On the speciation of metals in the water column of a polluted estuary

The dissolved (<0.40 γm) fraction of water samples from Newark Bay, New Jersey was analysed for Zn, Cu and Pb content by differential pulse anodic stripping polarographic techniques. In the dissolved fraction, non-labile forms of Zn, Cu and Pb exist as shown by differential pulse anodic stripping voltammetry analysis on acidified samples and acidified-UV irradiated samples. The particulate fraction (>;0.40 μm) contains metals in the form of sulphides (Fe, Zn, Cu, Mn), oxides and oxyhydroxides (Si, Al, Fe, Cu, Ni, Sn), phosphate (Ca, Ce, La), clay minerals (Fe, Zn, Cu, Ti) and carbonaceous material (Fe, Cu, Zn) as demonstrated by X-ray microanalysis. The solid phases are likely present in colloidal form in the dissolved fraction of the water column as well. The forms of the metals in the water column are partially due to the resuspension of bottom sediments by dredging and natural processes, to sewage outfall and to natural geochemical processes.     

Concentrations and transport of trace metals in the St. Lawrence River

Samples of raw water were collected in the St. Lawrence River during six sampling trips from August 1990 to April 1992. Water samples were analyzed for both dissolved and particulate phases for five trace metals. Partition coefficients (Kd) and metal fluxes were calculated in order to determine metal transport. A mass balance equation was used for the determination of the major metal sources to the St. Lawrence River and an estimation of metal loadings to the estuary was made. Average dissolved metal concentrations were found to be Cd 10 ng/L, Co 74 ng/L, Cu 64 ng/L, Fe 69 µg/L and Mn 700 ng/L. Particulate concentrations were (in µg/g) 1.68 for Cd, 31 for Co, 73 for Cu, 25 mg/g for Fe and 1.69 mg/g for Mn. Co, Fe and Mn were transported essentially in the particulate phase while Cd and Cu were predominantly found in the dissolved phase at 56% and 48% of the total metal concentration respectively. Log Kd values varied from 5.1 (for Cu) to 6.8 (for Mn). In the dissolved phase the major sources were found to be the Great Lakes and the Ottawa River whereas in the particulate phase Québec tributaries appear to be the most important. Industrial inputs are quite important in both the dissolved and the particulate phases for Cd, whereas other sources are very variable, especially for the dissolved phase.     

Dissolved trace metal distributions in Port Jackson estuary (Sydney Harbour), Australia

Concentrations of dissolved metals (Cd, Cu, Ni, Mn and Zn) were determined for summer and winter, under low-flow conditions in Port Jackson, a microtidal, well-mixed estuary in south-east Australia. Mean concentrations of Cd (0.04+/-0.02 microg/l), Ni (0.86+/-0.40 microg/l), Mn (20.0+/-25 microg/l) and Zn (6.47+/-2.0 microg/l) were below water quality guidelines. Concentrations of Cu (1.68+/-0.37 microg/l), however, slightly exceeded recommended values. Dissolved Ni and Mn behaved mostly conservatively, whereas Cd, Cu and Zn showed mid-estuarine maxima. Peaks in Cd, Cu and Zn concentrations were located in the upper estuary, independent of the salinity and suspended particulate matter loading, and were consistent with anthropogenic inputs of metals in the estuary. Concentrations of dissolved Cu were highest in summer, whereas concentrations of Cd, Ni and Mn were significantly lower in summer than winter (P< or =0.05). The increase in temperature and biological activity during summer explained the seasonal variation. The sequence of log K(d) values (20-30 salinity) was Mn>Zn>Cu>Ni. These results give unique information concerning the contemporaneous distribution of dissolved trace metals in the Port Jackson estuary and they provide a data set against which the long-term contamination may be assessed.     

Dissolved and particulate trace metals in a Scottish sea loch: an example of a pristine environment?

In the near pristine environment of a silled fjord on the west coast of Scotland samples were taken for the determination of dissolved and particulate trace metals (Fe, Mn, Cu, Ni, Cd, Zn and Pb), together with nutrient and hydrographic data, during 19 surveys carried out over a year. An indication of the pristine nature of the environment are the low concentrations of dissolved silicon, phosphate and nitrate which are considerably lower than those of coastal waters which are subject to larger anthropogenic burdens. Distributions of dissolved Cu, Ni and Cd were found to broadly reflect conservative mixing of freshwater and seawater with both end members having similar concentrations. The concentration of dissolved Cu and Ni in seawater entering upper Loch Linnhe (Cu 0.28 μg l⁻¹; Ni 0.26 μg l⁻¹) was consistent with the 1:1 conservative mix of Irish Sea water and North Atlantic surface water predicted from radio-caesium tracer experiments (Mackay & Baxter, 1985). Atmospheric input of trace metals to upper Loch Linnhe appeared to be a relatively minor term in the mass balance relative to fluvial inputs. Values of distribution coefficients Kd were similar to those previously reported for the coastal environment. Iron showed the strongest affinity for the suspended sediments; with particulate percentages of the total load usually greater than 80%. Lead and Mn showed a similar strong affinity to the particle phase. For Cu, Ni and Zn the mass of the element in the dissolved phase was generally greater than that in the particulate fraction. Cadmium, was least associated with the particles, with typically greater than 90% existing in the dissolved phase.     

Water quality, nutrients and the water framework directive in an agricultural region: The lower Humber Rivers, northern England

The water quality of rivers in the eastern part of the Humber Basin, north-eastern England is described from a baseflow survey (11–13 August 2006) of a wide range of water quality determinants, and long-term nutrient records of the Environment Agency of England and Wales (EA). The baseflow survey shows that the rivers are oversaturated with respect to dissolved CO₂ and calcite. They are sodium, potassium, lithium, boron, chloride, sulphate and fluoride bearing from a combination of atmospheric, within-catchment, agricultural and sewage effluent sources. Nitrate concentrations are uniform for rivers draining permeable bedrock but decrease for clay drainage areas. Soluble reactive phosphorus (SRP) concentrations are variable across the catchments reflecting the importance and variability of point sources and within-river processing. The EA data show annual oscillations for both NO₃ and SRP concentrations. Average NO₃ concentrations vary between 3.3 and 18.8 mgN/l and concentrations are low during the summer months. Average SRP concentrations vary between 23 and 1959 μg/l. Highest SRP concentrations generally occur when there is effluent input from sewage treatment works and agricultural point sources (e.g. overflow from slurry tanks, farmyard washings). Despite many of the rivers being nutrient rich, they are generally of good biological quality when point source inputs are not important.     

Measurement of soluble reactive phosphorus concentration profiles and fluxes in river-bed sediments using DET gel probes

DET (diffusive equilibrium in thin films) gel probes were used for sampling river-bed sediment porewaters, to characterise in situ soluble reactive phosphorus (SRP) concentration profiles and fluxes. DET probes were deployed in three contrasting rural streams: (1) a headwater ‘pristine’ stream, with minimal P inputs from low intensity grassland and no point sources, (2) an intensively cultivated arable catchment, and (3) a stream subject to high P loadings from sewage effluent and intensive arable farming. The DET results showed highly enriched porewater SRP concentrations of between ca. 400 and 5000 μg-P l⁻¹ in the sewage-impacted stream. In contrast, the arable and pristine streams had porewater SRP concentrations <70 μg-P l⁻¹ and <20 μg-P l⁻¹, respectively. Porewater SRP concentration profiles in both the sewage-impacted and arable-impacted streams showed well-defined vertical structure, indicating internal sources and sinks of SRP within the sediment. However, there was little variability in porewater SRP concentrations in the pristine stream. The DET porewater profiles indicated net diffusion of SRP (a) from the overlying river water into the surface sediment and (b) from subsurface sediment upwards towards the sediment–water interface. A mass balance for the sewage-impacted site showed that the influx of SRP into the surface sediments from the overlying river water was small (ca. 1% of the daily river SRP load). The DET results indicated that, in the arable and sewage-impacted streams, the surface ‘cap’ of fine sediment may play an important role in inhibiting upward movement of SRP from subsurface porewaters into the overlying river water, under steady-state, low-flow conditions.     

Influence of rural land use on streamwater nutrients and their ecological significance

Concentrations and loads of N and P fractions were examined for lowland rivers, the Wye and Avon, draining a range of representative agricultural land-use types in two major UK river basins. Data collected over a 2-year period demonstrated important diffuse agricultural source contributions to N and P loads in these rivers. Ground water provided a major source of total dissolved nitrogen (TDN) loads, whereas near-surface sources provided a major contribution to total phosphorus (TP) loads. In terms of aquatic ecology, concentrations of nutrients, at times of eutrophication risk (spring and summer low flows) were of key environmental and management significance. Agricultural diffuse sources provided the major source of long-term P loads across the two basins. However, the results demonstrated the dominance of point-source contributions to TP and SRP concentrations at times of ecological risk. Point sources typically ‘tip the balance’ of dissolved inorganic P (soluble reactive P, SRP) above the 100 μg l⁻¹ guideline value indicative of eutrophication risk. The significance of point sources for TP and SRP concentrations was shown by (a) the strong correlations between TP, SRP and B concentrations, using B as a tracer of sewage effluent, (b) the dominant contribution of SRP to TP concentrations and (c) the predominant pattern of dilution of SRP and B with flow. The clean Chalk streams draining low intensity grassland in areas of the Avon with sparse human settlement were oligotrophic and P limited with low SRP concentrations under spring and summer baseflows attributable to groundwater sources. The data provide important insights into the ecological functioning of different lowland stream systems. There was evidence of greater SRP losses and N-limitation in a stream which drains a pond system, demonstrating the importance of longer water residence times for biological nutrient uptake.     

Metals in Sediments of the Upper Laguna Madre

The Laguna Madre system is the largest hypersaline coastal basin in the United States. Surface sediments from 22 Upper Laguna Madre (ULM) sites were analyzed for grain size and metals (Cd, Cu, Ba, Cr, Mn, Ni, Pb, V, Zn, Fe, and Al) to assess the extent of contamination in the area. Sediments were found to consist mainly of sand texture. Clay and silts were minor constituents (<10%) of the sediments. Anomalies in metal concentrations were found at some sites and were related to probable sources, i.e., recreational and industrial activities. Concentrations of metals were normalized to grain size, Al, and Fe to distinguish natural and anthropogenic sources. Most metals showed positive correlations (p < 0.001) with Fe and Al, suggesting a natural variability of metal concentrations in sediments. Concentrations of metals, except Cd and Pb, at most sites were found to be below threshold concentrations thought to produce toxic effects in marine and estuarine organisms.     

Effects of Animal Size on Concentrations of Metallothionein and Metals in Periwinkles Littorina littorea Collected from the Firth of Clyde, Scotland

Different sizes of Littorina littorea were collected from four areas in the Firth of Clyde, Scotland. Their metallothionein (MT) and heavy metal concentrations were analysed using the Silver Saturation Method and Atomic Absorption Spectrophotometry, respectively. Concentrations of MT, Cd and Zn (as μg g⁻¹ dry soft-body weight) generally decreased with an increase in size of L. littorea. MT concentrations were better correlated with Cd than with Zn or Cu concentrations. Nevertheless, MT and the metals in periwinkles (as μg individual⁻¹) increased significantly with increasing size. Concentrations of MT and the metals among the sampling areas were compared at a standardised soft-body weight (10 mg). The results from Discriminant Analysis based on all metal concentrations indicate that Largs is different from the other areas and characterised by high Fe concentrations in L. littorea. The problems and differences in using either soft-body weight or shell length as independent variable for size-standardisation are discussed.     

Input and behaviour of linear alkylbenzenesulphonates (LAS) in a stratified estuary

Input and distribution of linear alkylbenzenesulphonates (LAS) in the highly stratified Krka River estuary were studied during 1990–1991. Determinations of individual LAS homologues in wastewater and estuarine waters were performed using reversed-phase high-performance liquid chromatography (HPLC) with spectrofluorimetric detection. The investigated municipal wastewaters contained LAS homologues with alkyl-chain lengths from C₁₀ to C₁₃ at total concentrations in the range of 285–1041 μg l⁻¹. Total LAS input via wastewaters into ibenik Harbour, the most polluted part of the estuary, was estimated at 12.6 kg day⁻¹. A relatively large fraction of total LAS, depending on the wastewater particle load, was found in particulate form (11–59%). Distributions of the individual homologues in dissolved and particulate fractions were rather different, the latter being significantly enriched in higher homologues (C₁₂, C₁₃). The LAS concentration in the estuary was quite low (0.2–23.9 μg l⁻¹) which was caused chiefly by a strong and fast dilution process. However, characteristic vertical distribution of LAS in the estuarine water column indicated that the wastewater plume spreads almost exclusively in the upper (brackish) layer. The concentration maxima were observed in the surface microlayer and at the brackish water/seawater boundary.     

Methane in the hyporheic zone of a small lowland stream (Sitka, Czech Republic)

Methane concentrations and selected chemical parameters in interstitial water were examined along subsurface flowpath in two subsystems (hyporheic and parafluvial sediments) in the Sitka stream, Czech Republic. Interstitial methane concentrations exhibited a distinct spatial pattern. In the hyporheic downwelling zone where the sediments are relatively well oxygenated due to high hydrologic exchange with the surface water, low interstitial methane concentrations, averaging 9.3 μg CH₄/l, were found. In contrast, upwelling sediments and parafluvial sediments (active channel sediments lateral to the wetted channel) had significantly higher methane concentrations (p < 0.05, and p < 0.01, respectively), averaging 43.2 μg CH₄/l and 160.5 μg CH₄/l, respectively. Dissolved oxygen was the highest where surface water entered hyporheic/parafluvial sediments and decreased with water residence time in the sediments (p < 0.01). Nitrate concentrations decreased along the flowpath and were significantly lower at downstream end of the riffle (p < 0.001). Sulfate concentrations also show a slight decline with the water residence time, but differences were not significant. Effect of both nitrate and sulfate on methanogenesis is also discussed. The interstitial methane concentration significantly increased with surface water temperature (p < 0.001) and was negatively correlated with redox potential (p < 0.01) and dissolved oxygen (p < 0.05).     

Chlorohydrocarbon pesticides in the Australian marine environment after banning in the period from the 1970s to 1980s

The chlorohydrocarbon pesticides (CHPs) (principally DDT, dieldrin, hexachlorocyclohexane (HCH) including lindane, chlordane, heptachlor and heptachlorepoxide, aldrin and hexachlorobenzene (HCB)) were used extensively in Australia during the 1960s and 1970s. These were progressively banned from the 1970s with DDT being totally banned in 1987. The levels in marine waters are believed to have progressively declined as indicated by water in the Brisbane River, which consistently fell from maximum concentrations of about 1.7 μg/l in 1972–1973 to not detectable in 1986–1987. In some urban areas sediment concentrations, up to 1700 μg/kg dry weight, were recorded in the early 1990s which may reflect the reworking and exposure of older more contaminated material. Fish exhibited consistent occurrence of DDT, HCH, dieldrin and HCB through the 1970s to the 1990s. Maximum concentrations of DDT and dieldrin occurred generally in the 1970s at 40.3 and 8.8 μg/g wet weight, and minimums in the 1990s at 2.4 and 0.041 μg/g wet weight respectively. Inputs of dialdrin from sewage and DDT from the broad environment, reflected by the calculated daily human uptake in the diet, were shown to decline with half-lives of 1.1–1.5 and about 3 years respectively. This evidence suggests that banning of the CHPs has greatly reduced the inputs into the marine environment with consequent reductions in concentrations in biota but there is the occasional occurrence of relatively high concentrations.     

Levels of heavy metals along the Libyan coastline

A coastal survey off the east part of the Libyan coastline which has no river inputs was initiated to measure the existing level and distribution of selected heavy metals (Cu, Zn, Cd, Ni, and Mn) in surface sediments along Benghazi Bay. In order to place the finding of this survey in proper perspective and for practical terms a Pollution Load Index (PLI) was used. The results establish that of the sites examined only the harbour site exhibited relatively high levels particularly in the cases of cadmium and copper indicating a detectable anthropogenic input. The range of concentrations were 1.03%–1.96% μg g⁻¹ (dry wt) for organic content and 64.7%–93% for calcium carbonate while the range of heavy metals concentrations (μg g⁻¹ dry wt) were 8.7–42 for Cu, 2.3–27.3 for Zn, 5.7–19 for Ni, 37–76.7 for Mn, and below the detection limit-1.73 for Cd. A PLI score of 2.93 was obtained at the harbour site and a range of 6.11–7.63 for the other sites. The PLI score for the control site was 9.51 and the score for the overall studied area was 7.51 indicating a very clean environment. The index would be, with some modification, a step in the right direction from the management point of view in Libya due to the absence of environmental expertise.     

Copper,zinc, cadmium,nickel, iron and manganese in the Southern Bight of the North Sea

Ni, Cd, Cu, Zn, and Fe were determined in water and suspended matter samples and Mn in suspended matter samples from the Southern Bight region of the North Sea. Four distinct areas, comprising the Rhine estuary, the Dutch coastal waters, the centre of the Bight, and the English coastal water can be distinguished on the basis of the distribution of the trace metals. The total concentrations of all metals were high in the Rhine estuary, decreased in a seaward direction to reach very low values, comparable to continental shelf water, in the centre of the Bight, and increased again towards the English coast. Elevated concentrations of dissolved Cu, Ni, Cd and Zn near the English coast are not only due to river inputs. Over the entire region, the fraction of Cd, Ni and Zn in solution was consistently more important than the fraction in the suspended particulate phase. In contast, the fraction of Fe and Mn in the solid phase was consistently greater than the fraction in solution. Total concentrations of Fe, Mn and Cu are more dictated by the suspended particulate matter concentration than Cd, Ni and Zn.     

PAH contamination of western Irish Sea sediments

Concentrations of the sum of 15 PAHs in 22 surficial sediment samples from the western Irish Sea ranged from below 100 ng g⁻¹ in sandy sediments to a maximum of 1422 ng g⁻¹ in the centre of the mud basin. The concentrations are typical of coastal shelf sediments, but greater than those observed for aquatic sediments remote from known anthropogenic sources. Organic C and the % sediment <15 μm were positively correlated with ΣPAH. ΣPAH was normalised to organic C and particle size (i.e. expressed as ΣPAH/C^org and ΣPAH/%<15 μm) and significant relationships were still observed with organic C and %<15 μm. The results are discussed in the context of using organic C to normalise concentrations of PAHs to assess contamination levels in sediments.     

Dissolved metal contamination in the East River-Long Island sound system: potential biological effects

A suite of dissolved trace metals (Ag, Cd, Cu and Pb), inorganic nutrients (NO(3), PO(4)), and chlorophyll a was measured along a 55 mile transect from the East River into western and central Long Island Sound. The main objectives of this study were to determine the relative levels of contamination from sewage, and to assess its possible biological impact on local waters. The East River-Long Island Sound system receives large volumes of treated sewage and industrial effluent as a result of the heavy urbanization of the area. Despite these strong environmental pressures, this study is among the first to report dissolved metal levels from that region. Consistent with the locations of anthropogenic sources, a strong east-west concentration gradient was observed for Ag, Pb, NO(3) and PO(4) with the highest levels found in the East River. In contrast, dissolved Cd and Cu were relatively constant throughout the area of study, suggesting that sewage sources have a more limited influence on the levels of those metals. Remobilization from contaminated sediments may represent the primary source of Cd and Cu to the Long Island Sound under low-runoff conditions in summer. Chlorophyll a concentrations, used as an indicator of total biomass, were also low in the East River. These low chlorophyll concentrations could not be explained by nutrient or light limitation, water column stratification, or to advection of phytoplankton out of the river during tidal flushing. These preliminary results suggest a potential toxic effect of sewage on the biological communities of the East River.     

Cadmium content, accumulation and toxicity in dog whelks collected around the Welsh coastline

The concentration of cadmium in sexually mature Nucella lapillus was investigated around the coast of Wales. Higher concentrations of cadmium were found in whelks located in the Bristol Channel compared with Cardigan Bay Nucella. Accumulation and toxicity of cadmium were investigated at 50–600 μg Cd l⁻¹ over a period of 21 days. At the lower concentrations of 50–100 μg Cd l⁻¹ all Nucella survived the 21 days of experimentation. The rate of cadmium uptake after exposure to cadmium was inconsistent and not related to the degree of site contamination. At all sites the concentration of cadmium in Nucella, after treatment with 50–600 μg Cd l⁻¹ for 21 days, was significantly different from controls, but after 7 days it varied with the site and cadmium concentration. Only samples collected at Fishguard, Dale, and Swansea survived 21 days at 600 μg Cd l⁻¹. Tissue cadmium concentrations did not significantly effect the whole body concentration of zinc.     

Zinc in Fish, Crabs, Oysters, and Mangrove Flora and Fauna from Cleveland Bay

Zinc in muscle tissue from 28 species of fish, three species of sharks and rays, two species of squid and cuttlefish, and three species of crustacea collected from Cleveland Bay, together with some measurements made on oysters and mangrove flora and fauna are reported. In the fish species the mean Zn concentration in muscle tissue was 4.83 ± 2.82 μg/g wet weight, slightly lower than zinc measurements (7 μg/g wet weight) made in 14 fish species collected from the bay in 1975. Zinc in sharks and rays ranged from 3.5 to 7.2 μg/g wet wt, in squid and cuttlefish 13–16 μg/g wet wt, and in crustacea levels ranged from 14 to 18 μg/g wet wt. Zinc levels in fish varied between species with concentrations well below the ANZECC Maximum Residue Limit of zinc in seafood (150 μg/g wet wt). Although lower zinc concentrations were found in oysters collected from the Townsville Harbour area (2080 μg/g wet wt), compared with previous measurements made 25 years ago at this site, these levels are still above the Maximum Residue Limit for zinc in oysters (1000 μg/g wet wt). Concentrations of zinc in oysters from Orpheus Island (2547 μg/g wet wt), about 74 km away from industrial and urban activity, are also above the safe guideline values. Horseshoe Bay oysters transplanted to Ross Creek accumulated zinc at a rate of about 100 μg/g of oyster tissue per week, suggesting that dissolved zinc levels at this site are elevated, and that oysters rapidly accumulate zinc. Highest concentrations of zinc in mangrove leaves (30–65 μg/g dry wt) occurred in Osbornia octodonta, Exocaria agalocha, and Aegialitus annulata, compared with Ceriops tagal, and Avicennia marina (5–10 μg/g dry wt). No significant difference in zinc concentration occurred between leaves and litter fall for most of these species, with the sole exception of Exocaria, which showed almost a twofold increase in concentration. In seven species of mangrove fauna levels of zinc were very uniform and close to 50 μg/g (dry wt) Telescopium telescopium from the mouth of the Ross Estuary was the exception with levels at 400 μg/g (dry wt).     

Toxicity of 1,4-dichlorobenzene in sediments to juvenile polychaete worms

Investigation of sediment contamination associated with a marine sewage outfall in Victoria (BC, Canada) found elevated concentrations of 1,4-dichlorobenzene (1,4-DCB). Juvenile polychaete worm (Neanthes) growth was significantly reduced at or near the outfall, roughly corresponding to elevated 1,4-DCB concentrations. There are few data on 1,4-DCB toxicity to marine organisms and no published literature on its toxicity to benthic marine organisms. To determine whether reduced polychaete growth (measured as dry weight) was due to 1,4-DCB exposure, a laboratory investigation was conducted. Uncontaminated marine sediment was spiked with 1,4-DCB and juvenile Neanthes were exposed in 20-d sublethal toxicity tests. There were no adverse effects on survival at any test concentration; mean survival was 80–100%. Statistically significant decreases in average dry weight only occurred at the highest 1,4-DCB concentration (19,900 μg/kg, dry weight); this represented a 1,4-DCB concentration more than 10 times higher than previously measured at the outfall (1710 μg/kg, dry weight). There were no adverse effects on survival or dry weight at the range of concentrations previously measured in sediments from the vicinity of the outfall.     

鄱阳湖流域赣江水系溶解态金属元素空间分布特征及污染来源

于2015年1月和7月在赣江干流和主要支流37个采样点共采集74个水样,分析赣江水系15种溶解态金属元素(Be、Al、V、Mn、Fe、Co、Ni、Cu、As、Mo、Cd、Sb、Tl、Pb、U)的空间分布特征和污染来源的贡献率.结果表明:多数水样的溶解态金属元素浓度符合水质标准,主要的超标元素是Fe,样品超标率为21.60%,其次为As(8.10%)、Mn(4.05%)、Tl(4.05%)和Al(1.35%).Be、Al、V、Fe、Co、Ni、Cu、U浓度在枯水期显著高于丰水期,其他元素差异不显著.依据溶解态金属元素的空间分布特征,赣江流域可分为3个区域:湘水、章水和赣江赣州市段(C1),桃江、袁水和锦江(C2),其他区域(C3);溶解态金属元素水平大小排序为C1C2C3,其中Be、Al、Cu、Mo、Sb、As浓度在C1最高,V、Mn、Fe、Ni、Cd浓度在C2最高.采矿废水、矿渣和农田土壤降雨淋滤、钢铁冶炼废水是赣江溶解金属元素的主要来源;Be、Al、Cu、Pb、U的污染源超过40%来自采矿废水,Cu、As、Mo、Cd的污染源超过35%来自矿渣和农田土壤降雨淋滤,V、Mn、Co、Ni的污染源超过41%来自钢铁冶炼废水.