correlation (R −0.25 to −0.5) was observed between sand and Cd.
In C2, a negative correlation was observed between sand and heavy metals, while positive correlations were observed among the heavy metals and between heavy metals and silt and clay (Fig. 8c).
A strong positive correlation (R 0.75-0.95) was observed between Pb and Zn, Cr, Co, Ni, Li, and V; Zn and Cr, Co, Ni, Fe, Mn, Li, and V; As and Fe, Mn, and Mo; Cr and Co, Ni, Fe, Mn, Li, and V;  A strong negative correlation (R −0.75 to −0.95) was observed between sand and silt and Cu. A moderate negative correlation (R − 0.5 to − 0.75) was observed between sand and Cr, Co, Ni, Li, and V. A weak negative correlation (R −0.25 to −0.5) was observed between sand and clay, Pb, Zn, Fe, Cd, Mn, Mo, and TOC.

Comparison of heavy metal concentrations in other regions
Comparison of the Mong Cai area with the surrounding regions, including Tien Yen, Cua Ong, and Cua Cam, revealed that the heavy metal concentrations in the Mong Cai surface sediments are relatively low (Ho et al., 2010(Ho et al., , 2013aQuy et al., 2012). In C1, only Zn, Mn, and V were higher than at Cua Ong (Ho et al., 2010), and the remaining heavy metals were lower than in Tien Yen (Quy et al., 2012) and Cua Cam (Ho et al., 2013a, b). In C2, Pb, Cd, As, Cr, Co, Ni, Mn, and V were higher, while Cu, Zn, and Mo were lower than in Tien Yen (Quy et al., 2012). Pb, Zn, Cr, Co, Fe, and Mn in C2 were higher, while Cu and As were lower than at Cua Ong (Ho et al., 2010). Only Fe in C2 was higher at Cua Cam, while other heavy metals were lower (Ho et al., 2013a, b). The surrounding areas of Cua Ong and Cua Cam are affected by anthropogenic impacts (coal mining at Cua Ong and industrial zones at Cua Cam) (Table 3).
Compared with the Red River estuary (Tue et al., 2011), the Thanh Hoa coastal area (Nhon et al., 2021), the Cai estuary (Koukina et al., 2017), and the Dong Nai River (Costa-Böddeker et al., 2017) receive large river influx and bring much material from the continent to the sea. The heavy metals in the Mong Cai area are lower than in these areas abovenot only those supplied by rivers but also those from industrial zones in coastal provinces, such as Thai Binh, Nam Dinh at the Red River estuary, Thanh Hoa at the Hoi estuary, Nha Trang at the Cai River estuary, Ho Chi Minh City, and Bien Hoa province at the Dong Nai estuary, where industrial development has recently accelerated significantly (Table 3).
The heavy metals on the Mong Cai coast were lower (Qiao et al., 2013) in both the surface and core sediments than those in China's Shantou Bay. This may be due to the fact that Shantou Bay has been significantly impacted by urbanization and industrialization. Compared with the area surrounding China's Leizhou peninsula, the heavy metal concentrations in the surface sediment and C1 at Mong Cai were lower, while those in C2 (Cu, Pb, Zn, Cd, As, Cr, Co, Ni, Fe, and Mo) were higher than those at the Leizhou peninsula (Bai et al., 2022).
The heavy metal concentrations in the surface sediments at Mong Cai were lower than those in the Gulf of Thailand. In C1, Cd and Fe were higher than in the Gulf of Thailand, while the remaining heavy metals (Cu, Pb, Zn, and Co) were lower. In C2, Pb, Zn, Cd, and Fe were higher than in the Gulf of Thailand, while Cu and Co were lower (Waewtaa Thongra-ar et al., 2008).
Surface sediments from the Malaysian coast had higher Cu, Zn, As, and Cr concentrations than those from Mong Cai, while the Pb concentration was lower (Ashraf et al., 2018). In C1 and C2 from the Mong Cai coast, Pb, Zn, As, and Cr were higher, while Cd and Cu at Mong Cai were lower (Ashraf et al., 2018).

Indexes of heavy metal pollution in sediments
Sediment quality was assessed using several indexes: Igeo, EF, CF, ER, CD, and RI. For the surface sediments, the Igeo indicated no pollution of any heavy metals (Fig. 9a). The EF showed no enrichment (Cr, Co, Ni, Mo, Mn, V, and Mo), minor enrichment (Pb, Cd, Zn, and Li), and moderate enrichment (As) (Fig. 9b). The CF showed low contamination (Fig. 9c). The ER indicated a low potential ecological risk (Fig. 9d). The CD indicated a low degree of contamination (Fig. 9e), and the RI showed a low ecological risk (Fig. 9f).
In C1, the Igeo showed no pollution for Cu, Pb, Zn, Cd, Cr, Co, Ni, Fe, Mn, Mo, Lo, or V, while only As showed unpolluted to moderately polluted (Fig. 10a). The EF showed no enrichment (Cu, Cr, Co, Ni, Mn, Mo, V), minor enrichment (Pb, Zn, Cd, Li), and moderate enrichment (As) (Fig. 10b). The CF showed moderate contamination of As, Pb, Cd, and Li, while the remaining heavy metals showed low contamination (Fig. 10c). The ER indicated a low potential ecological risk (Fig. 10d). The CD showed a moderate degree of contamination (Fig. 10e), while the RI indicated a low ecological risk (Fig. 10f).
The indices of heavy metals in sediments at the sampling sites (surfaces, core C1, and core C2) reflect the influence and interdependence between sediment parameters. The indices at the C2 ≥ C1 > surface (including Igeo, CF, CD, ER (Cu, Pb, Zn, As, Cr), EF (Mo)) depend on silt fraction, positive correlation between heavy metals, and silt fraction (Figs. 9, 10, and 11). The EF indices at the surface ≥ at C1 > at C2 (Cu, Pb, Zn, Cd, As, Cr, Co, Ni, Mn, Li, V) (Figs. 9, 10, and 11) were the result of the high content of Fe in C1 and C2 > surfaces, indicating that the oxidization of the environment at the surface was higher than that at C1 and C2 and that Fe accumulated in the C1 and C2 cores in the form of iron sulfide (reducing environment). The RI at C1 > at C2 > at the surface (Figs. 9, 10, and 11) was the result of supply near the Ka Long River and fine fractions (clay and silt) high content in C2 and C1, leading to conditions that were conducive to the accumulation of heavy metals in C1 and C2. The EF index at C1 > at C2 > at surface (Mo) reflects the positive correlation between Mo and silt fraction and other heavy metals.

Factors influencing the distribution of heavy metals
The FA of the surface sediments identified three factors (FA1, FA2, and FA3) that influence the distribution of heavy metals. FA1 accounted for 59.7% with sand, silt, clay, Cu, Pb, Zn, Cr, Co, Ni, Fe, Li, V, and TOC; FA2 accounted for 15.6% with gravel, Mn, and Mo; and FA3 accounted for 6.4% with As and Cd (Table 4). In addition to the correlation coefficient in surface sediments between heavy metals and gravel, sand, silt, and clay fraction, the positive correlation between heavy metals and silt and clay indicates similar sediment sources or favorable environmental conditions (silt and clay) for metal accumulation (Fig. 8a); the negative correlation represents an unfavorable condition for heavy metal accumulation, as demonstrated by the negative correlation between sand with silt, clay, and heavy metals.
In C1, three factors were identified as influencing the distribution of heavy metals: FA1 accounted for 80.0% with sand, silt, clay, Cu, Pb, Zn, As, Cr, Co, Ni, Fe, Mn, Mo, Li, and V; FA2 accounted for 6.4% with TOC; and FA3 accounted for 5.35% with Cd (Table 4). Correlations between sediment parameters were strong, accounting for the majority; only TOC was not correlated with other parameters (Fig. 8b). The positive correlations between the heavy metals themselves and silts and clays indicate that they share the same supply and favorable environment (silts and clays) for heavy metal accumulation. The negative correlation between sand and heavy metals reflects unfavorable conditions for heavy metal accumulation in sediments. Correlations between TOC and other parameters were non-significant (R < 0.25), reflecting local supply, regardless of supply from other sources. The sedimentation rate is related to particle size, and heavy metal accumulation in C1 is covariate and inverse. The covariate relationship shows that when the deposit rate increases, the sand content increases. An inverse relationship was identified between high sedimentation rate, reduced silt content, and reduced heavy metal content. This is evident when the parameters at C1 are compared; when the sedimentation rate increases, the sand content increases, the heavy metal content decreases (Figs. 2b, 3a, and 7a), the deposit rate increased from 2004 (at 17 cm), and the trend of heavy metals and silt also dropped thereafter.
In C2, FA1, FA2, and FA3 accounted for 60%,12.96%,and 8.15%,respectively. FA1 influenced sand,silt,Cu,Pb,Zn,As,Cr,Co,Ni,Fe,Mn,Mo,Li,and V;FA2 influenced clay and TOC;and FA3 influenced Cd (Table 4). Most of the positive correlations are concentrated between heavy metals; silt, TOC with Pb, Cr, Co, Ni, Li, and V indicate that the same source supplies are conducive to metal accumulation, as is evident in C2 at depths of 52-17 cm (before 1944) and 11-1 cm (1996-2019). Coinciding with the region's strong economic development, heavy metals increased over time (Fig. 7b). The negative correlation between sand and heavy metals, silt, and clay reflects conditions that are unfavorable for the accumulation of heavy metals. Clay and Cd showed almost no significant correlation with other metals, reflecting an independent supply from a local source (Fig. 8c).
The sedimentation rate in C2 is approximately three times lower than that of C1 (Fig. 2b), observed between 1944 and 2019 (0-13 cm). The deposition rate has a positive relationship with the accumulation of heavy metals. The deposition rate increases and then decreases, and heavy metals tend to behave similarly (Figs. 2b and 7b). Silt dominates in C2, which is mainly characterized by the accumulation of heavy metals from the silt fraction by showing the positive correlation between heavy metal and silt (Fig. 8c).
The distribution of heavy metals concentrated in the Mong Cai area generally depended on grain size and TOC in the surface sediments and C2. The TOC in C1 did not affect the distribution of heavy metals because it showed no positive correlation with the heavy metals, TOC reflected the in situ supply. The results from C1 indicate that the environment is relatively homogenous in dynamics and origin of sedimentary supplies, the FA1 with 15/17 parameters, the FA2 with TOC, and the FA3 with Cd (Table 4).

Sedimentary groups and their characteristics
Based on the CA results, the surface sediments were divided into three groups: those in C1 were divided into three groups, and those in C2 were divided into four groups. Each group was characterized according to heavy metal concentration, grain size, TOC, and dynamics (Table 5, Fig. 12).
Regarding the surface sediments, group 1 had high sand and gravel content, with the lowest concentration of heavy metals. Groups 2 and 3 were characterized by high silt and clay contents and thus had higher heavy metal concentrations than group 1 (Table 5). Group 1 was distributed in the Ka Long estuary and Ha Coi Bay, while groups 2 and 3 were distributed to the front of the Tra Co peninsula (Fig. 12a). The sediment parameters were divided into three groups: Group 1 contained gravel and Mo; group 2 contained heavy metals, silt, and clay (Fig. 12b); and group 3 contained sand. The parameter groups reflected two dynamics and one source of origin for Mo from gravel. The strong dynamic was sand; the weak dynamic was silt and clay with heavy metals. The environmental dynamics in group 1 are evident in the negative correlation between sand and heavy metals (Fig. 8a). Groups 2 and 3 represent the same sediment supplies, and favorable conditions for heavy metal accumulation by the positive correlation were together between metal and silt and clay.
C1 had three groups that reflected different dynamic conditions. Group 1 was distributed near the surface, middle, and bottom of the core; it had the highest sand content and low heavy metal concentrations. Group 2 was distributed in the middle and bottom of the core, with reduced sand content and increased silt, clay, and heavy metal contents. Group 3 was distributed in the core's middle and had the highest silt, clay, and heavy metal contents (Fig. 12c, Table 5). The three groups in C1 were characterized by three environment types. The distribution of the first type was strong at the bottom, middle, and top of the core, as characterized by group 1; the second type showed reduced distribution in the middle and bottom of the core, as characterized by group 2; and the third type was quiet, as characterized by group 3. The sediment parameters formed three groups: group 1 comprised sand; group 2 comprised silt, clay, and heavy metals; and group 3 comprised TOC (Fig. 12d). The three parameter groups showed different characteristics: group 1 was dominated by sand, independent of other parameters indicating a strong dynamic; group 2 was weak dynamics by characterized combination of silt, clay, and heavy metals; and group 3 was dominated by TOC, independent of other parameters indicating that TOC was provided in situ. Group 1 shows the environmental dynamics through the negative correlation between sand and heavy metals. Groups 2 and 3 indicated the same source sediment supplies and conditions favorable to the accumulation of heavy metals, showing positive correlations between heavy metals and between heavy metals and silts and clays. Independence from source supply is indicated by TOC, and correlation is insignificant for most parameters.
The sedimentation rate controls each group's sedimentary characteristics; group 1 has high sand content and low heavy metals, and the sedimentation rate in those positions is higher (Figs. 3c and 12c, Table 5). This is explained by a positive relationship with sand and a negative relationship with silt and metals (Fig. 8b); groups 2 and 3 increased the silt and heavy metal contents, and the sedimentation rate decreased (Figs. 3c and 12c, Table 5).
Four groups were distinguished in C2. Group 1 was distributed alternately at the surface, middle, and bottom and was characterized by high silt, clay, and heavy metal contents (Table 5, Fig. 12e). Group 2 was distributed near the surface, middle, and bottom and had high silt and sand contents but lower heavy metal concentrations. Group 3 was distributed near the surface (8-10 cm) and had the highest sand content and lowest heavy metal concentration. Group 4 was distributed near the surface (4-6 cm) and had a high silt content and a high heavy metal concentration. The four groups were characterized by four sedimentary environment types. Group 1 showed a weak dynamic associated with high silt and clay content, while group 2 showed a stronger dynamic characterized by high silt and sand contents. Group 4 was a quiet environment, characterized by the highest silt, clay, and heavy metal contents. Group 3 was the strongest environment, characterized by the highest sand content and lowest silt, clay, and heavy metal contents. The sediment parameters were divided into three groups: group 1 comprised sand and Cd, group 2 comprised silt and heavy metals, and group 3 comprised clay and organic carbon (Fig. 12f). Group 1 showed a strong dynamic, group 2 showed a weak dynamic, and group 3 showed a quiet dynamic. The negative correlation between sand and heavy metals reveals the environmental dynamics; high sand content is not favorable to heavy metal accumulation, as is shown in group 3. The environment is conducive to the accumulation of heavy metals. As indicated by the positive correlation between heavy metals and powder, TOC is represented by groups 1, 2, and 4. The correlation of Cd with other metals indicates no dependence on source supplies from elsewhere.
Although the sedimentation rate has a metal covariate relationship, in C2, the sedimentation rate is low, and so control is low (Figs. 2b, 7b, and 12b, Table 5). Silt content is high and controls heavy metal accumulation and shows a positive correlation between silt and heavy metals (Fig. 8c). Groups 1, 2, and 4 have weak and quiet environmental dynamics characterized by higher heavy metal content, silt predominates; group 3 at a depth of 6-8 cm (in 2000) is mainly sand with stronger dynamics and the highest sedimentation rate in the C2 (Fig. 2b, Table 5).

Conclusion
Fine sand was common in surface sediment along the Mong Cai coast. Very fine sand was abundant in C1, and very coarse and coarse silt dominated C2. Only the surface sediment contained a significant proportion of gravel. The average sand fractions in the surface, C1, and C2 sediments were 90.2%, 60.79%, and 28.9%, respectively. The silt fractions were 6.6%, 36.25%, and 64.1% in the surface, C1, and C2 sediments, respectively. The clay fractions were 0.5%, 2.23%, and 5.4% in the surface, C1, and C2 sediments, respectively.
TOC was highest in C2, followed by the surface, and lowest in C1, at 1.81%, 0.40%, and 0.31%, respectively. The chronology is 1877-2019 (142 years, 0-50 cm) in C1 at the Ka Long estuary, with an average sedimentation rate of 0.71 cm/year; 1944-2019 (75 years, 0-14 cm) in C2 at Mui Ngoc has an average sedimentation rate of 0.27 cm/year. The organic matter in sediment (ẟ 13 C, ẟ 15 N) indicated for environment surface sediments was sourced from marine and terrigenous sediment.
Heavy metal concentrations in sediments were surface < C1 < C2. Most heavy metals were lower than the ISQGs, with the exception of As in C1 and C2. In assessing metal pollution in the sediments, the Igeo showed < 1 (unpolluted) in the surface, C1, and C2 sediments, while only As and Li showed unpolluted to moderately polluted. The EF of Pb, As, Zn, Cd, and Li showed minor to moderate enrichment. The CF of Pb, As, Fe, Li, Cd, and Mo showed moderate to considerable contamination in C1 and C2. The ER of As indicated a moderate potential ecological risk, while the remaining heavy metals showed a low potential ecological risk. The CD indicated moderate (C1) and considerable (C2) degrees of contamination. The RI indicated a low ecological risk.
Heavy metals distribution in surface sediments is controlled by particle size, with TOC showing strong to moderate correlation, negative correlation between heavy metals and sand, and positive correlation between heavy metals and silt, clay, and TOC and between heavy metals together. Only Cd showed low content and no correlation with other metals, reflecting its in situ origin. Coarse sediments dominate, and so most of the heavy metal content is low. In C1, heavy metal distribution is affected by grain size and sedimentation rate but is not affected by TOC. Sedimentation rate and sand are negatively correlated with heavy metal accumulation. The sedimentation rate and sand are highly inhibited in terms of the accumulation of heavy metals, reflecting the decreasing trend over time in C1. No significant correlation was observed between TOC and the parameters, indicating that TOC is supplied in situ. In C2, the distribution of heavy metals is controlled by silt, high silt content, and positive correlation with heavy metals, although heavy metals are positively correlated with sedimentation rate (1944-2019) but at a gradual pace, so that heavy metal accumulations are small. Concentrations of Cu, Pb, Zn, As, Cr, Co, Ni, and Mn increased over time from 1996 to 2019, reflecting the impact of the local economy in coastal areas.
Finally, the accumulation of metals in surface sediments and sediment core is affected by sand on the surface and in C1, silt in C2, TOC at the surface and in C2, and sedimentation rate at C1. At the surface and in C1, sediments are sandy, and the sedimentation rate is not conducive to heavy metal accumulation. In C2, silt is the main factor favorable for heavy metal accumulation. The TOC at the surface and in C2 sediments is conducive to the accumulation of heavy metals. Although the CD coefficient ranged from moderate (C1) to considerable (C2), most was concentrated in the lower parts of the two cores, and the surface layer showed low (C1) and moderate (C2) degrees of contamination. Ecological risk coefficients (RIs) are all low. Sediment quality in Mong Cai does not currently affect the ecosystem or fauna in the coastal area. However, Cu, Pb, Zn, Cr, As, Co, Ni, and Mn levels should be monitored, as mentioned above, to mitigate the impact on the coastal environment.
Availability of data and materials Raw data will be made available on request.

Declarations
Ethical approval Not applicable.

Consent to participate Not applicable.
Consent to publish All of the authors agree with the present submission of this paper.