Metal accumulation in aquatic macrophytes from southeast Queensland, Australia

Metal accumulation in aquatic macrophytes from southeast Queensland, Australia

Chemosphere 48 (2002) 653–663 www.elsevier.com/locate/chemosphere Metal accumulation in aquatic macrophytes from southeast Queensland, Australia A.J...

410KB Sizes 57 Downloads 122 Views

Chemosphere 48 (2002) 653–663 www.elsevier.com/locate/chemosphere

Metal accumulation in aquatic macrophytes from southeast Queensland, Australia A.J. Cardwell, D.W. Hawker *, M. Greenway Faculty of Environmental Sciences, Griffith University, Nathan, Qld 4111, Australia Received 15 October 2001; received in revised form 18 March 2002; accepted 3 April 2002

Abstract To determine the extent of metal accumulation in some aquatic macrophytes from contaminated urban streams in southeast Queensland, plants were sampled from six sites, along with contiguous sediments. In all, 15 different species were collected, the most common genera being Typha (Cattails or Bulrushes) and Persicaria (Knotweeds). Before heavy metal analysis, plants were further separated into various morphological tissues, and five selected samples were separated into various physiological tissues. The cadmium, copper, lead and zinc content of the plants were analysed using flames AAS. In general, plant roots exhibited higher metal concentrations than the contiguous sediments. Of the metals of interest, only for zinc was there a relatively clear pattern of increasing accumulation in aquatic macrophytes with increasing sediment metal concentrations. Comparison between morphological tissues of the sampled plants found that roots consistently presented higher metal concentrations than either the stems or leaves, however unlike previous studies, this investigation revealed no consistent trend of stems accumulating more metals than the leaves. For Typha spp., metal concentrations followed the order of roots > rhizomes > leaves, while for Persicaria spp. the order was roots > leaves > stems. The submerged species Myriophyllum aquaticum accumulated the highest levels of metals overall (e.g. Zn 4300 lg g1 dry weight and Cd 6.5 lg g1 ), and the emergent macrophytes also exhibited relatively high metal contents in their roots. The leaves of the submerged and floating-leafed species collected contained relatively high quantities of the four metals of interest, compared with the leaves of emergent aquatic macrophytes. In the Typha rhizome and Persicaria stem samples analysed for internal variation in metal content, there was a pattern of increasing metal concentrations towards the external sections of the stem, both for subterranean stems (rhizomes) and above-substrate stems. For Persicaria stems, no clear pattern was observed for cadmium and lead, the two metals investigated that are not required by plants for survival.  2002 Elsevier Science Ltd. All rights reserved. Keywords: Metals; Aquatic plants; Morphological and physiological tissues; Translocation

1. Introduction Urban streams of the southeast Queensland region are typical of those in many other parts of the world in that they have historically received metal contamination

*

Corresponding author. Fax: +61-7-3875-7459. E-mail address: [email protected] (D.W. Hawker).

from a variety of anthropogenic sources, including treated sewage discharges, runoff from metal-related industry and urban stormwater runoff. Once in the aquatic environment, metals such as cadmium, copper, lead and zinc undergo many physical, chemical and biological transformations (Lung and Light, 1996; Moss and Costanzo, 1998; Semple and Williams, 1998). Most heavy metals in aquatic systems eventually become associated with particulate matter, which settles and accumulates in the bottom sediments. Empirical

0045-6535/02/$ - see front matter  2002 Elsevier Science Ltd. All rights reserved. PII: S 0 0 4 5 - 6 5 3 5 ( 0 2 ) 0 0 1 6 4 - 9

654

A.J. Cardwell et al. / Chemosphere 48 (2002) 653–663

studies and experiments using radioactive tracers have established that most rooted aquatic plants uptake chemicals primarily from the sediment pore water (Jackson, 1998). To the extent that metal species occur in the overlying water column, some uptake by above-sediment plant parts is also possible. For free floating macrophytes, this would be the only source of metals. Freshwater aquatic plants are those that are physiologically adapted to surviving in permanent or semipermanent freshwater ecosystems. The concentrations of metals in aquatic plants can be more than 100 000 times greater than in the associated water (Albers and Camardese, 1993). Certain aquatic plant species can thus be used as indicators of low level environmental contamination that might otherwise be difficult to detect. Of importance too is the fact that contaminated aquatic vegetation can be a source of food for a variety of herbivores and detritivores, leading to the possibility of bioaccumulation of metals in higher trophic levels of the food chain (Crowder, 1991; Devi et al., 1996). Research that undertakes to determine the extent of metal contamination in aquatic systems, and elucidates the pathways of metal accumulation in aquatic plants, is therefore of some value. These investigations can identify potential threats to natural aquatic ecosystems, as well as providing information as to which locally growing aquatic plant species might be of use in for example, constructed stormwater and mine-drainage treatment wetlands. Cadmium, copper, lead and zinc were chosen as the metals for this study for a number of reasons. Copper and zinc are trace metals required by plants for survival, whereas cadmium and lead are both toxic metals not required by plants. It was of interest to ascertain whether this would have a significant effect on plant metal concentrations. The presence of cadmium above trace levels in the environment is an indicator of contamination, and lead is a common pollutant from road runoff. Zinc is a common metal present in variable amounts, and if found in appreciable amounts can be an indicator of industrial pollution while copper is also an indicator of industrial contamination of urban streams. All four metals have been found in different quantities in urban creeks in the Brisbane area of southeast Queensland (Moss and Costanzo, 1988; Semple and Williams, 1998), and all are accumulated by plants in varying degrees. The uptake mechanisms are however of some debate. The primary objectives of the present investigation then were: (a) To quantify the extent of cadmium, copper, lead and zinc accumulation in aquatic plants growing in contaminated urban streams, and examine any relation-

ships between sediment metal concentrations and plant metal uptake. (b) To determine accumulation differences between plant species, and identify those species accumulating significantly higher amounts of the four metals examined. (c) To determine the extent of movement of the metals into different structural and internal tissues of the plants.

2. Materials and methods 2.1. Sampling site locations To investigate the extent of heavy metal uptake by aquatic plants in southeast Queensland, six urban creek study sites were selected in the southern Greater Brisbane area, as illustrated in Fig. 1. At each site, three samples of all representative species within a 25 m stretch of creek were collected. In November 1998, the water at all sites was generally less than 2 m in depth on average and relatively slow moving. Pertinent details of some of the sites are: • Site 1: A Melaleuca wetland upstream from a sewage treatment plant on Slacks Creek. • Site 2: Approximately 600 m downstream from a landfill and liquid waste treatment plant. • Site 5: Immediately downstream from a major road bridge and was surrounded by construction and scrap-metal industry. • Site 6: Approximately 1 km downstream from Site 5, and about 300 m below where the stream passes under a major five-road intersection. 2.2. Sampling, storage and analysis of plants Three specimens of each species were collected as whole plant samples from each site in November 1998, and transported to the laboratory in clean plastic bags. Following methods outlined in O’Halloran et al. (1997) and Sadler and Rynja (1992), plants were carefully washed using tap water then deionised water, to remove visible debris. They were then separated where appropriate into roots, stems, leaves and reproductive tissues (flowers, seeds, etc.). Some root, rhizome and stem samples were further separated into visibly differentiated internal physiological tissues. The washed samples were carefully dried of adherent water using absorbent paper and were stored in plastic ziplock bags (200  270 mm) and refrigerated at 4 C for not more than seven days. Samples were then cut into small pieces, placed in aluminium trays (140  160 mm) and dried to a constant mass in a fan-forced oven at 80 C (overnight). A tem-

A.J. Cardwell et al. / Chemosphere 48 (2002) 653–663

655

80 C and homogenised by grinding with a mortar and pestle. A sample of approximately 0.5 g of ground sediment was weighed into a 75 ml glass block digestion tube and underwent wet digestion following the same procedure as the plant samples (described below). 2.4. Modified wet digestion procedure for plant material Extraction of the metals from plant samples (0.5–1.0 g) prior to AAS analysis employed a modified wet digestion procedure involving concentrated nitric acid. The extraction method described above was modified from that outlined in Campbell and Plank (1998), and the accuracy of the procedure was assessed using standard reference material SRM 1571 (Orchard Leaves). Field and reagent blanks were routinely carried out. 2.5. Standard reference material Every series of 30–50 samples analysed included a sample of a SRM 1571 (Orchard Leaves), which was weighed (0.5 g) into a 75 ml digestion tube and underwent wet digestion along with the plant samples collected from the six sites. Mean measured concentrations of the metals of interest were within 1.7 lg g1 (or an average of 6.2%) of published values for the SRM. 2.6. Statistical analyses

Fig. 1. Map of the study location and sampling sites in the southern Greater Brisbane area of southeast Queensland.

perature of 80 C was used because below this temperature all moisture may not be removed from the sample, and above this temperature thermal decomposition may occur, resulting in a reduction in dry weight (Campbell and Plank, 1998). The oven-dried material was chopped finely then ground (to 6 1.0 mm particle size) to ensure homogeneity and to facilitate organic matter digestion. The finely ground material was stored in sealed plastic ziplock bags (75  140 mm) at room temperature for not more than one month, until wet digestion.

Metal concentrations of sediments and plant material are reported in lg g1 dry weight, and are the means of three replicates. For each sample, standard deviations were calculated using the statistical function available in the spreadsheet package Microsoft Excel 97. Confidence intervals (CI) were determined using Excel 97, which calculates CI for a given alpha value (a), a given standard deviation, and a given sample size n. In this study, a was chosen to be 0.05, which corresponds to a confidence level of 95%. Except where indicated, the sample size (n) was equal to 3. Where it was necessary to determine the differences between two sets of samples, Student’s t-tests were conducted, also using Excel 97. Results of testing (2tailed, unequal variances) were considered significant if calculated p-values were 6 0.05.

2.3. Sampling, storage and analysis of sediments 3. Results and discussion Sediment samples (three replicates) from the six sites were collected to a depth of 10 cm from the sediment surface using a small trowel, and were placed in plastic ziplock bags (200  270 mm) for transport to the laboratory. The samples were stored below 4 C for not more than one week. The samples were dried overnight in aluminium trays (140  160 mm) in a fan-forced oven at

3.1. Freshwater macrophyte species investigated Aquatic plants are known to accumulate metals from contaminated water and substrates (Garg and Chandra, 1993; Peverley et al., 1995; Rai et al., 1995; Noller and Parker, 1996; Wang et al., 1996). In addition to woody

656

A.J. Cardwell et al. / Chemosphere 48 (2002) 653–663

tree species found in freshwater environments (e.g. Melaleuca spp.) and macrophytic algae species there are typically three major categories of aquatic plant. Free-floating macrophytes are those that float on the surface of the water, and are not attached to the substrate. Emergent macrophytes have leaves and/or stems which rise above the water surface, and generally are anchored to the substrate. Submerged macrophytes are those residing below the surface, which may have emergent flowering bodies and may or may not be rooted to the substrate (Greenway, 1993; Sinha et al., 1994; Thomas et al., 1995). The extent of metal accumulation within aquatic macrophytes is known to vary significantly between species. For example, emergent aquatic vegetation usually accumulates lower amounts of metals than submerged aquatic vegetation (Albers and Camardese, 1993). Some species have been found to develop tolerant ecotypes that either are able to survive higher concentrations of metals accumulating within their tissues or have developed more efficient mechanisms to exclude metal ions from their tissues. Dunbabin and Bowmer (1992) for example noted that Schoenoplectus and Typha spp. (rushes) can be more tolerant than other species. Metal tolerance is a function of plant phenology, vigour and growth as well as metal speciation and aquatic chemistry (Dunbabin and Bowmer, 1992). Fifteen aquatic plant species as described in Table 1 were collected for analysis from the six study sites. On the whole, Typha spp. were the most common and

abundant plants found followed by Persicaria spp. These are widespread genera which are regarded as invasive weeds in many parts of the world. No free-floating species were collected from any of the sites (as noted in Table 1). This may have been because periodic flushing affects all of the creeks studied, particularly in the wet season. For some weeks before the November 1998 sampling, it had been raining consistently. Another possible reason for the absence of free-floating species may be competitive exclusion: some of the sites were congested with weedy grass species, leaving very little water surface available for floating plants. Submerged species were collected from Sites 4–6. The submerged plant M. aquaticum was relatively scarce at Site 4 compared to other species at this site, and while three replicates of leaf tissue were sampled, there was only enough stem and root biomass collected to combine into one pseudo-replicate. Emergent species including reeds, rushes, sedges and knotweeds were the most common type of aquatic macrophyte found, being present at all sites. 3.2. Sediment concentrations of metals Table 2 details the concentrations of cadmium, copper, lead and zinc found in sediments at the six study sites. The reference median range (RMR) is the range that includes 90% of sediment metal concentrations and is

Table 1 List and description of macrophyte species collected from six urban creek sites in southeast Queensland Species

Plant type

Sites

Colocasia esculenta Cyperus eragrostis Eleocharis equisitina Myriophyllum aquaticum

Taro, Elephant’s Ear––emergent plant found in hydric soils (native to Asiatic region) Umbrella Sedge––emergent tufted perennial sedge to 1 m tall (American) Spike Rush––Erect rhizomatous perennial to 1.5 m tall (native) Parrot’s Feather or Thread of Life––submerged and sometimes partly emergent feathery species (South American, potential weed) Waterlily––emergent species, attached to substrate with floating leaves and prominent flowers (native) Marshwort––emergent perennial with floating leaves and small white hairy flowers (native) Knotweeds, Smartweeds

4 3, 4 6 4

Nymphaea violacea Nymphoides germinata Persicaria attenuatum Persicaria orientalis Persicaria subsessilis Persicaria lapathifolium Potamogeton javanicus Rumex crispus Schoenoplectus validus Typha domingensis Typha orientalis

Willow Smartweed––emergent erect, branched annual found in marshy areas and creek banks (native) Rhizomatous perennial, mostly submerged with a few small floating leaves (native) Curled Dock––emergent, erect broad-leafed perennial to 1.5 m tall, found in seasonally wet areas (European, potential weed) River Clubrush, River Bulrush––emergent erect perennial with cylindrical stems to 3 m tall (native) Cumbungi––emergent erect perennial, extensive rhizome systems, flat leaves, to 4 m tall (Native, potential weed) Cumbungi––similar to T. domingensis, but with wider leaves and spikes (native, potential weed)

Submerged species are highlighted.

6 5 3 1, 6 3, 5 1 6 3 5 3, 4, 5 2, 6

A.J. Cardwell et al. / Chemosphere 48 (2002) 653–663

657

Table 2 Mean heavy metal concentrations in sediments and macrophyte roots at the six sites Site

Species

Cadmium

Copper

Lead

Zinc

1

P. orientalis

0.00 (AB)

P. lapathifolium

4.33 (C) 3.42–5.24 0.13 (B) 0.06–0.20

11.4 (A) 10.9–11.9 17.5 (A) 16.5–18.5 11.9 (B) 11.7–12.1

4.5 (A) 1.6–7.4 18.9 (B) 16.3–21.5 35.8 (C) 33.3–38.3

167.9 (C) 162.3–173.5 77.2 (B) 68–86.4 53.4 (A) 52.4–54.4

0.13 (A) 0.06–0.20 0.03 (A) 0.0–0.10

4.1 (A) 4.0–4.2 5.1 (A) 4.0–6.2

0.20 (A) 0.00–0.70 14.9 (B) 12.2–17.6

13.3 (A) 13.0–13.6 29.7 (B) 24.9–34.8

0.40 (A) 0.0–0.81 0.40 (A) 0.29–0.51 1.17 (AB) 0.70–1.64 2.6 (C) 2.49–2.71 1.57 (B) 1.50–1.64 0.27 (A) 0.20–0.34

29.6 (B) 26.5–32.7 96.1 (C) 92.7–99.5 75.0 (BC) 50.8–99.2 110.8 (D) 105.9–115.7 81.9 (C) 68.8–95.0 17.6 (A) 17.4–17.8

126.0 (B) 113.0–139.0 100.9 (B) 96.9–104.9 119.0 (AB) 84.7–153.3 431.5 (D) 417.6–445.4 201.6 (C) 179. 0–224.2 77.2 (A) 74.1–80.3

199.9 (B) 164.5–235.3 184.6 (B) 177.4–191.8 280.8 (BC) 217.3–344.3 621.0 (D) 619.4–622.6 355.5 (C) 328.7–382.3 93.4 (A) 90.5–96.3

0.57 (B) 0.50–0.64 0.60 (B) 0.40–0.80 6.50 (n ¼ 1) 1.47 (AB) 0.48–2.46 0.07 (A) 0.00–0.14

85.7 (B) 85.3–86.1 263.3 (D) 250.7–276.0 431.0 (n ¼ 1) 127.4 (C) 119.8–135.0 36.6 (A) 36.4–36.8

31.1 (C) 29.1–33.1 36.7 (D) 34.7–38.7 63.9 (n ¼ 1) 21.1 (B) 19.1–23.1 12.9 (A) 9.7–16.1

386.2 (B) 378.8–393.6 582.4 (D) 566.6–598.2 4296.1 (n ¼ 1) 497.5 (C) 485.1–509.9 128.1 (A) 122.0–134.2

3.75 (B) 3.67–3.83 16.1 (AB) 8.9–23.3 5.75 (AB) 4.71–6.79 2.57 (A) 2.44–2.70 1.53 (A) 0.97–2.09

26.7 (A) 25.6–27.8 136.8 (CD) 92.8–180.8 76.6 (D) 76.0–77.2 53.5 (C) 48.2–58.8 38.3 (B) 35.9–40.7

57.9 (A) 52.8–63.0 86.6 (AB) 48.1–125.1 126.4 (B) 117.6–135.2 131.4 (B) 120.8–142.0 72.5 (A) 66.2–78.8

782.2 (B) 779.1–785.3 1147 (CD) 985–1309 1568 (D) 1565–1571 1030 (C) 982–1078 514.1 (A) 482.2–546.0

2.20 (A) 2.09–2.31 2.50 (A) 2.27–2.73

47.1 (A) 43.0–51.2 49.6 (A) 47.4–51.8

92.7 (A) 90.2–95.2 168.7 (B) 155.6–181.8

764.2 (B) 743.4–785 0 675.0 (A) 631.7–718.3

Sediment 2

T. orientalis Sediment

3

Cy. eragrostis P. attenuatum P. subsessilis R. Crispus T. domingensis Sediment

4

Co. esculenta Cy. eragrostis M. aquaticum T. domingensis Sediment

5

N. germinata P. subsessilis S. validus T. domingensis Sediment

6

P. orientalis Sediment

All values are in lg g1 dry weight, with 95% CI (n ¼ 3, 2-tails) below. For a given metal at a particular site, mean concentrations followed by the same letter are not significantly different (Student’s t-test, 2-tails, p < 0:05). Submerged species are highlighted.

used as a benchmark to compare data from sites potentially affected by urbanisation or point sources. Effects range low (ERL) and effects range median (ERM) values (expressed as mg kg1 ) are derived from draft ANZECC Australian Water Quality Guidelines for

whole of sample concentrations in sediment. Guidelines relate to potential impacts on biota (derived from relationships between benthic community health and metal concentrations, based on overseas data due to lack of Australian data). Below the ERL there is a very low

658

A.J. Cardwell et al. / Chemosphere 48 (2002) 653–663

probability of effects on biota. Above the ERM there is a >50% probability of effects on the biota. Sediment concentrations between the ERL and ERM concentrations require biological testing to determine the potential impact on biota. RMRs have been determined by Moss and Costanzo (1998) for local freshwater sediments as 0.5–1.5 mg kg1 cadmium, 10–64 mg kg1 copper, 5–20 mg kg1 lead and 29–130 mg kg1 zinc. These RMRs for cadmium and copper contain values that fall above the ERL values proposed by ANZECC. For lead and zinc, the RMRs from 1970 to 1992 data are below the proposed ERLs. Of the four metals of interest then, cadmium and copper seem to be of particular concern for benthic community health. Only at Sites 1 and 2 were sediment metal levels below the ERL for all four metals. At all remaining sites investigated, concentrations of one or more metals exceeded the ERL threshold. In most cases, the mean heavy metal content for the sediment samples collected in this study was in the descending order of zinc > lead > copper > cadmium. This is consistent with many results from contaminated sediments elsewhere e.g. Yurukova and Kochev (1994) and van den Berg et al. (1998), but obviously depends on local conditions and input characteristics. The trend however does not reflect the RMRs for Queensland detailed in Moss and Costanzo, which are of the order zinc > copper > lead > cadmium. This disparity is likely due to the inclusion in the RMRs of data from mostly freshwater systems relatively distant from anthropogenic sources of lead contamination.

3.3. Comparison between metal concentrations in sediments and aquatic plant roots Table 2 also provides the results of metal analysis of root samples collected at the six sites, and compares them with the metal concentrations found in corresponding sediments using Student’s t-tests. All mean values were obtained from three replicates except M. aquaticum (Site 4), which was not found in high enough density to enable the collection of three replicate root samples. Data from Table 2 shows that at most of the streams investigated here, sediment metal concentrations were generally lower than resident plant root concentrations for all metals (not just cadmium and zinc). Even though some macrophyte species (such as Typha spp.) can have a very high subterranean biomass, any given area of a creek bed should contain a much higher total volume of sediment than plant roots. Therefore, while some of the plants in this investigation contained significantly higher metal concentrations than the corresponding sediments (on a dry weight basis), in

total this may represent only a tiny portion of the total mass of metal in the system. A review of the literature indicates that some of the highest plant root concentrations of metals determined in the sites studied here are comparable with some of the highest values reported elsewhere. For example, at a site in northern Greece examined by Sawidis et al. (1995) where sediments contained 3.3 lg g1 cadmium, some of the resident plants contained up to 5.4 lg g1 cadmium. In this study, roots of Schoenoplectus validus and Persicaria subsessilis at Site 5 contained cadmium concentrations at least as high. It should be noted that Persicaria lapathifolium at Site 1 also contained relatively high cadmium concentrations in the roots, but this site had relatively low sediment cadmium concentrations. A caveat with any comparison of metal concentrations in aquatic plants is that they are likely to depend on local sediment characteristics and the presence (or absence) of other metals. Attempts have been made to determine correlations between metal uptake by aquatic plants and sediment concentrations. However, in general, studies have shown only poor correlations, if any (Dunbabin and Bowmer, 1992; Jackson and Kalff, 1993). Of the four metals of interest in this investigation, zinc is the only metal that exhibits a relatively clear pattern of increasing accumulation in aquatic macrophytes with increasing sediment metal concentrations. For the other three metals investigated, the relationship is more complicated. Total sediment metal concentrations are a gross measure that does not distinguish between different metal forms including those which are bioavailable and those which are not. In addition, the limited sediment sampling undertaken may not be representative. 3.4. Metal concentrations in morphological tissues of macrophytes The general picture evident from Table 3 is that roots accumulate significantly greater concentrations of metals than stems and leaves, and this pattern is discernable for all four metals. Less clear is the quantitative translocation of these metals from the roots into the stems, then into the leaves. Consistent with observations from the literature (such as those reported in Biney et al., 1994; Ding et al., 1994; Sinha et al., 1994; Sawidis et al., 1995), the emergent aquatic macrophyte roots sampled in this study contained significantly higher amounts of each metal than the corresponding stems and leaves. Student’s t-testing of zinc concentration data from 13 different species collected did not result in significant differences between morphological tissues, primarily due to the high variation in zinc content from one species to the next. If the confounding data from the submerged species M. aquaticum is removed from the data used for t-testing, zinc root concentrations are found to be significantly

A.J. Cardwell et al. / Chemosphere 48 (2002) 653–663 Table 3 Mean heavy metal concentrations in roots, stems/rhizomes and leaves in 13 aquatic plant species Morphological tissue

Cadmium

Copper

Lead

Zinc

Roots

2.53 (B) 1.43–3.63 0.50 (A)

95.4 (B) 37.4–153.4 17.7 (A)

107 (B) 46.0–168 7.67 (A)

191 (A) 0–398 343 (A)

0.13–0.87 0.48 (A) 0.11–0.85

4.1–31.3 23.9 (A) 1.9–45.9

4.58–10.8 10.4 (A) 5.69–15.1

0–718 285 (A) 38–532

Stems/ rhizomes Leaves

All values are in lg g1 dry weight, with 95% CI (n ¼ 3, 2-tails) below. For a given metal in the various morphological tissues, mean concentrations followed by the same letter are not significantly different (Student’s t-test, 2-tails, p < 0:05).

higher than stems and leaves (p < 0:05). The samples of stem and root portions of this plant were combined into one replicate each due to the small amount of material collected. Some of the metal found associated with root tissue samples may be associated with iron plaque, a precipitate of iron oxyhydroxides which can co-precipitate other metals such as zinc and copper (Dunbabin and Bowmer, 1992; Peverley et al., 1995). In Peverley et al.’s study, SEM and X-ray microanalysis showed roots of aquatic macrophytes coated with foreign particles but rhizomes showed no surface debris. In addition, there was metal little transport to shoots and rhizomes. The data from this study do not indicate a clear pattern of stems containing higher metal concentrations than leaves. Concentrations in the stems as compared to leaves depend greatly on the individual species examined. For example, Persicaria attenuatum and P. subsessilis from Site 3, and Persicaria orientalis from Site 6, all exhibit a general trend in metal content that follows the pattern of roots > leaves > stems, as does Nymphoides germinata at Site 5. Most studies indicate that for Typha spp., the roots accumulate more metals than the rhizome (possibly due to iron plaque), and the leaves accumulate the least amount of metals (Dunbabin and Bowmer, 1992, Peverley et al., 1995). For Typha domingensis samples, the order of accumulation for the non-essential elements cadmium and lead was roots > rhizomes > leaves. For copper and zinc, while T. domingensis roots significantly accumulated more metal, the rhizomes did not contain significantly greater amounts of metal than the stems. From these results, it is likely that the plants are more readily translocating the essential metals copper and zinc from the roots and rhizomes into the above-sediment tissues for metabolic use, however have not developed specific pathways to transport cadmium or lead to these tissues.

659

As occurred with the emergent species, the sample of M. aquaticum displayed a general pattern of roots containing higher heavy metal concentrations than the stems and leaves. Only the leaves of this species were collected in significant amounts. Even so, this species overall contained the highest heavy metal concentrations found in any sample collected for this study, with copper and zinc concentrations being over twice the concentrations of these metals in the next most accumulative species. These excessive metal concentrations may be a function of the growth form of this species, which was truly submerged with no floating or emergent components. Typically, submerged species have been found to accumulate relatively high heavy metal concentrations when compared with emergent species in the same area (Albers and Camardese, 1993; Yurukova and Kochev, 1994; Gupta et al., 1995; Rai et al., 1995; Noller and Parker, 1996). 3.5. Metal concentrations in the leaves of aquatic macrophytes As shown in Table 4, the species that exhibited the highest heavy metal concentrations in the leaves in general was the submerged species M. aquaticum. Typha spp. at all sites contained relatively low metal concentrations in the leaves, with the possible exception of Typha orientalis at Site 6 (7.97 lg g1 lead). For emergent species, the primary source of metals in the leaves is likely to be the roots, however to a lesser degree leaves can take metals from the ambient water (Crowder, 1991). Leaves of submerged species (in this investigation, Potamogeton javanicus and M. aquaticum) are known to also receive metals from their root system, however they may accumulate metals directly from the water to a much greater extent than do emergent species (Dunbabin and Bowmer, 1992). As noted by Crowder (1991) however, elucidation of a single metal uptake pathway is difficult without the use of tracers, and comparison of several uptake pathways simultaneously (for example, sediment ! roots ! stems ! leaves, and water ! leaves) is even more complex. A notable pattern from the data in Table 4 is that for all metals except copper, the overall highest leaf concentrations were found in the two submerged species (M. aquaticum and Po. javanicus) and the two species (N. germinata and Nymphaea violacea) with floating leaves. While increased accumulation has been recognised for submerged and floating species, there has been little attention given to the likelihood of a similar phenomenon for floating-leafed plants. Interestingly, N. germinata at Site 5 accumulated zinc in its leaves to a much greater extent than in the stems and roots. This was the only instance from the samples collected in this study where the leaves of an aquatic macrophyte contained more metal than the subterranean

660

A.J. Cardwell et al. / Chemosphere 48 (2002) 653–663

Table 4 Mean heavy metal concentrations in aquatic plant leaves at the six sites Site

Species

Cadmium

Copper

Lead

Zinc

1

P. orientalis

0.17 (B) 0.10–0.24 0.03 (A) 0.00–0.10

8.80 (A) 8.35–9.25 16.2 (B) 14.9–17.5

2.27 (A) 1.87–2.67 1.67 (A) 1.60–1.74

70.3 (A) 66.4–74.2 52.4 (A) 45.9–58.9

P. lapathifolium 2

T. orientalis

0.17 0.10–0.24

2.37 2.24–2.50

0.07 0.00–0.20

20.2 18.8–21.6

3

P. attenuatum

0.33 (AB) 0.00–0.66 0:33  0:07 (B) 0.10 (n ¼ 2) 0.0 (A)

15.1 (B) 12.4–17.8 10:9  0:5 (B) 12.0 (n ¼ 2) 3.37 (A) 3.13–3.61

20.5 (D) 18.3–22.7 6:97  2:31 (B) 14.4 (n ¼ 2) 1.97 (A) 1.35–2.59

91.9 (D) 83.2–100.6 182:9  8:6 (B) 123.9 (n ¼ 2) 21.4 (A) 19.8–23.0

0:33  0:07 (A) 0.26–0.40 1.57 (B) 1.50–1.64 0.20 (A) 0.00–0.40

34.3 (B) 33.2–35.4 156.2 (C) 150.3–162.1 14.9 (A) 13.4–16.4

11.9 (B) 11.0–12.8 25.0 (C) 23.2–26.8 1.57 (A) 0.52–2.62

148.6 (B) 145.7–151.5 1458 (C) 1366–1550 83.4 (A) 76.6–90.2

2.27 (B) 1.99–2.55 0.20 (A) 0.00–0.59 0.07 (A) 0.00–0.14

16.3 (C) 14.4–18.2 3.20 (A) 2.49–3.91 8.80 (B) 8.57–9.03

20.4 (C) 18.2–22.6 4.33 (A) 3.58–5.08 4.53 (B) 3.57–5.49

1117 (B) 1076–1158 54.0 (A) 44.5–63.5 74.4 (A) 68.3–80.5

0.60 (ABC) 0.26–0.94 0.63 (B) 0.56–0.70 1.27 (C) 1.20–1.34 0.27 (A) 0.20–0.34

13.4 (C) 12.7–14.1 10.2 (B) 10.1–10.3 15.8 (BC) 5.7–25.9 4.93 (A) 4.86–5.00

37.6 (C) 35.0–40.2 20.1 (B) 17.9–22.3 23.2 (BC) 21.5–24.9 7.97 (A) 6.33–9.61

414.6 (C) 379.6–449.6 222.4 (B) 199.6–245.2 611.3 (B) 480.9–741.7 74.7 (A) 67.9–81.5

P. subsessilis R. Crispus T. domingensis 4

Co. esculenta M. aquaticum T. domingensis

5

N. germinata S. validus T. domingensis

6

N. violacea P. orientalis Po. javanicus T. orientalis

All values are in lg g1 dry weight, with 95% CI (n ¼ 3, 2-tails) below. For a given metal at a particular site, mean leaf concentrations followed by the same letter are not significantly different (Student’s t-test, 2-tails, p < 0:05). Submerged species are highlighted.

tissues. In this study, N. germinata at Site 5 and N. violacea at Site 6 virtually always contained higher heavy metal concentrations than did the emergent erect-leafed species at these sites. 3.6. Heavy metal concentrations in internal tissues of aquatic macrophytes A number of individual aquatic macrophytes were selected for further analysis of internal tissues. The plants selected had a stem/rhizome diameter of approximately 2.5 cm, and internal tissues were separated by hand. A clear differentiation between the tissue types was easily achieved, by direct visual observation. The differing tissue types readily peeled away from each other with the aid of tweezers. Fig. 2, an illustration of a

generalised rhizome, shows the physiological tissues analysed for this investigation. Fig. 3 presents the results of analysis of internal tissues of T. domingensis rhizomes collected from Site 3 (high sediment concentrations of lead only) and Site 5 (high sediment concentrations of all metals). For both samples, metal contents were lowest in internal tissues and highest in dermal tissues. The central area containing pith parenchyma contained very little analysable tissue, and consisted predominantly of lacunar spaces (aerenchyma). The cortex parenchyma contained greater metal content by mass than this central section, for all metals. The dermal tissues in the T. domingensis samples, which exhibited the highest metal concentrations, included the epidermis and the fibrous collenchyma layers.

A.J. Cardwell et al. / Chemosphere 48 (2002) 653–663

661

Fig. 2. Generalised internal structure of the rhizome of an aquatic macrophyte (in cross-section).

Fig. 3. Mean heavy metal concentrations in visibly differentiated physiological tissues of T. domingensis rhizomes from Sites 3 and 5. All concentrations in lg g1 dry weight. Error bars are 95% CI (n ¼ 3).

Further separation of the epidermal tissues may have revealed whether there is a continuing trend of increasing concentrations closer to the epidermis. One possible explanation for the patterns evident in Fig. 3 might be that the subterranean rhizomes are absorbing metals from the sediments, rather than receiving metals translocated from the roots. However, Peverley et al. (1993) observed through electron microscopy that rhizomes of Phragmites australis (an emergent reed) do not attract

significant amounts of surface debris, and visual observation during separation of tissues in this study showed that Typha rhizomes did not tend to have persistent debris coating the epidermis either. Peverley et al., also observed for Ph. australis that the rhizomes of that species do not absorb water to any extent. It may be presumed that the rhizomes of both species perform a similar physiological function and therefore Typha rhizomes may not absorb water either. If this is the case, metal distribution within these tissues is likely to be related to the amount of metal entering the rhizomes from the roots. Patterns of metal distribution within the plant rhizomes are most likely a function of physiological translocation by the plants themselves. However for the outermost tissues, the possibility of contamination by metal debris clinging to the outer surface of the sample cannot be discounted. In Fig. 4, a comparison between the four metals in Persicaria spp. stems from Sites 5 and 6 shows that at these two relatively contaminated sites, clear patterns of accumulation are evident for copper and zinc. The stems sampled from these emergent species had not been in prolonged contact with either water or sediment, so the metals present could only have entered the stems via translocation from the root system. It may be that the plants are physically moving metals from the vascular tissues toward the outer collenchyma (a very fibrous supportive tissue) and epidermis. As previously noted, both of these metals, copper and zinc, are required by

662

A.J. Cardwell et al. / Chemosphere 48 (2002) 653–663

stems indicated that these metals were translocated into the dermal layers and cortex parenchyma to a greater extent than the pith parenchyma.

References

Fig. 4. Mean heavy metal concentrations in visibly differentiated physiological tissues of Persicaria spp. stems from Sites 5 and 6. All concentrations in lg g1 dry weight. Error bars are 95% CI (n ¼ 3).

plants for survival. Patterns of accumulation in stem tissues of Persicaria spp. are less apparent for cadmium and lead, the two metals analysed that are not required by plants, perhaps because these plants have not evolved specific mechanisms for dealing with relatively high concentrations of these metals.

4. Conclusions On a dry weight basis, aquatic macrophyte roots can contain higher heavy metal concentrations than the corresponding sediments. Further analysis of a given area, which incorporates total root volume, total root wet weight, and perhaps sediment and plant densities, would reveal whether the plant roots can truly be considered a significant sink for metals in the aquatic environment. Sediment metal concentrations did not significantly correlate with root metal concentrations, except in the case of zinc and to a lesser extent cadmium. Consistent with the literature on heavy metal accumulation in aquatic macrophytes, the submerged species in this study exhibited significantly higher metal concentrations than the emergent species. For most emergent species, there was little translocation for metals from the roots into the above-ground tissues. However, the two floating-leafed species collected did contain relatively high metal concentrations in their leaves. Either these species actively translocate higher proportions of these metals from the root system, or they are accumulating metals from the ambient water into their leaves. The distribution of cadmium, copper, lead and zinc in the internal tissues of aquatic macrophyte rhizomes or

Albers, P.H., Camardese, M.B., 1993. Effects of acidification on metal accumulation by aquatic plants and invertebrates. 1. Constructed wetlands. Environmental Toxicology and Chemistry 12 (6), 959–967. Biney, C., Amazu, A.T., Calamari, D., Kaba, N., Mbome, I.L., Narve, H., Ochumba, P.B.O., Osibanjo, O., Radegonde, J.V., Saad, M.A.H., 1994. Review of heavy metals in the African aquatic environment. Ecotoxicology and Environmental Safety 28 (2), 134–159. Campbell, C.A., Plank, C.O., 1998. Preparation of plant tissues for laboratory analysis. In: Kalra, Y.P. (Ed.), Handbook of Reference Methods for Plant Analysis. CRC Press LLC, Boca Raton. Crowder, A., 1991. Acidification, metals and macrophytes. Environmental Pollution 71, 171–203. Devi, M., Thomas, D.A., Barber, J.T., Fingerman, M., 1996. Accumulation and physiological and biochemical effects of cadmium in a simple aquatic food chain. Ecotoxicology and Environmental Safety 33 (1), 38–43. Ding, X., Jiang, J., Wang, Y., Wang, W., Ru, B., 1994. Bioconcentration of cadmium in water hyacinth (Eichhornia crassipes) in relation to thiol group content. Environmental Pollution 84 (1), 93–96. Dunbabin, J.S., Bowmer, K.H., 1992. Potential use of constructed wetlands for treatment of industrial wastewaters containing metals. The Science of the Total Environment 111, 151–168. Garg, P., Chandra, P., 1993. The Duckweed Wolffia globosa as an indicator of heavy metal pollution: sensitivity to Cr and Cd. Environmental Monitoring and Assessment 29 (1), 89– 95. Greenway, M., 1993. Wetlands for wastewater and wildlife––an ecologist’s perspective. Water conservation and reuse grey issues, clear solutions. Australian Water and Wastewater Association (Qld Branch)., November 1993. Gupta, M., Rai, U.N., Tripathi, R.D., Chandra, P., 1995. Lead induced changes in glutathione and phytochelatin in Hydrilla verticillata (L.F) Royle. Chemosphere 30 (10), 2011– 2020. Jackson, L.J., 1998. Paradigms of metal accumulation in rooted aquatic vascular plants. Science of the Total Environment 219, 223–231. Jackson, L.J., Kalff, J., 1993. Patterns in metal content of submerged aquatic macrophytes: the role of plant growth form. Freshwater Biology 29, 351–359. Lung, W.-S., Light, R.N., 1996. Modelling copper removal in wetland systems. Ecological Modelling 93 (13), 89–100. Moss, A., Costanzo, S., 1998. Levels of heavy metals in the sediments of Queensland rivers, estuaries and coastal waters, Environment Technical Report No. 20 Department of Environment, Queensland Government. Noller, B.N., Parker, G., 1996. Design of wetland systems at mining projects in the tropics to control contaminant

A.J. Cardwell et al. / Chemosphere 48 (2002) 653–663 dispersion from waste water. In: National Engineering Conference ‘The Darwin Summit’ April 21–24, pp. 347– 353. O’Halloran, J., Walsh, A.R., Fitzpatrick, P.J., 1997. The determination of trace elements in biological and environmental samples using atomic absorption spectroscopy. In: Sheehan, D. (Ed.), Methods in Biotechnology, Bioremediation Protocols, vol. 2. Humana Press, New Jersey. Peverley, J.H., Surface, J.M., Wang, T., 1995. Growth and trace metal absorption by Phragmites australis in wetlands constructed for landfill leachate treatment. Ecological Engineering 5 (1), 21–35. Rai, U.N., Sinha, S., Tripathi, R.D., Chandra, P., 1995. Wastewater treatability potential of some aquatic macrophytes: removal of heavy metals. Ecological Engineering 5 (1), 5–12. Sadler, R., Rynja, G., 1992. Preservation, Storage, Transport, Analysis and Reporting of Water Samples, Queensland Government Chemical Laboratory Report Series No. 12 Queensland Government Publishers, Brisbane. Sawidis, T., Chettri, M.K., Zachariadis, G.A., Stratis, J.A., 1995. Heavy metals in aquatic plants and sediments from water systems in Macedonia, Greece. Ecotoxicology and Environmental Safety 32 (1), 73–80.

663

Semple, P., Williams, R.L., 1998. Metals in sediment data summary––Queensland, 1970–1992, Environment Technical Report No. 10 Department of Environment and Heritage, Queensland Government. Sinha, S., Gupta, M., Chandra, P., 1994. Bioaccumulation and toxicity of Cu and Cd in Vallisneria spiralis (L.). Environmental Monitoring and Assessment 33 (1), 75–84. Thomas, P.R., Glover, P., Kalaroopan, T., 1995. An evaluation of pollutant removal from secondary treated sewage effluent using a constructed wetland system. Water Science and Technology 32 (3), 87–93. van den Berg, G.A., Loch, J.P., Winkels, H.J., 1998. Effect of fluctuating hydrological conditions on the mobility of heavy metals in soils of a freshwater estuary in The Netherlands. Water, Air and Soil Pollution 102, 377–388. Wang, T.D., Weissman, J.C., Ramesh, G., Varadarajan, R., Benemann, J.R., 1996. Parameters for removal of toxic heavy metals by Water Milfoil (Myriophyllum spicatum). Bulletin of Environmental Contamination and Toxicology 57 (5), 779–786. Yurukova, L., Kochev, K., 1994. Heavy metal concentrations in freshwater macrophytes from the Aldomirovsko Swamp in the Sofia District, Bulgaria. Bulletin of Environmental Contamination and Toxicology 52 (4), 627–632.